PDA

View Full Version : Render Buffer - 16bit?



JeffrySG
01-01-2008, 04:35 PM
Is there a way to get the gray scale buffers, such as the depth buffer, to render as a 16bit/channel file? The 8-bit files are really prone to quantization after adjustments are made to them.

What's the best way to just have the render buffers to render and not the normal image? Let's say I rendered my RGB image already and now want to just render out the buffers, but the RGB takes a long time so I don't want to render it again.

(sorry for the double-post but I was worried it would not be seen in the other thread...)

--Jeff

Red_Oddity
01-01-2008, 04:42 PM
Use Deluxe RLA and use Deluxe RLA image filter to set the bit depth

JeffrySG
01-01-2008, 05:26 PM
Thanks RO! I'll look into those settings. Do you know if Photoshop CS2 can open RLA image files? If not, does anyone know of a way to resave RLA files as HDR files?

thisguy
01-01-2008, 05:31 PM
save it as a tif

JeffrySG
01-01-2008, 05:39 PM
I was looking through all of the image filters and found that the PSD layer export has 16bit option, and it's nice to have all the layer options as well. I think I'll just use that.

Thanks for all the help everyone! :)

Sensei
01-01-2008, 07:11 PM
LW and Kray are rendering internally 32 bit float. That's image saver picked by you which is truncating it to 8 bit per channel.

JeffrySG
01-01-2008, 08:51 PM
^I'm not sure I fully understand what you mean. Is there a better way to save the images. If I hit F9 and then save the rgba file as a PSD 32 bit file, when I open it, it still is 8bit/channel. But if I use the PSD process filter to save the image I can get it to be 16bit/channel. Is there another way to get the files to be higher than 8bit/channel?

Red_Oddity
01-02-2008, 02:40 AM
Make sure you use the FP viewer (which stores the preview images as float, so when you save using a 16 bit format it should, theoretically, be 16 bit)

Lightwolf
01-02-2008, 04:54 AM
A few savers allow for higher bit depths: SGI 48bit/64bit, radiance, flx.

Currently I'd recommend OpenEXR, since it supports up to 32bit float per image channel and is widely supported in newer apps.

Cheers,
Mike

JeffrySG
01-02-2008, 08:12 AM
Make sure you use the FP viewer (which stores the preview images as float, so when you save using a 16 bit format it should, theoretically, be 16 bit)
ahh... so that's what the FP viewer is for.... I get it now... thx RO!


A few savers allow for higher bit depths: SGI 48bit/64bit, radiance, flx.
Currently I'd recommend OpenEXR, since it supports up to 32bit float per image channel and is widely supported in newer apps.
Cheers,
Mike
Thanks Mike! I'll keep that in mind but now that I've have the 16bit layered file save down I might stick with that for a bit as most of the files I'm rendering are just stills. Does the OpenEXR save all of the different buffers as well?

Lightwolf
01-02-2008, 08:15 AM
Does the OpenEXR save all of the different buffers as well?
The LW one doesn't... except if you use the buffer saver and save them as separate images.
exrTrader does, but only the ones provided by the SDK. PS CS3 can only read multi-layer EXR files properly using a third party plugin though (http://fnordware.com - ProEXR).

Cheers,
Mike

JeffrySG
01-02-2008, 09:05 AM
^gotcha. Thx again Mike! :)