I try using ghostscript with device tiffsep1 to separate a pdf and produce 4 1-bit tiffs(CMYK) output. Everything looks like it's ok. But, When the /Frequency value setting too low that will produce a lot of black points(center of every cells) in white area or white background. What did I miss? Could you please give me some advice? thanks My ghostscript command with options ----------------------------------- gs -dSAFER -dBATCH -dNOPAUSE -sDEVICE=tiffsep1 -r300 -sOutputFile=output -c \ "<</HalftoneType 1 /Frequency 13 /Angle 75" \ "/SpotFunction { dup mul exch dup mul add 1 exch sub } >>" \ "/Default exch /Halftone defineresource sethalftone" \ -f text_graphic_image.pdf cyan 1-bit tiff ---------------- https://i.stack.imgur.com/V9ykK.png
(In reply to GSUser from comment #0) > I try using ghostscript with device tiffsep1 to separate a pdf and produce 4 > 1-bit tiffs(CMYK) output. > Everything looks like it's ok. > But, When the /Frequency value setting too low that will produce a lot of > black points(center of every cells) in white area or white background. > > What did I miss? > Could you please give me some advice? I'm afraid free users do not qualify for support. The obvious answer is not to use a very low frequency.
I tried: debugbin/gswin32c -r300 -dDisplayFormat=16#20102 -c "<< /HalftoneType 1 \ /Frequency 13 /Angle 75 /SpotFunction { dup mul exch dup mul add 1 exch sub }\ >> /Default exch /Halftone defineresource sethalftone" -f examples/colorcir.ps And it looks fine to me. Same with tiffg4. Since the halftoning appears as expected, the problem is more likely in your input file. Most likely what you expect to be a white area is not due to some aspect of color conversion (I notice you mention a Cyan separation (with a link I, and others should, REFUSE to click on)
Ray, I want Michael to look at this. This user opened a question on Stack Overflow where the images indicate some problem with the halftone and I asked them to report the bug here. If you *look* at the image in the screenshot, it has black pixels within the white halftone cell, that do not look correct to me. That's on every plate. Every halftone cell corner is set. You've closed the bug having apparently tried a totally different command line to that which was reported and then dismissed it because you couldn't see a problem. I've tried the supplied command line and it looks wrong to me. Even if the halftone is correct, then the colour is wrong, and its still something for Michael to look at. The file in question is one we supply as an example. The PDF file does not have a page Group, the page content stream does not appear to fill the page, so the background should be pure white. If it isn't then that seems to me like a problem.
In point of fact, you can use: gs -dSAFER -dBATCH -dNOPAUSE -sDEVICE=tiffsep1 -r300 -sOutputFile=output -c \ "<</HalftoneType 1 /Frequency 13 /Angle 75" \ "/SpotFunction { dup mul exch dup mul add 1 exch sub } >>" \ "/Default exch /Halftone defineresource sethalftone showpage" \ -f and get the same result. That doesn't even draw anything, so there's no way the page should be anything but pure white.
Oops, not the bug I thought it was...... This is really a duplciate of 701880 marking it as such. *** This bug has been marked as a duplicate of bug 701880 ***