[Matias Hansen] "Ok. I'll leave it. Thank you for the response"
You are free to work in higher bit depths (16bpc or 32bpc), even with 8bpc sources and destinations -- and this is often a good idea that will get you more precise results. Ae will dither your 16b or 32b work as necessary upon render to an 8b format, preserving the look (but not the data).
Working in 16b is pretty much just like working in 8b, but it is more precise (so it will have fewer rounding errors which lead to banding or posterization). Aside from this difference in precision, they are visually interchangeable. You could start in 8b and switch to 16b (or vice versa) without really impacting your work.
Working in 32b floating point is completely different -- values brighter than white and darker than black are preserved. As Dave pointed out, this is especially noticeable on blurs and glows, but it also affects blending operations significantly. Because of this huge difference in calculations, 32b and 16b/8b are not interchangeable, so if you change depths you may have to also change blurs, glows, blends, etc.
Walter Soyka
Principal & Designer at
Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog - What I'm thinking when my workstation's thinking
Creative Cow Forum Host: Live & Stage Events