Messages

1

[Max] General Discussion / Re: Time to ditch sRGB/Linear as default (?)

« on: 2017-02-24, 16:32:34 »

I tried Adanmq's LUTs on few of my scenes and i have mixed feelings about it. The results i got, is pretty unreliable - in some scenes these LUTs works good, in others, not so much. Besides, it seems that once i use Adan's LUTs, Corona's tonemapping controls becomes useless. I rarely feel the need to tweak my renders further outside of VFB, after i use Corona's tonemapping, but with these LUTs i have to send every picture to photoshop to tweak something. Maybe it's just my inexperience. I'll try to play with it more and see if i manage to get more consistent results.

Curious what others think about Adan's LUTs?

HI.
The LUTs that comes whit corona are just a color response and they are not standardized for any specific workflow so they are designed to be used in an "artistically" way and usually you need to do exposure, compresión and contrast separately to achieve final results.

LUTs are not a good way to grade a linear image to be honest, I personally only use contrast ones + highlight corona compresión in order to be able to better see the linear render on a sRGB monitor so i can work on lighting/materials whit something similar to the final post in mind but i usually do my post by hand. LUTs are great to color management between apps and platforms. LUTs needs some short of compresión before you apply them or you will kill your render and clamp the results.

The ones y post here are just a conversion from the original Filmic Blender thing and they only give you a contrast reference but you need to use them in LOG and help you pre-visualice your linear image on a sRGB monitor so you can better plan your materials/light.

I´m working on a new pack designed to be uses over compressed linear renders because even if LUTs are not designed for that purpose a lot of people are using it for it and it´s better if you have proper designed ones xD

Best
A

2

[Max] General Discussion / Re: Time to ditch sRGB/Linear as default (?)

« on: 2017-02-22, 23:51:28 »

Sure here you have the whole pack xD Just use Log and you will get the exact same result as in the famous video xD

Usually you need to increase the exposure in order to get proper midtones.

Best.
A

3

[Max] General Discussion / Re: Time to ditch sRGB/Linear as default (?)

« on: 2017-02-22, 22:56:00 »

This might be a silly question, but are you able to compile the above processes into one singular LUT file that we can use in the Corona VFB so we can test this out? Would make life a lot easier than having to fire up Nuke. Is this even possible using LUT's in Corona? I'm still kind of noobish with LUT's and their limitations.

Cheers

HI

You can´t. LUTs are only able to map 0-1 values that´s why you need to apply a "shaper" LUT (lin2log) before in order to re-map the linear values into a logarithmic 0 to 1 format and then apply the contrast LUT to this 0 to 1 logarithmic file. The OCIO "magic trick" apply this 2 LUTs in a row.

But you don´t need to do this in Corona, you already have amaizing tone mapping inside Corona, you can "skip" this lin2log because you can compress dynamic range using highlight compression for example, try this LUT but you need to apply heavy highlight compression in order to use it. But i recomend to undertand whats going on being before you start using this kind of stuff xD

Or you can tick "input LUT in Log space" and use the LUT from the OCIO package.

I edit to add the "Heavy contrast" LUT from the original Filmic Blender OCIO transformed into a Cube LUT so you can use (This needs to be used in Log space)


Best.
A

4

[Max] General Discussion / Re: Time to ditch sRGB/Linear as default (?)

« on: 2017-02-22, 21:39:16 »

Hi.

In my personal opinion this is a complex subject that can help Corona conquer an specific sector of the market (people that need/want easy/good results out of the box) but will leave Corona outside other markets like VFX (yes, most VFX studios use LWF). Looks to me that this is already your strategy and it´s working really good so far but i´m also sure you know about the "side" effects.

ArchViz studios probably will love this change because right now we are making this (not everyone) using compresión + custom LUTs (it´s not the best way to do it in my opinion) and can be confusing for a lot of people new to the industry. But i personally think it´s a mistake if you can´t choose, i prefer linear renders because they are important not only for multi pass compositing. You need linear data for comp in order to do a lot of things property like DOF, MB, Glints Glares, etc. and we usually use OCIO or LUTs just for pre visualice the end result but still working in linear during the entire pipeline. Comp operations still linear. If the user can choose i think it´s a good move xD

But i can understand why this can help Corona to conquer the ArchViz market since most ArchViz artist don´t really understand whats LWF or how to use tone mapping properly.

