Unity 3 announced

http://unity3d.com/unity/coming-soon/unity-3

It feels like GS is slipping lower and lower on the food chain...

Hm does not sound bad at all but still I don't know any great Unity game, which one did I miss?

@Toast: Ok that are some big titles, I did not know/see that.

Wow - Umbra Occlusion Culling for "free". That's the same tech of games like Dragon Age or Mass Effect...

@Superku:
Yeah it tends to get used by very niche games often although with the new Need for Speed World and Tiger Woods PGA stuff there are some more common titles in the works...

Well, I think they´ve got some well paying sponsors + many sold copies = much money -> nice features -> even more sold copies -> ...

I am not yet much into unity but will get more into it soon. So far, I like gamestudio a lot for its flexibility, by still being easy to understand and very comfortable in terms of coding most things.

However, those seem to be some nice features if one wants to create a modern game without having to care about the tech.
On the other hand, Unity costs MUCH more than gamestudio if you want to be able to have the same features...

I don't see a future for Gamestudio. Conitec seems to have given up already. They don't care about it anymore. Look at the mini-features that get added to the beta page every 2 weeks. str_setchr() and str_getchr()...

On the forecast page: WED & MED -> Revamping the user interface... It's kind of sad and it's a mockery of the customers. There was talk about revamping the user interface of WED & MED almost 10 years ago already but it seems all they did so far is moving some buttons around and making some new icons while keeping usability a nightmare that is messed up beyond all repair.

Where are the new developers JCL said he would hire?

The forum is dying too. A few years ago there were 10 times more posts per day.

Indeed. I can not argue with stated facts of amy, as I agree with them. But ah well, if 3dgs dies, we'll all step over to unity. Not?

i really don't want to switch from lite-c to another language -.-


but i really think 3dgs needs moar features like unity

untiy3d just keep getting better and better. Makes me want to stop fooling around with gamestudio and learn unity.

@JOOZEY-
i geuss so but like rei-ayanami said i really wouldnt want to.But what else could we do? unity is growing by leaps and bounds while gamestudio hasnt been changing much at all.

I think JCL will keep the Gamestudio project running with minimal effort as long as some sales still trickle in (like he already seems to be doing since a while) and at some point in the not so distant future it won't be worth it at all anymore.

Then he could open source Gamestudio.

Though it probably would be a lot harder to build a working open source community around Gamestudio (and its parts like the hopelessly outdated editors and a hard to understand Lite-C compiler with chinese comments) than to simply create a Gamestudio clone with already existing and well maintained open source projects like Ogre, Mono, Bullet, a customized Blender 2.6 for level editing,...

im gonna miss gamestudio

Yeah : Big pro Engine with big features.

But i assume the Indie License will not be any more for Unity 3.0 ?
OR perhaps they will remove some features like "Beast" lightmapping , "Umbra" Culling or Deffered lightening ?
They have now a complete renderer with great lightmaps , at same time with all shaders for a scene : AAA engine features.

I 'm not a team, and can't produce a complete big level of quality with AAA characters, i've got no AAA projet,i target Inidie games not big AAA.
Who can produce games like Bioshock 2 , Bad Company2, or even
Mass Effect 2 ??
So im'not that excited by the news, i'll play with it when it will be out only.
But cool news for Unity 3.0 that don't stop improving a lot each time !

seems it may be happening already, i mean, even a bug i mentioned weeks ago still isnt fixed and the forecast page is damn pathetic, theres NOTHING there, as if they think the engine is complete, a game engine is never truly complete, not with new technologies being released, everything from physics to rendering can and should be changed/optimized

Indie is gone a few weeks ago. The so called Indie became the normal Unity, and turned freeware. There will be a free version for 3.0 too. See this thread:

http://forum.unity3d.com/viewtopic.php?t=45787

Personally I think GS is a bit "year 2000". It never saw proper, rapid and dedicated development. It feels and looks like a side project. Since its stuck in a corporate setting, don't get your hopes up about it going open source. I don't think it will die completely, since its a nice intro to game development. But Im sure we will see less and less people using it for bigger projects. The pipeline for every facet of the toolset is severely outdated. It just isn't a serious toolset.

Unity has a free version with which you can make commercial games. Sure it lacks the advanced features, but you still get that amazing pipeline and ease of use. The only thing that will truly save GS is if they make everything up to the commercial license freeware, and start updating everything very quickly.

edit: As for the price, $1500 for unity pro is crazy good, if you look at what you get. Comparatively, GS PRO should cost about $200. Its overpriced.

While i think lite-c is a very good language(lite-c is why i didnot switched), but yes, features and editors are waay to old.

The next engine(A8? which i think they should start working on immediately) they should completely rewrite the rendering engine and the editors from scratch while keeping the lite-c functional. They should stop worrying about backward compatibility, and make a better engine with up-to-date features. Years ago when the unity was not around and when there was no T3D, GS was a good competitor to them. But now the editors are still same with that times, gamestudio is falling way behind them, soon it wont be able to keep up the race. -personal thoughts.

im gonna embrace Unity3 for my next project im affraid...
That would mean the end of 3DGS for me

Im still in love with game studio!

Same here. I don't see myself using GS ever again. I just hope I can sell my A6 pro license...

How is A7 overpriced o.O you get most features of unity pro already with the comm version. The outdated tools just kinda hide it.
While Unity allows you to create games, gamestudio is perfect to learn backgrounds about the technology.

Well, because you get most features of the comm version with the Unity free version too

What hurts a bit is of course the missing realtime shadow in the free edition. But GS Realtime Shadow is also not the best. So i can live with this odd blob shadow. Doesn`t look too bad when it`s combined with lightmapped terrain.

The outdated tools is the problem. Yes, you may learn the background about technology. But it`s an outdated technology in several areas. Dunno if this is really worth learning it then.

GS has waited too long with the jump to the next gen workflow in my opinion. And pays the price now. Given the pace of development and the number of involved people at Unity now, i don`t see GS coming back again. Counts for most other 3D development kits too by the way. They will all loose lots of customers in the nearer future. Even when they release free editions too, like Dark basic Professional. Unity is at its way to the top of the market at the moment.

Maybe on paper the features look similar at a first glance but in reality most things in Gamestudio (physics, shadows,...) are so sub-par that even the free Unity is superior to A7 Pro. Okay, you can extend gamestudio with plugins but if you want to work with C++ it's more worthwhile to spend your time with an open source engine.

Yes, but again, if you want to learn the backgrounds about the technology then better use an open source engine. This way you will learn much more skills that are useful in the industry than with an outdated closed engine with some proprietary scripting language like Lite-C.

I learned most of what I know about graphics programming with gamestudio and that is probably more than 99.9% of Unity users do.
There is no problem with postprocessing, deferred lighting, reflecting water and shadow mapping with A7 Comm.

Okay, seriously, I'll wait until Unity 3 is released. If it's released and Gamestudio didn't got some new awesome features, I'll get Unity and say bye bye to Gamestudio.

I know that stuff like shadow mapping can in theory be added to A7 by the user but I haven't seen a solution yet by anyone which can compare to Unity's (and also Ogre's) implementation in quality. There isn't anything production ready so far. This would be JCL's job actually.

a facelift to the editors (imo, completely drop the MED and write plugins for max/milkshape/fragmotion also drag and drop mdl importing for "new" world editor.) and a new rendering engine with better shadows etc. would give gamestudio a bump and shouldnot be too hard. Drop the qoole99 based WED, it's been 11 years. And write a new better editor from scratch.

Why not embrace Ogre, hire experienced users from the Ogre team/community , there are commercial engines already doing that. This could even remove the windows dependency of the 3dgs.

This is what gamestudio needs urgently, a new graphics renderer, better editor, platform independency. after that web-browser support.

Lite-c and the game programming logic/approach is pretty good, actually that is GameStudios best and last part that remains powerfull.

These are my actual wishes and suggestions for A8. For A7, it is already falling back, no need to add more viagra to the old man.

(p.s. i do not mean engine is bad or unusable, it is still very good and powerfull engine and final products that have been created on a7 can absolutely compete with the ones that have been created on the other engines.)

all GS needs is a better art pipeline, they rest too much of what can be done on the users shoulders, users should be focusing on making their games rather than dealing with the core technics of the engine, shaders, lighting and such, afterall, this is why developers license engines in the first place and so far it seems that alot of unity addons were made by its users then officially added to the engine, like the locomotion addon, after he wrote it, unity made it official for their engine and even hired him.. look how many awesome thirdparty tools we have, yet look at the the websites main page, its sad, or am to believe they dont even have as much as a web designer? which wouldnt matter because am sure ppl here would do it for free

Also Unitys shadow mapping is by no means perfect and the Ogre implementation is also not that awsome if you ask me. It is just that shadow mapping for the sun is not easy to archieve at a high quality and good performence. I had something working quite nice already, which thanks to jcl now also works without an engine extension and with view frustum culling. I just got a bit away from working on it due to other projects. But I think that it just needs one problem to be fixed until it is well usable. Unhappily I already messed with that problem for quite some time now and still don´t really know how to solve it.

The great thing about gamestudio is that you have everything you need to realize any feature you want as easy as it could be. I am pretty sure, that even with plugins you are more restricted in unity. Unity is just much more high level than gamestudio.

I am btw just trying to show you that gamestudio isn´t bad at all. It doesn´t mean that I consider Gamestudio to be the only engine everyone should use. I myself have already worked with A5/A6/A7, Irrlicht, Horde3D, ShiVa, Löve and will soon get into Unity. I just think that every engine has strength and weakneses and that one can´t say that gamestudio is overpriced crap while unity is god.

If you look at unities "biggest update ever" features, you may notice that from the unity side most of them meant nearly no work. There is absolutely no problem to use that lightmapper with gamestudio, same for the culling, deferred lighting and g-buffer means basicly the same and just needs multiple render targets to be fast and breakpoints should be a standart. From a coding side, there is no problem with that stuff in any other engine. The difference is that unity can actually pay it. And from the view of someone who wants to create a great game that update of course IS big and I mean really big.

I think that gamestudio will survive on the beginner market for quite some more time, at least if they really work on the templates, what conitec seems to do currently.
Another possibility to survive is in my opinion the way the C4 engine goes, which convinces with hightech and probably a flexible programming interface.
I see the biggest problem for T3D, but maybe just because I somehow always kinda disliked Torque...

Maybe we really should collect all things we want to see changed, and after we have a list, we post that to jcl, in the future forum.

I think if we ask him and not being harsh, i think he will change somethings


I mean, the whole community wants that, i think he will change things.

--> He will also earn more money


If you accept the idea, i will open a thread in morbius, where we can collect the ideas, and when we are ready i will make a list with all changes

I also always disliked Torque and still would use Gamestudio before Torque any day.

This exactly is the problem. Also the other user contributions always missed small but important last details and probably this always will be so unless someone gets payed properly and can work on it with full dedication.

I know that shadow mapping never is 100% perfect but still everything i have seen in Gamestudio so far was worse than in Unity and Ogre.

I really like this idea.

I am sure JCL already knows what should be improved and what is missing. I also think there are things and/or difficulities on JCL's side that we cannot see. I am sure that when this difficulities are overcomed, JCL and the team will start adding these things to engine.

i also completely agree with Slin's last post. A7 is still great.

Or even better, Lite-C wrapper for Unity

Gamestudio lacks manpower to do the needed steps ahead to catch up with Unity or other engines. Seriously have a look at Unity's lately doubled (!) team number. There's no way Conitec's small crew can compete here. Does anyone actually know the exact number of guys working on 3DGS? I always had the impression that it were around 5 brave little coders doing all the work...

When considering Unity's new features I'm also thinking of them maybe trying to get in a higher cost range or at least offer a new and more expensive edition at some point which offers all those fantastic 3rd party features. I at least seem to fail to understand how 1200$ or how much it was per licence should be able to pay for all that fancy stuff...

I don't think there are 5 full time coders for Gamestudio. They all (JCL included) only seem to work on it part time.

Unity probably won't get more expensive. Keep in mind that they don't only sell PC licenses but also iPhone/iPad, Android, Wii, XBox360, PS3.

LoL, i just tested the tropical demo.

I have to say: I am very very impressed...


I mean, this graphics are better as i saw them before - and that on a webplayer!


-.-

I keep hearing people say the toolset sux, but all else is OK... I'll agree.
And that Conitec doesnt have the manpower to resolve that... No comment.

So why cant WE do something about it?
George has a project running for the community to build a game,
is there anyone out there with the time and guts to co-ordinate a project
where we the community design and write our own editors?

With the varying skillset we have we should be able to build our editors
to satisfy most users, and have it written in lite-c...
Great way to PROVE the power of lite-c.

Let the flame-war begin ... I can take it ... Im a smoker anyway. :b

The key for success for anyone in any competitive market is to find a niche and fill it well. I think that the engine market is big enough for several players. Game Studio just needs a bit of tuning to a niche left open by other engine makers. My opinion is that it's the template system. I know seasoned programmers will say, bah, but a robust FPS or RPG template would probably do wonders to generate revenue from enthusiasts.

Making a spectacular game to showcase the engine and generate revenue would be wise as well.

not even close.... wow that is just so utterly wrong that I dont know what words to use to express how wrong that is.

I really like EvilSOB's Idea...


We do have great programmes like, evilsob, djbmaster and many more!

IntenseX was one of these things. Look what happened ...

We talk about the fact, that we, the community create editors

intenseX was not one of those things, intense x was basically a more advanced template replacement, and it does its job well so im not sure what your point is.

I'm trying to deter the use of WED with 3DGS Easy Scripter even though in its early stages.

I usually browse around the 'Future' forums and the things people ask for, like 'vertex' recently asked for a material editor and physics editor added to WED, they go on my list for easy scripter.

So as for 'why can't WE do something about it', i'm working on it.

hey DJB,

Is the level editor is a spreata project or a part of easyscripter?

Well, IntenseX started as a project from a community member. My point is, it is still not finished. A bug, and the fact that IX internal design was too easy to crack has more or less stopped development. Maybe you`ve read about it somewhere. Half finished, and it is on halt. No further infos available. No lifesign from Aris since a pretty while. Wouldn`t call this "does its job well"

Mh, maybe we need some more examples. Because this is no unique case. What about shaders? Shade C was another community member project. Is it finished? Or shadows. I think there is a promising shadow system called Shadows Plugin. In development since eons ...

