Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - Byteman3D

Pages: [1] 2 3
I had a crash on a scene which was imported into 3ds Max as an STP file. Just wanted to report. Can't open the scene directly anymore but merge some of the objects.

The scene consists of a seven piece robot arm, which I have linked each piece, top to bottom. Did not collapse the objects. Assigned separate materials to faces in Body Object form.

Then copied each piece one by one, applied a push modifier to slightly push the copied object outward slightly. This would serve as a "glowing" net surrounding the object by a material assigned to it. This net was supposed to somehow ripple upon touch, during animation.

Net material is a layered material -not yet finished- which had a base material and a second material with opacity and self illum. maps, containing Mask map to mask a red Corona Color map with a png file assigned through a Triplanar map. Same without the mask exists in Opacity. I guess I was working on a noise map at the time as a mask to the layered material's layer.

After not being able to assign Face IDs to one piece, (matl. was not showing correctly) saved the file and exited. File doesn't open anymore.

Could merge the robot pieces, but not the net copies. They are copies of the same objects only exception is their materials. They all have the net material.!AoFN063z_3UOgdNhZ7xB5Zg9FGNtOw?e=kV4xY7

Here is a link to the file. 90 mb max plus two maps. I thought you might want to check.

[Max] Feature Requests / Selection Render Improvement
« on: 2022-09-15, 14:21:34 »
I'm really disappointed to see the Render Selected option not helping reduce the rendering times as much as it can. Let me clarify what I mean and discuss what can be done:

I needed to change the texture of a small box object in one of my scenes, and I've had the chance to use selection render for the first time.
Whole frame rendering of 500 frames animation would last about 5.5 hours.
Selection mode with single object selection reported  3 hours 25 minutes, which according to my experience with other renderers, I found unnecessarily long.

I've set a region along with Selected option and now the predicted time is 1 hour 25 minutes.

The predictions are solid.

It seems a feature to set a border region around selected object's extents and render only that region would help reduce rendering times drastically.

This reduced the rendering time to almost a half.

In my case, as the camera was moving, I've set the region extents to cover the object's extents throughout the entire animation so this is even less efficient than what you can implement as a feature.

Feature Name : Auto Region Selected :D - or just go on with the Selected but set a region. ı guess it will work.

I have about 60 materials with 60 different maps of different pieces of an oil painting. The glossiness of each piece is the same. Textures will not change during the animation but I may need to change glossiness and reflection amounts from time to time.

These 60 materials may be sub materials of a Multi-Sub Object Material. I wonder if there is a method to change the glossiness of all 60 materials from one place. Wire parameters is not an option. A separate reflective layer like the clearcoat -as we often use it- is not what I want either, however I'd like to know if it would work like the primary glossiness channel to achieve the look of oil paint, and what numbers should I set for the reflection, roughness... values of the base materials?

[Max] I need help! / Corona MultiMap Usage
« on: 2022-08-13, 13:09:03 »
It seems Corona Multi Map randomizes textures. I wonder if it can be used to assign Maps by Material ID's but without randomizing. Map1 to ID1, Map2 to ID2 etc.

PS. I couldn't understand why there are several "Randomize by" choices, when eventually all are randomized?

[Max] I need help! / Clipping Issues with VR Camera Roll
« on: 2022-08-07, 09:18:31 »
I'm working on a flythrough animation between mountain like structures (Actually the details of an oil painting) and this is a stereoscopic VR scene.

This painting was initially in its original size (13x13 cms) and in order to achive the effect of moving inside some huge structure,  I've managed to reduce the eye distance to 1/100 th of its default and the painting became something like 13 meters, but there appeared a well known clipping issue in the viewport. As I come close to surfaces I see the interior of the painting.

I've reduced the camera Near Clip to 0 but the problem was still there. (Tried Viewport Clipping too) Made some search and found in Autodesk forums the exact same question but without a definite answer. Some clues made me think I'm close to the smallest size limit of my System Units Setup, so I decided to Scale up the objects. It helped. There were no clipping issues as I rotate the camera to any 360 degrees. Yet, I tried to be extremely careful because if there is a spherical near region and this is a 360 degree animation, any part getting closer than any near clipping value would be visible. I didn't move very close to faces.

This is a radial motion with a large radius - the paint tracks are like ripples-  so just to keep things natural I used some banking (By using the cameras Roll command), while taking this large rotational motion.

It seems the clipping issues are present again when the Roll amount reaches certain values.

I seem to understand but I don't have a lot of time to test. (Clipping is not visible in the viewports) Any ideas about the exact reason of this? Should I not Roll the camera or is it only because I'm rendering in Stereoscopic mode (This makes sense)
Would it help if I scale up even more?

Thank you

Edit: Just reduced my eye separation to 2.0 and there is no more clipping for my animation. I presume the technique behind stereoscopic panorama creation requires more clearance no matter what the clipping value is and there is no way to preview it in the viewports yet. I'm eager to see the end result.

