robcat2075

You question is not stupid. I don't want to belittle you for it. But you have asked it... it's going to get an answer. What I can tell you is what I know about A:M's development and how i know the lack of GPU rendering isn't because it's not been tried and how I see rendering time in the whole production process and what i know about other solutions to rendering time that can be done now. My next suggestion is a very serious one. It is not meant to brush you off... Blender has the render features it has only because they recruited very knowledgeable people to donate a huge amount of their programming time and knowledge to make it happen. Those are not trivial additions. Recruit a similar team to do that for A:M. That is what it will take. It will take an enormous programming effort. I don't see anyway to make it happen except to find the people who can do it. I mean that very seriously.

What does this mean? Does this mean 2D motion graphics like you do in After Effects with flat images but no 3D objects?

I'll tell you why the render time isn't a big hindrance for me personally... The amount of time I spend modeling, rigging, texturing and animating is so much larger than the render time that even if the render time were instantaneous it wouldn't change the duration of the project much. The things that A:M does without me, like rendering, have gotten to be the least worrisome part of it. The stuff where A:M is waiting for me to do something is the big bottle neck that no render speed up will change.

If someone can program it, I'm all for it.

It depends, of course, but in average 3-5 minutes for a 1280x640 9xmultipass frame without hair/particles. You should get in a time machine and go tell 1987 me that the day would come when it would not take 5-15 hours per 320x240 frame with no hair or particles even if you wanted them.

How long are your renders taking?

Well, I got taken! My Nvidia board has just 1GB and cost $85 about four years ago.

Sure it's 4? NVidia says it's 1 or 2 https://www.geforce.com/hardware/desktop-gpus/geforce-gt-630/specifications

It is impressive that it works at all since the GPU cards were never meant to do number crunching like CPUs do.

I'll note that Steffen did make a run at implementing GPU computing with A:M but it turned out to be not faster. Aside from the coding change that Fuchur discussed, a GPU renderer has to be able to hold all the data for what it is rendering in the GPU RAM so you'd probably need about a 4GB GPU card to do typical A:M stuff. A 4GB GPU card isn't wildly expensive anymore, I'm seeing prices of around $200-$500, but it would be an extra expense you'd have to accommodate. However, since we don't have GPU computing, you could take that same $200-$500 and get a 4 or more core computer to add to your rendering pile. I don't know much about the networking details but we have people on this forum who could help you with that if you went that route.

For Shaggy's "Bus Stop" Rope Trick I used a displacement map on a long cylinder. Because Shaggy's rope had to start out as a coil of simulated cloth that would fall on his shoulders, and because that was a difficult shape to decal i tried using the image as a "Patch Image" first. However I found that many patches would have to be individually selected to "Rotate" the patch image into the intended alignment. That was not undoable but I also found that when I set the image to Bump or Displacement the shading on many of the patches would appear inverted and I didn't have a solution for that. What i ultimately did was to simulate with the original coil and then constrain a long,straight cylinder version of the rope, with the image applied as a regular cylinder decal, to the coil which was then hidden for render. There were about 100 spline rings that were individually constrained across the two versions. (You can see several patches on the coil that did not orient right by default) Since you do not need to simulate flexible rope with Simcloth, my first idea is that you could model your ropes as straight cylinders, decal them, then use a pose to bend them into the shape you need them to be on the ship.

Because I don't like to paint in the Sun I envisioned this as a shady-side-of-the-house scene using Ambient Occlusion instead of directional lights. This is what the AO defaults do with Ambient Illumination and Occlusion set to 100% AO casts light from all directions but the Ground Plane is so large that it blocks all the light from below the horizon. That works well for most purposes but for something viewed from below it makes for some very dark surfaces facing the camera. In real life there is substantial light bounced up from the ground. To experiment, I progressively reduced the size of the Ground, to allow more light from below to reach the objects. You have to reduce it quite a bit to get a noticeable change. This is with 50% of the original dimensions... 20% 10%... No Ground Plane at all... The 20% version was about right for illumination levels but I began to get unhappy with the uniform grayness of it all. That led me to experimenting with Global Illumination, which I will show some tests of in the next post.

Yes, that would be essential when using "flatten" My preference, when painting decals, is to use cylindrical mapping which allows my to get good UVs with just one seam instead of two. However, if you are using decals from a photo, as for your jeans that may not be an option.

A case that produces a spline with odd curvature might be to take a beveled cube such as one out of the Library and decal it with an image using the spherical application method. The initial observation not having much to do with the application of the image itself so much as the layout of decal splines that appear via Decal>Edit. There should be at least one decal spline that has excessive curvature. i gave it a quick try. The result is not wildly unexpected. A cube is a difficult target for Spherical mapping. When you have time describe me more about what you are seeing that is odd.

My expectation is that you should apply a decal on splines in the state they would normally be in and I'm pretty sure that's true. Do you have a case you can show? I have often noticed distortions in realtime display that go away in final renders.

Fast motion blur and fast DOF are other effects that don't work against the default camera background. Put a card in back of them.

Initially, I put just numbers on the cards so it would be easier to be sure which one I was seeing in each window. Then I swapped in the proper images and fine tuned the placement.

Here is a problem I've made a report on. Anyone else encounter this shadow problem?

For this scene the global illumination/AO is the biggest cost of the render but in multi-pass that is distributed among the passes so more passes doesn't exactly equate to more render time 9 passes, 18 minutes 16 passes, 25 minutes 36 passes, 32 minutes

I really like the simplicity in that. The hints to the scene being part of a much larger world makes us believers that there really is a larger world there. My first thought was to use a dome with a 360° panorama on it but it's quite difficult to create such an image of high enough resolution so that you don't see pixels when you peer at just a small portion of it in the window reflections.

Here is a case where Multipass is potentially better The materials for the wood grain and paint chips make many fine lines and edges that are challenging for anti-aliasing. (click to see the full-res images) Regular render took about 76 minutes and has many dark spots... A 5 pass render took 19 minutes and gets a better appearance...

The reflections in the windows are photographs I took in my front yard and placed on some low-density grids that I positioned in back of the camera

Google has released a "Machine Learning Crash Course" https://developers.google.com/machine-learning/crash-course/ Someone learn it and tell me if it's useful.

If I rebuild the model I get sprites. Make sure you have Color Method>Additive Color>OFF

I'm not getting any sprites. What am I doing wrong?