No, i`m not really complaining. This is all great stuff - at a hobbylevel. No offence, that doesn`t mean that the involved didn`t do a great job. It`s just that a hobbyist never works at a project like a professional does. Quantitywise and because of that also qualitywise. Because time is always a big issue for a hobbyist.

Biggest problem: the community cannot fix the core product. That`s the job of the makers of GS. And when they stop caring then the community is lost too. And that`s what i see here. I hear complains about the ancient workflow since i know GS. For years. Nothing happened. As told, finally GS pays the price for that.

Maybe that makes my point a bit clearer

As far as Ive seen (which is small), isnt IntenseX waiting on an engine-memory-bug to be fixed?
(or am I thinking of a different add-on)

If we leave Conitec to concentrate on engine-bugs rather than editor-enhancements,
wouldnt that be good for IntenseX, and any other plugins, like say community-generated editors?

DBJ is a prime example. Just think how good Easy-Scripter would be if it was a community effort.
The community knows what it wants, so why not let the community build it?
Rather than wait for conitec to get around to it, and then find out either
it not what we actually wanted, or its not as useful as we expected. (ala Ged to me).

(Im flying a bit blind here as Ive never used Easy-Scripter, it sounded "too simplified" for MY design-style,
I dont build levels so I am very rudimentary with Wed, and Ive never gotten around to a serious playing with Ged))
But I was thinking of community-coded replacements for Wed (maybe, if Ged doesnt cut the mustard),
but certainly a seriously-enhanced Med with lite-c at its core.
This would enable TRUE preview windows, shader-support, etc.

A new Med is something I keep promising to do myself but its just
too daunting for me on my own, so Ive kept putting it off...

Mh. You forget something here. Unity has also its fanbase that does some work like writing shaders and stuff. And while GS needs this badly to substitute the own weak aspects, the Unity community builts on top of a very strong state of the art engine without ten years old ballast that not only gets engine fixes, but all the enhancemens the community is asking for, like editor enhancements. Plus next gen workflow, plus ...

At Unity they have 50 professionals working at it at the moment, growing. How much work at 3DGS? 3? 4? The gap gets bigger and bigger with every day

Well, I think MED not really needs a replacement since there are many better model editors out, like Blender, ZBRush, Cinema4D and so on.

The only really thing that need to be changed is WED.

I would like to have a material editor like in Cinema4D, and other features from blender and c4d

Here is a pic from what i mean:


The red parts are the material editor - really needed
objekt strucutre and the ability to add textures/materials - green
other features i miss in wed, like GOOD booles or Hypernurbs - yellow

@Quadraxas > Yeh it's built into easy scripter although it is built as a seperate .exe, and then embedded.

I found my old disc of 'FPS Creator' the other day, installed it and played around with the editor. That tile highlighting you helped me with was an idea on how it might work. I kinda like the idea of building a level, using predefined geometry like walls, floors, stairs etc, that you can snap together easily.

So maybe this 'tile' way and manual positioning is a good mix for a level editor?

Mh, yeah, but not for people who want to make profesionell(AA) Games...

Just to clarify my intentions Rei....
I dont mean to make a new Med as a full-blown 3D modelling APPLICATION,
there are plenty of good freeware ones around already, and many
modellers prefer very different interfaces when they work.
I myself use different apps for different 'types' of models.
(characters vs architecture vs terrain vs accessories)

I DO mean to turn it into a tool to allow you to IMPORT static(or even animated)
models from other formats (as it does already, but improved with more intuitive-ness),
and to improve its ability to "tweak" those models (like rag-dolling, clothing, etc),
animate them, skin/map them, add and edit materials and shaders(real-time),
and to preview what they will REALLY look like in-game
(possibly combined with other already completed models)
in real time, at each step...

but this would be a wed replacement, and not a model aplication:

Because I don't think clothing and rag-dolling or materials are needed in the model editor, but that is my thought...

maybe i am completly wrong and just telling stupid and senseless things...

No worries, I think there is a little confusion
between where your WED-replacement is going,
and where I would like to see a MED-replacement go...

Same here. I don't see myself using GS ever again.i'm really gonna miss but even though im comftorble with lite-c i have to move on at some point.Just like when i moved on from GAMEMAKER to A7, its the same thing.

Med does not need to be replaced. Just the import options should be opened up to more formats.
It's good to have a tool which can do rudimentary stuff in the engine's native format.
It gives you a certain chance to fix smaller import glitches.

i just think GED and WED shouldve been one tool, kinda pathetic to have to have WED, MED, SED and GED open, i mean, wth?

what i think GS needs:
-reworked lighting
-merge WED and GED
-more light types
-radiosity [dnt even tell me we already have that because its slow as HELL!]
-more material properties

thats just my list, i wont list 'optional' features, just listing what i think new members would like, i wont complain about the physics and all that because physics as never given me a single problem

I would not replace Med either. Getting things to work in any native 3D game format can be very difficult. Med is a plus. If you take out Med, then you have to confront many different ways of making a model and how they get in the engine. Just the differences between the way that each version of a single 3D art package exports a single 3D format is mind numbing.

UMMM appear gamestudio is go to dead rest in peace

Wow, this is amazing. I did not read every post here, but I read the feature list of Unity3. It is like a dream comes true.

First, my harshest critic on Unity was the scene management and they just bought an industrial proven tool to solve this:
http://www.umbrasoftware.com/
If you want to integrate this into your 3dgs game, then you have to pay $30,000. Unity has it build in now.

Then it renders with new shader system, deferred lighting, hundreds of lights possible with shaders, fantastic.

And shadows: I always dreamt of a solution to combine static and dynamic lighting including shaders in a working professional way, to fake global illumination plus having dynamic lighting. It works now. They just included another industrial tool, Beast:
http://www.illuminatelabs.com/press/newsarchive/state-of-the-art-light-mapping.pdf
This would cost you $100,000, but it is integrated in Unity. Amazing.

And I dont have to pay extra, I theoretically can make game scenes look as nice as a rendered image/video now.

Besides that I dont have to code in C++ like I have to do in Ogre, Irrlicht, T3D or C4. I can use the way more comfortable C#.

To me it is a no-brainer, I have to order this. And when it got released then I will make some of our preview-images at Dexsoft in Unity3

This will be an exciting future.

I wish you much fun with your preview images in Unity3.

regards,
gri

Thank you gri. I am sure I will have. I always liked the idea to create static shadow mapping and I enjoyed it first in WED, then in Constructor and later in UDK. Now, together with global illumination, post processing and shaders it will just be the next level and I am looking forward to this.

And besides that I am looking forward to C# coding.

My thoughts exactly! It's indeed a bit of a no-brainer.

Though I think i'll finish my project I have currently, and any next project will be done under Unity.
I'll be around for a year or so... but hey, that's probably when they'll release U3 as well

regards,

uh oh! it will get really hard for gamestudio. i also think it has to find some niche to survive but i don't see which one yet? the free unity actually covers most areas i can think of and it has the advantage that you aren't stuck with it (feature-wise and platform-wise) once you have created something nice.

i also look forward to working more with unity and improving my c# skills. i just would prefer if they used collada instead of fbx.

Well, Blender 2.5 can import Collada, and export FBX. Voila

Yeah, i know, it`s not this easy. But worth a try ^^

Yes, the niches are getting smaller now. Unity even covers Android-based mobile phones now. And the maker of Irrlicht just released an engine for the upcoming WebGL (natively running in browsers). So right from my head there is only Symbian left as another OS for mobile phones, that needs a nice game engine, could be even 2d in the beginning.

gamestudio is going to die soon. Unity3d is a beast now! it dominates this whole indie game developer business. Just the fact that when you create something awesome they give you the choice of publishing it as an iPhone app (very popular), Android app, iPad app, wii game, web browser game, or simply a cpu game is enought for me.Not to mention that you can you use shaders for free.The only "niche" gamestudio has is that its users dont really want to learn c# or any new language.i'm willing to sacrifice a month or 2 for all the features unity has.

uhmmmm why does it seem like everyone is over looking that a7 is much cheaper than unity, and yes im know im going to get the *oh well unity has a free version* well unities free version is about exactly the same feature wise and gstudios free version, and comm is only 200... unity doesnt have a comm level.

Unity is great I had an issue with it's state management system, which is seemingly going to get worked out. I might just switch back to complete my project.

hm... i see quite some differences actually...

some examples:

While reading the 3dgs free feature list, you could also add to unity free list: Decals, Multiplayer, Isometric Rendering, but substract realtime shadows.

And it doesn't really need one. Unity's free version just like Shiva's free version is a pretty perfect entry version. You can start to develop your games with those versions for free and once you see that things come along nicely you actually buy the full thing and in the case of unity e.g. get all the additional eyecandy like dynamic shadows & stuff. For Gamestudio with no shaders, bad editors and so on the start is by far not that great and complete...

I also do not think that Unity's cost are any problem for a serious developer. If you think of having created a game that sells well even if just on an Indie scale level you should have no problem in paying the bills for Unity. That also might reveal a niche 3DGS might fit in and that's the really low budget sector also aiming at those who just do it for fun and don't really plan to release a game that sells so well they even might live from it. That also lowers the bar for the needed features by a lot which is quite accomodating for 3DGS. You would need some easy accessible editors though and currently this fails horribly especially as WED never made its step past the cube code base (or whatever it was that got bought and used to be WED's base from the first days - it sort of was advertised as a feature in the A4 days ) from the last millenium...

Anyone wanna give a script example of a function from Unity that feels comfortable for a C-Script/Lite-C user?

For instance, how would I write things like camera looks at me, enemy follows me, state switches from attack to flee...

yeah i would also like to see such example.(just to get an idea)

Camera looks at me (all in C#):



assign it to the camera in the editor, and drag n drop the player into the "playerPtr" field. done.

Enemy follows player is simliar:



Apply to enemy, drag n drop player into the playerPtr field, done.


For statement switches, i actually always use ENUMS




It would run away backwards, looking at the player, but just to get an idea how you would make it.

Is it that what you wanted to know?

EDIT:
i left out the class definitions and library includings, as this is genereated automatically when creating a script anyway and looks the same for almost any script anyway.

it is just a plain:

Thanks a lot!
I still don't get the class things, but that's not your fault, your examples looks really easy compared to Lite-C.
Even me might have the chance to understand that.
Maybe, I get the chance to take a course in C# for a week in April.

Well the class thing isn't too relevant actually and you still could write in Javascript.

The class gets created for you by default when you create a script. You just have to make sure the class name is the equal to the files name.

And then you add all function into that class. There are default function like Start, Update, OnGUI, etc or you create your own ones (void foo(), int counter() ,etc).

You usually dont have multiple classes in one script file, so it doesn't really matter. Just see it as a border around your functions.

@Kihaku, you have any suggestions where to start with unity?(like you start with lite-c workshops for gamestudio)

Know the concepts and patterns of object oriented programming and have some just a alittle above beginner c# experience, i also know javascript.

Well i started with UnityScript which is basicly Javascript with some changes.
It is very easy to use and i think the only disadvantage is that you can't use the .NET features compared to C#.

So i basically used Javascript for few month, and then switched to C#. It's super easy to switch over to C# once you know all the stuff in Unity when scripting in UnityScript (JScript).

You obviously can start with UNITY here:
http://unity3d.com/support/documentation/Manual/Unity%20Basics.html
in particular here:
http://docwiki.unity3d.com/uploads/Main/Scripting%20Tutorial.pdf

And small off topic:
http://www.rockpapershotgun.com/2010/03/10/unlimited-detail-wants-to-kill-3d-cards/
:-)

Thanks Kiyaku, this looks really easy and convenient.

hi,

I checked the feature list of the HOLY free unity3d indie version.

Looks nice, but for me Render_To_Texture is a must-have!
So it have to pay 880,- euro for this feature.

And there are no Realtime-shadows, no Fullscreen Postprocessing.
So I come to the following conclusion:

Unity PRO
is very nice no question and for the guys who have the money its a big improvement compared with Gamestudio.

Unity Indie
doesnt realy compete with A7 com because the lack of features I want to have and get with A7 com !

So I wish all unity Indie fans much fun playing around with the included nice looking WorldEditor and coding in "modern" language like C#.

But what can you realy reach with this castrated version?

Building the 10000000-st Iphone App , maybe a puzzle-game in 3D? -useless

So I stay with A7 comm.
Maybe its Editors looks outdated but I dont what to marry them( I have affairs with Blender already).
I only need the potency of the Engine.


This is like the cars of 2 individuals.

The one likes the nice looking shiny varnish of his car and triggers the buttons of the stereo. He gives a damn that his engine was limited to drive only on Highways ....he loves to observing the automatic-antenna .

The other one doesnt care about the hard working steering control his car have. The car got cleaned 10 years ago but hey(!) he can drive wherever he want.

And yes I agree unity Pro is like a Porsche possible to swim,dive and fly.


regards gri,

@gri:
I wouldn't compare Unity Indie with A7 Com. Unity lacks anything in that area as it's either Indie or Pro and to my mind it doesn't really need anything inbetween. When now comparing both Unity's free version and the free 3DGS version Unity clearly wins. When comparing Unity Pro and 3DGS Pro Unity wins even more clearly. It may be 200 € more expensive but gives you more advanced features and in case of the upcoming 3.0 even saves you 130000$ per title if you'd want to use the same technology. As it was said before this really is a no-brainer...

So you're right - Unity doesn't have a good offer in the very low budget area (at least not if you must have good dynamic shadows no matter what - if that's not so important for your project Unity Indie becomes extremely attractive again). So maybe it's not the best idea for pupils with next to no money. At the same time one might think of trying to apply for an educational version in that case. I don't know when or for what prices you can get Unity then but I guess asking them won't hurt...

First official statement from Conitec:

http://www.opserver.de/ubb7/ubbthreads.php?ubb=showflat&Number=314815&gonew=1#UNREAD

Na..Na..Na..! Net so frech!

is there anywhere a date mentioned when the preorder pricing ends??

I agree with you. I am also convinced that the iPhone market is overcrowded and only high quality will survive now.

