[Scribus] cmyk workflow

[Axel Bojer]

David Eyre skrev:
> Just looking for some views from other users. At the
> moment, my workflow for placing images is as follows:
> 
> 1. Create image box
> 2. Import unedited image
> 3. Edit image to open in the GIMP
> 4. Re-size and crop image
> 5. Do CMYK-separation of edited image using GIMP
> Separation plug-in
> 6. Save separated edited image as tif
> 7. Back into Scribus to delete unedited image from box
> contents
> 8. Import edited image tif
> 
> Am I wasting time here? Do I really need to do the
> CMYK separation? Would the images be automatically
> CMYK-ed when I export the page to PDF?

As far as I have understood other advices on this list, you will not 
have to import images as CMYK, but just RGB and then, yes, Scribus 
should convert it to CMYK if you chose save as pdf *and* intended for Print.

Thus my workflow has been:

1. Import unedited image
2. Scale the image + its frame until it fits as I wish
3. Use Gimp to rescale my image to the size I found in 2)
4. Import the changed image
(3 and 4 are meant to save space, I don't think you *have* to do this)
  And when I am totally finished
5) Export the whole document as pdf -- then the image should be 
converted to CMYK by Scribus, or so I thought it was at least ...

But I have a problem concerning this:
Right now I got a telephone from my printer saying my pdf resulting from 
these steps are not usable, but greyish instead of grasgreen.

See the results yourselves here:
http://www.bojer.no/Ariadne2006/ariadnes_aarbok2006-omslag-ForSkjerm.pdf
http://www.bojer.no/Ariadne2006/ariadnes_aarbok2006-omslag-TilTrykking.pdf

("skjerm" means screen, and is the rgb variant, "Trykk" means print and 
is the one intended to be printed).

When I look at it in XPD or Kpdf it really looks greyish, as he said (he 
used Windows I suppose, at least he said he was using Acrobat), but on 
acrobat on Linux it show as it should, no greyish tone to it as in xpdf. 
  Having repeadedly read in thes thread, that acrobat is to be trusted 
over all the other pdf-programs, I trusted it to be OK, but it seems 
not. Or is he mistaken??

Now, what I have as a pdf-file is made out of many imageframes linking 
to the same picture. This gives a nice effect, and keeps the file pretty 
small. But the printer was in doubt wether his equipment would be able 
to use it, I am not sure, but said I think it ought to be ok (after all 
this function is quite the same as in other DTP programs, so I excpect 
this to be no different). But he wanted me to ?flatten the image?, that 
is merge all my layers into one. So the only reasonable way I found 
(without having to redo everything) was to export as png, put this png 
file as the only element in a new file with the same dimensions and then 
export to pdf again.
I looked at the png-file in Gimp, and it looks ok and is RGB.
The resulting file though is really greyish. Now I am not quite sure how 
to solve this, I had trusted Scribus to just solve this for me by export 
time ...

Am I doing something wrong here?

I ought to have it sent and proofed today, so every help is appreciated :-)

Best regards
Axel Bojer