iGen Customer has some questions

[Tasguy]

I am just about to have 50 copies of book printed using an iGen. Sheet size is 520 x 280, and the print run will be about 10,000 sheets. The file will be given to the printer in PDF format, already imposed, ready to print. Nothing will need to be done to the file. Nothing. It's as if I took the file to a local photocopy shop and said "Print". If it comes out wrong, there's no one to blame but me. I'm paying for the proofs up front, and by doing it this way I expect to get a pretty good deal.

In 2006 I had test prints run off, before I started laying out the book, to make sure that the iGen (and NexPress and Indigo) were of suitable quality. The iGen came out on top. My comments from the iGen tests will be posted as a reply.

I had a first proof run off in December 2008, but an hour or so before printing the printer rang me to say the price had doubled, through no fault of mine. He agreed to print the proof, which I would pay for, and then I was free to seek new quotes, including from him.

So I have a book now, but no printer. I have recently approached a few printers for indicative costs and asked them some detailed questions before I make a decision who to go with. I was disappointed at the lack of knowledge/conflicting answers, and so I thought I might pose my questions here to get some definitive answers. These are some of the questions I posed to the printers:


QUES 1
I wanted to know the print settings used on the first-proof, so that come second-proof and final-print, the settings could be reproduced exactly. I asked: was it possible at the end of the printing to save the print settings. The answers I received were split 50/50. Some said they could, some said they could not. The most knowledgeable sounding said: "I will make a Q with all the write settings attached to it."

So, can an iGen operator save print settings and, say, email them to a customer as a backup, so that next time the customer comes along with a repeat print, the settings are identical?


QUES 2
Some of the impressions for my book have colour (by impressions I mean one side of a sheet), some are true grayscale. The book is contained within one single PDF file. They will be no question of CMYK appearing on pages where there should be none. I will be thoroughly checking for colour on every page using Acrobat before it is sent to the printer.

I asked the question: "Will the IGen accurately report on the total number of colour impressions and grayscale impressions, or will I be billed as if every impression was colour. These are some of the answers I received:

1. "I am sorry to say, but when we print a particular set in color, the whole file is sent in color mode... If we try to separate pages, we will have to break the file in pieces and that can be a very lengthy process and is not recommended."

2. "The iGen does not report this information. However, in setting up the print run we can ascertain the number of B&W pages and colour pages and the total printed for each."

3. "The IGEN does not specify which pages are colour and which are B&W"​

Two other responses contradicted the above and said the iGen does keep track of colour/grayscale.

So, as a customer, should I expect a printer to offer me a price that reflects the actual number of colour impressions, and the actual number of grayscale impressions? Printer 1 is suggesting I have to pay the colour rate for all impressions, even though most are grayscale.


QUES 3
I asked about colour profiles. All my images have embedded profiles, but every one of the printers said they would ignore profiles. That surprised me, but I'm quite happy about not using profiles. The December 2008 proof, printed on Direct setting, was almost identical to what I see on my iMac screen. A surprisingly close match. I'm very impressed with the quality of the iGen and with the correlation between screen and print.

The first proof had a number of RGB images which I had not bothered to convert to CMYK. The printer said that was okay. The grayscale images would go through one stream, CMYK through another, and RGB through another.

My question is: when the RGB images are converted to CMYK by the iGen, what source and destination colour spaces are used? My images had been stripped of their embedded profiles, so the iGen must have assumed a source profile. What profile? sRGB?

If I want to convert the RGB images to CMYK myself, what destination colour space should I select? I want to mimic what the iGen does when it converts RGB to CMYK. Or is the destination profile a proprietory iGen profile that is not available to the public?


QUES 4
Do images that have resolutions higher than 300 ppi (say 600, 1200) have to be downsampled to 300 ppi? Do higher resolutions cause problems, or do they just slow down the ripping? I'd rather not downsample my images if it is not necessary to do so, given that the iGen effectively resamples anyway.


FINAL NOTE
I would appreciate any comments on the results of my tests (see reply to this post).

 

[Tasguy]

Summary of Test Prints using Xerox 7750 and iGens (July 2006)
The Xerox 7750 is capable of giving almost the same quality prints as the iGen. However, it is not colour-calibrated. I have tested four 7750s and the colour cast ranges from too green to too red. The iGen does not have this problem. I suspect the 7750 can give accurate colour rendition but it would need to be calibrated regularly. The colour casts I have seen are actually a good indication that the 7750 can give accurate colour – the variations says that the colours can be adjusted. It would be an indication of a poor machine if the colour was the same each time -- and wrong.

iGen Results
The half-tone frequency and angles for the C, M, Y and K toners, change for each resolution setting (150, 175, 200). The half-tone dots for each toner are laid out in a grid on two axes at right angles, and along those axes the resolution is not the same, even for the same overal resolution setting. eg for the black toner at 175 linescreen, the horizontal resolution is 175, but the vertical resolution is 150.

Frequencies and Angles
(Screen for the iGen is half-tone dots; for the 7750 it is lines)

Colour.....iGen 175 resolution........iGen 150 Resolution...........Xerox 7750
Cyan..........150 @ 40º, 175 @ 130º.......125 @ 73º, 135 @ 163º............190 lines @ 0º
Magneta....175 @ 58º, 150 @ 148º.......125 @ 13º, 135 @ 103º ............135 lines @ 135º
Black.........150 @ 90º, 175 @ 0º............135 @ 47º, 135 @ 137º ............135 lines @ 45º
Yellow.......Screen is not discernable - appears as blobs for the iGen and 7750​

Screen Problems at 150 resolution for colour
The screening became visible at 150 resolution because of an interaction between the C, M, Y and K screens. It is a similar problem to Moire patterns. What happens is that the screen frequencies that the iGen uses at 150 resolution interact to form visible lines at about 60 to the inch. Small regularly-spaced colour cells are formed by the C, M, Y and K screens. The effect was only obvious in the darker areas and was not visible when the image was magnified. This could be fixed if the iGen allowed user-selected line-screen frequencies; but the effect may then reappear in different areas of the image. This problem may not be correctable. DO NOT use 150 linescreen for colour images.


150 vs 175 resolution for Grayscale Images
Grayscale images showed a slight improvement in gradation in the highlight areas at 150 screen versus 175 screen, with no other discernable image quality loss. I could only detect the improvement in highlight areas where, for the 150 screen image, the gradation was smoother with less contouring. The improvement is very small, and difficult to detect, but it was definitely there. DO USE 150 linescreen for grayscale images, in preference to 175 linescreen.


Text Colour
For the iGen, 100% cyan text printed as about 90% cyan. For the 7750, coloured text printed as pure colour.


Low-Density Grays
The iGen appears not to be able to print densities less than about 3%, but it varied between the machines I tested. The clearest indication of this was the yellow gradient. The grayscale version of the yellow gradient cut out 80 mm from the 100% end on a 111 mm gradient. This equates to 28% yellow density. (A different iGen gave 13%.) A yellow level of 28% is equivalent to 3% grayscale (measured using the eyedropper on the original image), i.e. a level of 247. That means I should limit the range of levels when printing a B&W image in CMYK to no more than level 247.

Similarly, I checked the K-only gradient. It cut out 92 mm along a 96.5 mm gradient i.e. 4% density. Thus, to avoid the problems of highlights appearing, I should probably limit the white level to 5% density; ie level 242, say 245.


Low-Density Cyans
In the Blue Bird Cafe colour image (in CMYK, but actually a B&W image) there are highlights visible in the awning. These occur at low densities where yellow toner is primarily being used to achieve the low density required. (To retain the overall gray appearance, the yellow is balanced by magenta and cyan as you move away from the extreme highlight.) The highlights occur because cyan is introduced at higher densities and forms a boundary that becomes visible. Like the height contours on a map, if the cyan is introduced at a certain density (say 5%) it will form a contour where it first appears. And this contour is visible. This effect is only noticeable at low densities and occurs between where there is no cyan, and when the cyan is first introduced.

The highlight was not noticeable in the same area of the CMYK image converted to grayscale, which was covered by small half-tone dots.

This problem in the colour version of the image is not due to the low density (say less than 3%). It is caused by the CMYK mapping trying to achieve gentle gradation by introducing cyan. This causes a noticeable level jump – a contour – because cyan is significantly darker than yellow.


Converting to Grayscale using the iGen
Do not use the iGen to convert colour images to grayscale because it converts all gray densities above 90% to 100%. ie dark shadows end up as 100% solid black shadows with the accompanying sheen. I measured this figure from the K-only gradient. It appeared solid black for the first 9mm of a 96 mm black-to-white gradient. All densities between 100% and 91% ended up as 100% for the three iGens tested.

 

[X33]

wow, that's a lot of research to print some books.

answers :

1. Depending on the RIP being used with the iGen, I think the operator can export a "Job Ticket", it will have all the print settings.

2. Depending on the RIP, it shouldn't be a problem. The RIP should be able to differentiate between Color and Black pages in the same file and therefore the iGen will "click" correctly.

3. Again, depending on the RIP, it could be sRGB, or Adobe RGB, or EFI RGB, or maybe something else that the particular printer has as a default.

4. 300dpi is enough for digital printing. A higher resolution file slows down the processing and you might not appreciate the difference in quality at all.

 

[Craig]

Some people must have a lot of time on their hands.

If you need to be that exact then send it to an offset press or you'll never be happy.

 

[Joeutut]

Sounds like someone has a lot of school experience and very little actual experience. Should go work in a print shop and see how education and practical experience need each other. And, way too much research for a book. Hi end art repros don't require that much although the designer thinks so

 

[muminn]

Dear Tasguy,

I will recommend you a prepress expert. And because the digital is still "digital", your high quality book should be printed in offset for your satisfaction.
So, my point is that you can't ask for offset's printing standards in a digital printing! Simple, still it's not in that stage, after a decade of innovation.

Regards,

 

[muminn]

Try to test in HP Indigo

Try to test in HP Indigo

Try testing HP Indigo 5500. I'm interested in results.

Summary of Test Prints using Xerox 7750 and iGens (July 2006)
The Xerox 7750 is capable of giving almost the same quality prints as the iGen. However, it is not colour-calibrated. I have tested four 7750s and the colour cast ranges from too green to too red. The iGen does not have this problem. I suspect the 7750 can give accurate colour rendition but it would need to be calibrated regularly. The colour casts I have seen are actually a good indication that the 7750 can give accurate colour – the variations says that the colours can be adjusted. It would be an indication of a poor machine if the colour was the same each time -- and wrong.

iGen Results
The half-tone frequency and angles for the C, M, Y and K toners, change for each resolution setting (150, 175, 200). The half-tone dots for each toner are laid out in a grid on two axes at right angles, and along those axes the resolution is not the same, even for the same overal resolution setting. eg for the black toner at 175 linescreen, the horizontal resolution is 175, but the vertical resolution is 150.

Frequencies and Angles
(Screen for the iGen is half-tone dots; for the 7750 it is lines)

Colour.....iGen 175 resolution........iGen 150 Resolution...........Xerox 7750
Cyan..........150 @ 40º, 175 @ 130º.......125 @ 73º, 135 @ 163º............190 lines @ 0º
Magneta....175 @ 58º, 150 @ 148º.......125 @ 13º, 135 @ 103º ............135 lines @ 135º
Black.........150 @ 90º, 175 @ 0º............135 @ 47º, 135 @ 137º ............135 lines @ 45º
Yellow.......Screen is not discernable - appears as blobs for the iGen and 7750​

Screen Problems at 150 resolution for colour
The screening became visible at 150 resolution because of an interaction between the C, M, Y and K screens. It is a similar problem to Moire patterns. What happens is that the screen frequencies that the iGen uses at 150 resolution interact to form visible lines at about 60 to the inch. Small regularly-spaced colour cells are formed by the C, M, Y and K screens. The effect was only obvious in the darker areas and was not visible when the image was magnified. This could be fixed if the iGen allowed user-selected line-screen frequencies; but the effect may then reappear in different areas of the image. This problem may not be correctable. DO NOT use 150 linescreen for colour images.


150 vs 175 resolution for Grayscale Images
Grayscale images showed a slight improvement in gradation in the highlight areas at 150 screen versus 175 screen, with no other discernable image quality loss. I could only detect the improvement in highlight areas where, for the 150 screen image, the gradation was smoother with less contouring. The improvement is very small, and difficult to detect, but it was definitely there. DO USE 150 linescreen for grayscale images, in preference to 175 linescreen.


Text Colour
For the iGen, 100% cyan text printed as about 90% cyan. For the 7750, coloured text printed as pure colour.


Low-Density Grays
The iGen appears not to be able to print densities less than about 3%, but it varied between the machines I tested. The clearest indication of this was the yellow gradient. The grayscale version of the yellow gradient cut out 80 mm from the 100% end on a 111 mm gradient. This equates to 28% yellow density. (A different iGen gave 13%.) A yellow level of 28% is equivalent to 3% grayscale (measured using the eyedropper on the original image), i.e. a level of 247. That means I should limit the range of levels when printing a B&W image in CMYK to no more than level 247.

Similarly, I checked the K-only gradient. It cut out 92 mm along a 96.5 mm gradient i.e. 4% density. Thus, to avoid the problems of highlights appearing, I should probably limit the white level to 5% density; ie level 242, say 245.