But did you realize that Unity already announced support for Android mobile phones? That is a quite new market and will get some small indies some good money, I am sure.

And this also shows how fast and innovative the Unity team is. So I think investing there is a good future investment. It probably will not be outdated anytime soon.

But I also understand that students might not have enough money to enjoy the pro edition. And they want of course all these shiny features.
But then it might be more interesting to check Ogre or Irrlicht to have a nice renderer for free.

And then there is still C4 for $350 with everything in (supporting 3 platforms). And there is Shiva supporting even more platforms.
And there is even the FPS Creator as a drag&drop tool.

It looks like there is something at the market for everybody, including lots of shaders, render-to-texture and real-time shadows.

Yes, Unity Free lacks render-to-texture so Gamestudio Commercial has an advantage if this feature is important to you but Gamestudio doesn't come with good dynamic shadows in any edition. Like has been discussed in this thread already they theoretically could be added by yourself but it still seems like nobody has been able to achieve that. All attempts so far are missing something.

@ello: Preorder pricing will end the day Unity 3.0 gets released.

@Pappenheimer: Object oriented programming isn't very hard to learn if you use an easier language than C++ for learning it and like Captain_Kiyaku has mentioned it isn't really necessary to know much about it for programming with Unity.

WTF, i wasn't sure if i should write that i dont want to hear exactly that worthless answer.

What do you want to hear then? The exact release date I guess?

i think you do know what i mean.. one month or one week?? (so, just if there is a rough estimation)

Someone told me, the release is announced to happen around summer 2010.

thank you.. at least i have some time to get me a credit at the bank before paying more

I look forward to Unity 3 then I can complete my project, got an email about my bug report stated that the issue will be resolved with the new version, it is worth the money and the time to play around with.

Generally speaking, Unity3d is years ahead, out of question

However I am not yet confortable with its " Components " programming appproach even thouigh it seems to be the future of game programming
I still prefer the traditional "Entity" programming style
For this reason I am interested in Leadwerks
Is anybody using this engine on regular basis ?
Any comments ? bugs ?
For sure it is hard for a one man company to fight against Unity3d army

[just to all the posts that keep saying "dynamic shadowmapping"]
all am hearing is dynamic shadow mapping this and dynamic shadow mapping that. jeez come on, its not a NEED, it adds to a scene but if you cant make a scene look good WITHOUT it then enabling it is just sad, btw, most games now that support dynamic shadowmapping would still look great without it and there are ALOT of game engines that dont do dynamic shadowmapping and still look great:

unreal engine
source engine
the engine behind halo

all i see GS needing is a more professional approach to how it does work on the editors and these "updates" are quite crappy, 3 new script functions in 2 months and they call that an "update", quite pathetic, as far as i see it, graphics only come under 2 groups:

fail - ppl wont glass at your game unless it offers something truly unique otherwise
pass - quality is enough for people to make a glance, aiming above this is just unecessary for a indie because if after that first glance your gameplay didnt pull them in, no amount of shiny effects will make them love the game, theyll love how it looks but not necessarily get attached

you dont need dynamic shadowmapping to get a pass

i think this engine can pull off close to what the source engine can do, sure it may take some time and third party tools but am sure if you give a professional team 3dgs and the same basic tools we have, they can produce with it

i've seen phone games that look better than half of the projects i've seen by 3dgs

i still think all gs needs is better editors, material editors and pixel lighting

for example, my project 'scion', the only thing slowing it down now has NOTHING to do with the engine, we already sat and found the engine limits and new what needed to be done, workaround workarounds and more workarounds which is also what being a designer is about, only thing slowing us down now is probably lack of artists and my own laziness and i'm also quite sure that we'll be one of if not the best looking 'completed' projects to be done with this engine and i hope then GS will get some recognition

its more like:


so really be fair, if your going to show the limitations of the free a7 you really should show the many limitations on unity free.

you can't use c++ with unity free but your other points are wrong.

* you aren't limited to the included shaders
* you can integrate any .NET library

there are .NET libraries for networking, physics, databases, xml,...

this makes most c++ plugins unnecessary anyway and it is easier than c library integration in lite-c since lite-c isn't fully ansi-c compatible. this always causes lots of annoyances with incompatibilities in header files.



...
i have to admit though that i am no fan of physx. it isn't really better than some of the other physics engines (they seemingly aren't even able to do a cylinder collision shape :p) and it mainly gets used by AAA titles because nvidia pays the developers to use it. it is much better than ode though and at least unity doesn't need a driver since they have the source code and disabled this silly requirement.

isnt Extra now free? even if not as yet, wouldnt it be more like:

http://www.3dgamestudio.com/download.php

according to this comparison the free version is quite limited. plugins aren't even possible! but i though the free version would replace extra so i am a bit confused too...

I don't know what "pixel lighting" is meant to mean to you, but sounds to me like you need to either learn shaders, hire a shader programmer, or use Shade-C ^^
Perhaps all attempts that people are going to give away for free are missing something. You'll find that most of the best-looking games use very simple shadow systems -- from the Unity lighting demo's own completely unfiltered and low-resolution point lights to the shadow-mapping in any gorgeous PS3 game (Uncharted 2, anyone?) which is quite sneakily low in range, but the complexity and confined nature of most scenes hide it quite well.

What you miss is that render-to-texture and post-processing effects are required for some very, very simple stuff: bloom, HDR, depth of field, screen-warping effects such as heat-haze and refraction, real-time reflections, and the list goes on. And I've only mentioned the really simple stuff -- stuff that shouldn't be an issue to implement.

I think Unity has a very good balance -- multiplayer and simple shaders would be much preferred over free GS's features. And yes, "simple shaders" is appropriate, because even though you can write your own, almost any advanced shader effect needs render-to-texture. I think Extra would do very well with shaders but no render-to-texture. But if you already have Commercial, or especially Pro, I don't see a reason to move to Unity Free.

Jibb

lol.. oh whoops. per-pixel lighting, my bad.. lol, making me feel noobish and i do use shadeC, boh_havocs WIP shader package is being applied to every surface in scion, he did a great job with this pack in terms of speed

the fact of the matter is that most ppl here wont complete a game even if they get what they want, its just something i've seen, most games here dont progress pass a simple demo...

No worries All I'm saying is that we already have per-pixel lighting. You're using it, I'd guess, since you're using shadeC.
Far too true.

Jibb

the fact of the matter is that most ppl here wont complete a game even if they get what they want, its just something i've seen, most games here dont progress pass a simple demo...

Yeah , but that's normal caus it's indie and lot of beginners people and some passionnated ones like me here , not industry people.
That's true that asking lot of AAA features won't make a game for you.

while agree with darkinferno's #314872,

it could be easier(take less work) to get pass with unity. Though this highly depends on the person, if you know the limits and the workarounds of a7 you could end up with a better product with a7.

For a7 engine, even if it is not feature-rich as unity3, it is not bad - just the editors are too old. Lite-c is the main better part for me when compared to Unity, but having an editor like the one on unity can be more charming after some point.

MED is quite a good tool for importing and tweaking, it needs a interface polishing and more format support.They can focus on importing rahter than editing for MED. (more formats etc.)

WED could be completely dumped and a new editor can be written.

SED is not bad but looks old, i remember editors like LED on some threads it could be included instead.

For the engine itself, a better renderer can be better.

I donot know why JCL is not doing this but using open source tools or parts of them. it could decrase the need for the manpower. Like UV unwrap tools of blender(see Road Kill ) for med, or a proven Renderer.

yes i knw but i wasnt talking about the engine PLUS addons, i mean, not everyone wants to apply a shader to every model they wanna light nicely, basically am not worried too much because from what i see so far, GS can be great under the right user.. i just wish the engine was marketed more and given a facelift, starting with this dumbsh*t of a website, heard them mention something about wanting videos for a video gallery then what? nothing... nothing at all... and umm, to everyone thinking of A8? lol, oh stop.. it aint happening any time soon..

I don't want A8 anytime soon -- that's just time to shell out money.

The shader features can't be called an addon. That's like calling Lite-C an addon. And they're no more difficult than Lite-C.

As far as I know ShadeC is 100% Lite-C and HLSL -- all natively supported by A7.

So many people are just too scared to try the shader workshop.

Windows 7 took off really well -- the only reason I see to change the engine version from A7 to A8 will be when DirectX 11 gets supported.

Jibb

DX 11 will bring you only advantages if you really use it's features, if you support real-time tesselation, global illumination, deferreed lighting or ambient occlusion. Otherwise you can do like id-software and just stick to DX9.

And then you also have to adapt the architecture of the renderer to really support shader-based rendering.

oh, and that island demo by unity is easily doable in 3dgs, recall someone being impressed by it lol and btw nothing from unity surpasses the wavefield demo by hummel, the only problem is, jcl/3dgs developers[whoever they are] dont feature projects made by users, the website for example, should clearly have a screens gallery and vid gallery with some of the more awesome projects i've seen...

guess imma have to start uploading random pictures to the net and tagging them :-" so they showup in any google image search for gamestudio lol

My question concerns BSP/*.map based building.
BSP processes and related tools have been a tradition in game making for some time now. Unreal still features it. 3DGS has it. You can get blazing fast
interiors in 3DGS. How does unity stack up? I've noticed that most of the demos in Unity are exterior scenes.

Vertex, Darkinferno:
This is exactly the problem of many indie engines: effective scene management.

The BSP system of A6 was quite nice but took very long to calculate for bigger levels. And it was not able to handle shaders, created too many render batches.
UDK and Source probably have a more modern approach to BSP. They also have to place portals manually. This will help to calculate the BSP tree faster.
A7 got an ABT system that is quite helpful for outdoor scenes.

C4 has a more modern approach. You place your zones and portals manually but then it does not have to calculate any complicated BSP data at all. And it works really fast for indoors with shaders and real-time shadows. You can see real-time lights and real-time shadows without issues through portals.
C4 also has some occlusion portals and occlusion blocks (to block everything behind a certain area).

Unity3 will get this industrial proven Umbra system, it is an occlusion culling system similar to what I already mentioned for C4. It has been used for many big games like Mass Effect as an example. It uses modern hardware, multi-threading and different low level optimizations.
Unity also has some streaming system and it will probably have any data tree like quad or oct tree. I don't know for sure, but the island demo shows, that it works fine even on different systems.

The comment of Darkinferno, that all this can be done in A7 might be right for small scenes. But I am afraid it does not stand if you are looking for bigger scenes. Many of us know from experience that there are problems when you have lots of entities in your scene, there were many stress tests to show this. And there are also problems when you add shaders to bigger worlds. And shaders on BSP blocks are very slow.
And if you want to compare with the upcoming Unity 3 then it will be really hard. The deferred rendering alone will allow you to use way more shaders in a scene and way more lights than you can do with a traditional forward rendering approach.

But I agree, there are some nice small scenes around, made by some skilled shader programmers. I admire this.

Manual portals would give me reason to use WED ^^ Most times it's unnecessary to wait for a long compile time to have culling between a few rooms (unless they each have a LOT in them), but to be able to set out the really important occlusion portals by hand would be fantastic.
I agree -- DX9 is/was a milestone in graphics that can still be taken advantage of for more advanced effects than we've already seen (with shader model 3.0, practically the only thing limiting us is the speed of the graphics card) -- my interest in DX 11 lies almost completely in DirectCompute -- I went to a CUDA seminar last year and loved what I saw and heard, and it was all very straight-forward, but an enticing part of DirectCompute is compatibility -- presumably, a card that supports DX 11 supports DirectCompute.

Jibb

>But I agree, there are some nice small scenes around, made by some skilled >shader programmers. I admire this.
[referring to *.map/*.wmp/3DGS Wed-created with shader scenes...]


I suppose traditional map creation is part of the niche for Game Studio. There are 1000s of *.map based assets (many included as prefabs for free in Wed) that can be utilized in A7 and not in Unity, and the process of creating new ones is one that is easy to get into in Wed. Compare this with say... the inability to build structures in Unity...or building a level in Maya or Max for export to unity. Instead with 3DGS, build your BSP model in WED, light it, and populate it with entity props (with LODs). Make some functions/actions to attach to entities. Slap a skybox in if you have to. Set your clipping plane as close as possible and you've got a fast level in 3DGS/possibly a finished game. That will be good enough for many. 3DGS just needs to showcase the ease that this can be done.

I think a lot of what can be done with 3DGS simply hasn't. [I'm talking about we users here/3DGS potential to be used to make something cool. I know a lot of cool stuff has been done-- just talking about potential.] It does need better marketing and presentation.

I would be cautious about everyone jumping on the bandwagon of complex shaders
in PC games. The average PC user has a 5 year old computer. Shader tech limits your target market, increasing game dev time and skills required, increases file size reducing feasibility for digital downloads and taking dev time away from gameplay.
I've started work on a simple single player FPS, and I'm avoiding shader use. I hope to opt for a distinctive style. We'll see...just working on art assets now.

Switching gears back to scene management/culling:
Unity has a far camera and a near camera. It has the ability to leave a gap between the two. This has interesting implications for culling. If you're in a valley, you could see distant building tops peaking over the hills, but ignore the geometry in the valley just over the hill. Perhaps this would be easy to add to 3DGS...

I think for everyone a healthy attitude is to see the engines more as various tools in a tool box and less as racers trying to win a race. Is FPS Creator beinq squashed by Game Studio? No, they're just different tools. Ask yourself a similar question about 3DGS and Unity.

/signed

Although I must admit I messed around with Unity for a bit and I love the workflow: Having native access to .blend and .psd in the editor is great, doubleclick on a resource and the respective program starts or loads the file if it was running anyway.

There are many similar tools compared to WED. The Torque users have Constructor to create this kind of block geometry and calculate static shadows.
Torque 3D got a new "sketch" tool to draw 3d geometry.

In C4 you also can build primitives in the level editor, change mapping and add materials to them. The demo level is mainly build with the help of these.

And then there are tons of map-creating tools.

Regarding shaders:
This is the wrong approach to not include shaders to support older machines. C4, Unity, Torque 3D and other engines use a way to render on older machines with a downgraded version of the shader set. And you even dont have to care, you put in your diffuse texture, your normal and spec texture and the system will decide how or if they will be used on the machine of your customer.
The Source engine (Half Life 2) even has branches for different DirectX versions.