¿And what do you want to use instead linear? because i suppose Corona will keep making light calculations in a linear way so we are talking about some short of internal post-production i can´t control? Something like Shaper Log LUT + Filmic Contrast LUT?  will we have several contrast or camera response choices?

About the video from BG i think it´s not really accurate and can make some people misunderstand the problem/solution. Because most of the people are shocking about the "magic" button instead of really understanding what´s going on.

Of course this is only my personal opinion.


PD: if you want to replicate this "magic button" from the video just download the OCIO and use it to color management a Nuke viewer or if you really want to break your 32bpc file and transform it into a 16bpc tonemapped one transform your render into Log using nuke lin2log and apply the LUT from the OCIO pack you can also download OCIO profiles from Sony from the web site. This is nothing new to the industry as Rawalanche pointed, this is old stuff xD

Some people ask me for this on my channel and i make this examples to explain the process (they are in Spanish sorry):

https://www.flickr.com/photos/adanmq/32187170644/in/photostream/lightbox/
https://www.flickr.com/photos/adanmq/33032436735/in/photostream/lightbox/

Best.
A

5

[Max] Daily Builds / Re: Daily Builds 1.6

« on: 2017-01-05, 12:55:00 »

HI. i´m trying the 1.6 changes and everything looks great but I´m having a problem using MAX 2017 and Dec 13 version, no matter what, if i re-size the Interactive VFB always automatically goes back to 505x505px. Yes, i can zoom but i cant set the original size anymore. Tools panel it´s closed.

Anyone it´s having similar behavior? or just my?

Best!.

6

General CG Discussion / Re: Which lens to create hdri?

« on: 2016-10-03, 17:57:58 »

jejejeje. To be honest i think even 8K HDR it´s too big. But in archviz people want to have on click solution (light+background) and i start trying to make the "perfect" HDR just to better understand the difficulties. Finally i'm making 4K-8K ones and focusing on quality/precision (there is a lot involved on making really good HDR xD)

Best.

7

General CG Discussion / Re: Which lens to create hdri?

« on: 2016-10-03, 15:13:59 »

Hi.
jejejeje. Nope i´m not xD i just extrapolates the results i get using 50MP (8688x5792) 5DS, 12-24mm and 15ND to try get unclamped high resolution HDR and it was almost impossible to get good HDR (perfectly calibrated) because can take me 30min or more shooting the images i need and if you try the Unity "method" and want to mix ND shoots with non-ND ones making a single HDR can take even longer and the amount of problems/variables this workflow introduces makes it really un-consistent. You can make 32K HDR using 21-24mm lens and this camera but you need 32 different camera positions (24 can work also) (20%overlap) to make it and this can take a looot of time using NDs so it will fail sometimes. SO this is why most of the HDR you can buy are far from perfect usually they aren't even made from proper linear images.


I try every method/technique i can with my gear and after lots of testing i think you need to decide if you prefer unclamped HDR and sacrifice resolution or high Res HDR and sacrifice precision trying to get both makes me almost crazy xD. But as i say maybe i'm missing something. I will appreciate any suggestion ;)

Best

8

General CG Discussion / Re: Which lens to create hdri?

« on: 2016-10-03, 12:29:22 »

Hi.

ummmm, depending on a lot of factors you will be able to make the  HDR you need using fisheye or not. You first need to know what kind of HDR do you need: Really precise unclamped HDR just for lighting + Backplates? Really big HDR?, Quick HDR you can make on set in 2min? you need low Hemisphere?, etc...

