What's next?
Posted By: jcl
What's next? - 10/31/06 18:50
I normally don't use polls for determining our development planning. But as the new light system and the ABT are now mostly finished and in beta test, I'd like to know your opinion about which of the next planned features is most important for you. Please take a minute to fill in the poll. Thank you.
Posted By: broozar
Re: What's next? - 10/31/06 19:17
what does "more textures for terrain" mean? i guess it's not about textures, but about layers and the possibility to use more than 3 textures in the multiTextureTerrain shader, am i right?
"shadows for terrain" -> i hope the "soft shadows shader" will run on terrain as well, so i chose the shadow solution.
Posted By: Excessus
Re: What's next? - 10/31/06 20:20
It seems from these (preliminary) results that there is alot of interest in the soft shadows. I don't know how much research you have done regarding shadowmapping, but from my own experiments it turns out to be non-trivial. Not sure if you need it, but here is some advice:
For the last week or so I have been working on a
shadow mapping shader. They are not quite soft shadows yet, but as it turns out that is the least of the problems. When using the shader on any scene larger than the one you can see in that thread, serious aliassing problems appear, even when using a huge shadowmap. This is not just a problem of my shader, but a general problem of shadow maps.
Are you planning to create the shadowmapping shader for sunlight, or just small spotlights/omnidirectional lights? Spotlights are not so difficult when they are used on a small area and distance. The same goes for omni lights, but here you will have to use a cubemap or a
Parabloid map. Sunlight/directional lights are the most difficult.
Problems arise when a single shadowmap pixel is projected on multiple screen pixels. This happens when either the surface is (nearly) parallel to the light direction(so there is little space for that surface on the shadowmap), or when the shadowmap resolution is simply too low to cover the area. Another problem is that of 'surface acne': false self (un)shadowing. This happens because of limited numerical precision in the shadowmap or the calculations.
To solve these problems, there are basicly two solutions:
-warping: using a different projection matrix to make optimal use of the shadowmap space.
-partitioning: partition the scene/frustum in some way and using a seperate shadowmap for each one.
Here are some papers on warping and partitioning techniques that look promising:
Parallel-split shadow maps I'm currently implementing this, I can send you my code (although it's quite messy at the moment..).
Perspective shadowmaps First warping technique. Seems to work OK but also difficult to implement. Nevertheless, it's usefull to read this to understand the other warping papers better.
Trapezoidal shadow maps Looks like the best warping technique atm, but it's patented (just like the z-fail shadows implemented in the beta, btw!)
Light space persective shadow maps This is another goodlooking warping technique that is not patented.
There are tons of other papers adressing these problems, but many cannot be implemented in hardware/realtime rendering or simply don't seem to yield good results. Hope this was of some use..
Posted By: vlau
Re: What's next? - 10/31/06 20:20
Same here, I choose "More Textures & Shadows for Terrain".
Posted By: Blattsalat
Re: What's next? - 10/31/06 20:36
Very hard to decide. I picked "vertex weighting" because it is essential imo and we cant get to the next artistic step without them.
main reason is because all updates so far are focused on scene managament, lightning and code and i am afraid if this gets put asaid again we will have to wait another year for it.
Also a benefit would be: once it is done its done. It wont need lots of updates and fixes and shouldnt cause major errors. lightining, shadow aso sound more "fragile" to me and would also take longer to work on all systems.
as said, its very hard to pick from (kids in a candy store) and i wouldnt mind any of them.
a personal note:
After observing the last few weeks and months i have the feeling that conitec is investing a lot more power and energy into gs and especially into the updates.
it looks like they take the user complaints about long update-less times very serious and try to solve it. (reminds me of the first years i spent with gs)
I am sure i am not the only one who likes this new-old way.
cheers
Posted By: PHeMoX
Re: What's next? - 10/31/06 20:59
I don't want to be kiss-ass or something, but the last two updates have been great. Sure, one of them took a long time, but look at what it brought. I would like to see the shadows on terrain feature, especially since I'm leaning more and more towards a 100% model-only approach. With the required features for this almost in place, i think it would be a logical next step.
I like the other features about the same though, so choosing between them is very difficult.
Cheers
Posted By: lyingmime
Re: What's next? - 11/01/06 04:24
Vertex weights are essential for decent bones animation, but I'd also really love to have size prediction and speed-up for truetype text (2D improvements)! Imagine being able to automatically insert line breaks without resorting to a fixed-width font!
Posted By: William
Re: What's next? - 11/01/06 05:30
Quote:
Vertex weights are essential for decent bones animation, but I'd also really love to have size prediction and speed-up for truetype text (2D improvements)!
Basically what i'm thinking. 2d Improvments can really help speed up things(since i'll be doing menus eventually, eww..). So I voted for that, but vertex weights are very important as well. The rest is visual, which can wait a bit(if picking one of them, I choose softshadows).
Wow. That is hard to decide. As an artist I would like to see vertex weights and soft shadows.
But as a game creator I really need decals that bend around corners as well.
I hope that all that comes. I am optimistic and believe that these features can be faster created than a scene manager and a lighting system.
But besides the poll I am happy to hear that the scene and light manager will be finished very soon.
Regards,
Frank
Posted By: Matt_Coles
Re: What's next? - 11/01/06 08:15
decal manager and soft shadows for me
Posted By: Daedelus
Re: What's next? - 11/01/06 09:18
Quote:
decal manager and soft shadows for me
I'll second that even though I voted for soft shadows.
An option for Anti-aliasing wouldn't hurt us either
Posted By: Superku
Re: What's next? - 11/01/06 09:28
The decal management sounds most useful to me (basic feature + fps) or 2D improvements.
Felix
Posted By: Matt_Coles
Re: What's next? - 11/01/06 09:54
I second the anti aliasing option too
Posted By: DocJoe
Re: What's next? - 11/01/06 11:28
New engine functions for opening and closing a session as server or client.
Posted By: FBL
Re: What's next? - 11/01/06 11:49
If I understood correctly the last point includes proper light/shadowmapping for terrain to make it work hand in hand with level blocks when using the new mesh renderer. I think this is very important for using a mixture of both, so I voted for this one.
