Page 50 of 53

Re: Models by MaNGusT (AR)

Posted: 09 Mar 2019, 23:32
by Jorzi
MaNGusT wrote:
09 Mar 2019, 23:13
btw, why do we use inverted values for object-space normal maps? Spent a lot of time find out how to make it properly. In 3ds max I can't choose values for Z channel, only for x and y, so I export red +x and green +y and then invert colors in a 2d editor to get red -x green -y and blue -z. Different standards?
This is simply because blender 2.49, which I originally used, had its own slightly nonstandard implementation of object space normals. After that I ended up flipping the channels of existing maps, rather than just changing the code and converting the existing stuff. Sorry about that :P

Re: Models by MaNGusT (AR)

Posted: 10 Mar 2019, 12:45
by MaNGusT
when creating ao maps, I always think about ssao. Are there any problems to integrate it nowadays? The dilemma is should I bake basic shadows from propulsion, turret and terrain to universal models like bodies and weapons because in some combinations they are not needed and look wrong (vtol). SSAO, at this point, could solve this artist's dilemma.

update:
rebaked the ao texture, added some dust
Spoiler:

Re: Models by MaNGusT (AR)

Posted: 11 Mar 2019, 13:12
by MaNGusT
sсorpion body
Spoiler:

Re: Models by MaNGusT (AR)

Posted: 11 Mar 2019, 16:36
by Jorzi
Good stuff man 8)
I think the orange part looks a bit plain though. The roughness texture you put on the mantis was good. I think it would be a nice contrast to have the yellow panels have some surface roughness (perhaps the hardening process erodes the surface a bit?), while the details could be more shiny and metallic.

Re: Models by MaNGusT (AR)

Posted: 11 Mar 2019, 17:34
by WZ2100ModsFAn
Good models wish to see the v-tol remastered too.
Maybe when you guys have more time.
Mantis + V-Tol Remastered will be awesome

Re: Models by MaNGusT (AR)

Posted: 24 Mar 2019, 19:00
by MaNGusT
Basic colors and details of Bug body. WIP
Spoiler:

Re: Models by MaNGusT (AR)

Posted: 29 Mar 2019, 12:34
by MaNGusT
Broken shadows on rotated buildings in 3.3.0 :D Did It work this way before?
Image

Re: Models by MaNGusT (AR)

Posted: 30 Mar 2019, 14:07
by pastdue
MaNGusT wrote:
29 Mar 2019, 12:34
Broken shadows on rotated buildings in 3.3.0 :D Did It work this way before?
Image
Good catch - looks like this has been broken for a while. Fix incoming.

Re: Models by MaNGusT (AR)

Posted: 30 Mar 2019, 17:48
by Berserk Cyborg
I found another shadow related bug while I was testing your patch (however, it isn't caused by your patch). Stretched structures have broken shadows that look similar to what MaNGusT posted earlier. Other than that, everything else looked good with my tests.

At a glance, static shadows are used for some features like buildings, skyscrapers, and oil drums which otherwise suffer the same issue.

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 14:19
by MaNGusT
Bug body :ninja:
Spoiler:

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 15:55
by Jorzi
Excellent paintjob :) Just the right amount of roughness imo.

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 19:20
by MaNGusT
According to pie.md optional materials directive should be placed after connectors but it's wrong, it works as before and should be placed for every level after LEVEL # directive.
Also, looks like shininess parameter is not working anymore. I don't see difference between value 10 and 128, while it has to be. Shininess controls the spot area of specular effect. The higher the value is, the thinner area will be used.
I need it coz optimal value for my specular maps is ~30. Default value is 10, that makes shine's area very big.
Examples:
Spoiler:
in game it looks even worse( I think partially problem is that we use Phong's light formula, while Blinn-Phong's formula gives much smoother result in cost of little lost of performance.

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 20:00
by pastdue
MaNGusT wrote:
31 Mar 2019, 19:20
According to pie.md optional materials directive should be placed after connectors but it's wrong, it works as before and should be placed for every level after LEVEL # directive.
Also, looks like shininess parameter is not working anymore. I don't see difference between value 10 and 128, while it has to be. Shininess controls the spot area of specular effect. The higher the value is, the thinner area will be used.
I need it coz optimal value for my specular maps is ~30. Default value is 10, that makes shine's area very big.
It looks like support for the MATERIALS directive was removed many years ago in this commit:
https://github.com/Warzone2100/warzone2 ... ad5c48ccd8

That also seemingly removed support for providing a shininess parameter to the shader.

EDIT: What would be the best way to re-implement support for shininess in the pie files for your purposes?

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 20:49
by MaNGusT
I really don't know how it could be but it should be editable and optional at least per model or per mesh in ideal.
also, it would be good to return back the emitter(afair it's called so; update: LIGHT_EMISSIVE) parameter of materials to work per mesh(level), coz currently all meshes of a model are rendered in game. it's very useful to be able to make self-illuminated parts of models. ofc fake illumination but... it's better than nothing. :)

Re: Models by MaNGusT (AR)

Posted: 31 Mar 2019, 21:02
by Forgon
pastdue wrote:
31 Mar 2019, 20:00
MaNGusT wrote:
31 Mar 2019, 19:20
According to pie.md optional materials directive should be placed after connectors but it's wrong, it works as before and should be placed for every level after LEVEL # directive.
Also, looks like shininess parameter is not working anymore. I don't see difference between value 10 and 128, while it has to be. Shininess controls the spot area of specular effect. The higher the value is, the thinner area will be used.
I need it coz optimal value for my specular maps is ~30. Default value is 10, that makes shine's area very big.
It looks like support for the MATERIALS directive was removed many years ago in this commit:
https://github.com/Warzone2100/warzone2 ... ad5c48ccd8

That also seemingly removed support for providing a shininess parameter to the shader.

EDIT: What would be the best way to re-implement support for shininess in the pie files for your purposes?
@MaNGusT:
Could you edit the PIE.md file so that its information is correct?