For example: Let's say you want a digital "copy" of the day/hour you shoot your HDR. You need really precise and unclamped HDRi for lighting purposes, you want to be able to make the HDR inside a 5-10 min window tops (otherwise you can't make good sunset HDRi, clouds will move a lot, etc) since you need 10-15stops ND filter to get unclamped HDR and something like 20-24 EV you will be shooting at V 30seg+ a lot (this impacts drastically the HDRi). But using 10-15ND filter introduces a really big bias on the images you get (heavily colored vignetting) , so the post-processing part of this kind of HDR become really complicated (you can see it on the Unity document), time consuming and introduces random failures (some of the HDRs just wont work). In my opinion if you need/want perfect unclamped HDRi just for lighting (most of the ones from the internet are really, really bad to be honest) it´s better to use 180º Fisheye because otherwise you won't be able to capture all the images you need on time.

But if you don't mind if the HDRi it´s unclamped and you only need a beautiful/high resolution sky, don´t use ND filters and use medium format camera (or new generation DSLR) and 12-24 lenses, you can paint by hand the dynamic range you clamp on the sun during shooting to get a really high resolution HDR, it´s not 100% precise but if you do it right nobody will notice the difference.

You can use this calculator to decide which camera/lens you want to use: http://www.hdrlabs.com/tools/panocalc.html

Getting really precise/unclamped HDRi and 32k it´s almost impossible, even using medium format cameras or new generation DSLR, so keep this in mind if you are new to HDR capture.

About the "unity" method, they take +-30º because they want to be able to remove people/traffic from the horizon but in archviz we usually don't even want horizont so you can make it using just 1 rotation (3-4 positions) + Nadir and you can get unclamped HDR using ND inside the 5min window. This method it´s really good but will fail sometimes and usually requires more post-work so it's not rock solid, you can't do it on a set and risk the HDR for example.

About Pano Heads, if you need a versatile one you can adjust to be used with any lens i recommend you Nodal Ninja, if you want the easiest one to use you can get the 360Precision

This is just my recommendation after making lot´s of HDR but i can be mistaken ;) so you need to experiment a lot in order to be able to get the HDRi you want and be able to make it consistent.

Best.

9

Gallery / Re: Apoco77 House

« on: 2016-08-29, 19:15:00 »

One of those works. :O
Congratulations your work it´s amaizing i still remember the first work from you back on 2004-2006. ;)

10

[Max] General Discussion / Re: PBR kinda materials

« on: 2016-08-24, 17:16:21 »

Hi. Very interesting thread.

I´m doing some test using Megascans free samples and Vray/Corona. I don´t want to create a Vray Vs Corona conflict here i just need to do this kind of test. They are "visual" test, my objective wasn't´to match reference but to compare different shaders.

I´m getting best results using VrayalMtl GGX mode but i don´t consider this production ready because i don´t know if alMtl will be part of Vray someday and because the shader it´s far from complete.

In Corona i can get "similar" results but 1.5 PBR looks broken (Black halo) and using a falloff on reflection to fix it manually doesn´t look like the best way since falloff itself it´s broken and i don´t like to work "by eye" and i can´r match hotspot on Corona but maybe it´s because i didn´t spend enough time tweaking falloff.

Will do further testing next week after D2.

Corona Results:


Vray:



Best

11

Gallery / Re: bergere chair

« on: 2015-11-02, 16:35:20 »

Hi. Looks very nice.

Just a personal opinion, Chromatic Aberration is way too much.

;)

12

Gallery / Re: Moscow city traffic center (ANIMATION)

« on: 2015-09-19, 16:55:14 »

wow. Very impressive work, in some sequences i miss a little bit of MB.

can you share something about the making off the buildings?, they look very nice.

13

Gallery / Re: Deathclaw on the run (video)

« on: 2015-09-18, 02:43:12 »

Wow. Vegetation looks amazing. :O

14

Gallery / Re: MID-CENTURY MODERNIST SHOWROOM (free 3d scene)

« on: 2015-09-18, 02:42:04 »

Hi. Vary nice work.
 I like the materials and general look of this one. The Chromatic aberration looks to much to me. ;)

15

Work in Progress/Tests / Re: Interior experiment

« on: 2015-09-14, 19:01:45 »

Hi. Thanks Bluebox, you are right the concrete it´s to sharp i can improve it. The chairs are the other element i don´t like in my image, i was lazy to model the correct ones xD.