Thanks you for you input. I remenber now this behaviour from Luxcore. This is strange because you can't also set 0 input inside blendluxcore. it will always return 1 as minimum value so you end up with 2 intead of 1.Luxcore is always using whatever you input +1 bounce, that's how exporter is set up, Simon pls. Correct me if wrong.
Luxcore Vs Octane Preparation test
Forum rules
Please upload a testscene that allows developers to reproduce the problem, and attach some images.
Please upload a testscene that allows developers to reproduce the problem, and attach some images.
Re: Luxcore Vs Octane Preparation test
Re: Luxcore Vs Octane Preparation test
Yes, but that's only relevant if you compare speed at 1 bounce across renderers.
The output with 1 and 2 bounces is the same in LuxCore, just that the 2-bounce output uses MIS so you get less noise. This is exactly why this exporter behaviour is in place: to hide the MIS-specific implementation detail of the engine and make the bounces behave like the user would expect.
Re: Luxcore Vs Octane Preparation test
With further test i've the feeling that by using total path deph settings it is possible to bypass the n+1 behaviour.
Octane deph 1
Luxcore deph 1
Luxcore deph 2
Octane deph 2
As you can see Luxcore Deph 2 is different from deph 1. And only Octane deph look similar to luxcore deph 2 (same for deph 1 in both engine )
Octane deph 1
Luxcore deph 1
Luxcore deph 2
Octane deph 2
As you can see Luxcore Deph 2 is different from deph 1. And only Octane deph look similar to luxcore deph 2 (same for deph 1 in both engine )
Re: Luxcore Vs Octane Preparation test
are those same time renderings?Sharlybg wrote: ↑Fri Jan 31, 2020 10:08 am With further test i've the feeling that by using total path deph settings it is possible to bypass the n+1 behaviour.
Octane deph 1
octane 1.jpg
Luxcore deph 1
luxcore 1.jpg
Luxcore deph 2
luxcore 2.jpg
Octane deph 2
octane 2.jpg
As you can see Luxcore Deph 2 is different from deph 1. And only Octane deph look similar to luxcore deph 2 (same for deph 1 in both engine )
Re: Luxcore Vs Octane Preparation test
Not the same time. not also same quality. as you can see octane is clamping light very hard. at the same time you get this :are those same time renderings?
Octane 11sc Deph 1
Luxcore 11sc Deph 1
Re: Luxcore Vs Octane Preparation test
Interesting. Luxcore always introduces noise even in directly exposed areas, this is beyond me why it happens. How do we get noise even in the outside areas...
I think if luxcore could clamp indirect light separately from direct ligh, we might be able to get closer result... maybe...
Re: Luxcore Vs Octane Preparation test
Yes, because it actually is depth 3 and 2. They are supposed to be different.
The relevant part of the code is here: https://github.com/LuxCoreRender/BlendL ... ig.py#L276
There's also a comment explaining the reasoning.
Re: Luxcore Vs Octane Preparation test
Sure i think the same. And also i remenber Fstorm behave a bit like that with strong Clamping.I think if luxcore could clamp indirect light separately from direct ligh, we might be able to get closer result... maybe...
Re: Luxcore Vs Octane Preparation test
Ok so i've removed all the incremental value. Story is the same even if luxcore get now a bit darker Octane is even darker and clamped.Yes, because it actually is depth 3 and 2. They are supposed to be different.
The relevant part of the code is here: https://github.com/LuxCoreRender/BlendL ... ig.py#L276
There's also a comment explaining the reasoning.
Octane deph 1
Luxcore deph 1
Re: Luxcore Vs Octane Preparation test
it seems a huge difference...the main is obviously the noise.
Yes, also the lighting is different..but if you make a poll and ask "choose from noise and lighting what is the most different from lux and octane", I'm quite sure that 95% reply "noise".
I'm not sure it a just a matter of clamping...if you want to test this...why don't try to find the clamping value to have that kind of noise free on the surface that receive light?
In my opinion, in Lux it seems that we have always difficulties to clean not just the shadowed surfaces but the lighted too.
Yes, also the lighting is different..but if you make a poll and ask "choose from noise and lighting what is the most different from lux and octane", I'm quite sure that 95% reply "noise".
I'm not sure it a just a matter of clamping...if you want to test this...why don't try to find the clamping value to have that kind of noise free on the surface that receive light?
In my opinion, in Lux it seems that we have always difficulties to clean not just the shadowed surfaces but the lighted too.