I agree with you, that a developer should not care about shaders. We should not be forced to write shaders for each situation (1 light, 2 lights, with or without lightmap, with or without fog, with or without environment map, with or without emission map, parallax etc.). This should be generated by an intelligent shader system in the background.

And this is how C4, Unity, T3D and others work. You just plugin your textures and change some properties and the shader will be created and will run, no matter what platform you are running on.

And if you want to go past that, or mainly get to know how exactly it works in the background, A7 is perfect What I want to say, is that I don´t always want to create a great game, but to have fun while experimenting with technology and to learn new things.

That actually is a pretty bad argument. If you want to experiment and find out how things work and "learn new things" 3DGS probably is one of the worst pieces of software you could choose with "no source code" probably being the major flaw...

In order to learn how tech works etc. go for e.g. Ogre or Irrlicht to get a playground for free or buy something where you have access to the source code...

I hope you didn't want to point out that the behaviour described by Machinery_Frank is a downside because it actually saves every dev from reinventing the wheel in order to get some common effects...

No. Of course open-source projects are better for learning absolutely everything about engine development, but that doesn't make 3DGS bad.

A7 provides an excellent environment for shader programming. A7 suffers from the requirement to re-invent the wheel, but that's only if you want the wheel in the first place. Shader-wise, A7 gives a tremendous amount of freedom, including automatic stepping down for older systems.

I'm not ignoring the fact that A7 could definitely use a better automatic shader system for those who don't want to write any, and just want effects that are common today. That's a clear and definite downside to A7.

That's a clear and definite plus to Unity.

For us shader programmers, A7 is brilliant. I remember when I bought Commercial for the first time and thought to myself, "Shaders? Well if I've got this feature I might as well see if I can use it!" One shader workshop later and I'm making my own effects however I want.

These days I invent my own effects for the purpose of experimentation (as other users here do as well), and sometimes stumble across something that's useful in an actual game for today's/yesterday's hardware.

Jibb

Shaderprogramming with gamestudio is just a lot of fun actually
And no, Toast, I of course see the benefits of not having to always reinvent the wheel, I just wanted to mention that at least I sometimes love to reinvent the wheel. But as I also want to finish some projects, I of course also enjoy using other Tools, like for example Unity (I am currently getting into it a bit deeper due to some project). I also worked with ShiVa, even though I can´t really say that I would prefer it over gamestudio. And to reinvent the wheel on a none shader basis, I of course wrote a 2D iPhone engine with OpenGL ES and a 3D Linux/Windows Engine using OpenGL including an awsome shadersystem (just like the one of Gamestudio, but better of course :P).
But I am sure that I am not the only one here for whom the way often already is the goal. But if you are more an artist than a programmer, you of course want a game engine which makes it as easy as possible to create your game idea including fancy effect without having to care on how it actually works.
But well, I am just repeating myself. Unity is awsome for what it is, no question.

What I believe for 3dgs to shine really is DJBMASTERS work and a FAR BETTER HOMEPAGE that actually demonstrates the usefullness of this engine with pretty demos such as Hummels recent work right in the front page.

I've seen this homepage since f00king 2001 I think, it's a digitally bought software. If the homepage fails so does the interests, sales and the manpower.

About Unity, C# shines over Lite-C / C++ of 3DGS anyday, I feel alot more productive. This is ofc my opinion only.

And Unity is alot more artist and designer friendly and importing assets alone is amazing.

My primary point is that use of the best shader tech provides challenges from the perspective of a small developer trying to finish a product. If you make a million poly model to get your normal map for every prop (text book method), in a small team you're in for some trouble. The general public has no idea what a shader is. They're looking for game play that lets them forget about troubles.

I worked with Constructor using TGEA. GG dropped portals/BSP in TGEA (coming from TGE)-- not sure what happened or what is going on with interior rendering with current Torque 3D. Constructor is an app made not by the GG team and took a long long long time to come out.

C4 and Unity primitives might be useful, though they're not BSP.

My general point is that, you can easily select the tool that you need to do the job, but you must also be careful of the job that you are trying to do.

I think a healthy attitude too is that it's a great time in the industry-- so much going on.

@DarkInferno :
oh, and that island demo by unity is easily doable in 3dgs

Yeah , but you will nerver do it as fast.
- direct import in Unity for orcks and trees from Blender
- terrain and multitexturing , all in one : no need to swicth to another editor or search/apply shaders.
-simple shaders like normal map : just select the normal texture and apply it and select the normal map shader in the shader list : in 3DGS it's a headacke each time
- several camera systems to choose already made : just choose one and drop it in the scene : Yeha Unity drag and drop features are powerfull !
- You can use C# instead of their Javascript !

etc ...


Perhaps you can do it in A7, but even a newbbie with a step by step tutorial will do it easily in Unity , without having to use several tools (like MED, SED ,WED, GED) : all in one !


Well just my point of view after having played with these two engines.
Unity is so much advanced and today interface design, not comparable in terms of workflow.

I recognize A7 is great for programmers, little games and beginers, and skilled people can achieve some good things.

umm... i could do it quite quickly actually... once you have a pipeline for your art assets for gs then your set, setting up such a terrain would be just a few hours in photoshop and a very handy shader that we use, same goes for the other models, few minutes to handle texture and shaders... the water is quite easy also, ofcourse it'd take some planning to make the scene still look good, good models alone wont do it...

i often thought of doing a similar scene but then reality bitch slapped me and i knew that it would just be another scene that wont be featured and only get a few comments, then no matter how breathtaking it MIGHT look, fade out of existence.. if someone really needs to see such quality then lookup hummels wavefield...

so i didnt bother and probably never will.. any scene i work on will be from my game, senseless making another demo

and yes, a new member of gs probably couldnt do that as fast as a new member to unity, that much i knw is true...

as said, with 3dgs its never a question of IF it can be done, the question is HOW to do it and by WHEN

Maybe people should make up a Unity vs 3DGS - Gameproject Battle.

Two teams, with the same assets, making the same game, in the same time amount, team one in 3DGS, team two in Unity.

Dexsoft could provide the art assets muahaha (i know, shameless method to get some great assets, i'm sorry :<)

lol...theyd get pwned

Who "they"?
I am on Unity btw

And both sides will have the same assets, etc.

you will get pwned.. lol but this doesnt make much sense either way, wont prove much

I highly doubt that as i know both sides very well (7 years 3dgs, 1 year unity).

This would indeed be quite interesting, especially when we use both free versions, but i doubt people will actually participate in that

Sounds like an interesting idea. I have something like this in mind for a long time. But my idea was a bit different:
I want to make a scene like from a real game (indoor, outdoor or even both) and then I want to stress test it in different engines, compare fps, look and work flow. It would be interesting to see how much advantage a certain scene management brings, how shaders or models slow different engines down etc.

Maybe this could be the real battle: We prepare some ready made scenes and try to implement them as good as possible in each engine to get most power out of it.
I am sure there will be big differences.
My guess is, that C4 will win the indoor race but outdoor I am not sure at all.

A square off would be a great idea. It would test the efficiency of the pipeline, and the overall performance. On a slightly different note:

http://forum.unity3d.com/viewtopic.php?t=43733

I wonder how it would run under A7? The one guy reports 60fps in the webplayer, and 20-30 on the pc.

well one things for sure, GS will not be the fastest, some guy said he got 60fps for that unity remade crysis scene, i doubt you'll get that in gs.. then again, we arent aware of his machine specs also...

this would be interesting though, provided that we have the art assets and the type of scene to be done

Wow, thanks for this link, looks really pretty.
It renders very fast, about 120 fps on my pretty old system (Win7, 64 bit, 8 GB, GeForce 9800 GTX, Core 2 Duo E8500 3.2GHz).
And I have Modo, Firefox, Email, Avast and Messenger running in background).

Well, it is in no way comparable with the tech and details of crysis
I also hope that that isn´t the best possible setup for shadows, because those are much worse than what I already realized with gamestudio.
It features only few models, some strange brightness adjusting(?), a boring water shader and strange caustics. Nice is the LOD of the trees. But the trees are missing at least some basic normalmapping and where are the softbodies? Technically, it is actually quite boring and nothing fancy at all. But it shows well, how nice assets can make things look good.

3dgs shall not die. yes the new features are quite dull, but if you are happy with 3dgs then you should use 3dgs. o need to put it even further down the food chain.

come on guys....

regards,
michael

Of course it is not. Grass is missing, there are no buildings, no vehicles, no decals on terrain for wheel trails or other details.

But still this terrain looks good is huge, there are islands visible at the see and it renders fast.

It is huge, but has only few polygons and a very lowres texture. Just move a bit away from the start position and find out, that there is just one ugly lowres texture and a choppy terrain
The original island demo is much more impressive.

Since we talked about big terrains, here is a really big one with tons of foliage rendered in the Leadwerks engine:
http://www.youtube.com/user/pure3dviz

I believe that Unity and Leadwerks will attract a lot of people at the GDC while it will be harder for competition like Garage Games or Terathon with their old-school demos.

i hate trees popping up infront of me

Did you also see this WIP project featuring a fantasy setting?
http://forum.unity3d.com/viewtopic.php?t=38963



It constantly gets extended and there's a video of one of its versions:
[video:youtube]http://www.youtube.com/watch?v=svH6_fKULvw[/video]

You also can "play" it in the browser:
http://game.irilys.eu/olivier/zone01.html

I think it's very nice and it'll be interesting to see the final version which is supposed to also include stuff like animals walking around and so on. It "just" uses the free Indie version btw...

man i cant wait until i start creating my levels for scion so i cant try to compete with these videos am seeing, somehow i have a feeling i can do it all quite easily, or maybe am just underestimating them.. then again, i do notice that these arent just scenes but huge landscapes.. so, who knows, would take some time to set up

@DarkInferno :
That's the difference, you will put lot more time than Unity, caus you 'll have to use different editors.
In Unity , all tools are in the editors : you can edit anything directly, paint terrain ,deform it, import a model all incredibly easily : that's the difference of workflow.

But if you can create something outstanding in A7 , cool
Impress us !

actually... i just started something, giving myself... hmmm, say about 5 hours to see what i can get, i'll vid and post my results soon, wish me luck

and remember i truly am just gonna start, so.. we'll see

Wow, that Leadwerks clips are amazing. I hope they still are at a higher quality.
That Unity level on the other hand again istn´t that awsome. The terrain is very lowpoly and the trees are rendered using impostors already at a very short distance. On the other hand it shows that unity has a very well working and only hard to notice impostor system, even at that short distance used in that level. But I think that the creator should have faded out the massive foliage which is actually the only really amazing thing of that level and should instead have raised the LOD distance for the trees. It has a good atmosphere though.

ok, so, after a few hours of toying i only set up the terrain, sure i couldve just painted on in GED but am quite sure it would look great, so i did everything by hand.. took a while, i guess unity's workflow wouldve been easier since they couldve just painted terrain also but am thinking it'd be similar to what i'd get in GED

anyways, heres the terrain i achieved:



the terrain is huge, i had to double the cliprange, the water is the default water shader, theres bloom and the terrain had normal and specular applied...

now i guess i could just throw in some objects wildly all over but i wanna play with it and see what i can get, overall i would say, that the workflow isnt great but even though unity's workflow is faster, the quality of each object onscreen isnt as great as they could be i guess but in the end its the completed scene that matters and not some little terrain demo :)) lol...

You should just build something up first and then show it, right now it's just a randomly painted terrain that has too much specular for my taste, and a very ugly water shader.

I'm curious how it will look like in the end.

And yeah, you can make this in 5 mins in Unity, and no, the quality would be better.

And by the way, this terrain is small, no chance to make a performance comparison to what we saw in this thread.

Can we play?



http://www.youtube.com/watch?v=9S1Ef29o9lo&fmt=22

http://www.youtube.com/watch?v=A8HgYhZKq1Q&fmt=22

"the terrain is huge, i had to double the cliprange "the terrain is far from small... i stated that, its just not very high, i dont have art assests to try setting up what i want so it makes no sense really, my programmer art wont cut it and naah, i dont have enough interest in this to "build" something up, i have my own game to work on, but if i get art assets to create this scene then i'll give it a go

its mostly in the art assets, it'd be cool if everyone had the same resources and then had to create a scene, in GS, unity, unreal maybe? just to see how it goes

btw, i'm yet to see any great looking terrains from unity, it all seems to just be painted texture, i dont recall every seeing normal mapping or specular but as said its more about the art direction that the complexity of a single element

Well thats basicly what i said, about having the same assets and making the same scene/game.

And no, your terrain looks small. If it's suppose to look bigger, maybe less water or a different scale of textures would have helped.




Just to name one:

My problem with UDK are the required specs.
Sure, my system is old, but the systems of my expected customers are old, too.

I finally found a weakness of unity
The code editor isn´t integrated and there is no "Run" button That means, that I weather have to always switch between the code editor and unity or what I currently do is to have my screen splitted with Unity on the right side of my screen and the editor on the left. The disadvantage is that the previews windows format isn´t too great this way
But to be honest, it isn´t that bad actually, just not perfect

About big terrain... Are 10x10km big enough? As that is kinda the limit when using just one terrain in gamestudio and if you don´t want to loose all detail (which I of course already do a lot, what else do you expect of 1024 vertices streched over 10km...). And when putting it together from several single terrains, one notices the missing streaming feature as then my ram isn´t big enough (with about 16 terrains of that size). While trying to put some grass on my terrain the same problem occours, so that I will have to find a way around that. But what about a few polygons following the camera depending on the height and colormap?

However, this is what I´ve got so far (running with >150fps, depending on the viewing direction, as LOD chunks hidden behind the terrain aren´t clipped away, but on the other hand I can also improve it a bit more by messing around with the LOD distances, which are so far VERY far away):

And yes, it is ugly so far, but I am just not good at setting things up in a good looking way. And it shows well what I am writing all the time. Gamestudio is capable of everything, but you have to think about the techniques you use or even set them up yourself and can´t just use it like in unity. -> It takes longer, but is in my opinion a lot more interesting and fun

I agree with Slin on Unity's code editor, it seems a little bit of a pain to keep switching all the time. Also the intellisense is pretty awful, it kept overwriting the words I had or not auto-completing them.

