[Woolie Wool]

http://en.wikipedia.org/wiki/Blinn%E2%80%93Phong_shading_model

Would it be feasible to use a more sophisticated shading model in FreeSpace? Gouraud shading was perfectly adequate in FS retail, but I find that it often leads to odd results with specular lighting. Would this require an entirely different lighting  system or could we drop the Blinn-Phong model in place of Gouraud?

[taylor]

From the second sentence of the linked article:

Blinn-Phong is the default shading model used in OpenGL and Direct3D's fixed-function pipeline.

[Woolie Wool]

The behavior of the specular highlights on the borders of polygons reminds me strongly of Gouraud shading for some reason. Is this just because many models are so low-poly that a good-looking highlight is impossible?

[taylor]

Pretty much.  And some models just have issues anyway that work to hinder lighting quality.  But there is also the issue that the default specular exponent is not what I personally prefer, but what other people wanted since it better matched the D3D code effect wise.  You can play around with that using the -ogl_spec cmdline option, which may help improve the specular highlight effects for you (to some extent anyway).

[wolf]

The behavior of the specular highlights on the borders of polygons reminds me strongly of Gouraud shading for some reason. Is this just because many models are so low-poly that a good-looking highlight is impossible?

That's because of per-vertex lighting. What you want is per-pixel lighting, and it's being used in glsl builds. For the main light only, though.

[Woolie Wool]

Which builds are those? Unstable branch?

[taylor]

Which builds are those? Unstable branch?

My Xt builds, the ones with normal mapping support.

[Admiral_Stones]

Looking at the Wiki, I've never thought there are so many maths behind shadering.

[Koth]

Computer graphics are nothing more than visualized math.

[Flipside]

I read that as Bling shading, thought it must some kind of really heavy environment mapping shader or the like

[Admiral_Stones]

Phlame me if you wantm but I don't view a real difference between the three pics. A more obvious/larger example, maybe?

[Turambar]

the difference was in the highlights and they way they look in comparison with each other.

compare that to the standard fs2 shading method and you'll see a real difference.

[Woolie Wool]

http://en.wikipedia.org/wiki/Gouraud_shading <--This is the standard FS2 method. Compare it with the above examples.

[blackhole]

[Woolie Wool]

I'm using the XT build on a GeForce 8800 and mine doesn't look that good...is there a newer XT build than 10/28?

[Admiral_Stones]

http://en.wikipedia.org/wiki/Gouraud_shading <--This is the standard FS2 method. Compare it with the above examples.

That was published in 1971?

Did they use like 10 University Mainframes to calculate that for a sphere?  ?

[Woolie Wool]

Full-on Phong shading (not Blinn-Phong) was invented in 1973, and it wasn't feasible for computer games until very recently. A lot of stuff is developed for CGI rendering and trickles down to games.

[Admiral_Stones]

Back in 1973, I don't think they had CGI already.

[Nuke]

phong did do a computer rendering of a vw beetle back in 'the 70s i think. with their computing power it probibly took a few months to render.

[Admiral_Stones]

And how did he model it  ?