Maniphest T66642

The rendering result of hair is diffirent in the diffirent version!!!
Closed, Archived
Actions

Assigned To
Brecht Van Lommel (brecht)
Authored By
Renesmee (Renesmee)
Jul 10 2019, 4:33 AM
Tags
  • BF Blender
  • Cycles
  • Render Pipeline
Subscribers
Brecht Van Lommel (brecht)
Philipp Oeser (lichtwerk)
Renesmee (Renesmee)

Description

System Information
Operating system: Windows-10-10.0.17763 64 Bits
Graphics card: GeForce GTX 1060 6GB/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 417.01

Blender Version
Broken: version: 2.80 (sub 74), branch: master, commit date: 2019-07-10 10:48, hash: rBac8c795429f9
Worked: (optional)

Short description of error

untitled.7z54 MBDownload

[Hi,there is a question.The rendering result of hair is diffirent in the diffirent version（Blender2.79 and Blender2.80）.I really hope you can fix this bug.Because many projects of Blender2.79 need to rendered in 2.80. But now we can’t because this bug. (T_T)]

Exact steps for others to reproduce the error
[Please describe the exact steps needed to reproduce the issue]
[Based on the default startup or an attached .blend file (as simple as possible)]

Related Objects

Mentioned Here
T63534: Child hair particles set to "Interpolated" causes jittering when groom on alembic cache
rBcd023b6cecb7: Cycles: remove min bounces, modify RR to terminate less.
rBc9238e638fd5: Cycles: add back control to render first N bounces with path termination

Event Timeline

Renesmee (Renesmee) created this task.Jul 10 2019, 4:33 AM
Philipp Oeser (lichtwerk) added projects: Cycles, Render Pipeline.Jul 10 2019, 8:06 AM
Philipp Oeser (lichtwerk) added a subscriber: Philipp Oeser (lichtwerk).

I assume this is because of the more efficient Russian roulette termination from rBcd023b6cecb7.
In above commit Min Bounces was removed.
It has been brought back though rBc9238e638fd5 as Min Light Bounces / Min Transparent Bounces.

You can find these under Properties Editor > Render > Sampling > Advanced

Does setting this/these to something like Min Bounces in 2.79 give you a closer result?
(cannot render atm., but can check later, too...)

Philipp Oeser (lichtwerk) lowered the priority of this task from 90 to 50.Jul 10 2019, 10:52 AM
In T66642#716982, @Philipp Oeser (lichtwerk) wrote:

I assume this is because of the more efficient Russian roulette termination from rBcd023b6cecb7.
In above commit Min Bounces was removed.
It has been brought back though rBc9238e638fd5 as Min Light Bounces / Min Transparent Bounces.

You can find these under Properties Editor > Render > Sampling > Advanced

Does setting this/these to something like Min Bounces in 2.79 give you a closer result?
(cannot render atm., but can check later, too...)

Forget about this comment, I now have it rendered and this is not the look, it is about the particle distribution...
Will confirm on first sight, checking...

Brecht Van Lommel (brecht) added a subscriber: Brecht Van Lommel (brecht).Jul 10 2019, 11:38 AM

Note that the particle distribution random number generator was intentionally changed in 2.80.

Philipp Oeser (lichtwerk) added a comment.Jul 10 2019, 7:17 PM

Made a couple of test (without looking at code -- it might very well be the RNG @Brecht Van Lommel (brecht) was talking about): seems like only interpolated children are affected, simple children with all kinds of randomness seem the same in 2.79 and 2.8.

Philipp Oeser (lichtwerk) added a comment.Jul 10 2019, 7:22 PM

[which makes T63534 come to my mind -- might be unrelated though]

Renesmee (Renesmee) added a comment.Jul 12 2019, 10:10 AM

Note that the particle distribution random number generator was intentionally changed in 2.80.

Is there any solution? This problem has caused us a lot of trouble.

Brecht Van Lommel (brecht) changed the task status from Unknown Status to Unknown Status.Jul 14 2019, 7:11 PM
Brecht Van Lommel (brecht) claimed this task.

No, there is no way to get the exact same result in 2.79 and 2.80 with child particles unfortunately. Further I can't see a bug here.