Grimm

In general CUDA runs slower on Quadro cards, most people get Geforce cards instead. Cheaper too. That said I don't use the CUDA version of 3D-Coat as it needs a really old version to work. Other people here do use it and it does speed up the program. If you are interested in using other CUDA/GPU renderers, etc. than you would be much happier with a high-end Geforce card (780ti, Titan Black) with lots of memory.

Or maybe "PBR Materials" unless that is too generic, already taken, etc?

I concur with AbuRanger, I have a 460 and a 980. I use the 460 to drive my displays and just use the 980 for compute. Has Andrew updated the CUDA version for 3D-Coat? Last time I tried to run the CUDA enabled version it wanted version 3 I think. I'm currently running version 6.5, so it doesn't make the CUDA version of 3D-Coat very useful.

Thanks AbnRanger, I was thinking that was the problem. I will give the Strokes tool a try.

I made a landscape in Blender and brought it in to 3D-Coat in surface mode. I wanted to sculpt some details in and then paint it. Is there a way to autopo it? I have tried all sorts of ways to do it but it hasn't worked out very well. Or do I just have to manually topo it? It's not a volume so I'm not sure if autopo can handle it. Here is the landscape after the sculpt: After several crashes and really messed up meshes, here is the best of the lot. But it still has big issues and doesn't look right: Thanks, any hints on workflow would be awesome. Jason

+1 on "smart materials" as well.

Anything where the software is determining how the texture is computed is procedural. AO, cavities, noise, etc are all procedural processes. I do see that this shouldn't be called procedural because you can still build the material without using any of the conditionals or masks, so no worries.

Thanks for the explanation of the new tools Artman. So these new conditions would be the part that is procedural, very cool. I can't wait to try it out, Linux version?

Procedural paint ?

I suspect that the difference is that Cycles is an unbiased engine, whereas Substance Designer, like game engines, is biased. Vray should render a similar image as it's biased as well, at least you should be able to render it so it looks like Substance Designer's render. Personally I like the Cycles render better, it looks more real to me, but that depends on what your after.

Cool!

Pcie 2.0 should not be a problem, that is what I have on my motherboard and it works fine. Pcie 3.0 is backwards compatible. Would you be happy with one? I guess it depends on what you want to do. For games it's the fastest card on the planet, for GPU rendering not so much. It was a nice upgrade for me, 2 Gbytes more vram, many more CUDA cores = much faster rendering with less heat and power usage. Depending when and if the newer cards come out and are cheap enough, I might sell mine and get one of the new ones. Shouldn't be too hard to sell it as the gamers really love them.

No problem.

spacepainter, calling Andrew a good boy is, in english speaking countries, a term of endearment. It comes from the stories children are told about Santa Claus and how "good boys" get presents and "bad boys" do not. Look up "colloquialism" if your interested. In other words it is a friendly and benign statement.

I have a MSI 980 and it's very nice. It's only as fast as a 780 though and much slower than a 780ti. There might be some improvements though with drivers and software but I don't think that it will be the fastest card on the planet by any means. I use Octane and the devs have gotten the 980 almost as fast as a Titan Black with the path tracing kernel. The Blender/Cycles devs have been looking in to speeding up the 900 series as well. Compared to my old 460 it's blazing fast. The other advantages to these cards is that they use much less power and keep cooler than the 700 series cards. Even at full load my 980 has only ever gotten up to 76C.

Any news on this?

Awesome! This is going to be fun!

Abstrax, one of the Octane developers posted this today: Looks like the CUDA tool kit might be a lot of the problem, it's improving step by step though.

There is still a possibility that these cards can be speeded up some. They are talking about splitting up the kernels to get better performance in Cycles, the Octane devs have a 980 they are playing with. I guess we will see if they can come up with something. Also Nvidia might have some tricks they can put in the drivers. For awhile there Fermi cards were stomping Kepler cards until the later drivers have turned that around. It's a waiting game now.

I had seen those benchmarks but I don't know how they tested the cards. The results we have been getting with this testing here are much different: http://blenderartists.org/forum/showthread.php?350975-The-new-Cycles-GPU-2-72-Benchmark/page6 Scroll down for the spreadsheet of results. Mine is under Grimm of course. By playing around with the tile size I was only able to get the render down to 7 min. (256X256)

I'm really enjoying my 980, it's 5 to 6 times faster than my old 460. I still use my 460 for the display and keep the 980 for doing CUDA rendering. In my testing with Cycles and Octane it's almost as fast as a 780, but much slower than a 780ti. Hopefully these new versions will have a wider bus or they do some tweaking to the drivers, etc. so they are faster than a 780ti?

Thanks Javis, Issue1 - I really like the noise functions in live clay, it's so nice to use. PS doesn't run on Linux, although I could try Gimp or Krita. Issue2 - Yes exactly. The cavity functions do work great for cavity stuff, so I'm thinking that maybe the height functions might do what I want? The problem is that the height functions have never worked for me on the Linux version. If I'm remembering correctly (I need to double check to be sure) is that both of the height functions don't do anything no matter how you set the settings on them. If you use the preview window, one is always black, the other has some kind of strange inverse look to it. Jason

Thanks Javis, I put a note on the report. I will check and see if there isn't another report on the height option and either add a note or put in a new mantis report for it. I think my first issue is more of a feature request. Jason

Thanks carlosan, Yes, the latest Linux version 4.1.11B. I was thinking of putting in Mantis reports on these but wanted some confirmation first. Jason

Hey, I have been playing around with 3D-Coat for the last couple of days and have been stymied at every turn (at least it feels that way). I'm probably being stupid on most of this stuff so feel free to call me on it. Here is my list: 1) Surface mode using live clay with the noise functions. Is there any way to keep it from adding more depth but still have the noise imprinted on the surface? Do both a negative and positive displacement but keep the mid range depth the same as the original surface? 2) Painting by depth range. The height and cavity options don't work well, the height options don't work at all for me in the Linux version. I would like to be able to specify a depth range that the paint tools would be limited by. Is there any way? 3) Exporting displacement maps is very broken. Most of the options don't output anything but either a solid grey, or a completely transparent image (even if you do a PPP merge). I think that has been the most frustrating part as it seems that everytime I think I have figured it out it changes on me. I did finally get it to work by choosing "Grey - based, not normalized". If you choose "Zero level is grey" you get displacement above grey but loose all of the displacement below grey. Anyway thanks for letting me rant. Jason