Newbie question - Rendering 32 bit sequence
- Thread starter mark
- Start date
I've never changed the G-Buffer colour depth, but I do change the opacity and use ProRes for exporting the alpha channel. Does 32-bit give a cleaner result?
Awesome, thanks Mark. I'll be choosing 32-bit from now on.
ZooHead
0
I believe the 16 and 32 bit output is useful for single image renders only.
For animation don't bother with the G buffer at all just make the Camera Background Opacity 0% and save as ProRes4444.
For a png image you don't need 32 bit for alpha transparency so again don't bother with the G buffer.
Just make the camera background 0% and save as png.
When you check the save G buffer 32 bit, Cheetah generates exr and hdr files separate from the png tif and jpeg files.
A 32 bit exr image can not be viewed on any computer without being HDR tone mapped down to 8 bits.
Cheetah generates a preview image for this using the Exposure Gamma method
but like a Raw image from a camera you can change it later by tone mapping.
For animation don't bother with the G buffer at all just make the Camera Background Opacity 0% and save as ProRes4444.
For a png image you don't need 32 bit for alpha transparency so again don't bother with the G buffer.
Just make the camera background 0% and save as png.
When you check the save G buffer 32 bit, Cheetah generates exr and hdr files separate from the png tif and jpeg files.
A 32 bit exr image can not be viewed on any computer without being HDR tone mapped down to 8 bits.
Cheetah generates a preview image for this using the Exposure Gamma method
but like a Raw image from a camera you can change it later by tone mapping.
Thanks Eric. I think I have gone full circle, so it's back to the good old png with alpha transparency. It's good to know what to use 32-bit for should I need it one day.
Actually it's not that simple. 32-bit exr-files as video are actually common and it's in other 3d packages the standard output (linear).
The reason is the same why (most) professional photographers / camera operators rather use raw. In the end it's always toned down (tone mapped) to something we can see, but up to that point you would want as much information as possible. For example a (short) movie is filmed at different times with different lights (which changes the color), maybe even with different cameras etc. With the right color correction you get the same look for all of it. The simplest reason, though, is to get the style you want (or the customer wants) which is easier the more info the picture actually contends.
When you don't have such restrictions and your whole animation is just a single scene in Cheetah, you probably don't need it. So png is certainly enough most of the time for most of the users here.
For those who are interested in digging a bit deeper, see the wiki-links below:
Color correction
Tone mapping
LUT - Lookup table (scroll down to lookup tables in image processing)
The reason is the same why (most) professional photographers / camera operators rather use raw. In the end it's always toned down (tone mapped) to something we can see, but up to that point you would want as much information as possible. For example a (short) movie is filmed at different times with different lights (which changes the color), maybe even with different cameras etc. With the right color correction you get the same look for all of it. The simplest reason, though, is to get the style you want (or the customer wants) which is easier the more info the picture actually contends.
When you don't have such restrictions and your whole animation is just a single scene in Cheetah, you probably don't need it. So png is certainly enough most of the time for most of the users here.
For those who are interested in digging a bit deeper, see the wiki-links below:
Color correction
Tone mapping
LUT - Lookup table (scroll down to lookup tables in image processing)