Second choice would be soft shadows...
Posted By: Turotulco
Re: What's next? - 11/01/06 12:58
I'm also more interested in a greater use of models on terrain, so shadows on terrain are of great interest.
Posted By: Felixsg
Re: What's next? - 11/01/06 16:15
I vote for the decal system
by I really prefer all experimental features
(how the ABT and the light system) include
in the final version
quote: Why not the public beta are the same of the internal beta can that be changed?
Posted By: ello
Re: What's next? - 11/02/06 07:48
Quote:
I'm also more interested in a greater use of models on terrain, so shadows on terrain are of great interest.
from how i understand it, softshadow m apping includes shadows on terrain.. or am i wrong here?
Posted By: ChrisB
Re: What's next? - 11/02/06 13:25
I've voted for Decal Manager, because its a great and maybe must-have feature for an engine.
As far as i know the soft shadow would be just a template, therefor it could be "easily" done by the user.
I've chose the soft shadow, because I think shadows are one of the most important things in a scene. My second choice would be the Decal Manager.
Dark_samurai
All the voters: Do you see the problem of the poll? Most of you say that decal management is their second choice. But nobody votes for them. Since soft shadows have enough votes it could be prudent to vote for the second choice. At the moment even 2d-improvements have more votes than decals.
But anyway, I hope that all features will come more or less.
Posted By: Superku
Re: What's next? - 11/02/06 14:25
Maybe there should be a new poll with (in particular) "Soft Shadow Shader" and "More Textures & Shadows for Terrain" explained?
Posted By: Shadow969
Re: What's next? - 11/02/06 17:19
vertex weights because its hard to create realistic organic model animation without it
Posted By: Orange Brat
Re: What's next? - 11/02/06 23:38
Quote:
Maybe there should be a new poll with (in particular) "Soft Shadow Shader" and "More Textures & Shadows for Terrain" explained?
Soft dynamic shadow shader for models(like Sphere only it comes with 3DGS) and more textures for blending and static shadows for terrain.
Posted By: MaxF
Re: What's next? - 11/03/06 15:52
Hi All
Yes I would like "Soft Shadow Shader" for all models, static meshes, terrain, everything..
It would be nice in the GUI if you could control the softness of the shadow.
Posted By: testDummy
Re: What's next? - 11/04/06 06:44
Well, I see that "Soft Shadow Shader" seems to be winning in the poll. I would have placed that feature last in the list, in fact, that feature probably wouldn't ever be on any of my feature lists. (This bargain PC doesn't even support 'real' shaders and such a feature might be useless for users with similar PC configurations.) Actually, this is expected, I think the majority rarely makes the same choices I would make.
As for all such polls, I assume you are supposed to vote once for each user name you have.
well personally i think adding vertex weights will greatly imporve the quality of models and there animations and that feature will work on all pcs that can run game studio games.
Posted By: er_willy
Re: What's next? - 11/04/06 11:54
vertex weights is at old old old older request losing in night of the times, inclusive when my log was egf, iclusive before the invection to the "eye candy" shaders, incluse, maybe, before the creation to the mainkin, and the internet porn.
I think what the better opcion is create the vertex weight firts, and before put all jcl&co in create a full range the new shadow system, incluide shadows in terrain, soft shadows, and white russian shadows.
Posted By: LoneWolf
Re: What's next? - 11/04/06 16:48
I voted for vertex weights as i think this a very important feature that is missing. At the moment we use vertex animation for our models but it creates very large models.
Anything that could improve performance would also be on my list. Gamestudios performance is good but if there was anything that could be done to improve it (i.e Triangle stripping of models / levels, Occlusion tests etc) then i would like to see some experimentation done there.
Posted By: Joozey
Re: What's next? - 11/05/06 01:15
2D improvements!
Would be neat if gamestudio would support flash animations, gif animations, jpg, more video modes, maybe some more picture file extensions, rotating/sewing/distorting panels, different button system. I think this has a higher priority than any advanced superduper soft shadowing system or the like.
Basics first.
Posted By: beegee
Re: What's next? - 11/05/06 10:42
same here, i voted for 2d improvements
Posted By: JibbSmart
Re: What's next? - 11/06/06 09:57
ditto
that's the way to go!
I think vertex weight and "more of everything" for terrains are the most important (to me at least).
Posted By: TheExpert
Re: What's next? - 11/07/06 08:06
Terrain tools and shadows indeed.
Caus It's standard.
Posted By: zazang
Re: What's next? - 11/08/06 07:00
Even if its not in the polls,I'd love to see an A6 version of normal mapping shader on models...other than that soft shadows and more skins and shadows on terrain will be my vote
Posted By: Samb
Re: What's next? - 11/08/06 07:40
what exactly is the soft shadow shader?
i mean, if it is something like the useless stencil shadows, we'd have now,with just soft edges,then it will be a useless BUT soft feature
2D improvement like collision and c_move function for panels? that would be great!
Posted By: mannybass
Re: What's next? - 11/10/06 10:37
I voted for 2D improvements.
I voted for Vertex Weights and I'm proud of it
Actually I'm not sure this is the most right way to get the most wanted feature to add in next release, but -jcl- and the team will get a hinch of what most people of this forum wants. I buyed A6 commercial before I knew there wasn't any multiple bones per vertex animations, So my eager to start got stalled... I wont't start to put in great effort before this feature and some other ( split animations, multiple models only singel animation file) ideas suggested elsewere on this forum are in the pipe for release.
But after all what matters is what type of games do most people use GameStudio for? There would be the best place to put the efforts, and that would keep the GameStudio community happy.
You can't have the cake and eat it too, there must be a path selected, please the already devoted or add features that would add more new users, simple state but a hard golden quest to choose from
I follow the evolving of this great engine, but for the moment It doesn't have the features in the animations/model structures I would like to have to make things easy for me.
Posted By: ventilator
Re: What's next? - 11/10/06 20:37
together with vertex weights conitec should also implement a blend shape feature for facial animation which can be mixed with bone animation.
I'll let whatever comes, come. We're going to learn how to manipulate these things for other purposes either way.