As for that terrain, it looks pretty good. I'm no artist and so might be talking rubbish, but to me it doesn't look bad at all.

the upcoming monodevelop integration looks very promising though. this will make debugging much better too.

@ Slin:
Maybe, you can use something like "Mysty Mood" and/or "Seed it!" in combination with the weeds and trees of Loopix to achieve a huge and appealing and fast rendering landscape.

to be honest I would never do that. loopix's models were great 4-5 years ago when i first got into acknex, okay about a year or two ago, but nowadays, a pc game (unless its super casual) can handle much more than that, even slightly dated machines.

Yes, please do that! I think an empty terrain is not really something you can use for serious stress testing. There should be a decent amount of grass, rocks, vegetation and multi-texturing to see how it works in a real game environment.
And then there is still the option like you saw on udk demo to animate foliage, to show particles, moving clouds, animated water ect.

And it is not always needed to have the best and high poly vegetation. The lighting is way more important to achieve something more interesting. Just look at the UDK demo, the terrain and vegetation (though quite low poly) looks, like it really gets the light colour from the sky. But you will see from the A7-images of DarkInferno, that the terrain textures are way too saturated and do not reflect any colour and light from the sky. It looks not real, rather cartoony.
And shadows of course can add a lot to a good scene, like you saw from the Leadwerks video.

I already made populated terrains in Gamestudio and you will see, it is not easy to place tons of vegetation manually. After some weeks of hard work I just gave it up. But maybe some new user-made tools can help here.

a big plus for unity (in my opinion one of the more valuable) is that you can just open a website and join a funky multiplayer game. like paradise paintball for example..

I think it is possible to achieve with almost every engine a good looking outdoor level. The screen shows what i did in 3dgs a while ago.

However to get this environment running i had to implement a complex scene management and it was 2 years try and error to run with a
decent frame rate.

I also struggeled a lot with the 2GB memory barrier. Cause even if you move the same models around (what i did for grass and trees) you
still need a couple of thousands of them to fill a larger level.

In the leadwerks engine i get the same result with a couple of mouse clicks and everything is shadowed and animated...

So in my opinion the engines are designed for different purposes. And if you want to run a state of the art outdoor level i doubt 3dgs is
the right choice.

Wow, the pic looks really professionell

That is a pic what should be showed on the 3dgamestudio.com website...

3DGS is a JOKE nowadays as engines go. Soon Adobe Flash is going to have a more powerful 3d engine than gamestudio. It's the past , and hey , it got me into gamedev (which in my case I'm not too sure it was a good thing), and I've learned a lot using it. But it's over. Unless , somehow , Gamestudio soon releases A8 , a new super enhanced awesome version with super powerful engine and updated awesome editors , but then , we now have UE3 for free for huge projects , and Unity3d for smaller ones (though you can also make huge games with it), so , I can't see gamestudio surviving , less Conitec just wants to keep it alive without making any profits. Who knows , gamestudio will always hold a place in my heart , but as far as game engines go , it's but a mere memory in my mind.

LOL for someone who has been around as long as you, and who I'd assume even knows a little about the TECH BEHIND THE ENGINE and is not all "lolomg I want a point and click MMO with very little work, and oh its sooo hard to paint my terrains in a different third party program", I'd think you'd know how much complete bullshit that statement is.... I mean come on, comparing it to flash? really?, do you even know how difficult it is to get a decent speed renderer in flash?

Or are you just a super 1337 programmer that knows how to make his own software renderer that doesnt use dx or ogl but still has hardware acceleration....

or wait maybe its because your just a little to lazy and want an all-in-one paint by numbers art solution because you dont know a bit of programming and dont understand how a7 is a valuable programming asset and how its (despite popular belief) not that hard to use 3rd party tools to create art and levels in a7.

its fun to discuss technology with the retarded.

Hey now, boys, play nice...

I can see what WhydiDie is saying; looking at the last year or so (as far as 3d engines go), is making A7 look *very* long in the tooth, so to speak. A lot of people who are 'getting into game making' are just folks who love the games they buy and have a creative streak in them. They tend to NOT be programmers...but just average, every day Joe and Jane Pixel who play Oblivion, Warhammer Online, or Fallout 3 and thing "That is so-o cool! But, what *I* would have done is have that house kinda falling over, with a big hole in the side where you can look in and see a big crater under the house with water in it and stuff". ... ... and that gets them hunting for 'modding' software, which leads to looking at 3d game engines. They want to feed their artistic desire, but don't want to have to spend a year or more learning how to program just to get a decent looking mud shader into their game vision.

On the flip side, I also see what Lostclimate was on about; those same folks who look into buying a 3d game engine want to do all the cool stuff in their heads...and get frustrated when they can't because it "takes too long", or because they start to realize that in order to get something to look/act a particular way, they have to learn at least *some* programming (even if it is simply a scripting language). The time they have to spend figuring out what "while i<5;i++" really does is time spent not doing the 'cool stuff' (making levels, models, textures, animations, etc.).

So, when engines that can bridge the gap between artistic desire and the technical aspects of game design starts to be affordable to Joe and Jane Pixel, they will naturally gravitate towards them. Right now, A7 has the capability to be f'in awesome...but it's NOT going to happen because Joe and Jane are going to look at the two hours they spent in A7, and the two hours they spent in Unity/Torque3D/ShiVa/dxStudio/etc...and look at what they got done; as it stands, the non-A7 stuff they worked on will likely have a good terrain, hundreds of trees, plants and vegetation, a cool-ass sky with clouds, all populated with houses, towers, carts, vehicles, animated animals like horses, cows, pigs, a flowing river (with physics), sound, background music and who knows what else... ... and their foray into A7 netted them a low-res terrain with an even lower res single-texture, a couple of colored lights and a handful of level blocks with plane textures on them. Why? Because in A7, for Joe and Jane, it was just too confusing about how to get the .x models (some animated, some not) they bought into their A7 level.

I guess the end defining factor is going to be "Initial Ease of Use". I've seen people pop open Torque3D and in the space of a couple hours have something that looks like this:


When a 'newb' can get stuff looking like that in a few hours to a day or so of just farting around in an app...that's a selling point for others. "Hey, he's new too! If he can do it, so can I!"

Anyway, I have to run off to work now. *sigh*...

While it may be true that Unity is the better overall package, your post still is kinda pointless.

How about listing facts what 3dgs is missing, instead of just bashing?
The other users who switched to different engines were also capable of putting together a list why they changed...

3DGS is not bad. It is just outdated. It`s 20th century, but we live in the 21st. In the end it was the stoneage workflow of 3DGS that made me switch. I hated to even touch it anymore. Meh, again a dozen mouse clicks just to find out that the shader isn`t attached, the model doesn`t show, the texture is wrong. Even starting the game is half a point n click adventure with text output. So let`s fix this and try again, let`s fix that and try again, just to find yourself in the same situation again and again ...

In Unity i have everything in one editor. I apply shaders with one mouse stroke, see the result immediately, hit play and am ingame, have even wysiwyg in the level editor, and and and. Unity is Next Gen. 3DGS simply lacks a generation.

Time is a very important issue for me. I`m a hobbyist. And my lifespan is limited. So it`s not a "being a newb/noob" reason when i want the fastest workflow. I prefer the one click solution wherever possible. Making a game is unbelievable time consuming anyways. And as a hobbyist i am happy about every hour i don`t have to spend with stuff that could be done with one mouse click instead. Hours that easily adds to manyears over the time. And makes the difference between a hobby project that is impossible to finish and a hobby project that gets finished. This counts even more the more advanced games you want to make.

See, not only Joe and Jane Pixel are well advised to use the tool with the faster workflow. And we haven`t even talked about engine features. Here i clearly see Unity in front of 3DGS too. At the latest with the upcoming version 3

@ Paul_L_Ming:
I like your post very much! I started the same way, as an avid gamer (and I'm still an artist looking for the easy way for a lone wolf to get things up and running for potential clients/ideas), with 3DGS.
I see it as a monument, a pioneer of the evolution of game engines.
I stopped using it many years ago, but i still respect it and lurk here from time to time.
I think that an engine can be judged by its forums... its very easy to sense the general energy around the user base and very importantly the gallery and the WIPs... Unity3D's forum and gallery speaks for itself... we can flame all year long, but what we see and play there is a fact.
I was a big SEGA fanboy and when it died as a major console player i felt very bad... but i cant close my eyes to the 360 and the evolution in gaming. I'm saying this to illustrate that fanboyism doesn't help us... and the engines are just tools.

Btw, the workflow we consider bad actually is, what 13 to 17 years old users like the most about gamestudio, as you instantly understand their basics while they feel overwhelmed with those "next gen editors".
And in case you consider me fanboyish, I just don´t like people stating that gamestudio is bad and the worst thing one could work with. It is just that Unity is better in some ways and gamestudio is still a good tool. There is in my opinion no reason not to download Unity and start working with it (what I currently do, btw). And I also don´t get why people think that one could only use one engine and not several ones for different purposes. I love gamestudio for experimenting with shaders, but if I develop an dapp for the iPhone I am using the iPhone SDK and if I want to create an a bit more complex game I use Unity. Everything is great in its own way and I have equally much fun with all of them.

Just wanted to also express that Paul L Ming's post is not only very diplomatic, but quite true as well.

I was going to pop in and ask people why they even bother responding to Why Do I Die's posts anymore, but you handled it quite well.

From the beginning of my desire to get into game development, I wanted to be able to do as much of it myself as possible. I didn't want to re-invent the wheel, of course, but I didn't want my game playing and looking like just any other game. But when someone writes (for example) their own shadows in A7, they aren't re-inventing the wheel. Shadows are done in so many different ways, for so many different purposes.

As an example:
Shadowmapping goes hand-in-hand with good SSS. But if you want to use SSS in Unity, I'm pretty sure you'd have to write your own shadowmapping anyway. And this doesn't exclusively apply to fancy-pants SSS that only the latest computers will run -- such a method could be used for some VERY simple but gorgeous SSS that'll run silky smooth on any computer that can handle shadowmapping.

I know it doesn't suit everyone. But it suits me -- that's the way I like it, and I'd say that's probably how anyone else who knows how to write shaders likes it.

Jibb

Okay, then:
My specs are: Ati Radeon X300 (Shader 2.0), 1Gb RAM, Pentium 4, 3GHz
Any chance to achieve it for my system?

I've never had an ATI card, but I think it wouldn't have any trouble Just to be clear -- there's a difference between fancy front-scattering that'll put a lot of pressure on most systems, and some straight-forward back-scattering that'll look great for obvious things like foliage and light coming through the back of peoples' ears.

The back-scattering side can be done effectively with very little extra effort, while good-looking front-scattering requires a lot more shadow samples to be taken (although scattering on both sides would benefit from the extra samples).

Jibb

Unity rocks, but that is my point of view, considering I am not too good with C++ or Lite-C. I am better with JavaScript, which it runs. (However I am also trying to learn Java)

Hope, nobody minds me asking for some comparing the coding.

In GameStudio you have to think with while-loops and waits, some time ago, I still thought that that is essential to game programming generally, now I see other engines don't need that.
But, how do you manage that, and the order of functions?

Well, the often used alternative to gamestudios while loops with wait(1), which also Unity uses are event functions. in unity one is for example called Start. That is called in the moment the object is created. Then there is the update function, which is called each frame, directly before rendering And then there of course are some other events as well like mouseDown or what ever it is called, which is executed in the moment a mousebutton is pressed above the object and there of course are also other functions like that. It is actually very easy.
These function are always called in the same order of course. A bit more tricky in unity is in my opinion the way you can access other objects. But also that works fine if you once found out how to do it. What I like is that the code is in general very modular and quite clean, the disadvantage of the object oriented design are on the other hand the very long code lines which tend to be a bit messy. I for example have a script instance which has the pointer to another scripts instance which has the gameobject it handles as pointer, which has the transform member, which holds the position, which holds the x, y and z components. The result is something like this: scrobj1.scrobj2.gameObject.transform.position.x, but well this is an extreme case of course, just to demonstrate what I mean.

It appears to use a similar approach to lite-C's EVENT_FRAME, if I'm not mistaken. Of course, lite-C's frame event could more easily get quite convoluted since the same event/function will be used with any other events that the entity detects (EVENT_IMPACT and so on).

Unity appears to separate these, and it looks quite nice.

Jibb

EDIT: Okay, thanks for the explanation, Slin. (I had the reply window open for too long).

we are using while loops for repating same task in all frames(while(1) ... wait(1)) in Unity, there are functions that are automatically called every frame, so you are already writing the inside of the while.

Ah, great, that explains the start() and update() in the examples from Captain Kiyaku's explanations.

Now, I'm getting shameless :
How does one entity create another? F.i. a dying entity leaves a health pack...

(Hope, you don't mind my questions, for I don't have the time to take a closer look at Unity now, but I'm veeery curious. It is more a sort of entertainment, since there's not much happening at these forums at the moment and I enjoy your company... So, please don't feel forced to answer!)

Well there's more to it than that. The special thing about wait(1) in A7 is that function execution continues from that spot the next frame, with all local variables preserved.

Assuming Unity lets you extend its entity objects by adding your own variables to the objects themselves (which I'm sure it does, but I just don't know), this wouldn't be an issue anyway, I guess.

Jibb

there is a function smiliar to ent_create.

Instantiate(object/prefab, position, rotation);


edit: object/prefab can have several attached scripts to itself.

There is the function Instantiate(Load.Ressource("Prefabname"), Position, Rotation) which will create that prefab. A prefab is usually an object with script, collision hull and some other components, where in some way everything is a member of the script if it has one.

there also is something similar to wait() though:

http://unity3d.com/support/documentation/ScriptReference/index.Coroutines_26_Yield.html

Thanks for your answers so far!
The explanations from the link look quite clear.
Can't wait to read through the manual and take the first steps of scripting when I have time.

I find this all very exciting. I plan to do some more testing soon to compare the power and indoor scene management of different game engines.

I currently finished a new sci-fi pack and started to work on a few models / indoors for the next pack. Here is a first test in the C4 engine:



more images here:
http://www.terathon.com/forums/viewtopic.php?f=30&p=81005#p81002

I am really interested to see how other technology performs, when I add lights, shadows and shaders in a similar way and how I can optimize culling (zones and portals).
Since I have a license of Torque3D, I will do a test there also. I own C4 and I probably will get Unity 3 soon to better support our Unity customers.
But for testing A7 I probably need some help.

