I decided to do some detailed measurements for a baseline, going back to 10.5, before loading 10.6.3. I will then do a test for any changes afterwards.
First, The secret seems to be, as Jeff Schewe and others inferred earlier, to have the printer driver loaded so that its used instead of Apples drivers. In my case, its an Eps 4880, so when you click options for the printer, it should show driver ver. 6.12, which is the latest driver for 10.6 from eps.
Test profile- My own before snow leopard for Ilford Gold Fiber Silk from I1 TC 9.18 test charts, I1 Match 3.6.3, I1 Pro, rev. D. It has a few lumps but gives reasonable prints.
Test print- Eye ones lab_test_image.tif, which has the standard test chart imbedded.
From PS CS4, PS col. Mgt, rel. colormetric; Pr. col. Mgt. off. In case the post doesn't format correctly, I will try to correct it ( you may have to count the entries). In any case the important numbers are the dE's and the absolute comparison with the testchart orig. nos. Measured w/i1 share.
Patch no. Before SL (L*,a*,b*) After SL dE
1 38.2, 15.2, 17.7 38.2,14.8,16.2 dE1.5
2 66.9, 16.1, 18.5 66.2, 15.8, 17.0 dE1.7
13(blue) 29.4, 17.5, -53.4 29.7, 16.8, -52.8 dE1.0
14(gr) 56.0, -39.7, 36 55.0, -39.8, 34.4 dE1.9
15(red) 42.3, 56.6, 27.8 42.0, 55.7, 26.9 dE1.3
So- two possible conclusions- Getting the printer driver loaded correctly gets you past the snow Lep. problem, or------
Its uses the Apple driver no matter what it says its doing(which I don't believe to be the case).
As an aside, the dE between CS4 and LR2 prints goes up to between 4 and 7.
Scott McRae