I would like to see gamestudio integrate into blender and MED get disposed of.
i voted for soft shadow editor, because i believe shadows should be customizable by default, and to me it seemed the best choice
Posted By: MaxF
Re: What's next? - 11/15/06 00:50
Quote:
I would like to see gamestudio integrate into blender and MED get disposed of.
I would not want GS integrated into blender, just to add FBX file import/export to WEB/MED which is in development.
Posted By: PHeMoX
Re: What's next? - 11/15/06 01:56
Quote:
I would like to see gamestudio integrate into blender and MED get disposed of.
Eeeck! Heck no, hehehehe, please don't even consider that. Blender is awful for those who can't stand reading manuals and a big bunch of tutorials. The forum would be flooded with blender questions, besides if you like you can make games with blender too.. 
Cheers
Posted By: Galen
Re: What's next? - 11/15/06 02:03
I'd *love* to see an import from Blender that would retain materials, not just textures (there's an exporter plug-in now for Blender to MDL7, which just retains textures). Sure, Blender has a bit of a learning curve...at first, but once you get the hang of it it's actually pretty fun to use. I won't lie and say that I'm an ace who knows everything about Blender, but what I've used so far--namely the modeling and texture applications, is fairly easy. If 3DGS could import with materials too, then I'd take the time to use them in Blender, instead of and/or in addition to textures.
My 2 cents.
I don't mean get rid of WED, just enable lite-c/c-script(there is a text editor in blender) and a call to the Achnex engine frome within blender, besides I only use MED(I'm not the only one)to manage skins and frames..everything else is done in either max,lightwave,trueSpace, and since the mdl exporter...blender. blender also has great uv mapping and texture painting. though ditching med altogether could handicap/alienate some users.
Posted By: ventilator
Re: What's next? - 11/15/06 11:01
once 3dgs flawlessly supports arbitrary level geometry i will do a blender (and probably also modo) scene exporter (with support for setting skills, scripts, lights, automatic recognition of instanced models, directly calling the map compiler and the engine,...).
i also will work on material support once i have more time again. i don't know yet how it will work. maybe there just will be some default shaders which will use the available maps and settings of blender's materials. or a shader network to hlsl translator but i guess this would be quite tricky.
Sounds very interesting ventilator. I really support this ideas and if I can help you with models and textures then just contact me.
Material support is interesting but currently not needed. Maybe it makes sense in the future with new procedural shaders. But even in a rendering application a procedural shader is much slower than a texture with uv-map.
So my advice to all gamestudio users is to use textures with uv-maps instead. It looks better is more flexible and will be the standard for a long time.
Materials are an easy way in an application but look not very realistic in a game.
Posted By: ventilator
Re: What's next? - 11/15/06 11:39
i don't mean procedural textures (though probably they will become more and more useful for realtime too).
currently my exporter only cares for color maps which got assigned in blender's uv-editor. it would be nice if it could also recognize bump maps, specular maps,... and other settings in blender's materials and then automatically choose and set up the right shader template for exporting.
ok, in case i need some test models i will contact you.
Wow. That sounds great. A more clean pipeline for normal or parallax mapping could make much fun.
Your welcome. Just pick a few textures up from our sci-fi pack. I can even wrap them around a few models but I am sure you can do it yourself in Modo or Blender. They already have normal, specular and height maps.
Frank
Posted By: MaxF
Re: What's next? - 11/15/06 20:38
Hi JCL and ventilator
web page This looks good...
Abstract
We present a new method for real-time rendering of shadows in dynamic scenes. Our approach builds on the shadow map algorithm by attaching geometric primitives that we call "smoothies" to the objects' silhouettes. The smoothies give rise to fake shadows that appear qualitatively like soft shadows, without the cost of densely sampling an area light source. The soft shadow edges hide objectionable aliasing artifacts that are noticeable with ordinary shadow maps. Our algorithm computes shadows efficiently in image space and maps well to programmable graphics hardware. We present results from several example scenes rendered in real-time.
and check this out
web page
Posted By: MaxF
Re: What's next? - 11/15/06 21:04
Plus it looks like Crytek maybe use this:
Quote:
The method presented in this paper is also related to the method described in US patent application no. US 2003/0112237A1, filed by Marco Corbetta on behalf of Crytek GmbH in December 2001. The two methods were developed independently
thread theft is a crime, but I would gladly pay for a blender plug/py. I also use modo(102), but would rather have the features you described for blender..just think it is better suited and it's free, so no one gets alienated.
Posted By: ello
Re: What's next? - 11/17/06 13:05
what makes me curious is only 176 user gave their vote here. every other one dont care at all?
Posted By: SMichael
Re: What's next? - 11/17/06 13:18
Before I talk about future features I like to get a bug fixed working version.
Posted By: Frits
Re: What's next? - 11/18/06 08:36
I agree with you totally.
Posted By: Spirit
Re: What's next? - 11/19/06 17:03
I vote for 2D improvements. Why not make a special level format for 2D games by eliminating the z axis but allow all other functions, like collision detection?
Posted By: Helghast
Re: What's next? - 11/19/06 21:16
vertex weighting, would make import of models to med, who are rigged and modelled in maya/max easier...
also, would make ragdolls better possible if the bones system would be updated with a code in which you can directly set a bone point to a specified point
(
kinda like vec_set, but maybe bone_set then
bone_set(entity, string_bone_name, vector_position)
that would be very usefull!
)
regards,
Posted By: splashmaker
Re: What's next? - 11/19/06 23:15
We need vertex weighting along with importing bones animation for our project, this is the only thing holding us back from adding in good animations.
Posted By: FinalGod
Re: What's next? - 11/20/06 06:19
I voted for vertex weighting, becource:
1) Higher res models animations looks silly without.
2) If FXB support is implemented it should be able to handle this-for users of a better modelling packed it is actually crap to import bone animated models.
(later)
Ello, well most people are looking for the imediate reward(e.g. somthing they don't have to code IMHO). I think Useability, stability, cust service, and new engine features with new commands are the best thing...,but that's only my opinion.
