update documentation on saturation values for 10 bit

In reference to https://lsv.uky.edu/scripts/wa.exe?A2=AMIA-L;53d3ca3d.2109&S=.

This looks really good, however there's one thing I have a concern about in this section

A Saturation rate above 128 for 8-bit content or 512 for 10-bit content is considered illegal because data in this range can't be translated to/from camera data, and a videotape is technically incapable of storing values in this range (or beyond).

My first concern is that it should be clarified that saturation values above this threshold can be properly stored in a YUV file, so long as they aren't above the higher (latter mentioned) threshold. I think that this is overall a bit of a confusing topic, so it might be good to mention this concept in the discussion for each threshold.

My second concern is this: it might be worth mentioning that while a videotape cannot properly store saturation values above this threshold, it is certainly possible for a transfer to contain values above this threshold. Here's a few examples I can think of:

• Head switching noise in the underscan can often go above this threshold.
• High saturation levels during a crash record often go above this threshold.
• It's even possible for non-noise / error values to go above the threshold if the proc amp settings used have boosted the levels sufficiently.

I hope this makes sense!

Ya, it's tough to explain well.

Here, the saturation value is sqrt((U-2^bitdepth)^2+(V-2^bitdepth)^2), which is essentially the hypotenuse from the center of the vectorscope to the vectorscope plot. In 8 bit that's sqrt((U-128)^2+(V-128)^2). But since the range of U and V is constrained to 8 bit expressions of 0-255, then the maximum saturation value is sqrt((0-128)^2+(0-128)^2) which is ~181.02, the highest possible saturation value (for eight bit YUV at least).