Vray 1.5 Distrbuted Rendering Changes?

Hello,

• Rendering an animation sequence in distributed rendering mode is not supported. Use distributed rendering only for single frames.

• Incremental add to current map and Add to current map modes for the irradiance map are not supported in Distributed rendering mode.

Are both of these still the case with Vray 1.5?

Thank you

Re: Vray 1.5 Distrbuted Rendering Changes?

This is false. DR now works for animations. The help index needs to be corrected here.

• Incremental add to current map and Add to current map modes for the irradiance map are not supported in Distributed rendering mode.This is still true since we didn’t want to break the DR system at this point. However it is an important feature and will be implemented very soon.

Best regards,
Vlado

thats confusing. i always use incremental add to current with DR when doing animations

Well, it will do the same thing as single frame at the moment.

Best regards,
Vlado

Thanks for the responses – that’s good news! I am getting ready to render out my first animation. It is about 1800 frames, with only camera movement. I will use a pre-calculated IR map. I only have access to 3 rendering PC’s. Which method would render out these frames the quickest:

DR or Backburner?

Why?

Thank you for your help

well, i would imagine backburner would be faster. Of course I still dont see why people want to render animations with DR. Now sending DR stills to backburner is something that’s awesome.

I am wondering if its possible to send a GI calculation to backburner using DR instead of being able to use only a single computer to do that.
I mean, calculating the LC and the IM each 15 frames and saving it to a couple of files (one for LC and one for IM), but using all computers in the farm…Currently all computers have to wait with nothing to do until the GI calculation is done, that’s a huge waste of time since it can take hours!

Use QMC instead of LC. QMC will DR.

What? We can net render DR jobs with 1.5?

QMC is not the way to go with long -or short- -or microscopic- animations unless you have 1000’s of computers. And AFAIC, LC can be distributed in 1.5 like IM.

I don’t think we can netrender DR but using all computers to calculate the IM+LC would be awesome, dont u think?

Does the render still crash when u leave the DR option ON on backburner?

The current build (RC2) crashes, yes. However, it occurred to me recently how to deal with this issue so we’ll try to correct it - so that ideally you would even be able to queue DR jobs to backburner.

Best regards,
Vlado

It sounds pretty strange. In that case, should we assign the task to one computer and this one would try to use the other computers as DR nodes? I guess it would interfere with usual backburner per frame renders, because if you try to render via backburner at the same time as u try a DR the machine will crash due to lack of memory…The same would occur if 2 or more computers tried to use each other as DR nodes. Backburner should be aware somehow of the status of each machine (busy, idle), so that it didn´t assign more than one task to each node. Basically backburner processes and DR processes should be connected in some way…Does that make any sense?