Posted By: MaxF
Re: What's next? - 11/22/06 15:19
Well if I think about it more, I would want
1) FBX import/export in WED and MED working - love to HAVE it
(i know it's coming, and thats great! Levels can be imported from 3dMax, Modo, XSI and lots more...), but can we have it so it imports everything, textures, mesh, static mesh, models, lights and so on.
Number 1 would make a GREAT xmas gift
2) FAST Soft shadows, people have been waiting for this from about day one.
3) Finish all the 'experimental features' on beta page, I like to see all these done
4) Finish 'New Rendering Kernel' and make it public (download page).
Posted By: slacker
Re: What's next? - 11/23/06 17:23
Ragdoll physics system, where control of a boned model could be turned over to the physics engine. This would make the pro version more appealing to me.
Why is everyone voting for soft shadows - isn't this currently available with sphere? Some explanation here would be good RE: multiple dynamic lights/shadows..
Posted By: PHeMoX
Re: What's next? - 11/24/06 13:50
Quote:
Ragdoll physics system, where control of a boned model could be turned over to the physics engine. This would make the pro version more appealing to me.
Using 1 single model, and physics hinges you can already make ragdoll physics with the pro version.
Cheers
Posted By: Dyc
Re: What's next? - 11/24/06 14:33
"Using 1 single model, and physics hinges you can already make ragdoll physics with the pro version"
Yeah, how?
Posted By: slacker
Re: What's next? - 12/03/06 10:43
It would help to have a boned biped with physics attached as a template - and script examples on how to apply force for headshots, etc. and how to go from standard animation to letting the physics engine take over.
is it time for a new thread or to start posting into feature suggestions, since a lot of us have wandered off the initial listing of features. just my opinion.
I think a few more textures wouldn't hurt. And shadows on the terrain are needed anyways.
Posted By: Stansmedia
Re: What's next? - 12/23/06 13:29
I cant believe the shader has the most votes. Vertex weights and decal management should be a huge priority. Im sick of my blood splatters hanging over the corners of walls, it looks incredibly stupid. Its almost 2007 and engines from the late 90's still have better decals. In all reality, there is no decals in A6. Just oriented sprites... Cmon already
.
Posted By: sPlKe
Re: What's next? - 01/09/07 17:47
not only that. the shadow rendering is more than poor, so is the light engine in general. this would be my number one priority...
Posted By: ello
Re: What's next? - 01/10/07 08:45
well, i dont believe that "more textures and shadows on terrains" is having so much votes. a shadow shader works for terrains too and more textures?? for what. thats bulls*t.
i hope not only one thing is developed. softshadows are a must to be built in the engine as well as vertexWeights and better decal system. i'd voted for those three if it was possible
Posted By: Scorpion
Re: What's next? - 01/10/07 12:19
i think decals should come next, they can be used for blood, selections, holes in walls, shadows, also it would possible to add dirt or blood direct to a model without to change the skin.
i cant understand why everybody is voting for more textures there are so many in the web, for free. and shadows on terrain
would be possible with decals.
PS:there is a soft-shadow-shader in the aum if i am not wrong...
Posted By: ello
Re: What's next? - 01/10/07 12:30
Quote:
i cant understand why everybody is voting for more textures there are so many in the web, for free. ...
nope its about using more than 8 texturestages i guess. but again, for what?
Posted By: Scorpion
Re: What's next? - 01/10/07 12:38
what do you mean with texture stages?
i guess i dont know everything
Posted By: SMichael
Re: Improved collision detection - 01/10/07 13:19
Thats my greatest wish,yes. I tested a lot in the past with different
entities and the result is that the collision detection is still jerky and poor.
If you use only simple objects like a box or a sphere then it might be enough
but with complex models the engine still has problems with stucking, jittering
entities.
Posted By: PHeMoX
Re: Improved collision detection - 01/10/07 14:31
Do you mean the physics engine or normal collisions? Normal collisions work fine here.
Cheers
Posted By: SMichael
Re: Improved collision detection - 01/10/07 14:40
I mean the normal collision. It is difficult to write the problems
the engine has, but I can show you a lot if that was possible.
Posted By: PHeMoX
Re: Improved collision detection - 01/10/07 14:53
Collision problems might be very much related to scripts off course, but at the moment I'm not experiencing much problems with it, except problems caused by my own buggy programming style. The physics engine on the other hand, is somewhat of a different story.
Still, it's being worked on, no doubt.
Cheers
Posted By: laethyn
Re: Improved collision detection - 01/20/07 00:32
I've not updated in some time, so I don't know for certain if this has been implemented or not, but some sort of "target" function would be fantastic.
Something like
view 1stPerson
{
....
target = player;
}
or 1stPerson.target = player;
You get the idea.
This would be absolutely fantastic not only for views, but to point entities at other entities (ex. during combat, you set the player target to the monster, and the player constantly faces the target.
I realize this can be done via vec_to_angle et al. but having a .target handle the angles would be a lifesaver for a lot of people.
Posted By: slacker
Re:Decals - 01/30/07 16:40
Could someone better explain decals - could this be used to simulate cast lights on the ground - could they be animated to sway, or?
Can't you now just place an alpha sprite, or would a decal affect all geo in the direction it is cast - like a projection shader?
Posted By: Scorpion
Re:Decals - 01/31/07 13:37
a sprite is always flat, a decal can be also around corners or fit to terrain-surfaces
Posted By: TheExpert
Re:Decals - 01/31/07 18:02
decals ask , have been asked for more than one or two years !
nothing
, even some games on N64 have decals mapping around corners
perhaps using texture projection instead with shaders is more efficient and more fast ?
But yes this feature is really needed , for things like shadows of palm trees
or fake shadows to optimise levels
Posted By: slacker
Re:Decals - 01/31/07 18:12
I have been out of GS for a while - what is it the bright or flare flag that does an additive blend mode?
if so it seems like you could use this for lighting effects. If you could animate the coords, you could do swaying lights, rotating alarms - that kind of thing?
well, i would vote for dectals as well!!
Posted By: MaxF
Re:Decals - 02/03/07 07:06
Hi
What about "Cascaded Shadow Maps" for soft shadows, been reading on the net that its "offer the best performance/quality trade off"
Posted By: DARKLORD
Re:Decals - 02/04/07 12:47
I also vote for decals
Posted By: Yu_Une
Re: What's next? - 02/16/07 11:30
Something which can handle an imported pic from world machine.
