Skip to main content
deleted 6 characters in body
Source Link
brecht
  • 7.5k
  • 31
  • 50

Cycles uses Quasi Monte Carlo sampling:

http://en.wikipedia.org/wiki/Quasi-Monte_Carlo_method

That means it will try to evenly distribute samples so rendering converges faster. If you render multiple times with different seeds however then this even distribution is lost because it mixes different distributions without ensuring that the combined result is even too.

In principleSince Blender 2.78 there is a "sample offset" option could be implementedsample offset command line option to avoid that problem, where you would have to carefullycan set it so that each render uses a subset of samples from the same distribution. However Cycles does not currently support this.

Cycles uses Quasi Monte Carlo sampling:

http://en.wikipedia.org/wiki/Quasi-Monte_Carlo_method

That means it will try to evenly distribute samples so rendering converges faster. If you render multiple times with different seeds however then this even distribution is lost because it mixes different distributions without ensuring that the combined result is even too.

In principle a "sample offset" option could be implemented to avoid that problem, where you would have to carefully set it so that each render uses a subset of samples from the same distribution. However Cycles does not currently support this.

Cycles uses Quasi Monte Carlo sampling:

http://en.wikipedia.org/wiki/Quasi-Monte_Carlo_method

That means it will try to evenly distribute samples so rendering converges faster. If you render multiple times with different seeds however then this even distribution is lost because it mixes different distributions without ensuring that the combined result is even too.

Since Blender 2.78 there is a sample offset command line option to avoid that problem, where you can set it so that each render uses a subset of samples from the same distribution.

Bounty Ended with 50 reputation awarded by gandalf3
Source Link
brecht
  • 7.5k
  • 31
  • 50

Cycles uses Quasi Monte Carlo sampling:

http://en.wikipedia.org/wiki/Quasi-Monte_Carlo_method

That means it will try to evenly distribute samples so rendering converges faster. If you render multiple times with different seeds however then this even distribution is lost because it mixes different distributions without ensuring that the combined result is even too.

In principle a "sample offset" option could be implemented to avoid that problem, where you would have to carefully set it so that each render uses a subset of samples from the same distribution. However Cycles does not currently support this.