I could probably help you if you want as I am also courious to see the differences. I also think that hummel would love to help me . Even though I currently believe that in such a quite small(?) scene there shouldn´t be many differences. And I wonder where the shadows are on your screenshot. Are they static?

I made this corridor yesterday, so yes, it is still small. But the upcoming pack will have lots of parts to construct a level. So I want to have several rooms and corridors.

There are shadows as you can see behind the big pipes. The small lights at the bottom don't cast shadows to save performance.
And yes, there is a static flag for lights in C4 to not calculate shadow volume each frame. Such optimizations could make a difference. We will see.

a great example of my point. for instance right now I'm using unity for a web-based casual puzzle game. It's great for that, but on the other hand, once I actually find an apartment in Las Vegas (just moved) then i can get decent internet again and start working on my morpg again (its less than satisfying having the wifi around here cut out constantly).

all that said, I prefer the programmability of A7 (btw everyone talks about work-flow this, work-flow that, the work-flow is a piece of cake if you know how to do it, and stay away from wed.... I mean anyone who tries to use wed then complains has to have some sort of mental deficiency of some sort). Anyways, I need shaders, I need decent multiplayer and database functionality, and I need a decent programming language for my morg and I'd like to not pay 1.5k for a project that is the type that tends to fail in the end. I'd like it to not, but life might just bend me over like it does from time to time.

It's easier to drag&drop some image onto a script in Unity or simply load it into your assets folder than define multiple (hardcoded!) BMAP*s in your scripts.
Also the possibility to write your c# scripts in Visual Studio (with runtime debugging) and have access to most of the .NET functions is a nive to have.

In my opinion Unity speeds things up a lot!

drag and drop for assets is nice, i have to concede that unity definately wins there but blender's interface isnt too bad for assets if you know how to use it correctly, and thats what i use for my level design. and to be honest it wouldnt be that hard to set up a resource file system for a7 if someone spent a few hours on it. That would make it so that you dont have to hard code it.

Thats the point that makes a7 nice, for much cheaper you get the core features to work with and create cool features, JCL for some reason just wont make any standard libraries of those systems, unity admittedly does have a nice resource system, list of included shader, already made shadowmapping system etc..... but cost 7.5x as much for the same abilities, and if you want to customize them you'll probably still have to rewrite lots of stuff.

To make it very short:
If you are an indie game dev then you dont have much time. You will not earn much with a small game. You have to make them fast and efficiently and release them on many platforms if possible.

If you are a hobby user, then all this does not matter and you have all the time in the world to customize, learn, optimize, plug-in, add-on and trying to keep pace with the other ones while not really progressing on the actual game. But as Slin told, some of these users dont want to make games, they just have fun with technology. And I understand this, I love technology as well

@Machinery_Frank :

You resume very well the situation
People like me that would want to make a game and sell it, even
little games, don't want to waste time writing tools, plugins ,
or missing basic today standard stuff.
A7 is already good in it's actual state, and allow already to do
lot of things even if it needs more work.

well i didnt wanted to be the killjoy but after reading all of this i thought i had to share some thoughts too.

Reading all of those excitement posts regarding the new unity i still keep wondering one thing: Where the hell are those fabulous state of technology projects hiding that cant be finished and released because of some indy game engine limitations.
It sounds like all of us cant make the next mass effect, unreal or wow just because we cant get gs or unity or averagejoeengine to run some basic shaders.
we are all just waiting for this engine that will make it possible for us to beat high end high payed professional gamedeveloper studios out there.

What the hell happened to "design your project and pick the engine that fits your needs". not worth anything anymore?!

People here are complaining about tools and useability while at the same time there is not one gs project or even indy game project on the way that would require any better workflow then we already have.
There are no huge landscapes or levels out there for indy games. The reasons is simple: WE LACK MANPOWER

I totally disagree with the statement that indys dont have time. If we have something then its time. Compared to big studios we are Tiger Wooding in time.


The only reason why indys fail is very very very simple. It has nothing to do with shaders, or leveling tools or engines. Its all about STAMINA.
Finishing your project requires hard hard hard hard hard work.
And most of us here start gamedev because they wanna have fun and not hard work and sleepless nights.

You guys argue about the color of the hammer you want to use to hang up a picture while on the other hand you dont even want to paint it.


Unity is as good as any other engine. It offers a few more tools. It has a decent coding language but in no way will this make games for you.
As far as i am concerned i dont like the pricing politics of unity. Paying everything extra and having to upgrade the engine every 6 months aint my way of developer symbiosis (the engine was released barley a year ago and now its already version 3? how about some free updates here)


First game then game engine. At lest thats the way i see it.
if you like features and shaders get yourself a game with modding tools and create levels and mods for it. dont waste your time and health trying to create games. you will fail anyway.


Goooooo engines! The holy grail and main excuse for unfinished projects.
May you live forever!

cheese

Actually Unity's 2.x branch was released autumn 2007 and all the updates were free afaik. Just the free Indie version is pretty young so I don't think there's a totally short upgrade period but a pretty reasonable one (I don't know anything about upgrade pricing though)...

Concerning Unity itself my opinion is that it's the by far better choice when you actually have a small team and want to get serious. When your goal is a game you can sell those investments for a full license shoulnd't be a showstopper. For Lone Wolves or people on an extremely tight budget ... well I think those have to choose for themselves which engine they like better. I'd prefer Unity over 3DGS right now especially when having in mind that the 3DGS development seems to have about stopped except for some MMO capabilities which might make it into everyones engine due to their work for those guys hiring the Conitec devs to help them on that task. Without wanting to be dismissive I still don't really understand that move neither from the client side choosing 3DGS as basis for their MMO nor from Conitec thinking that this is what is the way to go for this engine but maybe it's a pretty profitable job. Anyway in my opinion 3DGS is designed for newbies, hobbyists and Indie's taking their first serious steps. MMO really isn't compatible with this as sort of the "champions league" of computer games both in terms of know-how or difficulty as well as money to run your servers. But oh well maybe they'll surprise us all...

this, i absolutely agree.(In fact that's what i have been trying to say for years, but could not put it this simple.)

so what we are arguing here is that is some engines require less MANPOWER in some parts of development, other engines requires less MANPOWER in other areaes. So yes, first project then the fitting engine.

Well put.

I absolutely agree here. But on the other hand I have read that 2009 was the year of independent game developers. Many people painted their pictures and used the fitting hammer to nail it on the wall.
There were thousands of them and many of them earned good money. They filled many gaps. Lots of good adventures came from small developers. Most of the mobile games were made by indies and there are web games from small and medium developers.

The indie market grew while big publishers lost ground with packed products.

And fact is that most of these indies switched to Unity. I have read about ten-thousands of flash developers who changed to Unity, many of them want to change.

I also have read that Unity sold more than 90 thousand licenses after they released their free base version. So this marketing tool worked fine for them.
I would call this a success.

Regarding the price policy:
When I keep in mind that they integrated 2 professional middleware tools like Umbra and Beast then it is absolutely fine to pay this little upgrade price for the technology. This is the right way to grow this technology, to let it become the best tool available.

If all this is not needed, there are enough alternatives like Ogre, Irrlicht, T3D, C4, Shiva or Gamestudio, though it depends on the platforms you are aiming for.

If all this is not needed, there are enough alternatives like Ogre, Irrlicht, T3D, C4, Shiva or Gamestudio, though it depends on the platforms you are aiming for.

It depends, but some engines need you to program tools or additionnal plugins or things like Ogre3D ou Irrlicht even 3DGS.
C4 need strong C++ and lot of line of code (juste look at the character animation tutorial).
Unity is the ready to go 3D Engine with all standard tools , shaders ready to go.

And still, without good coding skills, you won't be able to archieve your goals. When looking at questions and answers you will notice that unity lacks programmers and that many users there are maily doing trial and error which is time consuming and doesn't really result in good code. If you don't want to write code or pay someone for doing so, you should stick to modding your favourite games or using something like fps creator.

I'm not sure if this has already being discussed in this topic. But does Unity have a integrated Code Editor (like SED or better) or do you need some external IDE to programm for Unity?

I'm more a programmer than an artist (a lot more programmer actually ) and to me the workflow of unity looks "strange".

My workflow with 3DGS is the following:
- Create test level in WED if needed
- Setup script files, load level in script (if needed)
- script everything in SED, create most entities via script or place them in WED (if needed)

As you can see I use SED the most and run everything from there. How is the workflow of a programmer in Unity?

I appreciate any sharing of experience!
Thank you

They already did.
http://www.opserver.de/ubb7/ubbthreads.php?ubb=showflat&Number=304417&page=1

yeah they have a script editor, it isnt exactly integrated, but if you open a script from within unity's project list a popout script editor comes up.

But there is also a sync tool for Visual Studio, so you can use an industrial proven programming environment.

This is something that I really dont understand
It is a common opinion that 3DGS is for programmers while Unity for artists
Why ?
Unity3d use C#, JavaScript and Boo scripts
Last edition even supports Visual studio

C lite is not that bad but is far away from being a professional language
it may be easy, ok but it is not definitely a professional tool

I quitted with with 3dgs when it was clear that Conitec was not going to support C#, as promised

The only thing that looks strange to me in Unity are the " components "
I still prefer the traditional OO programming style, same as Leadwerks or C4 , but it may be just a a matter of habit

Many professional game programmers like " components "

Yes it's stupid.
I am 90% of my Unity developing time busy in a coding editor and not in the editor itself.
People don't have to use prefabs, or whatever but things like this make your life so much easier.

Maybe some people should just try out the free version, play around a while (more than 5 minutes) and then comment about workflow, etc.

I'm sorry that I'm interested in a toolset which I currently have no time for to explore. To me all this stuff is "just" a hobby.
Of course lite-C is not professional at all and Unity looks interesting because of C#.
But still thank your for explaining everything so detailed. I'll consider checking out Unity when I have some spare time.

I am not familiar in components, but one senior software head from headquarter said we have functions, later we group functions into objects (at this point I understood), then we group objects into packages (?), and findally we have components. I am not sure I am understand fully what he means.

In many programming languages you can find components. They are indeed some packed data and functionality. They often contain code but can also contain other resources like bitmaps, buttons or text. All this is like a black box, could even come without source in many cases. You get an interface to talk with such an component. Many of them even have real-time feedback. I worked with some very advanced ones like a complete word processor. So you can change parameters for behaviour and appearance. You can read values (like the currently selected text) or you can write values (insert a text at the current position). And then you have functionality like saving with just a single call of a method.

To make it short: A component can be very powerful while only exposing a structured and documented interface to you. It often allows to use lots of functionality in a couple of minutes.

I tried Unity , and now I see myself using Unity and leaving Gamestudio.

WHY? Because JCL is too lazy to hire new programmers

Your'e kidding ?




gri

it certainly is advantageous to work with non-proprietary languages like c#, boo, python or lua. they are usable in many other fields and so experience in them is a much more worthwhile skill to have.

with proprietary languages like lite-c, unityscript (it's not javascript!) or torquescript you are totally locked in. they also often can't compete feature-wise because they don't have the same manpower and communities behind them. in the case of lite-c i think the problems mainly are lacking oo and very weak debugging (not even a stack trace, crashes take sed down,...).

Close enough to give you some skills that you can use in normal javascript too. And it`s a good entry level into programming.



I want to make games only. And i want to make it as fast as possible. Javascript for Unity is easier and faster to learn and to use than C# or BOO. And speedwise there is no difference in the result to C# or BOO. So i`m fine with that

Well, SED doesn´t seem to crash anymore, at least it didn´t for me for a long time, but debugging could be improved of course. But doesn´t unity have nothing for debugging other than log messages at the moment? The language in my opinion also shouldn´t matter as you should be able to get the basics of any new language in a few hours if you know one quite well and have a basic understanding of how things work.

The component system of Unity is actually very nice if you just accept it, which is a bit hard if you aren´t used to something like that.
I am probably completely wrong, but currently my following explanation on how it may work was capable to explain me all behaviour and syntax for the work with components, I got to know so far:
The base component is called "GameObject". It has a list where you can add new components to, where each component is the pointer of the components classes instance. If you attach a script component, it always gets the pointers to the other base components set if they exist and otherwize null. Most components actually get this information so that you can easily access another component from each other. If you don´t have such a direct pointer, you can just search for the component in the components array attached to your object using GetComponent().
So you basicly just have those components, can link them together and get reach each other component from each component.
This system is clean, quite comfortable, well organized and basicly usefull for the one working with the editor But it makes you code quite modular and if you aren´t coding modular, it can easily get a bit messy But you often want moduls anyways to be able to reuse code.
I haven´t yet gone into unities shader system, but the editor seems to dislike it if you open windows forms. It crashes most of the time for me, but works well in the published version...

i don't think javascript and unityscript are very close. the syntax may look quite similar at a first glance but javascript is dynamically typed, oo is based on the concept of prototypes and there are many other differences.

WHY? Because JCL is too lazy to hire new programmers

No that's a matter of business, rentability.

3DGS is outdated in terms of interface and tools.
They need programmers indeed to redo all editors and make only one and a modern one.
And with Unity now being free , 3DGS feel very outdated also.
I like it, but i don't feel to spend money for the future commercial version of A8.
In Unity free version, even without shadows you can do lot of things if you use ligthmaps and some blob shadow for characters.

One way fo A7 (future A8) would be to make it open source like Ogre3D and perhaps with a special license to use it ?

Ogre 3D is very powerfull :
Ogre3D user project
It would need a all in one editor to compete with Unity easily.

I think A8 will remain an engine for programmers mainly , and programmers don't mind about workflow or a complete all in one editor !

Still not get the whole picture of component, too abstract for me.