Posted By: Trunks
Re: What's next? - 02/20/07 20:22
My vote is for decals
Posted By: HeelX
Re: What's next? - 02/20/07 20:27
Quote:
Something which can handle an imported pic from world machine.
What is this..
Quote:
once 3dgs flawlessly supports arbitrary level geometry
Hey! Cool ventilator, I think I understand what you mean, but to be sure-- can you esplain what is arbitrary level geometry? Like levels with scene managementī? (er ABT-Octrees, eg something better than BSP?)
A feature I would love to see is lightmapping on levels made entirely of MDL files.. or some way to turn those levels into the ABT format with all the different texture mappings intact... not having lightmaps on my mdl levels are kind of a bummer, and instead I either have to bake them in or use dynamic lights.
Posted By: ulf
Re: What's next? - 02/21/07 17:00
i cant understand why so many people want to see a softshadow next... template games will not look better with softshader or bloom. i would rather see any real improvement in the engine than anything that can be scripted with c or put in as shader already.
Posted By: TWO
Re: What's next? - 02/21/07 17:49
Indeed, a better multiplayer part wouldn't hurt
p.s I just saw my SED publish request was implemented, great
Posted By: Orange Brat
Re: What's next? - 02/21/07 20:13
Quote:
i cant understand why so many people want to see a softshadow next... template games will not look better with softshader or bloom. i would rather see any real improvement in the engine than anything that can be scripted with c or put in as shader already.
Soft shadows and bloom are a couple of options which probably need to be implemented as an in-engine feature. They do add a lot of extra oomph to even mediocre looking scenes, and they're standard in most titles these days.
I voted for 2D improvement, however I'd change my vote to vertex weighting, now. It's fruitless, though given this is in beta test, now. In fact, it looks like most of the features on the list are either in development, in beta, or finally made the forecast page.
Posted By: TWO
Re: What's next? - 02/21/07 20:34
You can have bloom easily since one and a half year, throught sylex3. Itīs very easy to implement the build in things. And with a 'litte' dx programming, you can have all you want, just look at sphere.
Ok, from my point of view everything can be implemented, I know, but that would need time, and most of us have a litte time for there hobby.
Ed: I just reread your post, orange. Yes, you are right, a ready-to-go solution wouldn't hurt, even more when this solution is aviable through open source.
Posted By: EX Citer
Re: What's next? - 02/22/07 11:22
Damn, found this so late... In the last days I am finding a lot of stuff very late :/ Strange.
Anyway, I donīt understand that softshadow thing too. Itīs nice to have soft shadows, but I think they can only be used rarely because they wonīt be very fast and need a good computer to run on.
And too bad I could only vote for one. Otherwise I would have voted for decals AND vertex weights. And I think many would have done that.
Because now it's not like, if all the people who wanted decals or vertex weights (for example clones of me
) would have voted for the one with higher votes. I mean if I have only one vote and I see that vertex weights is almost leading, and decals have no chance, then I would give my vote vertex weights instead of decals to make at least my other favourite leading.
I mean then wouldnīt it be 50 votes decals, and 50 votes vertex weights and 51 votes soft shadows. It would be 100 votes vertex weigths and 0 votes decals and 51 votes soft shadows. Ok, thatīs the extremest possible case. And this example shows that this way of voting would make decals look unwanted, but that isnīt intended by the voters. THatīs why I like polls where I can vote for everything I like/dislike.
Last but not least a nice idea, to make a poll. I like it.
Posted By: slacker
Re: What's next? - 03/02/07 02:21
Quote:
Anyway, I donīt understand that softshadow thing too. Itīs nice to have soft shadows, but I think they can only be used rarely because they wonīt be very fast and need a good computer to run on.
Isn't there a 3rd party soft shadow solution now - or...? (sphere)
Posted By: EX Citer
Re: What's next? - 03/02/07 18:58
Yep, thatīs also a point why I donīt understand the urgent need of softshadow. I realy donīt understand it.
I donīt totally know how soft shadows could win.
Well, softshadows are visually nice, but as far as I can they run pretty slow.
I mean the shadows right now are slow and they have no smoothing. When they now get smoothed they will logically run even slower. I mean most users playing my games are complaining that the shadows are slow so I guess most people turn them off.
And now for some reason I totally can not understand why users voted for softshadows. Stencil shadows are a nice gimmick, but nothing I would build on... Can someone of the people who voted for softshadows why she/he has voted for softshadows, with the knowlegde in the background that the stencil shadows are slow and difficult to use because of the camera issue with the shadow volume.
EDIT: Same for the terrain thing. ISnīt there already a multi terrain shader thing? And there is a simple and well working application which renders static shadows on a terrain. Of course colored shadows would be nice. But itīs very useful as it is now. I think most people just donīt know that tool.
I think that lighting and shadows make the most important point for creating visual worlds. Without light and shadow it looks flat and can only create cartoon worlds.
Softshadows can be rendered on the graphics card and therefore don't have to be slow with the right technique. Besides that they probably don't have the problems with clipping errors when within the camera volume.
With the fact in mind that current stencil shadows are quite useless with those problems I would say: softshadows (or any other new shadow technique) is our only hope to get realistic worlds.
We need shadows reacting to dynamic lights (maybe even optional more than one light) in indoors or to sunlight (outdoors). They should be casted correct. The current perspective is wrong (probably because it comes from the very distant sun).
So I really think this is a very very important point when you want to create 3d-worlds. Otherwise you can paint a 2d-background and make an adventure game.
Posted By: Ghost
Re: What's next? - 03/03/07 12:59
We need good soft-shadows and lighting(and wieghted vertex animation) to be able do more realistic looking worlds and characters, so the engine and our games look up-to-date visually.
Like it or not our customers' visual expectations are driven forward by the big commercial games. Farcry was 3 years ago now and pretty much all other indie engines offer soft-shadows already or are implementing them soon, 3DGS needs to keep up with the market.
