• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: about negs and ROMM / Pro Photo RGB
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: about negs and ROMM / Pro Photo RGB


  • Subject: Re: about negs and ROMM / Pro Photo RGB
  • From: email@hidden (Bruce Fraser)
  • Date: Sat, 4 Aug 2001 14:39:57 -0700

At 10:48 AM +0000 8/4/01, neilB wrote:

OK, I've obviously got this one wrong. Unfortunately the
effective intent is difficult to test by trying the alternative
selections since, as expected, the results are all the same - but
which intent is it I'm seeing??

It's Relative Colorimetric unless you ask for Absolute Colorimetric


There has been a lot of discussion here of late on this point and
I thought the final consensus was that a conversion [input >
workingspace, workingspace > workingspace] was always Perceptual
regardless of choice apparently made.

This theory was reinforced by opening a few workingspaces in
Steve Upton's Colorthink [I love it]. There I noted that, for
example, both Adobe RGB and ProPhoto are set up to default to
Perceptual - [and to add to the confusion - apparently have no
rendering tags at all.]

The tags say Perceptual, but the rendering is in fact RelCol. You can test this by taking an RGB gradient in a large space, then converting it to a smaller one. You'll see that the gradient gets clipped, not scaled into the smaller space. If it were a Perceptual rendering, one would expect the gradient to be preserved. For the kicker, if you look at where the clipping takes place, it's at the gamut boundary of the smaller space.


On negatives:
Unless you're planning on making fairly large moves in 8-bit space,
ProPhoto is probably OK. I make all my big moves in 16-bit, but I
often downsample to 8-bit to do things I can only do with layers, and
so far ProPhoto hasn't been a problem.
thanks, it's certainly going to get a run here.
When you said keep data away from the extremes in ProPhoto are
you able to give any numerical guides please?

Not really, but here's a tool you may find helpful. Start out with a square image, say 600x600, and fill it with a spectrum gradient from left to right. Add a layer, and fill that layer with a black-to-white gradient from top to bottom. Set the blending mode for the layer to Luminosity. You now have a file that contains pretty much the whole range of RGB values.

Set the color picker to read RGB and Lab. Anything that returns a value beyond about 120 or -120 in a or b is in science-fiction territory, particularly if it has a luminance of 0-10 or 90-100.

Bruce

--
email@hidden


  • Prev by Date: Monitor Profiles Missing
  • Next by Date: Re: Conversion Steps/PRO/Observations
  • Previous by thread: Re: about negs and ROMM / Pro Photo RGB
  • Next by thread: Re: about negs and ROMM / Pro Photo RGB
  • Index(es):
    • Date
    • Thread