I wonder if there is a way to match the darkness of the shadows of the 3D objects and the background image? (In the 3D software- not in post)


There is a warning about motion blur on objects with changing topology which is pretty much acceptable. The message also says   those objects will be excluded from motion blur calculations, which is what I want but it doesn't seem to work this way. Objects with changing topology are not excluded and create artifacts. I've had these artifacts on an object with animated Slice modifier and an animated Symmetry modifier. I need to turn off motion blur for these objects. Standard Max motion blur switches on the Object Properties dialog don't help. Should I completely disable moblur on geometry from Camera settings?

I need two displacement maps on two separate materials of a Corona layered material.

In the image I have a base material of a human body and two different layers of scar tissue applied by radial gradient masks. Diffuse maps seem to combine well with masks but I can't manage to assign a displacement map along with other maps.

Is there a limitation or an I missing something?

I need to fade out a Shadow Catcher floor plane and I've tried lowering the visibility or blending the material with some transparent material. None of them seems to work. (Shadows disappear instantly) Is it not possible within 3ds Max with Corona Renderer?

This is an old feature I used to benefit a lot with Mental Ray, named as "Render Subset of Scene/Masking"

There are times we need to add people, plants or other objects after an animation rendering is completed. The Render Selected feature can be used to render objects separately with alpha, but this way you can't have shadows, reflections, GI effects of the object in relation to the existing surrounding objects. Adding Shadow Catcher Material to the rest of the objects can help render shadows but it requires extra work and poses a risk of saving the scene with Shadow Catcher Materials assigned, unless you save it with another name.

I'd be more than happy to have this feature at hand.

While I use UHD Cache as secondary algorithm, with,

Animation (Flicker Free)
Try to Load +Append
Save to File


rendering time is improved compared to when each frame is rendered from scratch.
The file size normally increases by time but the rendering time, first decreases gradually but increases as the rendering progresses without any noticaeble quality difference compared to the same frame rendered with a new UHD Cache file.
In a current scene I left the machine to render during the night and after about 200 frames, the rendering time was 2.09 at frame 2048.
I chose a new cache file and restarted. After the first frame is complete, rendering time is 1.07. Almost a minute of performance gain doubling the previous.
Normally rendering from the new UHD Cache file presents a very slight difference in overall image brightness but luckily it is not noticeable in my current scene and I moved on with the new cache file.

Previously it was reported that rendering would take about ten hours (which was stretching towards a clean 11-12 hours. It is now 6.30 hours left.

I believe there is no resource problem in the machine rendering this scene. With a totally fresh 2 TB's of SSD drive and a total of 128 GB's of RAM.

I would love to draw a performance/time graph by but I'm kind of busy right now. It seems to go down a soft slope.

ps. The Captcha window in the ticketing system has a problem in my browser (Chrome) its vertical size is limited to the Captcha area. You have to move a slider to see the blocks. It doesn't verify anyway.

Best regards.

I've been trying to find a way to render a 1750 frame medical animation which is crashing during render, sometimes after rendering 5 frames and sometimes not even a single frame. I have tried literally everything, and consumed all options.

Is it possible to send over a link to my Max 2022 archive scene file in private? It is about 60 mbytes.

Edit: I'm still trying to find a workaround. May be speaking early, too, but as a last resort, I turned of motion blur and depth of field on the Corona Camera object. DOF F-Stop parameter was animated. No crash yet after 4 frames. If it is related to an animated F-Stop parameter, what can be done as a workaround? DOF adds realism to medical scenes.

Thank you

I'm working on a medical animation consisting of a camera travelling inside the human vessel, while several particles are passing by. Furthermore, this is a 360 degrees panoramic animation.

Wherever I put the light, there is always a harsh light explosion on the particles passing close by. Highlight compress helps to some extent but still not enough- plus the image looses contrast.
I need the Light exclude function to exclude the particles but illuminate only the vein walls from the center of the vein (which is possible), while with separate light objects, illuminate the blood particles from outside the vein, which is not possible.

I need a soft illumination without harsh hotspots. Any recommendations?

As you might have guessed, I need to align a target reticle plane one by one to a number of particles.

I moved the pivot of the reticle plane to the exact same position of my 360 degree panoramic camera. Assigned a look at controller to the plane with appropriate settings to look towards a SDeflector.

I moved the deflector one by one to each individual particle. I thought this would do the trick but particle positions change during rendering and I can't achieve targetting effect.

I know this may be due to several reasons related to the particle settings, but I did the alignment by moving the deflector into each particle, viewport displays %100 of particles.
I have attached the scene file, too.

Can you please help?

I have a spare machine and thought it can contribute an animation render.
Communication seems ok but I get the following messages on slave node and master machine's Error Message(s) window.

I checked for the Map names but they don't correspond to any material maps in the scene.

I am rendering with Path Tracing+UHD cache, and enabled Load and Append UHD Map Option. Thought this could be the reason but tried even without GI. (And Cleared the UHD File name and saved the scene) - Still the same.

What is the problem here?

Edit: I tried to set the slave machine as master and  I 've got the errors on single machine rendering, too but this time Error Message(s) window presented options to select objects with missing maps. Missing Maps are from another renderer which is only installed on the first master machine.

This problem is solved but I still have a problem.

Two machines are not making the render faster. Instead two machines doubled the rendering time. There is this "collecting any extra data..." message which increases the rendering time with a long pause. What is that about?

Pages: [1] 2 3