Soft-shadows are a different technique from stencils and do not have the camera volume problems. Links to several papers describing methods to do soft-shadows with good performance were already posted on here, including a patented technique that can be licensed very cheaply, so it is wrong to assume soft-shadows must be slow or poor.
Sphere 2.0 has very nice soft shadows, but doesn't suit all projects as you lose out on the perfomance gains from A6/A7's native culling methods, the option to have pre-calculated light mapping as well in your scenes and you also lose the other light management features added for A7.
Having pre-calculated light mapping will remain important for visual quality and performance, especially on lower end hardware but even at the high-end too, take a look at Gears of War.
Posted By: Damocles
Re: What's next? - 03/03/07 13:07
The shadow Rendering on terrain should be part of WED, as the Light/shadow Rendering on blocks
is at the moment. This is totally the same concept: Create static lightning onto
the unmovable Structures of your level. The terrain could get a flag in WED: "light rendering"
and WED alters the skin of the terrain and applies the shadowmap onto it, reqacting on
the sun and Lightsources of the level.
This is a "soft shadow" then, but baked onto the terrain.
Posted By: EX Citer
Re: What's next? - 03/03/07 17:24
Allright. THanks for clearing that up.
I donīt want to make anything bad. I also think that soft shadows for dynamic model rendering are a good thing.
But here is why I am so afraid of dynamic shadows. This is a demo of crytek, I think one of the best real time render engines, and I also think it has the best dynamic shadows. The point is they look awful compared to static shadows because they are flat. There are no light in the shadow. I mean a real shadow is created by much light which gets reflected by all "walls". It doesnīt matter if itīs sunlight or if itīs a lamp. A shadow has almost never the same color in itself. Well, maybe even never. I mean the reflected light will reach the shadow somewhere more somewhere less.
Thatīs why I think soft dynamic shadows are no advantage for static models. Anyway, here is the link to the movie which made me think dynamic shadows are bad for enviroment:
http://www.gametrailers.com/umwatcher.php?id=44386http://www.gametrailers.com/umwatcher.php?id=44370I have seen good dynamic shadows in Severance for example. But as far as I remember they used dynamic shadows only for characters and items. But for enviroment static shadows look better as far as I can see. By static shadows I mean also textures drawn by hand inclusive shading.
Posted By: Ghost
Re: What's next? - 03/05/07 22:15
@Exciter
The point is we are asking for A7 to allow pre-calculated static soft shadows AND dynamic soft-shadows too, then you as the developer use those techniques seperately or combined as best suits your game or particular scene. FarCry and Gears of war are good examples of this, they do not just use one technique.
As far as the Crytek engine example you are talking about, I think it's pretty clear that this engine has been developed with the intention to alter the environment and lighting dynamically in mind. I.e. knock down trees, destroy buildings, have burst of light from explosions, swinging lamps etc. etc. In both cases the lighting and shadows need to change dynamically, so obviously static shadows would look wrong. Having said that, the engine definitely allows static lighting as well, so this is just a creative choice in the example you picked out. i.e. Their creative decision in this case is having dynamic environment damage and day to night cycles etc. compensates adequately for the dynamic shadows required not looking quite as good as pre-calculated ones.
Posted By: slacker
Re: What's next? - 03/07/07 16:38
Crytek is crytek, but it strikes me that if there is a capable soft shadow solution as 3rd party software, wouldn't decals or vertex weights make more sense because these kinds of things can't be addresssed by the community?
Posted By: broozar
Re: What's next? - 03/07/07 21:05
as 90% of the users are using 3rd party modelling and animation tools, exporting their anims as vertex animations, there's rarely someone who would profit from it. a general visual enhancement would benefit us all.
Posted By: Ghost
Re: What's next? - 03/07/07 22:23
@Slacker
I see your point, but in the end as we really need ALL of these features, maybe a better way to decide the best priority is to think about how each of these affects development.
I would say we need the lighting/shadows first, as you generally need to develop the look of your game through several iterations over time. Also if you want to make a demo to raise funding or attract team members, the look of your graphics is going to be very important, so again lighting/shadows is key to have early.
It's not fair to say there is an off-the-shelf method for soft-shadows if you are referring to Sphere as I pointed out previously it's unfortunately not suitable in a lot situations.
Animation in contrast can be fully created in Max or Blender etc. and just exported as vertex anims or bones without weighting, until weighted anims becomes supported. Same for decals, they are perfectly workable as is until a better system arrives. In short it's going to be much easier to have your art style/lighting finalised from early on and then updating animation and decals later on. Trying to do the reverse will be problematic.
Posted By: William
Re: What's next? - 03/09/07 08:44
"A Map renderer for using a view as a map or radar screen, Scaling Truetype and Bitmap Fonts. Font installer for Truetype fonts. Size prediction for Truetype TEXTs. Proportional Bitmap fonts. Speed-Up of Truetype text drawing by preloading font textures. "
These are actually quite important for those looking for a good menu/hud system in their games, trust me when I say it will make things much easier, it's not flashy but it's backend stuff.
Vertex weighting is important if you plan to have the characters programmed, I've done a fair amount of bone programming for weapons, which are not organic so they don't need the weighting. However, characters, whether you want their heads to turn to enemies, arms pointing at enemies, feet bending on the ground, ect. vertex weighting is important, and in the end, just makes things easier too.
Shadows and lighting are also very important, stencil shadows don't work in my game as none of my mesh's are closed(I like to optimize models/worlds). Old school 2d shadows are my only option right now, as Sphere, if I remember right, requires your entire worlds to be normal mapped and set within the engine, also very slow(for my project type) in a large outdoor scene. So far the plans are to use FBX import and lightmap levels, and use 2d shadows for the main dynamic objects. Optional(so user can switch on/off depending on computer) softshadows for these dynamic objects would make me very happy.
Posted By: EX Citer
Re: What's next? - 03/12/07 21:08
You canīt do with vertex animations what you can do with bones. Almost every game uses nowadays bones animation because the games are more and more interactive. That are improvements.
