-
Final Bit Depth in Broadcast work?
I work for a broadcast company where I create motion graphic spots and promos using C4D and After Effects. My current project workflow is to render out my images linearly using 16 or 32 bit images and composting my project in a linear workspace in a 16 or 32bit comp. My final render is an uncompressed MOV. Recently, I was told that I should be rendering out all my files as 8 bit and comping the same since Broadcast HD signal is only 8bit. Now, this doesn’t make sense to me but I can’t seem to find anything that states which way is correct. Can someone point me in the right direction? I can’t imagine that broadcast TV is limited to only 8bit color depth especially with such great TV shows being aired primary on broadcast TV (i.e. “The Flash”).
-
8 Replies
-
[Christopher Sosa] “I can’t imagine that broadcast TV is limited to only 8bit color depth especially with such great TV shows being aired primary on broadcast TV (i.e. “The Flash”).”
It is 8b, but…
[Christopher Sosa] “Recently, I was told that I should be rendering out all my files as 8 bit and comping the same since Broadcast HD signal is only 8bit.”
… that’s not a reason to comp at 8bpc. Both 16bpc and 32bpc will give you greater precision while you work and can visibly improve the quality of a render destined for 8bpc output.
Why? Because bit depth is a project-level setting that affects every single mathematical operation performed inside your composite before the final 16/32bpc image is resampled at 8bpc for output. More bits gives you more precision, which means fewer compounding rounding errors that show as steps in your image.
Comping at 16bpc and rendering to 8bpc will force Ae to dither [link] to reduce visible quantization (banding) in subtle gradients.
Comping at 32bpc and rendering to 8bpc also engages dithering, plus it allows you to exploit floating point mathematics, where the value of a pixel can be brighter than white or darker than black during compositing. This can dramatically alter the appearance of a composite. Lighting and blur effects in particular benefit from floating point, but you’ll want to work in floating point from the start, or else you may have to re-work effects and blends.
While friends don’t let friends comp at 8bpc, I’ll note that there are downsides to 16bpc and 32bpc: they require double and quadruple the amount of RAM and cache storage per frame-layer. If your computer is limited on RAM, this can cut into your preview length.
Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive | RenderBreak [blog] | Profile [LinkedIn] -
Okay, well that clears up a lot thanks so much for the awesome response. Now when I render out my MOV as Animation codec does it render out the mov with a 8bit color depth? Essentially tone mapping my 32 bit comp into a 8bit video file? Not sure if I said that correctly.
-
[Christopher Sosa] “Now when I render out my MOV as Animation codec does it render out the mov with a 8bit color depth? Essentially tone mapping my 32 bit comp into a 8bit video file?”
Yes.
Tone-mapping might be overstating the case, though — any value above 1.0 (floating point) will be clamped to 255 (8b). It will look in the render as it does on your screen, but overbrights and superblacks will not be preserved. Does that make sense?
Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive | RenderBreak [blog] | Profile [LinkedIn] -
Yes, that does make sense thanks again. Now I do have one last question and I think I’m good. I’ve noticed when I take my uncompressed MOV back into After Effects it recognizes it as being a 32bit U color depth. But if I’m understanding what’s going on it should be a 8bit color depth. Does this have something to do with the metadata retaining what the original comp settings were or am I missing something when I’m rendering. And thanks again Walter I really appreciate you clearing this up. I can’t seem to find any information online that explains it.
-
[Christopher Sosa] “I’ve noticed when I take my uncompressed MOV back into After Effects it recognizes it as being a 32bit U color depth.”
Can you post a quick screenshot showing what you mean?
Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive | RenderBreak [blog] | Profile [LinkedIn] -
Here you go! Actually, in this example it shows 16bit U but this is the exact file I output for from my end.
-
Oh… so just realized that’s talking about the audio channel. Can we pretend like I didn’t ask that questions?
-
Your correct, I was just confused on the end results bit depth rate vs the bit depth of my source material.
Reply to this Discussion! Login or Sign Up