However if programming language evolves from function -> object -> package -> component, that imply lite-C just like the editors of GamesStudio

Lets make concrete examples

#case 1 OO programming style - A Race game

You declare a class Pilot() which contain generic variables and generic methods
Then you instance objects : Schumaker, Alonso, Massa etc
In the game play you call

Schumaker.Move(); Massa.Collide();

There are 20 or so pilots in the race who share the same methods and the same variables
However this is not alwayes the case

#Case 2 - Components , an RPG game

You have a lot of different type of characters in the game
In case of 00P , You should

a) To declare few complicated classes and sub classes which instance many objects
b) To declare many simple classes which instance few objects

Or something in between , in any case it is evident that OOP is not an optimal solution

Using components you create your objects and indipendent components, as well
Components are basically a set of variables and functions performing a certain task
Then you attach the components to your objects, just the ones which are stricly needed

Apparently it is an optimal solution for many games
However , I came across some issues

a) It is hard for me to think of generic components which are suitable for a variety of different entities
On the other hand if you code specific components for specific objects you lose the advantage of having generic re usable components

b)The functions of the same class share the same variables and they work well together, having been created togheter
Components, of course, do not, being created as indipendent items
A component must call , for example, an other component even though they are attached to the same object


May be it is just a matter of habit

Actually, I'm serious. But I won't leave the community. I just can't.

I'm gonna renew my credit card so I can renew my Paypal account so I can sell my Gamestudio so I can buy a new iPod Touch instead of that broken one (It fell from my pocket, when I got back to that spot, a car ran over it )

What???!

I hope your aware that scripting languages linked to engines is not only professional but considered a norm in the game programming industry, why do you think lua has lasted so long.

Thanks for the explanation about component. It that similar to interface in Java?

While this is true, you have to keep in mind, that all professional game engines (including Ogre, Torque and C4) use scripting as an additional tool besides their real programming language. They use scripting for prototyping and for level design or for fine tuning (like adapting an AI). But all core elements get programmed in the end. The reason is optimization, speed. AI as an example can "think" multi-threaded in background while the script can change parameters of this behaviour.

Scripting an entire larger scaled game can lead to problems like you can see when you watch current problems of the very ambitious and excellent Intense-X.
But it will be sufficient for smaller prototypes and casual games, as you can see from all these flash projects.

what is the the difference between scripting and programming? between scripting languages and "real" programming languages? compiled vs. interpreted? this all is very fuzzy. "scripting" often gets used deprecatingly by guys who think only c++ is a real programming language. hm... but there are interpreters for c/c++ too for example. what is it then? there also are more and more jit compiled languages (c#,...) where the line between compiled and interpreted is very blurry.

i don't think that lite-c is unprofessional. to me it's more a matter of proprietary vs. non-proprietary like i already mentioned above.

Yes, I agree. Keeping C# in mind, there is a blurry line.

There might be another reason why someone prefers C#, Delphi or Java over Lite-C, LUA or JavaScript. The reason might be how comfortable you can work with it (debugger and documentation only as 2 examples).

Never used Java
However it is not so important the programming lamguage by itself rather the game engine architecture

Using a traditional OO game engine the developer must reason in term of entities
The engine supplies classes such as : Camera, Light, Player, Scene etc
Each class contain the basic functions which you may expect from the class name

Using a Component game engine the developer must reason in term of tasks
The engine supplies component such as : Transform , animation etc
Each component contain functions which you may expect from the component name
The functions of the component Animation can be used to animate anything which can be animated in a game
either a player or a light or a camera etc

To me the traditional approach comes more natural

Scripting languages such as Lua are very popular in game programming due to the fact that they run on virtual machines thus you dont need to recompile anytime you make a modification
This is very important for example for a " trial and error " tuning of your AI

Well Lite C is not that bad but I would not say it is also an elegant programming language
I am so happy to get rid of those " void ponter " bloody messages , just as an example

c# and java run on virtual machines too. why is lua supposed to be a scripting language and c# and java not? there even is a very fast jit compiler for lua. this scripting/programming distinction is a bit silly in my opinion.

i agree about lite-c. i often find it way too low level for writing gameplay code. it can be very cumbersome...

[quote=Cowabanga] ... and now ...I leaving Actually, I'm serious. But I won't leave the community. I just can't.
[quote]

i agree im done with lite-c but not the community

Oh I think I get it. So a component is like a class without instances, and its methods don't refer to itself -- each method is called by referring to the component name?
I like low-level for things that need to be really efficient -- such as game programming (although the CPU load is perhaps less an issue now in games). Perhaps not so much for event scripting and some things, but I still don't have a problem with it.

Jibb

A8 has just been announced.

It was about time for A8 but the feature list didn't knocked me off my feet. For me Unity has risen the bar.

Also the prices for the pro versions at the moment aren't that different:
A7 Pro: 798.49 EUR
Unity 2.5 Pro: 880.00 EUR

I'm somewhat curious if the A3/A4 textures are being replaced in A8

Yes, this is the time i get to decide which way i am going to go. If i dont get what i expect from a8, i will switch to unity.

(altough if the price to upgrade from is same with a6->a7 comm. upgrage, i will still upgrade )

Yes,In Unity3d you have for example a class Transform which includes the functions : Translate,Rotate,RotateAround,LookAt...
and the Inherited Variable : transform
If you want to move your object you can attach a script to it with the command : transform.Translate(1,1,1);

In TrueVision3d you have a class named TVActor which include the method MoveRelative()
You instance your object and you write :

myObject.MoveRelative(1,1,1,);

In my opinion the latter approach is more intuitive

Same here
Once I have placed in the scene my characters, buildings, trees, lights etc I wonder what else shall I do more in the editor ?

I spend my time in tuning my game , by trial and errors
I change a variable , I run the game , I observe , I tweak my code etc
Such exercise take 90% of my time

I suspect that many people speak of games but they actually mean virtual reality

Anyway me and you are lonely wolves
Everybody seems to be interested in graphics only

Is there btw a reason for the mouse position being given based on the bottom left corner, while all 2D elements have their origin on the top left? o.O

I agree. I thought that was the point of OO ^^

Jibb

hm... i think there is some confusion about components.

unity uses objects just like other engines.

this.transform.position = ...

"this" is the current object. i haven't used unityscript yet but probably "this" is implicit there and you can leave it away?

in gamestudio the pendant is "my".

my.x = ...

Now thats a picky non related question which should belong to unity forum lol.

But you can always use Screen.height - Input.mousePosition.y

Though i don't know right now if it's really inverted...


@Ventilator, yes you can leave the "this" away. I never use it.

It's just useful if use it like this way:



But this is not Unity related and you probably know it by yourself anyway.

yes, i just think that leaving away "this" can lead to confusion.

the zen of python:
http://www.python.org/dev/peps/pep-0020/
"explicit is better than implicit."

Kiyaku, it of course is picky, but it took me a lot of time to figure out why rect.Contains() returned wrong values for my mouse position. The worst thing was that it was only a problem for one rect, as all others where kinda mirrored...
What I basicly wanted to express was, that even Unity isn´t perfect even though quite close, as far as I got to know it until now
I also had a bit trouble to figure out that network view system, as I didn´t find much information about it (using the search through the answers stuff and the reference manual thingy), there are some more things which aren´t documented too great. I for example just searched for a way to adjust the speed of an animation. It is actually very easy, but the information on how to do it for just a single animation with example wasn´t really were I expected it to be.
But well, there really isn´t much to complain about

You want to have a character walking on the screen

3dgs

you load a model in the editor
You create an Action in SED :
ent_Animate(my,...);
c_move(my,..);
you assign the Action to the model

Unity3d

you load a model in the editor
you create a script :
transform.Translate(...);
animation.Play("Walk",..);
you assign the script to the model

Truevision3D

The engine supplies a TVActor() class

you instance your model
TVActor myModel;

You load the model
myModel = TVActor.Load(".."):

In the game play you write
myModel.Move(...);
myModel.PlayAnimation("Walk",...);

Are exactly the same stuff for you ? I dont think so

In my opinion TrueVision only allow a true traditional OOP
Unity3d is on Component side, Components are also classes, if it is what you meant
3dgs is something in between
Actions in my opinion are components created by developer
3DGS is probably more flexibile than other engines but I find the code style quite ugly...personal taste of course

yes, it's the same stuff to me. yes, components are just objects attached to other objects.

in unity it's:

this.transform.Translate(...); // "this" can be left away. internally "this" still will be used then.
this.animation.Play("Walk", ...);

or:

someotherobject.transform.Translate(...);
someotherobject.animation.Play("Walk", ...);



in gamestudio:

c_move(my, ...);
ent_animate(my, ...);

c_move(someotherobject, ...);
ent_animate(someotherobject, ...);

if gamestudio used object oriented syntax it would be (that's how it works with my python wrapper):

my.move(...);
my.animate(...);

someotherobject.move(...);
someotherobject.animate(...);



i don't see the big difference to unity there? and i don't think gamestudio is more flexible. the lack of oo rather makes it less flexible in my opinion (or maybe not less flexible but certainly more cumbersome in many situations).

C# together with unity is probably the most powerful solution.

Unity3d and 3dgs are about the same stuff but TrueVision3d game architecture is different
The formers are component oriented engines
The latter an OO engine

And wtf is the difference between

and

?

I quote from " The complete guide to Torque x " by John Kanalakis

Quote

Torque x promotes a new approach to game programming.
Unlike the classic inheritance model,objects share common components that provide reusable functionality instead of sharing a common base class.
For example when an object needs to process collisions, you simply add a collision component.
The aggregation methodology results in more code reuse and less copy and past code sharing "

and again

" Object oriented languages such as C++ and C# promote the practice of inheritance.In this case , new functionality is added to an existing object by deriving a new class and adding the functionality there.
....over Time it becomes increasingly complicated to add new functionality to an existing class hierarchy "

again

" Components are simply packaged element of game functionality.
In Chapter 2 we used the physics, Collision, worldLimits components...
In the following chapters we wil create several additional components from movement components to weapon components to AI components.
Each of this components will be designed to be as generic as possible to add functionality to a wide variety of games "

unquote

I do not know how to explain it better, to me the different game engine architecture is evident

In classic OOP you focus on objects and you add functionalities to your objects
C++ has not been designed for games
Suppose you want to develop a database for a Company
You declare a Class Employee which contains variable such as : id , name , salary etc and functions such as : CalculateSalary() etc
You can instance hundreds or even thousamds employees objects: Mr Smith, Mr John etc

This may not be the case of games

In games it may be better to focus on functionalities , I called them "task" in my previous post
Functionalities can be : Movement, Collision,Physics etc

Of course you must link the components to the objects thus you must use pointers such as "my" or "this" or such pointers may be even implicit

ah, now i understand a bit better what you mean.

i think this is a matter of what programming style you prefer. some people like to work with complicated class hierarchies (i don't) and some prefer to keep the oo design simple and work more with classes that don't rely on inheritance that much but simply are aggregates of objects of other classes.

but i still would call both styles object oriented and i think with most engines you can decide yourself how you want to work. i don't have any experience with truevision3d though and i am still a beginner with unity.

Yes both are object oriented programming

I meant that an engine such as Truevision3d implement a traditional OOP style while Unity and Torque x propose a new approach to OOP

3DGS is in my opinion a precursor of the component phylosophi

I remember when I switched to 3dgs some years ago, from TrueVison and blitz3d mainly because they lack an editor, that I found 3dgs programming style a little odd

Which is better?
Well probably it depends on taste , not to mention that you can have hybrid solution in the same engine

John Kanalakis is of course on Component side but I tried to use Torque X , believe me it is not that easy
Traditional OOP may have some drawbacks but also the new approach has its issues
Go through Torque x tutorial projects , they are far away from being intuitive

Unity3d is a little easier because it supplies a lot of prefabbricated components while Torque x supplies only the basic ones but even Unity is not that easy to use

However getting back to my first post

I wonder why people consider Unity3d an engine for artists
Unity3d is not an user friendly engine, maybe it is a powerfull engine but I would not say that it is an engine for "non programmers "
Maybe they tested the editor only

The truth of course is just in between. A pure artist will not make a game and a pure programmer also will not sell something. If you are making a game on your own, then you need to be both of them or you need help.

But Unity sold 90 thousand licenses after they released the basic edition for free. The new licensees had enough time to check it, to read tutorials and to get an idea how hard programming really is:
http://unity3d.com/support/resources/tutorials/2d-gameplay-tutorial
http://unity3d.com/support/resources/tutorials/fpstutorial
http://unity3d.com/support/resources/tutorials/3d-platform-game

But with tutorials like this it is feasible. And this might be the big difference to other engines: the learning curve is not so steep in the beginning. This attracts way more users.

And with this in mind you should not underestimate ambitious artists. A skilled artist already knows most of these 3d concepts. They know about the concept of instantiation, 3d-coordinates, transforms, uv-maps. Often they create custom shaders. Many of them even scripted in their 3d modelling tools.

But in the end it does not matter where you come from when you are willing to learn and when you have fun doing it.

right , but posts about game engines are not something in between
It seems that people choose an engine just because of the editor
Take for example Unity3d and Leadwerks
The former use the component approach the latter the traditional OOP
Sometimes someone express some preference about a programming language but C++ , at a basic lavel, is neither mor powerful nor less user friendly than Lite c
It is about the same stuff
It is the game engine architecture which makes the difference
The Leadwerks, C4, Unity , 3dgs... programming style is quite different but I have never seen a forum discussion about the programming topic
On the other hand in my opinion a one man developer should spend 90 % of his time in the code rather than in the graphic editor
the other way around he makes a Virtual reality project , not a interactive game

I absolutely agree with you.

But there is a little difference, when I personally rate these different tools. I also have my preferences and I personally love some Delphi-like style, a mixture of OO, components and a powerful library with a very good documentation.

But nevertheless I would even program in C++ (I dont like it because you have to deal with so much technical issues of the language instead of writing your code), when this is the only way to get my hand on a good renderer with a good scene management.
In the end I see the language as an interface to a very powerful technology. I would not use a Delphi-Game-Engine, only because I love this language. I am first checking how powerful the technology is. Can it render the world I want to build?
I love sci-fi interiors, but when I dont have zones and portals together with shadows and shaders, then I never can create interesting sci-fi interiors. So I have to check, where I can render something like this.
The same counts for people who want to make a large terrain game. They have to check, make stress tests and then use what works best for them.

I would read a comparison of languages and code styles of different engines. I would take this into account but in the end it would not be the most important feature to make a decision.

Are you guys Unity spies trying to sway 3DGS users away from Conitec!
Why is this post here?

BTW, I am a Torque user, and while I believe there is no perfect engine. I do think that Torque and 3dGS comes close to the mark.
I do not yet own a copy of 3DGS, but it is on my 'must buy' list. I prefer this engine 'authoring system' as conitec calls it over Unity.

Discussions around competitor products are allowed on this forum, so all is fine.

http://arstechnica.com/apple/news/2010/04/apple-takes-aim-at-adobe-or-android.ars

looks like apple is going crazy.

maybe unity (and other similar technologies) will be forbidden on iphone.

http://blogs.unity3d.com/2010/04/10/unity-and-the-iphone-os-4-0/

Yeah everyone is talking about it on the unity forum too.
But as far as i know, apple wants to force people to develope on an apple when the want to release an iPhone game. So Unity might still be supported afterwards. But we will see.

This is a step into the wrong direction. I hope, that more people will change to Adroid and other more open environments as a clear answer to this restrictive policy. Maybe even Windows on mobile phones might become an alternative for some game engines in the future.

To develop a game in C# in Unity is a great advantage. You can use the same code for different platforms. It simply would be a step back to write iPhone games in Objective C instead.

In the end the most interesting question is:
Will it work? Will Apple win developers with this strategy and get rid of cross-platform to harm Windows, Android, Symbian whatever with Apple-exclusive applications?
Or will it work the other way around? Will Apple lose developers? We have to keep in mind that there is no good development tool from Apple, there is no game engine, everyone has to write its own now.

This is a big step back into the middle age of software development.

Well, if it means that a lot of competition in the appstore disappears, I only see an advantage in it for myself. But I guess that the "big players" will very soon find workarounds, so that it doesn´t change much.

How do you want to find a workaround? You have to use an Apple compiler with C, C++ and ObjC.

But there are alternatives like this one:
http://www.edgelib.com/

It is a pure C++ multi-platform engine and supports almost all mobile phones plus Windows, Linux, ect.

C4 is told to be ported to iPhone also and it is pure C++, so it also will be not an issue.

But Unity, Flash and maybe Torque will suffer a lot.

theoretically they could do a c# to c compiler.

Not only on the unity forum, on the Apple developer Forum are also many people (maybe the same?) who wants some clarification from Apple (I can't go into detail because I'm bound to the NDA, sry).

So the question is; Why should Apple do this?
I think that it has some legal reasons, cause when a third party framework developer fucks the whole system, Apple gets bad press and must wait for the third party dev to fix the thing. But if there is something wrong with an framework from Apple, they can just fix it in very short time with an OS update. (Thats my personal opinion and nothing from the dev Forum and/or the Beta SDK!)

Yes, this is what Steve Jobs wrote in some of his emails. But is there a good framework from Apple? Is there any game engine from Apple, physics layer, sound engine, collision detection, scene management? Or is each developer now doomed to program its own system from scratch? Is this the step back to the middle age of programming? Back to the pointers, memory allocators and pointers to pointers to manage stacks, get rid of good garbage collection and powerful function libraries?

Weren´t you the one posting for example edgelib above?

Yes. Although edgelib is C++ (with all known advantages and disadvantages), it at least offers some function libraries.

i just got the free version of unity. i must say that it is good. i still will work with 3dgs though. unity will be more of a fun thing.

http://blogs.unity3d.com/2010/04/24/car-tutorial/

finally they have posted their car tutorial. looks very interesting.

i am sure a lot of stuff can be learned from it that can also be useful for gamestudio projects.

Yeah I already got the car tutorial
It's really amazing!

An mmo made by 3 guys in 3 months with Unity !

Unity mmo

Yeah MMO is possible with Unity and few people , but if you use the right network lib and really skilled artist.
This mmo is done i think just to demonstrate the network lib capablity , if i have read well.
For this game they used Badumna Network Suite.
badumna

if you use the right network lib and really skilled artist, you can make mmo with almost any indie grade engine.

I really love the style of the game

Another engine is going to this direction, you have the choice of using C/C++/Objective C/LUA scripting to C translation.

http://developer.stonetrip.com/index.php...p;limitstart=12

come an guys,3dgs is great tool,but sure it need some small improvements,especially MED WED,lite-c is OK,I agree with you,but don't forget
"the ease of use",and they already developing a new engine(A8),
plus...wait a minute who am i kidding,,,the glorious old days of almighty 3dgs
are gone..with no return..
sound's poetical lol.

we'll all miss gamestudio , when I got into game dev gamestudio was the best there was, well with torque but 3DGS was hands down the best for indie dev because of it's power and ease of use. But as the other engines have started to vastly progress , gamestudio seems to be staying farther and farther behind. But they ARE working on A8 , so who knows , maybe , just maybe , they might surprise us and release a kick ass GameStudio. But in all truth , I HIGHLY doubt , though only time will tell.

March 27, 2008:
"The A7 version sold more copies since its 2007 release
than its A6 predecessor in its whole lifetime."

I would call this a success.

correct,but A6 misses alot of features relatively to A7,even though A7 is also misses alot of features relatively to other competitors Engines such as unity 2.6 not "unity 3!!",torque,...etc,
don't forget that when A6 was in it's glorious days it did'nt support SHADERs,this technology been around for long time even before A6 is out,sure A7 got alot features than A6 ,but again comparing to other competitors Engines,whell..you got the picture.

take alook at unity 2.6,great shaders,real time dynamic soft shadows,the editor is amazing,fallbacks to really old GPUs....
i think the only bad think about it is: those so many damn .DLLs created when publish

I love A7,but it need's alot of work,I hope A8 Won't disappoint us.

How can Unity be as fast as 3dgs even to old GPUs ?
3DGS use ABT / BSP scene management algo's , i.e a huge amount of data are precalculated
I can understand that this feature may be not that important for new GPUs capable of rendering milion tris but for old ones it should make the difference

@AlbertoT I think what he meant was that shaders have good fallbacks to older hardware. So you can have your game with shaders, and if somebody plays it with a pc that doesn't support those specific shaders it will fall back to an older version of the shaders that the pc supports.

Exactly!.




why should 3dgs users always aim at OLD"even extinct"machines???come an man take alook around,forget unity for now,every game in the markets today"or even 3-4 years before"uses technologys that 3dgs A7
could'nt dream of.all of this AAA games need's medium2fast machines,and they sold too many copies of them.you can't consider 3dgs ability to run on old machines as feature,even though 3dgs not that fast after all.put some shader's in the scene and some ENTITY's
with stencil shadows"not even blured" and the engine will become a monster eating all your mashine's resources,if not use this tow neat features you'r game become as old as quake"even quake uses some sort of decals shadows!!!".

again i love 3dgs,but it definitely need's some improvement,

It is a matter fact that Unity3d game engine is well optimized
I tested Unity, dxstudio, Leadwerks
Even though I would prefer Leadewerks workflow over DXSTUDIO and DXSTUDIO workflow over Unity I finally went for unity, Unity being much faster on modern PC
However i doubt that , generally speaking,Unity can be faster than 3dgs on old PC'a unless , as you said, he was talking about shaders only

I even can get more FPS in Unity then Gamestudio. Shaders in Unity are waaaaaay faster than Gamestudio's.

Unity wins. Unity kicks Gamestudio's ass away behind the game engines. Face it, it's the truth.

lol

@Cowabanga: Shaders in A8 are fast

Have seen it myself

Ok first, it's Cowabanga.
And second, is it faster than Unity?

Sorry bout the e/a ^^

Don't know if its faster but it is at least 6 times faster than before...

No problem

Well, since Unity's about 10 times faster (Well at least on my PC) than A7, then it's probably 4 times faster than A8.

Unity wins, deeply.

this discussion is pointles, since all this shader performance thing is highly dependant on GPU.

...Do you enjoy ruining my discussion?

Strange math.
A8 = 6 * A7
=>[1] A7 = 1/6 * A8
[2] Unity = 10 * A7
[1] in [2]=> Unity = 10/6 * A8 => Unity = 1.6666 * A8

Did you even test it with two identical shaders? Or are you just guessing?

The point is, that you are posting bullshit cowabange .
As I highly doubt that you are comparing the exact same shaders, within the same scenery, which doesn´t gain any profit of LOD, culling or something similar.

The whole discussion about Unity vs Gamestudio is senseless, as there are pros and cons for both of them. Gamestudio is great for what it is and the same goes for Unity. And the biggest problem is, that there are so many people trying to run this discussion, without knowing even one of them well.

Well I know both of them , and feature wise Unity wins , period. However , if you are working on purely a PC game , and have no problem dealing with the workarounds needed for some of GameStudio's problems , then , yeah , Gamestudio is sufficient. But who would prefer to pay for a subpar outdated goofy looking engine when you can get a professional engine for free ?

Unity free is by todays standards "outdated".

Jibb

Actually, I have a point Slie.

I tried lots of shaders on Unity AND Gamestudio, with one entity, the SAME SHADER, and I get more FPS in Unity, see what I mean, Slie?

@ChrisB: I know, I know. But I'm just too lazy to do all this.

same shader?

exact same shader algorithm? exact same entity?

I highly doubt that YOU (Cowabanga) are able to get the exact same shader running in both engines. Normalmapping does not have equal normalmapping and multitexturing does not have to equal multitexturing. It depends on the shader code and not what the shaders are called.

Same shader: Yes.
Same shader algorithm: Kinda.
Same entity: Yes.
@Slin: Not %100 the same lines of code (Since Unity uses another shader language), but yeah, they do all the same thing, they have a very close algorithm.

then your argument is pointless unless they are using the exact same algorithm and giving the exact same results.

And just for your information...
Unity uses the same shader language as gamestudio...

Cowabanga, what is the shader?

Jibb

Bumped Diffuse and Toon.

No , it's not pointless. If you apply a normals mapping shader to a gamestudio entity and test the fps , and then you load Unity and apply a normals mapping shader to it and test it and it's faster than Gamestudio's , then Unity is performing better , PERIOD. Unity also has fallbacks for it's shaders , so that's another plus. So let's once and for all answer the question , is Unity better than GameStudio , YES , IN EVERY SINGLE ASPECT as a game engine. But now we even have Unreal Engine 3 for free , which is of course better than Unity, GameStudio , and Torque combined, so , while Unity is an awesome engine that kicks gamestudio in the face when compared to it , it isn't the best engine out there , as even Unreal Engine 3 has some steep competition with the Crysis engine , however , UE3 is free, which to this day I can't belief.

I agree,definitely.

is that true??

Yeah, Google "UDK".

WDID: Yes, it is pointless. Do you have any idea what sort of a difference it makes supporting extra dynamic lights? As well as different cards supporting OpenGL better than DirectX and vice versa.

Jibb

EDIT: I've just been looking at the way Unity handles lights, and there is NO WAY you automatically get better performance. Unity adds an extra pass for every light. It's a great way to simplify their shader/material/light system, but since it's forward rendered this will generally be less efficient than a decently-written shader in A7.

What's the point? Let Unity 3 be 1.xxx times faster than A7/A8, it doesn't really matter when you're an indie. Stop making playable graphic demos and start making games again, not the ones that have hundreds of characters and big open worlds (except for Harry Potter ). Unity won't make your games fun to play.

"is that true??"
Yeah, but if you make a game with it you have to give them 25% profit, but other than that , you can download Unreal Development Kit and use it to make a game for free.

http://www.udk.com/

Such discussions always make me smile, discussions will never find an end when there are obviously big differences in knowledge about the discussion theme, like shaders and so on.

Shaders are not an engine. They are little programs running on the 3D card and their speed only depends on your 3D card when the code is similar.

I have tried Unity shaders with the pro trial version and found, A7 kicks Unity's ass into orbit shader-wise. Not because A7 shaders are faster, they are exactly as fast as Unity, but they are better to write. You see the effect immediately while you write the code. Unity can not do that. And for any decent shader effect you must pay 5 times the money that you pay for A7.

I understand shader programmers, that they like A7. But at the same time keep in mind, that engines like Unity or C4 use a different and more closed approach to be cross-platform and to have a good fall-back strategy.

If you fiddle around manually on your shaders, they might work better for a certain situation, but they will not work on other systems in most cases.

Just take C4 as an example. There is a visual shader editor. The reason is that each shader will work under different lighting situations, with or without fog, at different gaming platforms. Each shader will work together with other shaders. They will be compiled from this graph in real-time on each platform just as needed.

You cannot achieve this when you write some HLSL code.

So it comes to the same point like always: A7 is great for learning and prototyping. But for a flexible cross-platform system, working on different targets (weak and good hardware) it will not work as good, or not at all, as competition does.

yes i know that"i read UDK license agian",but imagine what you can do with such an engine,it's first class AAA engine,alot of great games were developed with it"you know that of course",BIO SHOCK-GEAR OF WAR-SPLIENTER CELL-etc..,check out it's features,the only bad thing about it is UNREALSCRIPT"i mean for beginners",iam not really big fan of Java or Javastyle programing.

Well, C4 is a completely different approach. And actually a very good one if it really works without restrictions.
But Unity is different. They just use CG and you have to code shaders and their fallbacks the same as in gamestudio. The only difference which is quite nice, is that you can link to another shader for the fallback, so that different effects can share the same fallback without the programmer having to copy and paste the code. This results in more shaders, that include fallback, which is nice but doesn´t really make a big difference to gamestudio.
There are probably a few more differences, but they should all be small... At least I forgot them, after writing a shader in Unity a few weeks ago .

No, you're mixing two different concepts together. A visual editor is just for visual editing. For lighting, most engines use lighting modules from a shader code library. A7 does it this way and I bet C4 also. A7 shaders also work under all lighting conditions, fog, and so on. You only write the specific code for your shader. Writing code can't be done with a visual shader editor, thats why I prefer the A7 method.

It does work in C4 as far as I know from checking it some time ago.

It is still the same process of shader editing:



You have a lot of processes (basic, mathematical, complex and interpolant). And you connect these pre-defined processes in a similar way as you would program it in your code. It is just another kind of programming, a graphical approach just like Kismet is for Unreal or the graphical scripting approach in C4 (similar to the shader editor).

But the big difference is that it works on more than one platform. It does not matter what is behind these processes (CG, ASM or HLSL). Actually in C4 it is even more low-level than CG.