Mirror Communication (No.006 in my Caustic Series)

Our forum gallery for displaying your finished renders.
User avatar
DM3
Posts: 11
Joined: Thu Dec 17, 2020 12:45 am

Mirror Communication (No.006 in my Caustic Series)

Post by DM3 »

Some of you are familiar with this scene on the Discord, but wanted to be more active on the forum as well!
Here's a scene I built as my 6th installment of my Caustics Series, which is mostly rendered in Keyshot 10 and you can find out more about that series here if you want to see more. (I think a few of my renders from that series have already been posted by someone else on this forums as an example of spectral caustics). https://forum.keyshot.com/index.php?topic=27629.0

In this particular composition, I was trying to understand more about caustics/SDS rays/Volumetrics. One thing I wasn't able to solve for is the 'light beam' won't show up in the mirror. I was able to get this to work in Keyshot 10- but not in Luxcore, despite some help from the discord server.

I tried everything from Path engine, environment caches, photon caches, caustic cache, etc. Not a big deal, learned a lot along the way, and since I am still relatively new to Luxcore, I am very happy with the result. What are your thoughts?

Are there other scenes from my Keyshot exploration that you'd like seen done in LuxCore?

Specs;

LuxCore 2.5 (Blender Addon)
CPU - BiDir + Environment Cache + Denoiser (OIDN)
Originally rendered at 2160x2700px (HD shown here) - 2 hours - 991 Samples.
Attachments
MirrorCommunicationRev2_LuxCore_HD_DENOISED.png
User avatar
Dade
Developer
Developer
Posts: 5672
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: Mirror Communication (No.006 in my Caustic Series)

Post by Dade »

DM3 wrote: Fri May 14, 2021 5:09 pm I was able to get this to work in Keyshot 10- but not in Luxcore, despite some help from the discord server.
It is an SDS path so you need to use caustic cache. Caustic cache is supported in BiDir only from v2.6 but, in general, don't use BiDir (especially if you have a fast GPU), path+light tracing are more than enough and a lot faster.
Support LuxCoreRender project with salts and bounties
User avatar
DM3
Posts: 11
Joined: Thu Dec 17, 2020 12:45 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by DM3 »

Hey Dade - thanks for the tip -

I had heard this from the guys on the discord as well - and I actually had started with Path tracing and Caustic Cache - and still didn't get the beam to show up.

This was an earlier exploration I had done with Path + Caustic Cache. I am also fairly new to LuxCore, so there might be other elements at play here that I am just missing the ball on. (A few major changes had occurred since this initial scene) - originally I was using a world volume, and since changed it to a cube (with Photon GI cache turned on).

Luxcore on the right
Attachments
KSvsLXCR_MirrorCommunication_Rev1_DM.png
User avatar
Dade
Developer
Developer
Posts: 5672
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: Mirror Communication (No.006 in my Caustic Series)

Post by Dade »

Try to post a simple test scene so we can fix the rendering parameters. LuxCore can renders SDS path since v2.3:

Image
Support LuxCoreRender project with salts and bounties
epilectrolytics
Donor
Donor
Posts: 790
Joined: Thu Oct 04, 2018 6:06 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by epilectrolytics »

I think this is a bug with PGI volumetrics, so I opened a GitHub issue (#533) with a simple test scene.
User avatar
DM3
Posts: 11
Joined: Thu Dec 17, 2020 12:45 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by DM3 »

Tried a 'simplified' scene here - see attached.
Screenshot 2021-05-17 143603.png
Seems I got 'half' of the beam to show up here - the beam coming from the spot directly to the mirror is showing up (green arrow) - but the bounce off the mirror to the Suzanne is missing in the reflection of the mirror.

I tried to capture the render settings in the right side there, but let me know if there's something else I am missing.
User avatar
DM3
Posts: 11
Joined: Thu Dec 17, 2020 12:45 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by DM3 »

Quick 'update here' - which might support the theory that there's a bug involved.

When I removed the 'Blender Clouds' node from the scatter input on the heterogeneous volume, there is some verrrrrry faint reflection of the bounce 'beam' happening now in the mirror, it's super subtle - especially with the denoiser, and after 14 mins of letting it go, doesn't seem to be getting much brighter and doesn't come close to matching the intensity of the other 'beams'.

Again - maybe there's something I am missing in the settings here (I'm still a novice when it comes to Luxcore settings). If anyone has ideas to try let me know. Also if it's helpful to have this test file .blend - I can provide that too if anyone actually wants it
Attachments
Screenshot 2021-05-17 150826.png
User avatar
Dade
Developer
Developer
Posts: 5672
Joined: Mon Dec 04, 2017 8:36 pm
Location: Italy

Re: Mirror Communication (No.006 in my Caustic Series)

Post by Dade »

epilectrolytics wrote: Mon May 17, 2021 3:35 pm I think this is a bug with PGI volumetrics, so I opened a GitHub issue (#533) with a simple test scene.
I should have fixed the problem: https://github.com/LuxCoreRender/LuxCor ... -846407561
Support LuxCoreRender project with salts and bounties
epilectrolytics
Donor
Donor
Posts: 790
Joined: Thu Oct 04, 2018 6:06 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by epilectrolytics »

DM3 wrote: Sun May 16, 2021 3:26 pm This was an earlier exploration I had done with Path + Caustic Cache.
You might try to render your scene again with latest BlendLuxCore because Dade has fixed the issue!
User avatar
DM3
Posts: 11
Joined: Thu Dec 17, 2020 12:45 am

Re: Mirror Communication (No.006 in my Caustic Series)

Post by DM3 »

Thanks Dade and Epilectrolytics, I will give it a shot!
Post Reply