Manfrog

An update: I have managed to get the occlusion to work on the first object, by renaming the automatic UV map to something other than "default" Unfortunately now when I try to get the second object to Autopo, 3D Coat crashes to the desktop when it is calculating occlusion. The workaround for now is just to bring in one piece of the mesh at a time, but it would be great to be able to get the whole object at once.

I still can't get it to work. I may just be screwing up something with my workflow. Is there a documented best practice workflow for doing this kind of thing? I need to bring in a high poly model made of several pieces, retopologise them, bake textures especially occlusion then export them as LWO. If you could point me in the right direction it would be greatly apreciated. Thanks!

Thanks Javis! That's good to know, I'll try changing the materials. In the meantime, if anyone cares to give it a go, here is a link to the model: https://dl.dropboxusercontent.com/u/97214604/NMDA-to-3dc.zip

Hi Carlosan, Thanks for the response! Unfortunately, I still get black maps with 4.1.17D. I need to do more experimenting I guess.

Hi, I am bringing in high poly Lightwave objects, cleaning them up a bit with voxel brushes, then using autopo for per-pixel to retopologise them and then export the meshes back to Lightwave. It's worked well for the most part, and I love the occlusion pass that happens on the fly as part of the process. The problem I have is that sometimes the color map just comes out black. It seems to be worse if I bring in a multi-layered object. Sometimes the first layer that I autopo works fine, but then the other layers come out black. Is there a better workflow for this sort of thing? I'm using 3d-coat 4.1. Is there a newer workflow in 4.5? Any guidance would be much appreciated. Thanks!

Hi Andrew, I'm getting very slight stickiness with multicore on, no stickiness with it off. changing the number of processors had no effect. This is on the dx 3.2.12B 64 bit cuda, as well as the dx and gl versions of 32 bit cuda on my other machine. I did not test gl on the 64 bit machine because the "serial number is being used by another machine, exit immediately" warning popped up even though I had shut down 3DCoat on the other machine. I thought it was allowed to install on up to three machines, as long as I am only using it on one at a time? If it is allowed, how long must I wait after shutting down on copy, before opening another? John

Hi Andrew, I tried build_3.2.12A on both of my machines, the results were the same as 3.2.12 both were smooth in per-pixel painting until i turned on multicore optimisation then they were lagging again. Attached are my log files. Please let me know if I can do anything else to help. Thanks! John 32-bit_Log.txt 64-bit_Log.txt

OK, here's what I'm getting with 3.2.12: i7 950, geforce gtx275, Win7 64 home premium, 3DCoat 3.2.12 64bit cuda, per pixel painting brush lags in both dx and Gl versions. turning off multicore optimisation makes it nice and smooth. Athlon Quad core, geforce gtx275, WinXP professional 32bit, 3DCoat 3.2.12 32bit cuda, per pixel painting brush lags in both dx and Gl versions. turning off multicore optimisation makes it smooth. Thanks for working on this Andrew. You are definately making progress!

Thanks for playing geo. Can I ask about your machine specs? I'm going to keep posting about this problem on this thread: http://www.3d-coat.com/forum/index.php?showtopic=5589 Andrew posted there yesterday, and it's probably good to try to keep everything in one place. Thanks! John

I tried 3.2.11 on my old computer last night, and here'es what I got: 32bit DX version, the per pixel brush lagged and jumped, the same as on the 64bit dx and ogl versions on my other machine. BUT, the 32bit OGL version worked smoothly. So at least I can work that version. To sum up: i7 950, geforce gtx275, Win7 64 home premium, 3DCoat 64bit cuda (and non cuda) 3dcoat per pixel painting brush lags in both dx and Gl versions. i7 laptop, geforce gtx260m, Win7 64 home premium, 3DCoat 64bit cuda (and non cuda) 3dcoat per pixel painting brush lags in both dx and Gl versions. Athlon Quad core, geforce gtx275, WinXP professional 32bit, 3DCoat 32bit cuda 3dcoat per pixel painting brush lags in dx version, works fine in Gl version. In all cases where the brush lagged, turning off multicore optimisation had no effect. Please let me know if I can test or try anything else. Thanks! John

Hi Andrew, thanks for your input, but turning off multicore optimisation didn't fix it. I tried 3d coat on my wife's laptop, and it has the same problem. It is running Windows 7 64 bit, with an i7 processor and an nvidia gtx 260m for graphics. Let me know if I can give you more information to track this down. Thanks! John

This is with Per Pixel Painting. Micro vertex painting is much smoother, so that's what I used for my last project, but I think per pixel is the best method for my workflow - if I could get it to work correctly. I haven't tried ptex, I'm not even sure how that works, I need to do some reading

Hi Geo_n, what video driver version are you using?

Anyone find any solutions for this?

Has anyone who has experienced this problem found a solution? Is there a place that I can download older builds of 3dcoat, to see if they are any better?

Hi Andrew, Thanks for responding. This is the cube preset, per pixel painting, with the default settings. The texture resolution is 1024, no subdivision, no smoothing of UVs.

Here is a screengrab. What happens, is this: I draw a nice curvy stroke, but the cursor sort of jumps from point to point, leaving straight line segments between jumps. The effect is exaggerated if I paint quickly, but also happens if the brush is moved slowly.

I have been having this problem as well. I am using 3.2.11 and the brush lag is still happening for me. My machine is an i7 920 with 12 gigs of ram, nvidia GTX275 graphics card windows 7 64 bit. It happens if I just open the program and open one of the standard models (the cube for, instance) in per pixel painting. Microvertex is nice and smooth.

I'm getting the same thing. The brush lag makes per pixel painting pretty much useless except for very rough work. I'm using 3.2.11 on a similar system 64 bit windows 7, i7 920, 12 gigs of ram except my graphics card is an Nvidia gtx275 with the latest drivers. Is anybody using per pixel smoothly? Thanks! John www.MeshAndBones.com