Digital BW, The Print

I'm trying to get started with the MIS VM Sepia-Neutral (SN) inkset 
in a CIS I just installed in my 1160. In the first nozzle test, there 
were a few gaps in the second set of black stairs and several gaps in 
the third set of black stairs. (And the non-black colors were of 
course a little weird.) But after only a single cleaning and no 
waiting, the second and subsequent nozzle tests look good in all 
steps in each "color". I hope this means that the SN inks are loaded 
properly and the cartridges are correctly seated. So I'm trying to 
get ready to use the step test with Paul Roark's curves from the MIS 
web page (InkSupply.com).

If you go to the home page for this list

http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint

and then to the Files page, "Image processing and workflows", the 
first entry is "21stepNew.jpg, Paul Roark's New & Improved 21-step 
Wedge, 122 KB".

21stepNew is the most complex of the 21step images I've seen. So I 
thought I'd try to use it. But I'm concerned to understand first a 
difference I see in how 21stepNew looks in Internet Explorer 5 (via 
QuickTime's PictureViewer 5.0.2) on the Yahoo page and in how it 
looks in after it's downloaded by Explorer 5.

In Explorer, to the left of the 21-step wedge, I see a column of 
thirteen vertical bars of graduated shades of gray. The bars are very 
distinct; the column looks almost fluted.

Having used Explorer to download the image, when I try to open it in 
Photoshop 6.0.1, I first get the Embedded Profile Mismatch dialog 
box, which gives me the usual three choices, all of which result in, 
instead of a fluted column, a perfectly smooth one with no bars, just 
a smooth gradient. If I close Photoshop and double-click on 
21stepNew.jpg, the Mac OS (9.1) uses PictureViewer 5.0.2,  but from 
Apple's QuickTime instead of from Explorer, to open 21stepNew, but 
this time with the gradient instead of distinct bars. So the download 
seems to cause the change.

I'm curious to know both why the bars turn into a gradient and, if 
this reduces 21stepNew's usefulness, whether I should try to 
compensate for it somehow.

Thanks.

Sam