In an article unironically titled Don’t repeat these mistakes about color spaces and HDR: insights from the demo “Debunking HDR” by Steve Yedlin (Part 2),” Daniel Bañuelos Cuéllar, a lead editor and post-production supervisor who specializes in HDR workflows and color management, shared some takeaways after watching Yedlin’s presentation. We unpack just a few of the outrageous claims and glaring inaccuracies.
“Extreme brightness does not necessarily improve the image.”
There is nothing extreme about a well-graded show with carefully deployed 1,000-nit highlights. The major studios don’t dictate the peak brightness level of films anyhow.
“When we talk about HDR, what we do use is Rec.2100”
No, we do not. Netflix/Disney/Apple mandate P3-D65 ST2084.
“This [visual anchoring] is key to understanding why some HDR systems end up looking weird or ‘crushed.’”
No, it isn’t. If a grade looks crushed, it’s the fault of the colorist – not biology!
“The eye doesn’t see in absolute nits (nor should it).”
The Claim: “PQ’s absolute values are unnatural because vision is relative.”
The Reality: PQ ST2084 is designed to honor human vision by reproducing the actual luminance levels (absolute nits) found in the real world, rather than relying on the relative percentage-based approach used in SDR.
The author makes numerous dangerously misleading statements about gamuts and transfer functions, of which we’ll highlight just a few.
“Does Rec.2020 look better because it has more colors? Nope.”
Commercial work is mastered at P3-D65, not Rec. 2020.
“Display P3: the color space of Apple monitors, which in many cases simply maps from Rec. 1886 [sic]”
Display P3 is a color space. BT.1886 is a transfer function. One cannot simply “convert” BT.1886 to Display P3 in the manner described.
Bañuelos makes similar blunders in the two following definitions.
“sRGB: it is almost the same as Rec.1886 [sic], but with a slightly different transfer curve.”
sRGB is a color space. BT.1886 is a transfer function. Therefore, sRGB ≠ BT.1886 + “a slightly different transfer function.”
“Rec.2020: it is Rec.1886 [sic] with a wider gamut.”
Rec.2020 is a color space. BT.1886 is an EOTF. It has no primaries. Therefore, Rec.2020 cannot be “BT.1886 + wider gamut” It’s like saying that a bicycle is a toaster on wheels.
FACTS DEMAND CORRECTION
The Daejeon Chronicles 