Page 2 of 6

Re: nakaging capsule

Posted: Sun Jan 28, 2018 3:55 pm
by kintuX
Dade wrote:
Fri Jan 26, 2018 10:09 pm
Dade wrote:
Wed Jan 24, 2018 9:15 am
It is in the know bugs of the release notes: "PATHOCL+METROPOLIS output difference with PATHOCL+RANDOM/SOBOL" (https://wiki.luxcorerender.org/LuxCoreR ... Notes_v2.0)

I guess it is time to look again in this problem.
All the day to fix this damn bloody bug ! I have the proof, some bug is sentient, only an intelligent form could hide so well.

Anyway, fixed !
This is great. LuxCore is coming to life... :D

Re: nakaging capsule

Posted: Sat Feb 03, 2018 9:31 am
by Sharlybg
Just a 1.5 k test of interior:

GPU ==> HD 7970
RENDERTIME ==> 8h
POLYCOUNT ==> 2192k
RENDERMOD ==> Path OCL Sobol
EDIT ==> none
EFFECTS ==> none
RENDER SET ==> difduse/glossy = 8 ... specular/total = 16
RESOLUTION ==> 1337 X 1540

NK 15k.jpg

Re: nakaging capsule

Posted: Sat Feb 03, 2018 9:50 am
by Dade
Use variance clamping :idea:

Re: nakaging capsule

Posted: Sat Feb 03, 2018 10:32 am
by Sharlybg
Dade wrote:
Sat Feb 03, 2018 9:50 am
Use variance clamping :idea:
Of course ;) . but this one is for defferent purpose want to judge the raw speed and directly compare with the upcoming apha 3.0 without clamping.

Re: nakaging capsule

Posted: Sat Feb 03, 2018 8:03 pm
by david57
Looks good.

Re: nakaging capsule

Posted: Sat Feb 03, 2018 9:26 pm
by Sharlybg
david57 wrote:
Sat Feb 03, 2018 8:03 pm
Looks good.
thanks. I also hope the final output will be better than that :roll:

New test for the BIDIR engine in alpha 2( i think it is the same in alpha 3).

___first of all big 5 ***** to you DADE for the work on the glass 2 shader. it is unbelievable the way glass get fast clean now.would never believe that glass shader could be cleaner after few minute than diffuse or glossy.good job.

GPU ==> i7 6700k
RENDERTIME ==> 7h
POLYCOUNT ==> 2192k
RENDERMOD ==> BIDIR
EDIT ==> none
EFFECTS ==> none
RENDER SET ==> eye/light deph ==>> 12
RESOLUTION ==> 859 X 990
LIGHTING ===>> 1 HDRi nothing more


Wip bidir 7hr alpha 2.png
Strange thing the ceilling and the wall share the same shader but ceilling look like hyper white diffuse will check.

Re: nakaging capsule

Posted: Sun Feb 04, 2018 1:22 pm
by Dade
Sharlybg wrote:
Sat Feb 03, 2018 9:26 pm
Strange thing the ceilling and the wall share the same shader but ceilling look like hyper white diffuse will check.
The ceiling looks quite strange, are you sure it uses the same wall material definition ? If it does, can you upload the light+walls+ceil scene so I can check what is going on ?

Re: nakaging capsule

Posted: Sun Feb 04, 2018 1:26 pm
by B.Y.O.B.
Sharlybg wrote:
Sat Feb 03, 2018 9:26 pm
Strange thing the ceilling and the wall share the same shader but ceilling look like hyper white diffuse will check.
Are you using a glossy shader for the walls?
Maybe the normal of the ceiling is wrong (pointing away from the camera).

Glossy materials are rendered as matte on the backface (I don't really understand the reason for this and it has bitten me a few times in the past).

Re: nakaging capsule

Posted: Sun Feb 04, 2018 1:38 pm
by Dade
B.Y.O.B. wrote:
Sun Feb 04, 2018 1:26 pm
Glossy materials are rendered as matte on the backface (I don't really understand the reason for this and it has bitten me a few times in the past).
True, it is something I have always found a bit strange.

Re: nakaging capsule

Posted: Sun Feb 04, 2018 1:44 pm
by B.Y.O.B.
Can we change it? :D