Author Topic: Suggestion for noise evaluation to prevent premature renders  (Read 1597 times)

2018-08-29, 20:39:57

cgifarm

  • Active Users
  • **
  • Posts: 55
  • Your Brand New RenderFarm
    • View Profile
    • CGIFarm
Hey guys,

I run Corona on our farm and some times it happens that a client uses just noise limit to stop a frame and we get some frames finishing faster than
they should be.

I've discussed with Thomas M. Grimes a few days ago at D2 in Vienna about this issue and he mentioned that he was able to reproduce and that's a known bug
which has not been addressed so far.

My suggestion is that instead of sampling every X passes (5 is default) is that you sample every pass for noise limit, and every X passes you do an average
of noise collected from the sampled data. In this way if there's a big difference between sample 4 and 5, it can be filtered out by doing math on curves, so it should give less false
stopping signals.

Not sure what is the cost of sampling every pass, but at least it will not stop frames unexpectedly and people can rely on using noise level in their animations
or still renders.

Let me know what you think.

Thanks!

Alex
Working on a Renderfarm Platform - checkout our website cgifarm.com and our cost calculator : https://www.cgifarm.com/renderfarm-cost-calculator

2018-08-31, 15:57:44
Reply #1

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 6102
    • View Profile
Not so much "not addressed" as "not pinned down as to how / where / why it happens" :) Definitely has been addressed in that it has been checked in quite some detail, with facts and figures collected and recorded :)

As an fyi, a couple of days ago I added the suggestion you mentioned into the internal bug tracker on the issue as being a possible solution - ideally we'd be able to identify why the noise level sometimes drops extra low, and prevent that from happening; but if not possible (or maybe possible but would just take longer), the idea of taking 3 measures of adaptivity (at 4,5,6 rather than just 5) and throwing out an outlier could work - though of course then you'd be doing three times as many adaptivity calculations which would introduce some extra calculations to the render time (which is why option 1, find and prevent why noise limit sometimes calculates lower than expected would be best).
Tom Grimes | chaos-corona.com
Product Manager | contact us

2018-08-31, 18:39:59
Reply #2

cgifarm

  • Active Users
  • **
  • Posts: 55
  • Your Brand New RenderFarm
    • View Profile
    • CGIFarm
Hey Tom, thanks for the answer! I just wanted to write it down because I forgot so many times about this and I always remember when a client has an issue.

Alex
Working on a Renderfarm Platform - checkout our website cgifarm.com and our cost calculator : https://www.cgifarm.com/renderfarm-cost-calculator

2018-08-31, 18:43:00
Reply #3

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 6102
    • View Profile
I agree, it's good to have it written down :) I added a link to this thread to the bug tracking task too, so we can reply here with any updates or current thinking on the idea. Cheers for posting!
Tom Grimes | chaos-corona.com
Product Manager | contact us