|
|
Re: What are you working on?
[Re: HeelX]
#461858
08/25/16 22:41
08/25/16 22:41
|
Joined: Aug 2009
Posts: 1,438 Spain
painkiller
Serious User
|
Serious User
Joined: Aug 2009
Posts: 1,438
Spain
|
More experimenting with Scifi environment and Shade-C EVO   (there are some artifacts on the floor and ground, not sure why)
Last edited by painkiller; 08/25/16 22:48.
3D Gamestudio A8 Pro AMD FX 8350 4.00 Ghz 16GB RAM Gigabyte GeForce GTX 960 4GB
|
|
|
Re: What are you working on?
[Re: rayp]
#461871
08/26/16 15:01
08/26/16 15:01
|
Joined: Aug 2009
Posts: 1,438 Spain
painkiller
Serious User
|
Serious User
Joined: Aug 2009
Posts: 1,438
Spain
|
Wow. Great looking screens you've there painkiller! Nice! thanks! just some models and blocks with textures put pogether, not a real game level for now (there are some artifacts on the floor and ground, not sure why) , maybe shadow artifacts (see left pic here to compare https://i-msdn.sec.s-msft.com/dynimg/IC340448.jpg) or some other shade-c (lod) settings not set correctly? (just some wildquesses) looks like it's related to shade-c's normal mapping, maybe because of having small details that cause that effect when seen from a certain angle
3D Gamestudio A8 Pro AMD FX 8350 4.00 Ghz 16GB RAM Gigabyte GeForce GTX 960 4GB
|
|
|
Re: What are you working on?
[Re: Kartoffel]
#461884
08/27/16 13:28
08/27/16 13:28
|
Malice
Unregistered
|
Malice
Unregistered
|
^ From what I member looking through it's the same as No-back-Culling or the two_sided standard 3dgs material.. **However, if you are picking up fps and can see on screen transparency disabled than Kartoffel is more likely correct. It seems to set bool PASS_SOLID which is used in sc_lights.c set(shadowView.material,PASS_SOLID); set(shadowMtl,PASS_SOLID); set(screen.views.sunShadowDepth[i].material,PASS_SOLID); which seems to be a standard view material flag for 3dgs here http://www.conitec.net/beta/material-flags.htmOpposite of TRANSLUCENT: enforces rendering on the solid pass, i.e. together with non-transparent objects. For entities with alpha channel textures that are used for other purposes, such as a height map in a shader. Entities in the solid pass are not sorted, and thus render faster than in the transparent pass. This flag can also be set by defining a bool variable within the .fx file (bool PASS_SOLID;) A And PotatoMan saves the day Again... He's correct!! Selective uses in object shaders may be to force that object alone into the solid pass. I could also be a switch for alphatest or alphablend(???). Sorry I can't recall now, I may have know more in post past when we talked transparency before, if you can find those..
Last edited by Malice; 08/27/16 13:51.
|
|
|
Re: What are you working on?
[Re: ]
#461885
08/27/16 14:14
08/27/16 14:14
|
Joined: Dec 2011
Posts: 1,823 Netherlands
Reconnoiter
Serious User
|
Serious User
Joined: Dec 2011
Posts: 1,823
Netherlands
|
It is weird that the quote says that with solid pass it should be faster cause of entities are not sorted (which makes sense), but when I comment it out the game runs actually faster. What kind of weird Acknex wizardry is this?!
Last edited by Reconnoiter; 08/27/16 14:17.
|
|
|
Re: What are you working on?
[Re: Reconnoiter]
#461886
08/27/16 14:59
08/27/16 14:59
|
Malice
Unregistered
|
Malice
Unregistered
|
If I had to take a wild guess it would be... -- The system and rules that 3dgs was built under may no longer hold true with newer GPU design and on-board programming - This may mean that you GPU gen or even more current ones now use better systems to run the transparent pass, as this would be a logic place for each gen to improve optimization.
Other notes begin that using transparency and solid pass are pointless opposites and my lead to more work as the system or shade-c attempts to set-up alphatest/alphablend and then drop down to a solid pass. It's like saying "You must be transparent" but then saying "you must be rendered none-transparent. "
But hey guess is fun right... I wonder what it feels like to actually know... hehe
Last edited by Malice; 08/27/16 14:59.
|
|
|
|