Originally posted by vlado
View Post
Announcement
Collapse
No announcement yet.
Fstorm render
Collapse
X
-
-
Originally posted by Donfarese View PostAlso I'm fairy new to Redshift so I'm still learning how to set it up, I just found a better way to set things up so below are new renders from Vray and Redshift
Best regards,
Vlado
Leave a comment:
-
Originally posted by Ivan1982 View PostCan you save vray test scene for max 2015-2016?
Leave a comment:
-
Originally posted by Moriah View Post@Lele, i agree with what you're saying, but on the other hand, from an artist's perspective, data doesn't matter THAT much, what matters is the final visual representation, and if you can get something "similar" (doesn't matter if it has GI clamping, or 432432 bounces, or whatever..) within a shorter render time and having flexibility regarding settings and quality, that's what matters the most sometimes (at least for me since i'm a user with access to limited hardware power)...
You're saying that RS has heavy GI clamping, and Vray has not. So ideally, we could have the ability to clamp the GI in Vray GPU, so we can get a faster render time even if we have to sacrifice quality... Same in RS, don't quite know if they have that option, but if they don't they should give us the option to not clamp even if it takes longer but in the end looks better.
When you're using IES lights and trying to sell an accurate lighting distribution in your scene, you're (by default) screwed, or if the intensity of something has to be above 1k float (lights, probes, you name it), you're screwed, and so is your lighting distribution.
If you need reflective GI caustics well resolved and contributing to your lighting, all you'll get will be fireflies, unless you clamp hard and clamp early.
When you want to comp, you're by default outside of linearity within the radiometric equation itself because of all the clamping, so you guessed it, you're screwed as far as LWF comping is concerned.
It's not faster, it's ridiculously more messy to set up and get results from, it's severely more limited in features, it's broken in a number of places, but you like it.
What can i say, go ahead.
peace out. ^^
Leave a comment:
-
Originally posted by Donfarese View PostI'm trying to figure out how to get Redshift to use 100%.
@Lele, i agree with what you're saying, but on the other hand, from an artist's perspective, data doesn't matter THAT much, what matters is the final visual representation, and if you can get something "similar" (doesn't matter if it has GI clamping, or 432432 bounces, or whatever..) within a shorter render time and having flexibility regarding settings and quality, that's what matters the most sometimes (at least for me since i'm a user with access to limited hardware power)...
You're saying that RS has heavy GI clamping, and Vray has not. So ideally, we could have the ability to clamp the GI in Vray GPU, so we can get a faster render time even if we have to sacrifice quality... Same in RS, don't quite know if they have that option, but if they don't they should give us the option to not clamp even if it takes longer but in the end looks better.Last edited by Moriah; 20-07-2016, 02:48 AM.
Leave a comment:
-
Well, the vray scene loads with redshift here.
However, the above are NOT valid comparisons, and you can tell visually.
RS is clamped to 5 bounces of GI, and to a 2.0f clamp of subsamples and secondaries.
If you used the LC for V-Ray , you're asking it to calculate 100 bounces of GI, and if you left it at defaults, you only clamp secondaries at 20f (ten times higher).
Add speculars, of any of the BRDFs they implement, without clamping, and you will see the reasons why they default to such an aggressive clamping: their actual scene sampling sucks, and that's the ugly patch all over it (perhaps V-Ray 10 years ago. Maybe. But no, never quite to this degree, not even at the times of switching from qmc do dmc.)
I wanted to write this yesterday, but refrained: please, ensure you are doing it right before you post something for others to look at.
It's not a contest of offended pride to be healed (i know to many of you i must represent the definition of fanboy, but it's not my job description, believe it or not.), it's an analysis made on strict scientific bases, or nothing at all worth wasting time on.
In fStorm, i cannot get ANYTHING to work unless i clamp it as it comes by default (it literally blows up in my face), so i don't know what all the ravings are about.
What's unfathomable is personal bias, what i follow is data.
Leave a comment:
-
Originally posted by vlado View PostCan you also post the hardware used for these renders?
Best regards,
Vlado
Win 10
3dsmax 2017
3930k
32GB Ram
Titan X only card
Vray RT is using 100% of the card, I'm trying to figure out how to get Redshift to use 100%. Gonna have to ask them in there forums. Also I'm fairy new to Redshift so I'm still learning how to set it up, I just found a better way to set things up so below are new renders from Vray and Redshift
Vray 3min 59sec
Redshift 3min 20sec
Leave a comment:
-
Did a new test that you guys can download if you want and try yourselves. Just for the record, there is no speed problem with VrayRT, it's really fast and I actually like the way it's set up better with just the "Max Noise" setting. I'm hoping Redshift adapts a more "Max Noise" approach. Redshift is much more complicated to set up, think of old vray. But it's complication is good in certain situations for large or complicated animations. The thing that Redshift does so well is it's super stable, and has tons of features and support for max shaders...etc. I couldn't find one area of working with it that I had to say, oh damn I can't do that.
Below is the tests, You'll see that Redshift actually took 3 seconds longer, but is much cleaner even though it's hard to see in the compressed versions on the forum.
Vray BF + LC 4min 9sec
Redshift BF + IPC 4min 12sec
Redshift IC +IPC 54sec
Since it is really hard to tell on the forum (Makes the Redshift render look a lot more noisy than it is) I included a zip with the actual renders, called "Render Scenes.zip"Last edited by Donfarese; 19-07-2016, 09:41 PM.
Leave a comment:
-
Hmm... That seems i unbelievably long for RS render.Mind sharing scene.
Just got home an looked at those images, the RS ones don't seem like the GI was set up right knowing how it acts with correct settings. Due to the incredible noise in certain areas.Last edited by Donfarese; 19-07-2016, 06:15 PM.
Leave a comment:
-
-
Originally posted by Donfarese View PostDon't get me wrong I loved RT but in the past two projects it couldn't handle the jobs so we switched. I'm not one of these people that switch and go back and trash there previous render, I'm just giving my experience with both.
Best regards,
Vlado
Leave a comment:
Leave a comment: