CMYK Export

[Klaus Nordby]

That Charles Moir - a good businessman - and Jens Benthien - a dreamy designer - are happy with XaraX's non-customizable defaults for outputting separations means zilch to _me._ That just tells us they have lower standards than I and other print professionals have.

I have seen separations from Xara X. To anyone with trained pre-press eyes they are obviously second-rate.

Xara X is still the greatest vector illustration tool in the world. It works flawlessly with RGB color space. But for outputting high-quality color separations - forget it. That DEMANDS customization of GCR/UCR, etc. There is no point in arguing over this.

K
www.klausnordby.com/xara (big how-to article)
www.xaraxone.com/FeaturedArt/kn/ (I was the first-ever featured artist in the Xone)
www.graphics.com (columnist, "The I of The Perceiver")

[jens g.r. benthien]

all I can say:

Amen Mr. Hair Splitter.

And best regards to the Swedish forest. BTW, how do you actually print - carving the trees in your woods with your well trained pre-press eyes, drowning them into ink pots and then stamping a rag cloth with it?

Ok, don't need to answer, I'm just curious ;-}

jens g.r. benthien
designer
http://www.sacalobra.de

----------//--
If you don't know how to dream you'll never be a designer.
----------//--

[Klaus Nordby]

Some separation tests I just did with Xara X shows that _flat_ colors defined as CMYK seems to separate fine: the exact CMYK values come out, unchanged. But as soon as _gradients_ are used, XX's infernal internal (sic!) algorithm messes royally with the CMYK values, using its own GCR/UCR methods.

My guess is (but I don't know this for sure) that XX's gradients only work in RGB space - hence the CMYK values you input are always inevitably distorted in the converson process.

So if your art never uses any gradients, you will probably be fine in separating from XX. As for me, I doubt I have ever made any piece in XX which doesn't use gradient colors - and so I would never gamble on using XX for separations.

K
www.klausnordby.com/xara (big how-to article)
www.xaraxone.com/FeaturedArt/kn/ (I was the first-ever featured artist in the Xone)
www.graphics.com (columnist, "The I of The Perceiver")

[jens g.r. benthien]

...that I gambled and always won.

Not only the printing process, but the clients money as well.

And I use simple, plain RGB Tiffs without a well trained pre-press eye, without the fuzz of dot gain, dot angles, screen alignments, UCR...

Fiesta, I like gambling in this Xara win-win shoppe. In the meantime you can get goin' carving your trees - but watch the dot gains! And if you will come up with some results some day in the future - of course scientifically founded - we'd like to hear from you about the easy way of printing with Xara.

Happy dot-watching! And care for your underware removal ;-}

jens g.r. benthien
designer
http://www.sacalobra.de

----------//--
If you don't know how to dream you'll never be a designer.
----------//--

[Klaus Nordby]

Some further tests seem to strongly corroborate my hypothesis from earlier today that XX's color gradients are in RGB space only. Now I did separation tests by:

a. printing a PDF composite file
b. printing a PDF separations file
c. exporting a "Xara X EPS" file

In all three cases the _flat_ CMYK colors were output without any changes, as determined by viewing/rasterizing with both Acrobat 6 Pro and Photoshop 7.

And in all three cases _gradient_ CMYK colors were output with substantial GCR/UCR changes, again as determined by viewing/rasterizing with both Acrobat 6 Pro and Photoshop 7.

In addition, I printed with a PDF/X-1a specification in Acrobat Distiller 6. Contrary to what Gary found, this passed validation fine - when there were only _flat_ CMYK colors in the artwork! But when I printed with CMYK gradations, the PDF/X-1a validation failed. Guess which error message I got? Anyone?

If you guessed that Distiller's error message was "RGB colors found" . . . you would be damn right!!!

PDF/X-1a is a CMYK-only standard, which does not allow RGB colors in a PDF file. So most likely Gary's PDF/X-1a test file had gradations (or transparency) in it. Right, Gary?

In sum, it seems my previous verdict holds up very well: only flat CMYK colors separate from XX in a predictable way. With gradations (and transparency) all bets are off and nothing becomes predictable - because these are in RGB color space and XX's infernal internal separation engine kicks into gear.

So - apart from all-flat-comic-book-colored art - my earlier argument against XX's ability to produce _professional_ separations comes down to this simple syllogism:

Major premise: professional separations produce predictable color values
Minor premise: Xara X does not produce predictable color values
Conclusion: Xara X does not produce professional separations

QED.

Now, Charles will probably "personally find" that my reasoning here is "offensive" - and Jens will probably just evade this with more dishonest sarcasm, to avoid admitting he's both wrong and ignorant - but I hope that the rest of you TalkGraphics folks will approach my findings with a more rational, scientific and just simply _decent_ attitude. As far as I know, no one in the Xara X community has published findings on exactly _these_ separations issues before. I certainly learned something today - and so, most likely, did you.

K
www.klausnordby.com/xara (big how-to article)
www.xaraxone.com/FeaturedArt/kn/ (I was the first-ever featured artist in the Xone)
www.graphics.com (columnist, "The I of The Perceiver")

[jens g.r. benthien]

so many words for what?

You know what? I've printed many flyers, info brochures etc. I didn't care about the hair splitting you try to sell us as a necessary science for high quality printing. And the results have been great - be it from RGB TIFFs or Xara X *.eps.

Maybe you will be able to explain to me why my print job turned out to be fantastic without all the scientific arguments...

I think I've got RESULTS because I just did it instead of discussing some weird scientific issues about dot gain, UCR, over, side. horizontal, vertical printing for weeks or even months. Or is this discussion already running for years by now? Sorry that I can't remember. I just printed the stuff I needed.

My personal conclusion: if you want results, go for Xara. If you want to split hairs, go for Adobe products and join a discussion group or forum to exchange formulas, test runs, rumors, speculative ideas, fantasy - you name it.

Simple formula, eh?

http://www.talkgraphics.com/images/smilies/biggrin.gif

jens g.r. benthien
designer
http://www.sacalobra.de

----------//--
If you don't know how to dream you'll never be a designer.
----------//--

[FAx]

Klaus, you right. I make some tests and see no way make PDF without RGBcolors. It so strange.

Im use FlightCheck 5.02 for check Xara EPSAI - test passed, no RGB. I try to make PDFX - Distiller yelling about wrong color space. I make standart PDF and go validate it - he full of RGB colors.. Whats wrong?!!

Today im no have Illustrator, tomorrow im try to make native Illustrator EPS (gradient) and resave Xara EPS in Illustrator. Probably, im understand where that problem..

Jens, rest in peace and don`t worry about cmyk, eps, gradient troubles. http://www.talkgraphics.com/images/smilies/wink.gif

So!, FAx
www.artin.ru | fax@goody.ru | icq 39779414
www.livejournal.com/users/faxenoff

[jens g.r. benthien]

great that you want to prove some useless theory in the pre-press process.

It's obvious that you only use software to check the Xara output. Software can and does have bugs, even if it carries the brand 'Adobe'.

I doubt that you guys have a Linotronic Hell imagesetter with a full fledget RIP in your livingroom or home office to run the 'tests'. Unless you examine the final film or printing plates you can't tell anything about the result. As long as you guys use a software to analyse the output, you only **think** you are on the right track.

It's hilarious if not amusing for me to see how you try to tell us some buggy truths. I've sent your posts to the printing company I'm currently working with, and all they did was LOL, LOL, LOL. And the same question I put up here: please explain why we can print and get a result if it's doomed to fail based on theory?

Right they are. Monday they will run another file from me (made with Xara X, including bitmaps, gradients, shadows, bevels and all the nine yards) through the image setter and produce another great set of CMYK films for printing a flyer.

A flyer of course that can't exist because you guys know better. So I will deliver a non-existing flyer to the customer.

Really interesting how you guys deliver a proof without a Lino equipment.

Good luck with your theoretical knowledge. It feels so good to be more practical oriented and to be able to deliver results instead of beating a dead fly.

jens g.r. benthien
designer
http://www.sacalobra.de

----------//--
If you don't know how to dream you'll never be a designer.
----------//--

[FAx]

Jens, make simple experiment.
1.Open Photoshop-Create New File(any size)
2.Set Mode CMYK
3.Fill with C:0 M:100 Y:0 K:50
4.Convert to RGB
5.Convert to CMYK
6.Look at color. (C:29 M:96 Y:35 K:33)

First and second colors not equal!
Print with reconverted CMYK and real CMYK look different! First color can printing with two inks (150% ink total), second color - 4 inks (~190% total) and look dirty. Some papers just killing with more 250% inks. Show this topic your friends in print bureau and listen what they say. 8-)

You print flyers on thick paper, and some color difference not critical. If im try to print Bank brochure (where background in corporate color MUST BE C:100, K:20) as is, without checking - bank not accept it, i lose HUGE AMOUNT OF MONEYS!!

That real situation, and not forget about this, if you planned work with big corporates.

Klaus - my test show me:
- Xara EPS contain good CMYK gradient. (tested by Photoshop).
- Distiller make wrong gradient on distill EPS.
- Corel EPS distill to PDF/X without problems.
- Im look into Xara EPS code and see MANY rgb2cmyk, cmyk2rgb and etc commands! (worry, am not know PS language well)

Summary:
Xara have wrong EPS header or special field generation. Colors of objects generate ok.

Workaround:
1.Export to CMX. Open with Corel. Publish to PDF.
2.Export to AI. Open with AI. Publish to PDF.

Im try:
Im download PS Reference and learn PS more.
Try to fix EPS header in Xara DLL resourse file.

Dear Xara Ltd - if that hack worked well, that be illegal of legal action? Can i distribute it?

So!, FAx
www.artin.ru | fax@goody.ru | icq 39779414
www.livejournal.com/users/faxenoff

[jens g.r. benthien]

I don't use Photoshop, I don't use Illustrator. The only product I use is Acrobat to distribute files and samples.

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR> - Xara EPS contain good CMYK gradient. (tested by Photoshop). <HR></BLOCKQUOTE>

Hey, Hey, read this! Klaus stated that Xara X eps can NOT produce gradients.

You know what? Keep discussing and testing forever, this doesn't make sense anymore. If Fedor will pinpoint the bug in the eps header he should inform Charles immediately.

However, rest assured that the PS bug is not in the header, it's in the converted objects after the bounding box definition.

That's what my PS analyzer showed me.

jens g.r. benthien
designer
http://www.sacalobra.de

----------//--
If you don't know how to dream you'll never be a designer.
----------//--