Re: Proof Colors in Photoshop
Re: Proof Colors in Photoshop
- Subject: Re: Proof Colors in Photoshop
- From: Ben Goren <email@hidden>
- Date: Fri, 02 Aug 2013 11:28:39 -0700
On Aug 2, 2013, at 7:35 AM, Mark Stegman <email@hidden> wrote:
> However, just because someone uses the sRGB colour space DOESN'T mean they don't understand colour.
I hope I didn't imply as much! I was trying to suggest that, when people don't know what they're doing, sRGB is the right choice. It's still the right choice for lots of people who *do* know what they're doing -- publishing for the Web, for the example I gave.
> I DON'T understand why you would recommend NTSC as a video space rather than PAL or Rec 709 unless you want to restrict your work to those parts of the world whose broadcast systems are based on this old (?) standard from a time when full HD flat screen displays of the quality that are available today did not exist.
Actually, I just tossed that out as a random example of where there could be a specific instance of an ``unusual'' color space you'd use for some (unspecified) specific reason.
> You seem to think that this so-called "absolute" colour can be captured which ignores the fact that you are using a device. I fear that you may need to rip my head off and stick it on a tripod with my visual system and brain intact, attach a myriad of senses to my cerebellum and somehow convert the array of signals into some visual projection in order to achieve what you are looking for. Of course, this begs the question... how will you capture it? In my opinion, the price is too high anyway.
Actually...we already have all the hardware necessary -- and have, for some time. Every digital camera records exposure information along with the RGB values, and that gives us the absolute brightness. The problem is that all our working spaces assume you're viewing a print in a viewing booth with standard illumination; R=G=B=255 is L*=100 a*=0 b*=0. But L*=100 is actually pretty dim, especially compared to daylight...and RGB values are integers that would overflow if you tried to make them more than 255 (for eight-bit images -- and higher-bit images are still limited to L*=100).
We can record, within a suitable margin of error, the actual exact tristimulus value for anything a camera can capture. And we can, with an even smaller margin of error, define the gamuts of our displays and printers. What I'm suggesting is that we should be working in an absolute color space rather than still fiddling bits in device driver space, which is really what RGB values amount to.
> Many applications allow you to 'work' in these colour spaces.
No, they don't -- with some rare exceptions mostly used by the movie and animation industries, and, even then, they're using them as conversion spaces and using traditional tools for editing with traditional working spaces.
> Even in Photoshop you can work in Lab
Though Lab itself isn't limited to L* <= 100, Photoshop doesn't. And Photoshop's implementation of Lab makes it as much of a relative color space as the various RGB spaces.
> This is why standardised viewing conditions and all the other standards are so important.
If we were working in an absolute color space and did gamut mapping, we'd have tools much more suited to doing gamut mapping. You could increase the brightness of the backlight on your monitor but the image it displayed would remain exactly the same. You could then decide how you wanted to use that expanded tonal range; do you want to stretch the original image, or just translate it to the brighter image, or do you want to keep it as it is and use the extra headroom to display the highlights that had previously been compressed because of the limited gamut of the dimmer monitor? You could also specify the viewing conditions for the print and, so long as those viewing conditions weren't brighter than your display, you could see exactly what the print would look like. And with something like the always-on ambient measuring of the i1 Display Pro, you can potentially have your computer control your ambient conditions as well to ensure proper adaptation.
Hopefully that gives you a better idea of the sort of thing I'm dreaming about.
Cheers,
b&
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden