@ komaokc - thanks! the bake took a couple of tries.
@ tynew - thank you. yeah the bloom came out a little strong. I used transparency when I captured my images from Marmoset and it made the bloom look more like "outer glow" PS filter I guess the alpha can't quite handle the transperency of the bloom and it turns more opaque.
Looks great Howcome you used a Spec map over Metallic though? I would've thought Metallic would have made your life a lot easier as you could've assigned a pure white to the metal surfaces in the metallic map and then focused solely on the roughness, (or at least that's how I think it works, need to learn more about this.)
This is awesome sauce! I've been still trying to wrap my head around some of the PBR stuff and actually seeing these labeled textures and examples from a few other artists trying it out is pretty helpful. Considering I'm coming from a background of lowpoly and hand-painting textures, the concept of really minimal diffuse/albedo maps is still super strange for me. Great work!
@ Torch - Thanks man. My understanding is that metalness maps are simpler but less versatile. In the initial phase of texturing I briefly tried them both and found that I had more control over the end result using Spec. I also noticed that a lot of the amazing-looking PBR renders were using spec maps rather than metalness maps.
@Rocket Brian - Thank you! Yes I agree, the theory itself is pretty straightforward but I found that the best way to learn was to reverse engineer other people's maps and experiment endlessly with greyscale values. Yes you spend much more time in the microsurface / specular maps however the thing which I initially overlooked was that the brightness of the albedo channel interacts with the specularity, it doesnt just add colour.
@Nerf Bat Ninja - Thankyou very much. This was pretty much the sole aim of the project, I don't think I did a perfect job but with a lot of study and trial and error it didn't turn out too bad.
Thanks for the explanation, I'll definitely try out the method with Spec maps instead. Also for anyone interested, in this (or last) month's 3D artist magazine Etienne from Slide did a pretty interesting article on PBR workflow and values for various materials, definitely worth a look!
Looks good! You have AO in your metaliness which obviously doesn't make any sense. Switch it to solids and keep your scratches, you'll see the difference..
@sltrOlsson - Thank you! It is a specular map (different to metalness) but yes, I was really torn on this, and decided to put it on a very low opacity. I saw it done this way on Joeri Vromann's KSG Rifle, although he is using it much more subtly, which is probably more sensible:
This was before I had overlaid the AO/Cavity in the designated AO channel so it probably became reduandant after this was done. You are quite right though, it does not really make sense, next time I will try it without, as is recommended by many.
Well, I beg to differ. He slightly darkened the specular fresenl/metaliness/material map (what ever you want to call it, it's all the same) with the dirt mask. The dirt is in the cavities, that's why. It would make sense for you to base your dirt from the AO and then put some of that in the spec. Like 3-5% difference..
I took another look at it and you are right. The areas which looked like subtle AO are only in subtle areas where there is dirt (notably the top left corner where the grip is). Poor observation on my part.
Replies
@ tynew - thank you. yeah the bloom came out a little strong. I used transparency when I captured my images from Marmoset and it made the bloom look more like "outer glow" PS filter I guess the alpha can't quite handle the transperency of the bloom and it turns more opaque.
@Rocket Brian - Thank you! Yes I agree, the theory itself is pretty straightforward but I found that the best way to learn was to reverse engineer other people's maps and experiment endlessly with greyscale values. Yes you spend much more time in the microsurface / specular maps however the thing which I initially overlooked was that the brightness of the albedo channel interacts with the specularity, it doesnt just add colour.
@Nerf Bat Ninja - Thankyou very much. This was pretty much the sole aim of the project, I don't think I did a perfect job but with a lot of study and trial and error it didn't turn out too bad.
http://www.joerivromman.com/KSG.html
This was before I had overlaid the AO/Cavity in the designated AO channel so it probably became reduandant after this was done. You are quite right though, it does not really make sense, next time I will try it without, as is recommended by many.
Was checking these posts as a newbie to PBR here.
Workflow breakdowns are pretty informative, Thanks for sharing.