DLSC test

Use this forum for general user support and related questions.
Forum rules
Please upload a testscene that allows developers to reproduce the problem, and attach some images.
User avatar
FarbigeWelt
Donor
Donor
Posts: 1046
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: DLSC test

Post by FarbigeWelt »

Interesting note from second or your link, about indirect light though.
„IMPORTANT
Icon
When an emitter is excluded and does no affect a certain object, what is excluded is the direct contribution from that emitter, but its indirect illumination as it bounces off other surfaces will still affect the object, as once bounced it cannot be attributed to any particular source.“

Well, to be honest, I wished severeal times mureley, material should have the option to increase number of bounces, or change their count depending on counts (of course there should be a safety feauture to kill beams or certain types, or set all to default after a time limit)).
I can imagine these additional material features is easy to use, quite helpful for complex scenes and feasible with reasonable effort. Its not really a tweak its just changing probability of beams next step. 😎
Light and Word designing Creator - www.farbigewelt.ch - aka quantenkristall || #luxcorerender
MacBook Air with M1
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: DLSC test

Post by lacilaci »

That second part I am actually not big fan of. Inclusion/Exclusion should be a light property not object.. well in theory this could go both ways? Bounces or samples amount should always be handled by the renderer though/globally not per object.
But DLCS should work hand in hand with gi visibility of object, and this goes for future secondary gi caching too.
In theory it is very simple, but of course I have no idea how hard these things are to implement. One way or another, the more luxcore will be popular the more requests for this there will be.
User avatar
FarbigeWelt
Donor
Donor
Posts: 1046
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: DLSC test

Post by FarbigeWelt »

In my opinion light interacting is a matter of objects and therefore related to material.
Surely you may add some attributes to ray like number of bounces since emiitter and a stack with references to last object/material and a stack for last vectors. Vectors and object references may be used for denoising, removing fireflies.
Light and Word designing Creator - www.farbigewelt.ch - aka quantenkristall || #luxcorerender
MacBook Air with M1
User avatar
Sharlybg
Donor
Donor
Posts: 3101
Joined: Mon Dec 04, 2017 10:11 pm
Location: Ivory Coast

Re: DLSC test

Post by Sharlybg »

Struggle to render this with DLSC but metropolis + power give good result. Saddly DLCS with cache entry higher than 1cm ( at 5cm DLSC computing start to e to long so imagine 1cm ) lead to black artifact on the chair.

metropolis + power
The attachment nj.png is no longer available
metropolis + DLSC
nj1.jpg
Support LuxCoreRender project with salts and bounties

Portfolio : https://www.behance.net/DRAVIA
User avatar
Dade
Developer
Developer
Posts: 5672
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: DLSC test

Post by Dade »

Sharlybg wrote: Thu Oct 18, 2018 4:47 pm Struggle to render this with DLSC but metropolis + power give good result. Saddly DLCS with cache entry higher than 1cm ( at 5cm DLSC computing start to e to long so imagine 1cm ) lead to black artifact on the chair.
As I wrote in another thread, the problem is likely to be the warm up samples count for cache entry, not the cache entry radius. Try to increase the warm up samples count (you can probably even increase the radius without problems).
Support LuxCoreRender project with salts and bounties
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: DLSC test

Post by lacilaci »

Sharlybg wrote: Thu Oct 18, 2018 4:47 pm Struggle to render this with DLSC but metropolis + power give good result. Saddly DLCS with cache entry higher than 1cm ( at 5cm DLSC computing start to e to long so imagine 1cm ) lead to black artifact on the chair.

metropolis + power

nj.png

metropolis + DLSC

nj1.jpg
Those artifacts look worse than shadow terminator...
User avatar
Sharlybg
Donor
Donor
Posts: 3101
Joined: Mon Dec 04, 2017 10:11 pm
Location: Ivory Coast

Re: DLSC test

Post by Sharlybg »

Try to increase the warm up samples count (you can probably even increase the radius without problems).
I think we need tooltips for DLCS setting ?
Here are my test and it seem that i've to decrease samples count to have good result. high samples count + small Entry radius lead to white dot on chair (look like extremelly converged area) . don't really know impact on DLCS performance :roll:


DLCS 10cm t 1000.jpg
DLCS 1cm t 16.jpg
DLCS 1cm t 1000.jpg
DLCS 1cm t 1000000.jpg


Here is my light bulb setup and it work pretty well ;)

bubl setup.jpg
Support LuxCoreRender project with salts and bounties

Portfolio : https://www.behance.net/DRAVIA
User avatar
Dade
Developer
Developer
Posts: 5672
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: DLSC test

Post by Dade »

It looks like BlendLuxCore is not exposing the warmup property I was talking about. You need that one.
Support LuxCoreRender project with salts and bounties
User avatar
FarbigeWelt
Donor
Donor
Posts: 1046
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: DLSC test

Post by FarbigeWelt »

what is your light count for the scene in info tool?

Nice idea to use a 2D approach for a still. :idea:
Light and Word designing Creator - www.farbigewelt.ch - aka quantenkristall || #luxcorerender
MacBook Air with M1
User avatar
B.Y.O.B.
Developer
Developer
Posts: 4146
Joined: Mon Dec 04, 2017 10:08 pm
Location: Germany
Contact:

Re: DLSC test

Post by B.Y.O.B. »

Dade wrote: Fri Oct 19, 2018 10:56 am It looks like BlendLuxCore is not exposing the warmup property I was talking about. You need that one.
I added support for it: https://github.com/LuxCoreRender/BlendL ... 190fa44cd1
Post Reply