Numerical Error

[flipbook]

Hello

Getting a 'numerical error' at random times which halts the sim. It is 15th Feb nightly Max 2014 Vray 3. Below is a link to the log right after it produces the error.

https://www.dropbox.com/s/xvf9xxvb31...ver16.log?dl=0

https://www.dropbox.com/s/d59jq382fr...rors_.JPG?dl=0

Thanks!

[Svetlin.Nikolov]

Hmm, for this kind of error the log cannot help - would it be possible to send over the scene?

[flipbook]

I don't have the exact scene anymore as deleted everything and started again lol. I had 2 body force helpers (seperate objects) first active from frame 0-20 and the second active from 15-100 (animated strength values). The error pops up during the crossover frames. It did not do it when only 1 Body Force was being used. If that helps!

[Svetlin.Nikolov]

Ahh, okay - I think this was related to a fix that goes into today's nightly regarding body forces - it should be gone since today, but in case it happens again - please bump the thread and keep a scene so we can track it down and fix it Thanks!

[Truffi]

Hi
Using PhoenixFD3.00.01 inside 3dsMax2015 I have a new one on the numerical error side. Sending you the link to the scene..
My guess it has to do with the turbulent force..

[Truffi]

In case you want to have a look, it occured at frame 186 in the AUR sequence (so +50 in the timeline).
https://www.dropbox.com/s/9lg77hetfr...Small.rar?dl=1

[Svetlin.Nikolov]

Hey, thanks very much for the scene. We are still working to simplify it and make it reproduce the effect faster in order to be able to debug it.

[Truffi]

I recreated the scene this time without a force. Turning off UNIFORM DENSITY for the flames helped me getting the twirly flametrails to work without encountering this error.

[Truffi]

Hi there
This video should show another example of the issue:
https://www.dropbox.com/s/0irj6z1m1f...scene.mp4?dl=0
There is no numerical error message, but the same phenomenon like with forces in the simulation: the fire and smoke gets sprayed apart at one point in the simulation (after 7 seconds in the video clip). Up until then everything burns just fine. There is no force or any other apparent source of this strange behavior.
Restoring the simulation at the point before the problem occurs didn't help, it results in the same unnatural instant spray-effect. It is pretty severe and I am almost sure people brought it up here before. If you know a fix or work around, please let me know. Thanks!
Find the scene attached here, in case you want to have a look:
https://www.dropbox.com/s/mrvjz0vleo...ssues.rar?dl=1

[Truffi]

same setup, but the inertial force is slightly lowered:
https://www.dropbox.com/s/p4783vcb11...82%29.mp4?dl=0
again, the fire should only be affected by the movement of the tire and the gridcage.

[Svetlin.Nikolov]

Hey,

I'm still simulating both scenes, but for the second and third problem so far my only suggestion is to not raise the inertial forces multiplier above 1. Haven't found the exact reason for the issue yet. Would it be possible to share the settings for the latest video?

Thanks!

[Truffi]

The inertial force in the first video was 1,5 and in the second video it was 1,25.
https://www.dropbox.com/s/dq8xa6si26...ation.rar?dl=1

Does it affect the simulation, if I make a check at "Motion Vel." in the helper?
From my understanding it shouldn't but could be wrong. I just checked it to make sure I could render a 3d motion blur.

[Truffi]

The older SLOG file was created at the start of the sim in the last video file, the newer SLOG file is dated to the time it was stopped. So I sent both. Let me know if I can provide anything more to clarify things..

[Truffi]

A new sim finished with inertial force set to 1.0, the strange behavior did not occur anymore. Very Good!
I'll continue messing around and let you know once something decent comes out (or a new issue). Thanks for the support.

[Svetlin.Nikolov]

Roger that, I keep simplifying the scene with the turbulence, because it seems to me like the issue there is of a different nature.