Displacement issue ? - corona1.3 dailyBuild Aug 31 2015

[Alexandre Besson]

Hi!

The calculation time of displacement is so long (over 60 min, i don't know, i stopped 3ds before)
Same thing with 15 px in screen size.  (CPU is running at 50 % in displacement pass).

Regards.

[johan belmans]

Same overhere (Build 17/08/2015)

[Ondra]

displacement passes our automated tests... can you post a problematic scene?

[johan belmans]

Hi Ondra,

I have more issues with displacement, for example very high poly scene + Hig res renders 5K (PT + UHD cache) +  displacement = screws up my GI. This is not the case with PT + PT
Some images even won't render, they do with displacement settings from 4.0 to upwards. But I need lower values.
RAM = 32 GB.

I am planning to write a Mantis report, but unfortunattely I do not have the time right now.

[Alexandre Besson]

http://we.tl/umPaltpj1P

[Ondra]

note to self: scene name: DISPLACEMENT ISSUE.zip

[Nekrobul]

Most probably the problem is with tesalation.

1 - too dense wireframe.

2 - too many points on a single poly (mesh\plane created from editable spline with bend edges)

3 - too sharp displacement map ( it should always be blurred 2-3px gausian blurr in photoshop)

4 - ???

[hubrobin]

Hi,

there is a lot of non-manifold geometry in your scene, which is messing up the displacement algorithm. It would actually never finish on one of the objects. Watch out for the next daily build as it should pass through in a reasonable amount of time.

Cheers,

Robin

[hubrobin]

I didn't realize that today's daily has already been released. That means the fix will be in the tomorrow's one.

Sorry,

Robin

[Alexandre Besson]

No problem now (DB 06.09.2015) you are fantastic!

Thanks.