[PRODIG] 1290 problems/answers

[Giles Stokoe]

Title: 1290 problems/answers

Firstly, I am very thankful for any advice that anyone can throw my way. I don't mind whether advice stems from personal experience or professional - if its free and it helps, i'm grateful. If someone is truly helpful and they incidentally offer a commercial service that I can use, then I am pleased to now about it and much more inclined to use them than someone whose helpfulness is untested (my database of suppliers has a pleasure-to-do-business-with-ness ranking).

I totally understand that the only way to get a truly accurate and predictable result is to have monitor and printer calibrated, as they will all be unique to a greater or lesser degree (if something can be 'less' unique?). What I expect from a generic profile is similar to what I would hope for in a proof print from my own black and white darkroom: black at black, white at white (after all 0%C 0%M 0%Y 0%K is pretty unambiguous), and everything else roughly in the right place in between. Then I know whether it is worth spending time (and money) refining the output... in this case seeing the potential of the setup and deciding whether to spend money on perfect calibration.

Now, results so far, using Mac OS9.0.4, Photoshop 6.0.

Using Thomas Holm's advice, converted to profile first using Adobe ACE engine, then printed using no conversion. This produced (bingo) no unwanted background colour, but a rather unpleasantly coloured print that was nonetheless rather similar to the converted image as displayed by p'shop. I guess that this is perfectly possible as (from what I can tell) any CMYK gamut is going to be within that of an RGB screen. Have tried some other engines (Apple CMM and Heidleberg CMM) and the result appear the same.

Armed with this info, it appears that the conversion to profile can be done.

The problem is... the final image (as appears on print and screen) is not the same appearance as the original scan, with which I was happy and had required very little adjustment. Some weird things have happened (e.g. clipping of black and white points shown in 'levels') as well as significant shifts of red and green. Do I have to manipulate the converted image back to the original appearance (prob not as easy as theory suggests, and a time consuming pain in the butt), or can someone suggest how to get the computer to do the conversion properly (knowing now that the generic printer profile can be used to generate a roughly accurate print of what appears on screen)?

Rather ignorantly perhaps, it seems to me that the previous (normal) method was bunging a layer of magenta and cyan over the whole printed image (whites and all) to get the converted (red and green shifted) image to look like the Adobe RGB 1998 rendition.

(By the way Neil, I have '1290-Fton-manualv4.pdf', which I guess is the manual you are referring to. Its all out the window using the above method though... apart from printer res etc. Also my 1290 refers to itself as a 1290S... is this an indication of mods, or has this always been the case?)

Thanks to all so far

Giles