Is there really practical need in such huge samples number for animations?

I'd guess with use with indirect light , complex lighting conditions and more bounces, this could be reasonable for single image.
Here I do not really see the reason why it was brought so much up.

---

The reason I borought up the number of samples up was because of the emission shader on the sparks. If the sample number was at say 750, the reflected light off the plane would be grainy and not look like reflected light. The higher the number of samples, the more accurate the bounced light was on the reflection plane.

It's look like 1 part is broken because 246 of 250 frames are ready and 996 of 1000 are ready but BURP shows that 997 of 1000 are ready.

It looks like it. The part count says 1000/1000 parts completed but the frame count says 249/150 completed. I can't find the missing part or frame in the details. I think this is a job for Janus.

Issue is in frame 43. I'll have a look at it when I can get to the server later today.

The WU in question was restarted.

What's the story with the motion blur? Sorry I'm getting to this so late. I've been busy.

Is this vector based motion blur? Pildanovak was mentioning something odd about that in his DNA/RNA sessions too.

I do not think this is vector based motion blur. all I did was hit the motion blur check box in the properties panel.

sorry I am getting the to this late. I haven't had a dependable internet connection long time enough to post this.