Path Space Regularization (aka the solution to SDS paths)

Discussion related to the Engine functionality, implementations and API.
User avatar
Dade
Developer
Posts: 3402
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: Path Space Regularization (aka the solution to SDS paths)

Post by Dade » Fri Sep 20, 2019 2:21 pm

Vutshi wrote:
Fri Sep 20, 2019 1:25 pm
BTW, I tried the smallpt code from the Supplemental Material and it is broken :). After fixing it seems to work but produces a bit more fireflies as compared to the figure in the SM. It does not darken as far as I can see.
And have you seen how many samples it takes just to have a vague caustic and few fireflies ?
Support LuxCoreRender project with salts and bounties

Vutshi
Posts: 15
Joined: Tue Jul 17, 2018 8:24 pm

Re: Path Space Regularization (aka the solution to SDS paths)

Post by Vutshi » Fri Sep 20, 2019 2:31 pm

Dade wrote:
Fri Sep 20, 2019 2:21 pm
And have you seen how many samples it takes just to have a vague caustic and few fireflies ?
These are 128 and 2048 samples per pixel, respectively.
Attachments
image_new_128v2.png
image_new_2048v2.png

provisory
Posts: 224
Joined: Wed Aug 01, 2018 4:26 pm

Re: Path Space Regularization (aka the solution to SDS paths)

Post by provisory » Fri Sep 20, 2019 2:41 pm

Is this metropolis? Doesn't look like.

provisory
Posts: 224
Joined: Wed Aug 01, 2018 4:26 pm

Re: Path Space Regularization (aka the solution to SDS paths)

Post by provisory » Fri Sep 20, 2019 2:47 pm

And in the SM there is a 65536 samples per pixel image of this scene.

Vutshi
Posts: 15
Joined: Tue Jul 17, 2018 8:24 pm

Re: Path Space Regularization (aka the solution to SDS paths)

Post by Vutshi » Fri Sep 20, 2019 2:50 pm

provisory wrote:
Fri Sep 20, 2019 2:41 pm
Is this metropolis? Doesn't look like.
No it is regularised path tracing inspired by the code in the paper.

My initial plan was first to understand smallpt with Path Space Regularization and then apply it to smallmmlt. But smallmmlt is above my poor understanding of all this stuff. They don't even have glass material there :)

EDIT:
Here is 2^17 samples per pixel. Looks noisy to me I guess I messed up something :)
Attachments
image_new_small 217v2.png
image_new_small 217v2.png (16.84 KiB) Viewed 592 times

User avatar
Dade
Developer
Posts: 3402
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: Path Space Regularization (aka the solution to SDS paths)

Post by Dade » Fri Sep 20, 2019 3:11 pm

Sharlybg wrote:
Fri Sep 20, 2019 1:42 pm
lacilaci wrote:
Fri Sep 20, 2019 12:27 pm
starting to look good. (except the overall darkness underwater)
what about sun+sky or HDRi
This is sun+sky (with the same exposure so it is a bit over-exposed due the additional light source):

psr-sky.jpg

Note: the water volume of the scene has (a bluish) absorption (so it does intentionally take away some light).
Support LuxCoreRender project with salts and bounties

User avatar
Sharlybg
Donor
Posts: 1574
Joined: Mon Dec 04, 2017 10:11 pm
Location: Ivory Coast

Re: Path Space Regularization (aka the solution to SDS paths)

Post by Sharlybg » Fri Sep 20, 2019 4:04 pm

Now it is clear that with sun+sky this water should be a loooot more clear than that. There is surely a terrible wrong thing going on here :shock:
Support LuxCoreRender project with salts and bounties

Portfolio : https://www.behance.net/DRAVIA

User avatar
B.Y.O.B.
Developer
Posts: 3017
Joined: Mon Dec 04, 2017 10:08 pm
Location: Germany
Contact:

Re: Path Space Regularization (aka the solution to SDS paths)

Post by B.Y.O.B. » Fri Sep 20, 2019 4:06 pm

It would be interesting to see this scene without absorption.
Support LuxCoreRender project with salts and bounties

User avatar
epilectrolytics
Donor
Posts: 605
Joined: Thu Oct 04, 2018 6:06 am

Re: Path Space Regularization (aka the solution to SDS paths)

Post by epilectrolytics » Fri Sep 20, 2019 4:16 pm

If this scene is derived from my test pool here it probably contains a volume setup error because I used only a water (ocean modifier) plane and no solid block.
The underwater pool tile object then has not the absorption volume properly assigned as exterior.
But with just killing all volumes it should render properly (I guess).
MBPro 15" 16GB i7-4850HQ GT750M, MacOS 10.13.6 & Win10Pro PC 16GB Ryzen 2700X, 2 x RTX 2070

User avatar
B.Y.O.B.
Developer
Posts: 3017
Joined: Mon Dec 04, 2017 10:08 pm
Location: Germany
Contact:

Re: Path Space Regularization (aka the solution to SDS paths)

Post by B.Y.O.B. » Fri Sep 20, 2019 4:28 pm

epilectrolytics wrote:
Fri Sep 20, 2019 4:16 pm
If this scene is derived from my test pool here it probably contains a volume setup error because I used only a water (ocean modifier) plane and no solid block.
The underwater pool tile object then has not the absorption volume properly assigned as exterior.
I think this is not a problem, because rays keep their volume when the material they hit doesn't have one assigned (if I remember correctly).
Since all rays pass through the water surface on entry and exit, it should work. It would be a problem though if you were to dip the camera below the water surface without assigning the water volume as exterior camera volume.
Support LuxCoreRender project with salts and bounties

Post Reply