PDA

View Full Version : Framebuffer or tediousness?


Hakushinkan
07-23-2010, 04:49 PM
I've been reading a lot of articles and thoughts about linear workflow in maya, and eventually I understood the basic concept.

The only thing I just don't understand right now, is what's wrong with using the framebuffer.

Based on tests I have done, it gives the same results. If considering bump mapping, they show up even BETTER when using the framebuffer method. More smooth and clean.

Either way, I saw on DJX's blog that he doesnt recommend even going near the framebuffer gamma.

Why is that? what's the different in the end of the day?
And one important thing about it - is linear workflow something that exists in Max and XSI too, or is it just one of those annoying things only maya has?

Redsand1080
07-23-2010, 06:43 PM
I would stay away from the frame buffer gamma correct. It's an older control that's still just hanging around, at least that's the case for 2010 and older. Not sure about 2011 because I don't use it. Also, it will gamma correct everything, even normal maps and bump maps. Generally not a desirable thing.

A linear workflow is a methodology, a way to approach rendering and compositing so that the math is done correctly. So it's not limited to any application. It applies to image manipulation and compositing in addition to rendering.

Hakushinkan
07-23-2010, 07:09 PM
I would stay away from the frame buffer gamma correct. It's an older control that's still just hanging around, at least that's the case for 2010 and older. Not sure about 2011 because I don't use it. Also, it will gamma correct everything, even normal maps and bump maps. Generally not a desirable thing.

A linear workflow is a methodology, a way to approach rendering and compositing so that the math is done correctly. So it's not limited to any application. It applies to image manipulation and compositing in addition to rendering.

Thanks for the reply.
I'm a student, so I use the 2011 student version. The framebuffer's still there and it look like it did in previous versions...

What I understood about linear/non-linear is basically this: our eyes work in a non-linear way, but in order to maintain accurate image data throughout a pipeline (and maya is a pipeline), it has to be calculated linearly in order for the results to be consistent and predictable. So basically I have to tell maya to calculate everything linearly (1.0 gamma), and only show me the final result in a non-linear way (gamma 2.2, done via lens shader).

So - what happens is that basically anything and everything I do within maya that's in non-linear mode (an 8-bit image applied on a material for example), has to be tweaked to be linear. That's done by applying another gamma correction on the image, bringing it down to 1.

Basically everything except for bump maps, normal maps, and hdr textures needs to be corrected, so wouldn't it be less work to add a 2.2 gamma node on those and just use a 0.454 frame buffer?

3dlight (http://3dlight.blogspot.com/2008/09/linear-workflow-for-maya-mental-ray.html) recommends the framebuffer method, while DJX recommends staying away from it. I first found out about the framebuffer's validity in Floze's 6 tutorials, but since then I heard a lot against it too.

Basically I'm not educated enough about digital image sciences to understand the whole math behind it, so I can't make a decision on my own... so this thread's here to help me (and others) understand the whole thing better.

TaKIKO
07-23-2010, 08:05 PM
Check out the links below, imo they best explain LWF..


http://vimeo.com/8119194

http://averpil.com/fredrik/articles/article.php?id=linear_workflow_pt2

Redsand1080
07-23-2010, 09:51 PM
Inserting gamma nodes can indeed be tedious. That's why I wrote that script I posted in this thread. It may help you out. I created it to handle many of the situations I've run into personally, and others gave great suggestions on how to improve it further. You may want to give it a shot if you still do things the old-school way like myself. Meaning, not using the new LWF features in 2011.

Hakushinkan
07-24-2010, 12:05 AM
Inserting gamma nodes can indeed be tedious. That's why I wrote that script I posted in this thread. It may help you out. I created it to handle many of the situations I've run into personally, and others gave great suggestions on how to improve it further. You may want to give it a shot if you still do things the old-school way like myself. Meaning, not using the new LWF features in 2011.

I didn't know any new LWF features in 2011.
Could you tell me? or where can I read on it?

And I downloaded your script. It's great, thanks a lot. very useful :)

CGTalk Moderation
07-24-2010, 12:05 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.