[red/]You can import and export bones models with ms3d and fragmotion, so there is no reason why anyone shouldnīt use bones animation. [red]
"It's not fair..." You are talking about workarounds which reduce the quality drastically/donīt work, while there is a tool which produces soft shadows as well as you can draw shadows on the texture. With the dds format the texture is tiny. Anyway, thanks for trying, but these arguments didnīt satisfied me very much
Posted By: slacker
Re: What's next? - 03/14/07 23:40
I animate in max, and export mesh animation because I want good joint deformation - this works good, but file sizes get huge if you have many frames, and you don't have the flexibility to program in movements. This as well as decals have been standard in games for some time, no?
I place sprites for blood spray, and ornimental graphics, so a decal system would be great.
As far as soft shadows - I am not sure what Sphere doesn't do well, I have adapted my own shaders from forum posts - nor am I 100% on what a conitec solution to soft shadows would be.
I feel Conitec should address some of the issues that have become standard in other games - they are standard in other games for a reason. Gloss is great, but gloss is gloss.
And there is much eye candy to be had with shaders right now.
One last thing on this - who wants more textures? There is tons of great textures to be had on the net:
image afteris one of thousand free texture sites. Besides art development is a huge part of game design - stock textures will get you... well a stock game.
Posted By: AlbertoT
Re: What's next? - 04/06/07 21:39
Vertex and weight
decals
The following is an example of why vertex weights are important to have good looking animations
the third deformation is what we have now the first is the ideal
WE NEED THIS!
Posted By: Joey
Re: What's next? - 04/22/07 14:47
that animation looks ugly in either case. i don't understand what that has to do with vertex weighting? you can achieve that without vw, can't you?
No Joey. The middle cut of the first animation is influenced 50% by bone1 and 50% from bone2. You cannot do that in GS. But you can do version 2 and 3 of that image.
Posted By: broozar
Re: What's next? - 04/22/07 21:15
as an LW user, it shouldn't bother you, does it? what does it bother a blender, c4d, xsi, max, ... user? what's a good animation worth if the model can't be lit properly and can't throw usable shadows?
Broozar: I agree with you. Lighting comes first. Light and shadows make a virtual world go round.
Posted By: FinalGod
Re: What's next? - 04/23/07 23:23
Quote:
as an LW user, it shouldn't bother you, does it? what does it bother a blender, c4d, xsi, max, ... user? what's a good animation worth if the model can't be lit properly and can't throw usable shadows?
What does matter that shadows are a more simple part than rework 50-200 Character-Models of a game for us developers? Its a s_h_itt task to rework that amound of models if Gamestudio once get Vertex-Weighting, and it is since longer time a standart in most Games than things like softshadows, which are anyway for many not useable becource the performancebrake. Gamestudio should not get features that come in the last 3 years, it should get that features that are long time here but conitec was not able to integrate it right into Gamestudio. They did bring simple Boneanimation where vertex-Weigted Bones are on most other engines standart-if bringing a "feature" than should it be up-to-date and not like Conitec it does.
So why wishing for new things and not rework the old "errors"?
Its offtopic but thats the same with MED-drop it Conitec, better make for Gamestudio optimized tools to export from REAL modeling packages. MED is a joke, so why waste development time with it-we need this time more in the Engine itself, becource Gamestudio lacks a lot of features. C-Lite-for what was that needed? For sure it was not that strongly needed like other things. Decals for bulletholes or blood on an Characters? Other Engines can do it, when will Gamestudio be able to handle this? Shaders on levelblocks? -Again they did implement something like usage of shaders, but somehow not complete...
And monthly updates would also be a nice "feature", its [censored] to wait what in the next update was how (often bad) implementated-anyway its usualy not the things we could really need.
... but who needs today softshadows, working shaders, vertex-wheigts, decals, better lightining if you can run gamestudio from visual studio with a new scripting language. Maybe A7, for A6 i was waiting for years to get such things and got a bone system in which Charactermodels with above 5000Tris looks more ugly when they move, if you can have VisualStudio compatibility - for what ever...
A dummy question. What is vertex weighting?
Posted By: FinalGod
Re: What's next? - 04/24/07 06:51
Posted By: broozar
Re: What's next? - 04/24/07 11:00
may i ask you why you need to rework your models? which tools and which workflow do you use?
the MAX2mdl plugin converts the bones animations to vertex ones automatically, the blender exporter offers you both possibilities(bones and vertices), about lw i don't know, but c4d does the same with it Xport tool, and even as a last resort, you can export your animations as a series of 3ds models and merge em together im MED, works as well. there is no need to rework anything, it's just a matter of converting the (weighed) bones animations from you favourite tool into simple vertex animations without losing the look.
there is no way to do anything similar with shadows.
besides, there is no need to use such kind words as you do to describe your problem.
Quote:
..about lw i don't know
With the Greebriar's plug-in for Lightwave you also have the choice to export bones- and vertex-animations.
Continuous LOD please for model entities!
Posted By: FinalGod
Re: What's next? - 04/24/07 15:14
Dunno how you work, or how you models look or what your are aiming for a polygon count on your models, but making the movement parts of a model with 1 vertex weight ist completly different than that with multiple weights.
Maybe i have to repost the picture from alienheretic so that you can see the difference, or show me one of your models how they look with that kind of animation, but if you go for 5000 or more tris i would REALLY be interested how you would animate this with only 1 weight. Hard corners on every arm, knee, or every other modelpart or do you use for every vertex 1 bone? (joking)
And of cource i dont use the max2mdl plugin becource its outdated. And iam talking here about BONEANIMATION and not VERTEXANIMATION. That it is working very well with pure vertexanimation that know we all (i hope), but tell me 1 (one) game from the last years which did not use boneanimations for character models. Hope i have not to go and tell the advertages of bone vs vertex - dunno how good your knowlegde in this area is. Mine is very well and everyone knows that on character starting from 2500 tris upwards, weighting is more and more needed. Even on fingers it is today usual to weight the vertexes to get smooter movement.
Posted By: Doug
Re: What's next? - 04/24/07 19:18
Quote:
...MED-drop it Conitec...why waste development time with it-we need this time more in the Engine itself, becource Gamestudio lacks a lot of features.
