Search the Community

I'm building a railroad indicator lamp, essentially a red/green traffic light. I wanted to simulate the look of the optics of these lights which were designed by mounting an incandescent lamp at the focal length of a coloured glass fresnel lens (like in a lighthouse) such that its light is only visible when viewed almost on-axis as one would from a long distance. I think I've come up with a believable result using a hierarchical klieg light mounted inside the enclosure. It's aimed to shine inside the case and illuminate a small internal disc, standing-in for the lamp filament, sitting near the focal length of the lens. Viewing the indicator while it turns in front of you produces a result I'm satisfied with. semaphore_test.mp4 The problem is that, as far as I can tell, hierarchical lights (sun, bulb or klieg) do not handle well the shadows of other parts of the model. I set up a test model that's similar to my indicator lamp. Inside the open sided box is a klieg light aimed at a target disc that in this case has a hole in its centre. In any render from the modeling window there are no shadows generated by any surface regardless of whether the shadows are ray-traced or Z-buffered. Strangely a chor render does have translucent ray-traced shadows but the light still passes through the entire model and falls on an external model. My work around was to make an inside screen that was totally black so the light hitting the rear of the case is absorbed and only the target disc is illuminated. As for the light leaving the back of the enclosure, I'm fortunate that there will be about 4 ft. (1.2 m) between the rear of my indicator lamps and the nearest surface so I should be able to set a fall-off value that minimizes the spill. I suppose since hierarchical lights are typically used for things like automobile headlights this has never been an issue before. However you'd quickly notice this glitch if you were trying to shooting a night scene using the dome light of car as illumination. The passengers would have shadows while the other parts of the car would not. Obviously the ultimate work-around would be to add the lights in the chor but it would be nice to have this convenient feature work.

Hello AM, I explore AM's extremely efficient and powerful mapping system ! I have a question about surface displacement : The definition is not excellent and we often end up with a rather coarse physical aliasing if we approach the 3D model. Is there an option to achieve progressive aliasing depending on the distance to the camera? thanks in advance greetings

I guess you are stuck. Since no program presents everything you want and since you have to make some accommodation for what it doesn't, you will have to choose the one with the work-arounds you can live with having to do. The grid in the modeler can be set and re-set to any increment you want for precise CP placements with Snap to Grid. The "Measure Distance" menu option finds the distance between any two CPs.

By "particles" you mean hair? I can get hair toon lines by setting Toon in the Hair surface props. It doesn't seem to obey color and thickness unless those are set in the render settings. (Notice that toon lines only render if they have patches behind them.)

If I craft a displacement map for a wrinkle line it has an unappealing grayness to it on an SSS surface

I don't think Displacement maps work right in SSS. When this rectangular map is applied to a surface it seems that the 50% gray is being interpreted as negative displacement instead of zero. Ouch! Edit: this appears to be problem even without SSS Edit edit: It's a problem with PNG! If that decal is saved as a JPG the 50% displacement works correctly. Must be something to do with the weird way PNG makes its alpha channel.

I was aware that that is the case for quick renders, but I haven't seen it in a full render. I also haven't checked earlier versions for quite a while. Here are some quotes from Yves from this post. I'm not sure if anything changed following that post. Considering the skin texture would be different on the lips, decals for the actual texture and one Group for the surface settings (SSS, etc) would be the way to go. I got some (what I think are) improvements by dinking with a few things. Here are some settings I've tried: Specular size: 3000 Specular intensity: 30 SSS: 3.9, 2.55, 1.68 (Skin1 settings multiplied by 3) I'm currently doing some flattening in order to paint the textures (I've been messing with 3DPaint and Krita along with a drawing tablet) and I've made the eyebrows sub-dermal. Hopefully, I'll have an improved render in a day or two.

A few experiments toward getting a nice human Sub Surface Scattering result. "Loyd" model by Steve Shelton. Default surface (no SSS) render... Sub Surface Scattering renders. Pink and less pink... I've turned on some Surface "noise" to suggest the pores in the skin. SSS with Ambient Occlusion... Separate AO render... Curve-Adjusted AO... SSS render composited with adjusted AO... Separate specular render... Composited SSS+ AO +Specular...

