V-Ray 2.0: VRayBPTracer

[vlado]

Some images with the bidirectional path tracing in V-Ray 2.0. Obiously more work is required to make it useable (noise levels are far from optimal) but some of these images are very hard to do with the traditional V-Ray GI methods. Especially the first two - the ashtray one and the interior that is lit predominantly with reflective caustics.

Best regards,
Vlado

[sandykoufax]

Is this method work similar to PPT ?

[vlado]

Is this method work similar to PPT ?

Yep, sort of, but it traces rays from the lights as well (not just the camera), so things like caustics and multiple light bounces clear up much faster. It is mostly better for interior scenes (not much point to use it for exteriors). It is also purely an unbiased method (doesn't use the light cache like PPT, at least not yet).

Best regards,
Vlado

[sandykoufax]

Thanks. looks very promising.

[Dariusz Makowski (Dadal)]

Humh that VRayBPTracer is on back of my head for a long time now...

I was wondering even tho I could spend 2 days and test it myself but anyway, how do you think Vlado this BT would work with heavy caustic images? I mean car Headlights 10-20k+ renders 20 lights casting caustics and loads of reflection.

Will it look better than standard or better? &time too...

[vlado]

I was wondering even tho I could spend 2 days and test it myself but anyway, how do you think Vlado this BT would work with heavy caustic images? I mean car Headlights 10-20k+ renders 20 lights casting caustics and loads of reflection.

Will it look better than standard or better? &time too...

Well, it won't be fast for certain For such large resolutions, photon-mapped caustics might be more efficient. I do not know exactly though... would be good to test it in reality. If you do get to the testing, make sure you get the 2.00.02 build from our site as it has some fixes in that regard.

Best regards,
Vlado

[Dariusz Makowski (Dadal)]

Will give it ago, is the 2.00.02 up already ?

The photon mapped-caustics well they are pretty...bad atm. I'm going up to 20k samples for each light and so on. Entire process is a nightmare atm to be honest. I kinda pray to u guys that at some point u'll do some work on caustic work-flow, maybe grid alike(phopnix fd) or something like it...

[vlado]

Will give it ago, is the 2.00.02 up already ?

Yep, it is on the site.

The photon mapped-caustics well they are pretty...bad atm. I'm going up to 20k samples for each light and so on. Entire process is a nightmare atm to be honest. I kinda pray to u guys that at some point u'll do some work on caustic work-flow, maybe grid alike(phopnix fd) or something like it...

I see. If you can get us some examples of the kind of work you need to do, it will be helpful in trying to find ways to optimize it

Best regards,
Vlado

[Dariusz Makowski (Dadal)]

Sure np vlado@chaosgrouop.com ?

I'll be doing a close up of BMW rear lights soon before/after Christmas I'll share with you the pain ! ^^

[vlado]

Sure np vlado@chaosgrouop.com?

It is vlado@chaosgroup.com

I'll be doing a close up of BMW rear lights soon before/after Christmas I'll share with you the pain ! ^^

Ok then

Best regards,
Vlado

[Paul Oblomov]

Would be great to have ability to reuse the data from bdpt pass

[PIXELBOX_SRO]

perhaps a stupid question, but whats the BPT good for at the moment (judging the quality)?
should i take it as more or less experimental feature so far?

[duke2]

The first time I heard about was when ChaosGroup pimped it at some conference as being something that was good at faster caustics, so that's the benefit. Other than that I don't know.

[vlado]

perhaps a stupid question, but whats the BPT good for at the moment (judging the quality)?
should i take it as more or less experimental feature so far?

Partly, yes. For some cases, esp. when rendering jewelery and stuff like that it may be able to produce images faster and with less tweaking than other methods.

Best regards,
Vlado

[Morne]

ok so how do I enable this "BPTracer" ?