The problem with this logic is that the person that works on MED isn't the same person who works on the Engine. So dropping MED will not make other features appear any faster, it will just damage the "Studio" part of 3DGameStudio.
The better solution is to hire more people to work on the engine.
Posted By: FBL
Re: What's next? - 04/24/07 19:33
When do the reinforcements arrive?
the problem is not getting smooth deformations with vertex animation but rather getting smooth deformations and using bones at the same time!
Another advantage of bones vs. vertex animation is the file size.
Posted By: FinalGod
Re: What's next? - 04/25/07 05:11
Quote:
The problem with this logic is that the person that works on MED isn't the same person who works on the Engine. So dropping MED will not make other features appear any faster, it will just damage the "Studio" part of 3DGameStudio.
The better solution is to hire more people to work on the engine.
Hi Doug,
but that is that what iam talking about. Give all possible resources to missing engine features.
How old is this topic? -its VERY ok if you ask the users what to integrate next, but in this time this feature (Softshadows votet most) must now already be here, so jump on to the next part of requested features. According to the Betapage there is done a lot-but not that what is really needed, for modelling, texturing, animating,...,... is MED for sure not best/used by the most. For the most users it just a fileformat converter or needed to integrate more skins.
Posted By: FinalGod
Re: What's next? - 04/25/07 05:20
Quote:
the problem is not getting smooth deformations with vertex animation but rather getting smooth deformations and using bones at the same time!
True, with pure vertexanimation you get on "highres" Characters even smoother animations (according to the limit of blend vertexes the engine could handle, and as i did read conitec whant to inegrate only 2 vertex blending), than with bonebased animation, but it lacks on "interactivity"
Quote:
Another advantage of bones vs. vertex animation is the file size.
You, but it would be even greater if it would be possible to seperate the animated bones from the file with the rigged one. As far as i know Gamestudio actually need it that its in one file, well it would be possible to animate the bones by hand but... lol
Posted By: broozar
Re: What's next? - 04/25/07 08:00
@FinalGod: i do not need interactivity, so i animate my models with BONES in any of my modelling applications, let it CONVERT to vert anims and then use it. so my WEIGHTING is done in my modelling software (which is obviously not MED). and yes, i am working on models that have a higher polycount than 2500 tris. if i had to do this in MED, i'd rather quit it.
Quote:
For the most users it just a fileformat converter
exactly. and i do not expect more, as long as i can do anything else by script or in my preferred modelling package.
btw, there is no reason to argue as the decision has already been made.
Posted By: FinalGod
Re: What's next? - 04/25/07 10:25
Quote:
broozar
i do not need interactivity, CONVERT to vert anims
With other words you dont have that a need for a good boneanimation system in Gamestudio, but i and other seams to need it. We was happy to get it and after we see it was only primitive 1 weighting system we say give us a real thing-maybe the same you would do if softshadows comes and it would work only on static meshes for example=not fully integrated
Quote:
broozar
btw, there is no reason to argue as the decision has already been made.
Why should there be no reson? Its always worth to talk about something you like or dislike, and talking is important that conitec know whats up...
@FinalGod: First, get some experience with 3DGS and show us what you can do before comming here and b*tching about sh*t. conitec works very hard to ensure that everything is working fine, providing support for the thousands of users, and making large progress on the new A7 engine, which is a major upgrade from A6.
Yes, there are some features that we're all hoping for (soft-shadows, better static shadows, bone weighting, etc), and conitec works to add these features when they can. Nothing is promised, the package is already done, their just adding more features for us.
Your infactual posts and numerous spelling errors make you seem very uneducated...
Please feel free to STFU...
Posted By: FinalGod
Re: What's next? - 04/25/07 17:20
Quote:
First, get some experience with 3DGS
still got my first white sybex 3.5 zoll floppy discs here with them i started for many years with gamestudio to now to my Com 6.506. I would more say that iam b*tching (like your seams language is to be used) becource my knowlegde of 3 versions
Quote:
conitec works very hard
no one says anything else, or did one???
Quote:
large progress on the new A7 engine...to ensure that everything is working fine...
I hope A7 becomes useable then shader on levergeometry is still not working in A6 what makes the usage of the bsp based A6 engine with the shader feature real questionable...
Quote:
providing support for the thousands of users
how about givin support in shadow system, or are the dynamic shadows acually not more drawn from one romm to the other threw walls?
Quote:
which is a major upgrade from A6.
we will see, actually the main feature is new renderer, if they fix the other bugs was never answered
Quote:
and conitec works to add these features when they can
I hope so, the question is WHEN do they bring this. Is it coming for A6 or in some month/years in A7...
Quote:
numerous spelling errors make you seem very uneducated...
at least i got some education, maybe you get it once...
Iam from Germany and not from an english based country. Would be the same if i go out and try to correct posts from others who try to write german. And it is childish to come with that "arguments" or with a complete post in a topic with a post which have nothing to do with the theme of it.
Quote:
Please feel free to STFU...
I dont let me from someone say that i have to shut up, if iam talking about product related thing-your spelling sounds a little that you have a hang to rassism...
Next time post somthing related to a topic-here is discussion about whats next and not how to go out and spam. education-you used that word?
I am German as well, but raised in America. I even have the blue eyes/blonde hair ideal going on. [National Socialism, aheh...]. Anyways, I am simply posting this becuase your posts seem very uneducated and are insulting many users. This is NOT appreciated!
Quote:
I hope A7 becomes useable then shader on levergeometry is still not working in A6
This is completely incorrect. Shaders do work on level geometry. However, the way that level blocks are managed [to increase performance in standard rendering mode] makes this process slow [texture lookup].
Quote:
majot upgrade from A6. we will see, actually the main feature is new renderer, if they fix the other bugs was never answered.
Yeah, an entirely new renderer is knid of a BIG feature, don't you think?
Quote:
I am from Germany and not from an english based country
You are right here, I am sorry. I didn't realize this until AFTER I posted. Once again, my apologies. Ich spreche sehr kleinen Deutsch. Bitte...
Posted By: Doug
Re: What's next? - 04/26/07 00:41
And now we've devolved into personal attacks.
Thread closed.
Feel free to open a new thread for civil conversation about new features.