The ALT snap bias aim feature is such a pleasure to use, I've regained my enthusiasm for an architectural model I set aside in 2009. It's my version of a Hollywood set of the grand old Pennsylvania Station built in New York in 1910 and demolished in 1963. I'm calling it a Hollywood version for three reasons. 1. Since my reference images from books and the Net only have a few readable dimensions I have to make a lot of "looks good to me" guesses. 2. The amount of detailing required for a 100% accurate copy is too staggering so I've simplified it while trying maintain a similar look. 3. The section of the building where people board a train is called the Concourse. While it was a spectacular work of architecture, it was essentially a large room where passengers would congregate before descending down stairs to tunnels in the lowest level to board coaches pulled by electric locomotives. To me that's a subway station not a rail station. So my Concourse will be a complete fiction but it will have steam locomotives. Here's four views of the real thing.

Here's a clue... If I add another surface (the large cylinder) outside the surface I want to target (the vase shape), the CPs I select and move (the 4 point ring) will go to the intended surface and properly slide on it when i move them. It is as if it always ignores the first surface it encounters and goes to the second surface. snap Test002.prj

When I test with a simple Prop it works as I would expect. I can draw a new spline and it is created on the front surface. I can slide points I have made on the surface and they remain on the front surface. But if my target surface is a regular MDL only the first creation works. if I touch them after that, they either go to a far surface or some place that is not either surface. That happens in both the modeler and in Model Mode in the Chor.

There have been attempts by Steffen to solve that I think, but in the end: If you can, hide the other side of the surface. If it is a prop, move the point after you have put it in. In most cases it jumps to the right one. It might have to do with normals, but I am not sure about it. Best regards *Fuchur*

Is there a way to keep Snap to Surface from jumping point to the opposite side of a model?

Excellent suggestion! Attached are three sets of four images with descriptive names for clarity. 9x9_noblur are the originals; 9x9 oversampling with no motion blur 9x9_blur are the same oversampling but with 100% motion blur 16x16_blur are sampled at 16x16 with 100% motion blur It certainly helps the chaos in the reflections but ultimately these grazing angled reflections of closely spaced vertical lines in a very wavy surface will be pretty noisy. That chaos is a good simulation of reality. If I really want less noise, I need to reduce the bumps in the sheet metal material. As for replacing the normal map with a bump map, also a great suggestion since it's trivial to generate and replace with no re-modeling required. I'll just have to get over my obsession with the improved (IMO) look of a normal map. Thanks, Robert. piggyback_walkaround_9x9_noblur_00.tga piggyback_walkaround_9x9_noblur_01.tga piggyback_walkaround_9x9_noblur_02.tga piggyback_walkaround_9x9_noblur_03.tga piggyback_walkaround_9x9_blur_00.tga piggyback_walkaround_9x9_blur_01.tga piggyback_walkaround_9x9_blur_02.tga piggyback_walkaround_9x9_blur_03.tga piggyback_walkaround_16x16_blur_00.tga piggyback_walkaround_16x16_blur_01.tga piggyback_walkaround_16x16_blur_02.tga piggyback_walkaround_16x16_blur_03.tga

If I export a model that is not a single surface to STL and import that STL as a prop it is definitely not a single surface.

The problem that is solved is the limited resolution of the image file used by the decal. A way to paint a texture using a 2d software is to unwrap / flatten the model, make a decal and save that decal out using a screen shot or use take snap shot in A:M. Both is great, but it has limitation that the resolution of the taken snap shot is limited by the resolution of the display you are working on. This feature can now create "the same" result / something similar but with a more or less unlimited resolution. Best regards *Fuchur*

