Reading your question carefully, you've asked for the widest gamut you can get assuming a Lab source. Have you evaluated how much of Lab your catalog actually requires? Roger's (et al) point about value of sRGB assumption, even though somewhat restrictive, carries the day in my view. But I sense you already know and have a reason for why that's not sufficient... So what are your rendering assumptions? I'll take a chance and offer a perspective on displays: Given the commodities display market and the status of wide gamut today, where new TV displays are reaching for but usually missing Rec.2020, that would seem to be an appropriate max. For desktop computer displays, Display P3 (Adobe Image P3) which is sRGB gamma is a practical compromise between a significant additional (and balanced) extension to gamut with not-too-risky backwards compatibility to sRGB w3c. It won't be outlandish if viewed wrong but sRGB viewers will notice dull saturation. DCI-P3 is to be avoided due to gamma 2.6. Graceful handling of tonal response should be considered as important as gamut, do that narrows the field quite a lot for backwards compat. You've gotta think of both. Rec.2020 is nominally 2.4, pretty close but significant variance from w3c sRGB The safety of sRGB I consider very desirable. While it leaves some real-world gamut on the table, no one is likely to miss it unless they know exactly what to look for. An image in a wide-gamut space that is incorrectly rendered due to lax ICC support is very likely to be noticed! I don't intend to be purely contrary when I regard Adobe RGB as a one-off space that can't be justified in any generally hygienic view of display color. It was an error when it was created that was never corrected, and is being forgotten with Display P3. If you haven't reviewed Lindbloom chart, it helps get space performance re Lab into a bit of perspective http://www.brucelindbloom.com/index.html?WorkingSpaceInfo.html The critical question I would consider: Will your target be a display? Or another app with subsequent processing? If former, sRGB is safest If latter, then pick the largest space that contains your catalog, but no larger, and tend towards G2.2 TRC. ECI RGB v2 with L* TRC, a D50 whitepoint and a wider gamut might make sense. It not likely to waste bits on color that will never be used, minimizing noise injection. FWIW Yours /wire On Wed, May 6, 2020 at 5:00 PM Jeff Nova via colorsync-users < colorsync-users@lists.apple.com> wrote:
Hi friends,
We have a web application and want to display the best previews we can for Lab images, which are not supported by any browser. The Lab we expect is the default Lab from Photoshop, which I think is D50. So, what is the best choice for RGB space? We are tagging the files, and prefer v2 profiles for more browser compatibility, but will do v4 is needed. I don't need help in the transformation - I just want to know the best RGB space to present the widest gamut resembling Lab as possible.
I assume one of the best choices is Don's MaxRGB?
http://www.hutchcolor.com/profiles.html
I welcome any input!
Very best, Jeff
-- Jeff NovaChief Executive OfficerColorhythm jnova@colorhythm.comMain: +1 415-399-9921Direct: +1 415-399-9921 x407Mobile: +1 510-710-9590Fax: +1 415-399-9928
1015 Battery Street, Suite CSan Francisco CA 94111http://colorhythm.com
Please consider our environmentbefore printing this email Colorhythm is a certified Green Business:
This electronic mail and the documents accompanying it are considered trade secret, confidential and/or proprietary by Colorhythm LLC. This information is intended for use by the individual or entity to whom this e-mail is addressed. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/wire%40lexiphanicism...
This email sent to wire@lexiphanicism.com