News for hackers. Stuff that matters.
Re:Apple's Software Quality Crisis: When Premium Hardware Me... by meindnoch on Tuesday March 04, 2025 @06:16AM (#43253160)
It depends on the exact meaning of "YCbCr" and the meaning of "RGB". Is it BT.601? BT.709? BT.2020? Adobe RGB? Display P3?

Also extra fun is guaranteed if one end of the video cable is encoding with e.g. BT.601 primaries, while the other end is decoding as e.g. BT.709, or vice versa.

Re:Apple's Software Quality Crisis: When Premium Hardware Me... by WithinReason on Tuesday March 04, 2025 @09:18AM (#43254894)
RGB to YCbCr conversion and back is well defined, and RGB->YCbCR->RGB should give you back the original RGB values, no matter what the colorspace was.
Re:Apple's Software Quality Crisis: When Premium Hardware Me... by meindnoch on Tuesday March 04, 2025 @11:47AM (#43257115)
Lossless roundtrip conversion is only true for schemes like YCoCg-R [1]. It is not true for Rec.601, Rec.709, etc. because these standards require quantizing to 8bits (actually even less than 8bits due to the 16-235 thing).

[1] https://www.microsoft.com/en-us/research/wp-content/uploads/...

Re:Apple's Software Quality Crisis: When Premium Hardware Me... by WithinReason on Tuesday March 04, 2025 @03:11PM (#43259354)
I'm not talking about mathematically exact reconstruction, I'm talking about not being able to see the difference.
Re:Apple's Software Quality Crisis: When Premium Hardware Me... by karmakaze on Thursday March 06, 2025 @06:36PM (#43286123)
I first noticed it around 2014 on a 1080p Dell display (maybe ~24") with an i7 MacBook Pro of the day.

It was the poor non-crisp pixels that led me to even knowing there were even RGB/YPbPr modes.

Re:Apple's Software Quality Crisis: When Premium Hardware Me... by WithinReason on Friday March 07, 2025 @02:26AM (#43287950)
was it maybe with 4:2:2 or 4:2:0 subsampling? at 4:4:4 there should be no difference