Installers: Windows: Windows 32Bit Windows 64Bit Mac OS X 10.13.6 or earlier: Intel Mac v19.0 Intel Mac SSE4.2 SDK: v19.0 SDK UpDate Change Log: Fixed All 7081 PNG transparency not working Fixed All 7074 Chain of bones scales erratically Fixed All 7077 Bone color not displayed only OpenGL affected, not OpenGL3 (recommended to use this driver) Fixed All 7057 Crash during A:M Backup it should have now a better error message for this (both as messagebox and in the logfile), let me know if the crash still occur. Fixed All 7073 OBJ export or Prop import not working 5 point / hook export fixed Repeated decal export not fixable, due obj file format doesn't know anything about repeatable decals Fixed All 7075 5 point patches render with gaps Fixed All 6976 crash on Paste Mirror on a Pose Fixed All 7072 Progressive renders are blank Fixed All 7070 Shift-cursor key doesn't move rotoscope. Shift + Cursor moves now the rotoscope/bitmap 10 pixel Ctrl + Cursor scales around center Fixed All 7071 CFA creates unnecessary copies of Materials Fixed All 7062 Copying patches dosen't copy decals on patches Fixed All 7060 Copying a patch (only 1) with a decal does not copy the decal Fixed All 7054 Decal exporting A new menu entry for single stamps / a opened stamp view is added. "Render Stamp to File" This renders the actual stampview to file with a user defined resolution. If You start from the stamp entry in the PWS and the stampview isn't open, a new stamp view is opened. In the dialog , which opend after hit the menu entry, You can change the Path and Filename (the button on the right side) fileformat is selected trough the filename extension (supported OpenEXR, TGA, PNG and PSD) the "Width" and "Height" for the newly created file limited to 16384 x 16384 for Win64 version, 8192 x 8192 for the others the limitation is the result size, means a resolution from 32768 x 8192 is also possible. Warning: On the highest resolution around 14 Gigabyte is needed at the creation process. The width and height are always changed to a even size. (1025 becomes 1028 , always modulo 2) "Save layered" If this checkbox is checked, two layers are created, one for the decal self and one for the splines. For all formats ,excluding OpenEXR, in this case two files are created with the naming scheme "Filename"_Decal".Extension" and "Filename"_Splines".Extension" OpenExr is saved as one file with two layers. "Render with CP's" rendering also the CP's on the spline layer Fixed All 7061 Snap Group to Surface snaps to wrong surface. Fixed All 7068 Rename Bone renames wrong bone Fixed All 7066 Crash on Radiosity render Fixed All 7052 Environment Map materials not working Fixed All 7059 Tooltips only show for keyboard-shortcutted buttons Fixed All 7058 Progressive Render of shag crashes

Looking a bit further... when I look at just the cross section of the shape I see that the portion between the big ridges is only a single surface with no enclosed volume (see arrows). The 3D printer will need a model that is entirely enclosed. If you lathe an outline that is a continuous enclosed shape and use that as the basis for your model I think that would export and print successfully.

Hello, I had an idea about an animation I'd like to do which I'm trying (probably unsuccessfully, knowing me) to keep fairly simple. It involves basically one thing that I want to make sure is feasible before moving forward. If not, then I need a Plan B. I have a round hole in a flat surface. For part of the animation, I want the round hole to morph into a square hole, and then back to a round hole. Is that something that can be animated? Thanks

The latest addition to my railroad needs long sheets of diamond pattern steel plate. I wanted these plates to look well used, slightly rusty and to have varying amounts of dried dirt on them. I wasn't looking forward to building the material because while it's relatively easy to build show-room new surface finishes, I had never been happy with any of my attempts at old steel and procedural dirt looked too much like soft edged clouds. I thought the only way I could get hard edged dirt would be a decal. However after much experimentation I've got a material that's more than adequate. At this distance it's fairly convincing. The diamond pattern is a tiled normal decal. And it almost holds up at arm's length. The best part is that since it's procedural you can have a lot of dirt. Or just a few random patches. The materials (one for colours and one for textures) use a combination of native A:M and Enhance:AM combiners. If anyone's interested, I can post a schematic of the materials showing how they work.

