8 bpc 16bpc 32 bpc and render quality?

[Chris Wright]

I guess I’ll be the first one to take a stab at this. Since AE still hasn’t updated all their plugins to 32bpc, we’ve been stuck with a conundrum. Some effects give warnings but can be somewhat protected if you use utility->hdr compander.

Think of bracketing your 8,16 warnings with 1 at the very top set as-compress range and one at the bottom set at -expand range. You can take a screenshot and compare how the compander protects color.
As for why we should ever use 16 bit if we only output milllions+ 8bit anyway, is because AE will preserve posterization calculations then spit out correct 8 bit again.

As for 32bpc, you can push,pull pixels to your hearts delight and some effects are even redundant and can be replaced, like shift instead of set channels, hue/sat instead of color hls, transfer modes instead of calculations are 32bpc.

Should I go 32bpc all the time no matter what? I haven’t seen any reasons not to if you use compander as needed. This way you can use Color Finesse onto hundreds of precomps and not worry about posterization.

The only downside I can think of is really old plugins may not work even with compander and really large images, if you don’t have enought RAM, could give you a image buffer error. 8 bit is so 1980’s.:)

https://technicolorsoftware.hostzi.com/

[Anita Sancha]

gee thanks

wow … I had a look at the “hdr compander” effect, in time I am sure I will get to grips with this all. So thanks. Never heard of it before.

I feel that by just looking at the final tagas outputs that I am getting, on a full size detail, they look fine at 16bpc in project and 32 bit output. I was just confused as to why if I output at 32bit and set the project at 8 or 16 bpc seems somehow these numbers should be the same.

Its all a bit “bitty”

thanks

Thanks for all your help
Anita Sancha.

http://www.anitasancha.co.uk