Gradients print as solid colors

[[email protected]]

Has anyone got some idea of why the following would happen:

File designed in Illustrator CS3, and put to PDF, has gradient orange to white, prints from PDF onto DC252 with what appears to be a solid dark red where the white was supposed to be in the gradient.
Same file on the Heidelberg QM Di 46 prints with a dark GREEN instead of the light orange gradient.
Another file done originally in CorelDRAW 12, produces a similar problem on the DI.

I have 4 cases like this, and i've attached the files, files come from different sources and all of them have gradients that print incorrectly.

Is this a...
RIP issue?
PDF creation issue? (PDF's are all done through Enfocus Instant PDF 3.6 thru Acrobat6 or 7, or Enfocus Instant PDF 4 thru Acrobat 8 - btw, the files all passed through Instant PDF's certification using our standard inhouse PDF Queue)

Please help us shed some light on this...the attached 4 files printed with gradients all incorrect, either as solid colors or as other colors all together.

Thanks in advance!

 

[mschilling]

I've seen this many times when printing from a mac/acrobat8 with ps gradients to our Ikon color copier. It seems to be a problem with the rip parsing the gradients, and I usually cure it by printing as an image from acrobat, but I think the rip likes different levels of acrobat, i.e., 1.3, vs. 1.4, etc.

 

[Soilworker]

This isn't going to make any sense because, quite frankly, I can't see why it would matter. But, if it is the problem it will make even less sense.

I noticed that the gradients have clipping masks on them. Instead of just adding a gradient to a vector shape the gradients appear to have been added to a box (in one case rotated) and the desired shape is used as a mask. I suppose this could be a result of the pdf or flattener settings but I'm not sure.

The question is if you have to the create the mask as a vector shape anyway why not just add the gradient that instead? I can see how this could potentially do something weird but I have no idea why it would do what it's doing for you.

What RIP are you using? Perhaps it's outdated? I know that when we were using an old version of Nexus we had issues with gradients. To fix it we just expanded them in Illy.

I can't see anything else wrong with the files. They seem really simple. Any chance of us seeing the PDF setting or the original files?

 

[[email protected]]

Well, the files come from a variety of designers and some come straight from customers. The clipping path you mention, soilworker, is a possible suspect i agree. why the designers would use a clipping path to do the grad shape is beyond me, but so it has been done...
as for the rip - its prinect metadimension.

mshcillng - that would also be my first port of call, in fact i have some tests running with rasterized versions of some of them.

thanks

 

[ian]

I ripped the files using two different fiery rips and didnt see any problems with the file at all. You might not have the correct version of acrobat

Ian

 

[rich apollo]

Looking at the files I can't see any reason why they shouldn't print correctly, unless the RIPs in question are not PostScript v3 compatible. The clipping paths you see are standard for defining the boundaries of a smooth shade region.

 

[Soilworker]

Rasterization might work as well. That's another step we take when an expanded gradient is so large that the sheer number of vector objects and points chokes our poor rip to death (we're working on an upgrade).

Apollo, that did cross my mind when considering the clipping mask so I ended up running a test in which I created a custom shape and added a gradient to it, then I flattened it (in case it was a side effect) in a couple of different ways (directly in Illy and in the PDF settings). I then opened the PDF in Illy (Oh my god, the horror!) and there was only the shape I created whereas the given PDFs have the custom shape as a clipping mask and a square/rectangle behind it with the gradient applied to that instead. That's why I questioned it... I couldn't recreate it.

I'm not doubting your knowledge... I truly enjoy your posts/answers - I just thought the discrepancies between the test files and the given files were worth mentioning. Again, seeing the native files or the PDF settings might help.

I'd have to agree with you, Apollo, and say it must be the RIP then. I just ran the files through Nexus and proofed to an Epson 7800 and an archaic Dupont iG4... no problems on either. If it is the RIP causing the issue and you can't seem to get around it then either expanding or rasterizing the gradients should do the trick. At least it's a plausible work around for problem files until a more concrete solution can be reached.

 

[[email protected]]

Would this RIP issue be prevalent in both the Xerox' Creo RIP and the Heidelberg's Prinect Metadimension? Ok... I suppose stranger things have happened, eh?