That is a little strange formula for BRDF.
Second part look like Beckmann distribution, but were is 'e' it thats case? And whole top part is seems like oversimplified, isn't it should be e^(((n.h)^2-1)/C^4*(n.h)^2)? Assuming that C^4 = roughness^2?

And the third part kinda looks like Cook-Torrance geometric function (because using 'min'), but also look strange.
So, i'm just curious, what distribution/geometry term did you use?

"dynamic dirt levels (older ships showing their age?)"

YES PLEASE! I can also think this would allow dynamic damage on the hull, right?

On the dirty ships idea, I've always thought about what EVE would be like with a system that allowed a ship to "get old". Sort of like that trusty Corellian freighter that looks like crap but can still make the Kessel run in only 18 parsecs! It's not a gamechanger, but I think it would add a whole new dimension to ship ownership. It's the little things that make me excite! :)

Dynamic hull damage is unfortunatly something else and not included but i might be able to convince ccp mankiller to think of a way to use the dirtmap for stuff like that.

(I've also asked this on reddit...)

I've taken a look at the resource files on Sisi and I have a few questions:

• What is the *_ap.dds texture? An albedo map + some kind of mask?
• The *_pgs.dds files are now *_pgr.dds files. What is the "R" channel that replaced the specular map? (Roughness?)
• The *.black files are gone, so where are the material definitions?

Once upon a time there was a Dev Blog that gave a nice breakdown:
http://community.eveonline.com/news/dev-blogs/art-blog-improvement-of-textures-and-a-new-scorpion-model/

I'm just wondering. There are a lot of people who like to use Eve assets in 3D projects and the new models are too beautiful to pass on!

so where are the material definitions?
[/list]

Once upon a time there was a Dev Blog that gave a nice breakdown:
http://community.eveonline.com/news/dev-blogs/art-blog-improvement-of-textures-and-a-new-scorpion-model/

I'm just wondering. There are a lot of people who like to use Eve assets in 3D projects and the new models are too beautiful to pass on!

I guess you can say that the new light modeling looks pretty... Schlick.

First, the _ap.dds could you mean _ao.dds? If so it is an ambient occlusion map
Second, the R represents the roughness map and
last all material definitions are in the data.red file.

I mean the *_ap.dds files, just like I said...

http://i.imgur.com/0wSxcG1.png

I do understand that large models (capital ships, etc) have ambient occulsion maps in the *_ao.dds files. But, smaller ships do not. All ships using PBR have the *_ap.dds files.

Those files have greyscale texture data on the RGB channels with some colored data for "weathering" and fine painted details (as seen in the image I linked), and a mask for said fine painted details in the alpha channel.

That is a little strange formula for BRDF.
Second part look like Beckmann distribution, but were is 'e' it thats case? And whole top part is seems like oversimplified, isn't it should be e^(((n.h)^2-1)/C^4*(n.h)^2)? Assuming that C^4 = roughness^2?

And the third part kinda looks like Cook-Torrance geometric function (because using 'min'), but also look strange.
So, i'm just curious, what distribution/geometry term did you use?

In that case its the albedo map + a new mask map

last all material definitions are in the data.red file.

(I've also asked this on reddit...)

I've taken a look at the resource files on Sisi and I have a few questions:

• What is the *_ap.dds texture? An albedo map + some kind of mask?
• The *_pgs.dds files are now *_pgr.dds files. What is the "R" channel that replaced the specular map? (Roughness?)
• The *.black files are gone, so where are the material definitions?

Once upon a time there was a Dev Blog that gave a nice breakdown:
http://community.eveonline.com/news/dev-blogs/art-blog-improvement-of-textures-and-a-new-scorpion-model/

I'm just wondering. There are a lot of people who like to use Eve assets in 3D projects and the new models are too beautiful to pass on!

And finally the black files for ships are all gone because all of this data moved into the SOF, as explained in this devblog!

Sooo, i had just looked at those same ships on SiSi moments before i read that blog....

While i was looking at the new ships i kept saying, that looks nothing like it...the one i'm seeing has no color minus a few spots...

Then my borther looks at them on his PC...his looks MUCH different....so we compare graphics settings...

As it was, most ships looked absolutely terribad on my PC...with these settings (note, not exactly 'everything at lowest possible' )

A-A = medium
P-P = Low
Shader = Medium
Texture = High
LOD = High
Shadow = Low
Interior EFX = Medium
Interior Shader = Low

---
Now once i changed everything to max did the ships look decent. Brutix still looks like i'm flying a chrome Christmas tree ornament.

I'm sorry, but did you even consider those of us that don't run our clients on full max GFX settings before you made the non-full versions? Very disappointed.

--
Okay, so i narrowed the biggest "ugliness" culprit down....its the "shader quality"

With everything else set on "high"...here is what happens with the model when changing "shader quality" from high-medium-low....something is off here....

High
Medium
Low

Maybe someone (i don't know, maybe the Imbedded QA????) should test their stuff before saying its 'done'. Good form CCP.

I suspect it is like this for numerous other ships as well. But since i'm paying you (CCP) to test these sort of things while you are developing them...and not the other way around...i have other things more important to do.

Would you folks care to give us a list of all the maps (and specific channels/masks) and other files you guys are using to texture with?

Hmmm, I would have explained that in the devblog, but we have some brand new idea here for how to pack these channels and we are investigating this right now. So as soon as we have something solid, I'll let you know!

I smell a min spec increase on its way.

SoonGäó

So, the first part of the shown equation is the fresnel term and it's based on Schlick's approximation, nothing special here.

The second part is the distribution model and this one is called GGX, it's good for all non-organic surfaces like metal, plastic, etc. Please find a good explanation here, it's equation (3)

The final part then is the Schlick-Smith based visibility function, a bit optimized for performance.

I hope that helps a bit.

Hi Lil' Brudder Too,

We most certainly consider and go to great lengths to investigate the visual and performance impact of our graphic features on a wide range of hardware, specifically using our minimum & recommended specification machines as control points.

You have discovered yourself that the ships don't look as good on Medium shader quality, this is because we do not include the dirt map in that shader model. This reduces the instruction count by roughly 50%, this was done purposefully to ensure players with mid-range hardware that cant cope with high settings can have improved performance at the cost of reduced visual fidelity.

I can see your GPU should be more than capable of running the client with high shader quality, if there is a reason such as performance issues that is forcing you to run on medium shader quality then we'd love to see a bug report from you so we can investigate.