Absolutely. As with most things in A:M how you model depends on things like; 1) What's your personal tolerance to patch creases 2) How is the model to be used; a standalone "prop" or something whose shape is going to be animated - a billiard ball versus Mike Wazowski 3) Does the camera get close enough to the model to see any patch creases Speaking as someone who mostly models inorganic hardware from the middle of the last century (i.e. things that are assembled using many, many rivets) I want a rivet head (essentially a demi-sphere) that uses as few patches as possible but still looks good at arms length. Here's a screen render of four medium grey (33% reflective) demi-spheres using the same 3cp cross-section spline, lathed 3X, 4X, 8X and 12X. The surface creases in the 3X and 4X are especially obvious in the top views. In my opinion 8X is the acceptable minimum while 12X is a waste of patches. Here are the same models rendered "outside". Once again the 3X and 4X would be fine if they were in the background of the scene. Not so much if a main character uses them to gaze at their reflection. However there are more tools at your disposal; manually adjusting spline magnitudes and the groups "normal weight" which is a material property you can apply to control how patch normals are averaged. Here's four demi-spheres again but the third model now has only four, 3pt patches. All its' magnitudes have been tweaked for circularity and a material has been applied to it whose only attribute is a normal weight of 67%. Here are the same models "outside". If you import this image and the previous, similar one into an image viewer and switch back and forth between them, it's clear the 16 patch model is more accurate. But for use as a rivet head, being able to replace a 16 patch model with a more than adequate 4 patch model really reduces patch counts.

The voting is closed. Now we await the results video! While you are waiting, consider... A compelling collection of creative computer compositions commendably completed for... Finish the Unfinished! You may still peruse the contest entries below You may still click on any image to see it full screen. You may still see the entire collection as a slide show by clicking on the first image and using the arrow tool at the right side of the screen to advance through the images. (I'm not sure if this works on a phone but, of course, you're not looking at these wonderful images on a tiny phone anyway!) Wireframes are still included with each image for your splinesmanship edification! Competition Images Dream Weaver: Wireframe: Dragula in Peril: Wireframe: Evolutions: Wireframe: Awkward Moon: Wireframe: Super Pfleu: Wireframe: Titan Surface: Wireframe: Quest for the Cup: Wireframe: LIAM: Wireframe: Metamorphose: Wireframe: Cow at Rest: Wireframe: Exhibition Images These images are not in competition but are submitted for your enjoyment, none-the-less! Planes Trains Automobiles Medal: Wireframe: New Year: Wireframe: Goldfish: Wireframe: Salt & Pepper: Wireframe: Bosch!: Wireframe: Thanks to everyone for all the fine and creative entries!

hello, this is the image generated with AM or your creative template ?? Modeling this with patches seems very complex to me. Especially to maintain a regular curvature of the surface of the back! A bump map or a normal map should do just fine. Would a displacement card work? Perhaps that would perhaps lengthen the computation times !? salutations

Hey. I selected the splines on the center and then did the steps below. It worked quite nicely. AND THEN CHOOSE THE X plane and keep the number at 000 for X axis. THEN CLICK "OK" and WALLAHHHH!!! ALL THE CPS SNAP TO CENTER.

I'm having issues in v19o with rendering normal maps on non-rectilinear patches using anything thing other than a final camera render in a chor. At the top of the attached image are shown five flat surfaces. The left-most four patcher is similar to the surface that I was modelling when I discovered this. The next three surfaces have the same outline but each is built with a different patch layout to see its effect. The right-most surface is a reference square 4x4 patch plane. All patch normals are aligned. The same normal map image is decaled to each plane. Isometric views of these five surfaces are rendered as 5 pass progressive or 16 pass final in a modelling window, in a chor window (not thru the camera) and with the chor camera view. In all cases, the decal renders properly only on the 4x4 square surface or when it's a final render thru the camera. Naturally since everything renders properly in a final render it's not a show-stopper but it would be nice if a normal map decal always rendered properly. I've attached the project used to generate the illustrative images. nrml_map_test_embed.prj