XYZ Monitoring on Win/Mac

Hi Guys,

I'm looking at building a grading/screening room when the RedRay projector hits, and am hoping to be able to monitor in 12-bit XYZ color space right out of Resolve in a Windows environment. My hopes had been set on getting a Decklink 4K, but now that it seems to have become the 4K Extreme it's only 10-bit and no longer supports the XYZ colorspace.

I don't think there is an answer to this that I am going to enjoy hearing, but I'm posting more to show BMD that there is interest in a product to solve this problem. Currently the only way to monitor in XYZ is on the Linux side, and it's only 10bit as far as I can tell.

For those interested the RedRay Projector supports the XYZ space (in 12bit) natively, and is locked onto the spec and doesn't need recalibration since it uses lasers for its light source.

Anyways, if there is no product on the roadmap currently BMD, please take this as a big +1 to there being one

I don't see how using lasers means it doesn't need calibration, unless the lasers know exactly what screen you're using and it's spectral characteristics (hint: they don't).

Yes if you want xyz or even 4:4:4 4K then Linux is your only option, and it does do 12bit sampling in xyz.

There aren't any laser products on the market right now that allow you to evaluate this, but laser light is VERY spikey, not broad light like that preferred for color critical projection (think xeon vs UHD, but waaaaaaayyyyy worse) so it remains to be seen how the new laser projectors hitting the market will fair in the color critical space

That's assuming the RED Laser (REDray is their playback product, not the projector) ever gets to market, red said it would ship in 2012(obviously didn't happen), but has been mum on the projector for 7 months since nab, also has scrapped various demos of odemax, and REDray. So I wouldn't hold my breath at this point for the red laser to ship anytime soon.

On top of it, when you have spike light (like laser...) the metamerisms(1) you get make the appearance of the image to your eye different of anybody else on the planet.

if you are doing a real side by side comparison between a broad light source element (film projection) against laser projection, even when the instrument (spectral color meter) will say that the images are the same, to your eye will look not the same, and the older you became, the more shifter to yellow the perception will be. The cone power reading distribution will be emphasize by the pulse narrow spectrum of the light source.

Salvo is also right when it keep in account the light path: glasses and screen. If you will make a silly experiment and add a colored filter in front of the light path, I'm pretty sure that even a Laser will tint the final image that color, hence you need MCGD/TCGD correct representation.

for the XYZ, can you create a x33 LUT that goes from the color space you're using to the final color space? (I use LightSpace CMS for this type of conversions)


reference:
(1) http://en.wikipedia.org/wiki/Metamerism_%28color%29
LightSpace CMS :http://www.lightillusion.com/lightspace.html

There is an xyz lut built in to resolve. Problem is signal path, not color space.

Thanks for the info guys.

Wasn't meaning really to turn it into a discussion of whether laser based projection systems are suitable for color grading. Thought it is relevant, I'm keen to wait until the RedRay Projector lands and then see what Red's recommendations are as far as ideal viewing conditions (screen types, glasses etc).

Also Salvo, I probably didn't speak as clearly as I should have. When I was referring to the lack of a need to calibrate the projector, I should have said "re-calibrate". Obviously different screen types are going to result in different light bouncing back into the eyes (not to mention different types of 3D glasses). What I was referring to was that since it's laser based, it's not going to need constant re-calibration that a typical projector might as the bulb ages over time. Red's claim is that no calibration is needed - how this is achieved remains to be seen - but I'm certainly interested in seeing what they manage to come up with.

Regarding 12-bit XYZ output on linux, what card is that running through? I was under the impression that XYZ on Linux used Quadro cards, but thought these cards were limited to 10bit?

There is an xyz lut built in to resolve. Problem is signal path, not color space.

I see.

My bad!

Doesn't use quadro cards. That was the old config for 2k stereoscopic.
http://www.blackmagicdesign.com/support ... e&os=linux