Low-Density Cyans
In the Blue Bird Cafe colour image (in CMYK, but actually a B&W image) there are highlights visible in the awning. These occur at low densities where yellow toner is primarily being used to achieve the low density required. (To retain the overall gray appearance, the yellow is balanced by magenta and cyan as you move away from the extreme highlight.) The highlights occur because cyan is introduced at higher densities and forms a boundary that becomes visible. Like the height contours on a map, if the cyan is introduced at a certain density (say 5%) it will form a contour where it first appears. And this contour is visible. This effect is only noticeable at low densities and occurs between where there is no cyan, and when the cyan is first introduced.

The highlight was not noticeable in the same area of the CMYK image converted to grayscale, which was covered by small half-tone dots.

This problem in the colour version of the image is not due to the low density (say less than 3%). It is caused by the CMYK mapping trying to achieve gentle gradation by introducing cyan. This causes a noticeable level jump – a contour – because cyan is significantly darker than yellow.


Converting to Grayscale using the iGen
Do not use the iGen to convert colour images to grayscale because it converts all gray densities above 90% to 100%. ie dark shadows end up as 100% solid black shadows with the accompanying sheen. I measured this figure from the K-only gradient. It appeared solid black for the first 9mm of a 96 mm black-to-white gradient. All densities between 100% and 91% ended up as 100% for the three iGens tested.

 

[GusG]

QUES 4
Do images that have resolutions higher than 300 ppi (say 600, 1200) have to be downsampled to 300 ppi? Do higher resolutions cause problems, or do they just slow down the ripping? I'd rather not downsample my images if it is not necessary to do so, given that the iGen effectively resamples anyway.

Rule of thumb is that the image dpi should be 1.5 to 2 times the screen (lpi) you are printing at. Anything higher than that does not add anything to the quality of the printed image. On the contrary, the downsampling in the RIP doesn't only take up processing time it is also very likely that you will loose some sharpness in the process. Much better to do that yourself where you have full control.

 

[billyinlasvegas]

1. Depends on the RIP but it can be done. Now giving these settings to a customer might not be the easiest thing. The good news is that most operators try to use the same settings on all jobs unless something has gone wrong.

2. In the print for pay world I don't know of anyone (locally) that will charge you a lower price for b/w pages run on a 4 color digital device even though it costs them less money.
Also the iGen can keep track of b/w vs. CMYK pages but it will also depend on the RIP and the operator.

3. This also depends on the RIP. If you want they can use your embedded profile.
Also since an iGen has a different color spectrum than traditional CMYK I know lots of people that prefer RGB files and let the RIP do all the work. The iGen color profiles are easy to get just ask your service provider and they'll give them to you.

4. Since this is a digital press 300ppi is fine anything else is overkill and don't worry about slowing down the RIP it can handle almost anything you throw at it.

Like many have said if you're this concerned with quality and want that much control over how it's printed go with offset printing. Also try an Indigo.

Also for most shops this is a nightmare scenario, just trust your print shop.
If they own an iGen (they are VERY expensive!) they probably have a good operator that can give you good results.

 

[chiefy]

We often run short runs of high end art books on iGen while a large offset run is in production in order to meet a public launch or such. The offset and particularly binding process may take some time so the customer will also require say 50 items printed digital on the iGen and bound. On normal jobs the igen operator controls the profiles based on stock types and jobs but in order to match our offset standards we have customised an ISO profile to create a 'press matching' colour profile that we have tested against our XL105, CD74 and our 102 perfectors. Its not perfect but its acceptable for the product and the coour is close but the image quality and screens are no match for 250lpi hybid screening.

Like the previous reply i dont know many operators that will charge bw rate on the igen for a colour job that contains bw. If you want bw we would run that on one of our mono digital printers and they have their own charge/click rate for quoting or run mono on the igen. The iGen would be quoted at colour. It does have the different click charges for bw/colour but if the job was mixed throughout it would all be run as colour, otherwise the collating time and cost would go up significantly. If your book was split simply it could be done, ie pages 1-32 colour, 33 onwards bw.

 

[leetaylor]

We often run short runs of high end art books on iGen while a large offset run is in production in order to meet a public launch or such. The offset and particularly binding process may take some time so the customer will also require say 50 items printed digital on the iGen and bound. On normal jobs the igen operator controls the profiles based on stock types and jobs but in order to match our offset standards we have customised an ISO profile to create a 'press matching' colour profile that we have tested against our XL105, CD74 and our 102 perfectors. Its not perfect but its acceptable for the product and the coour is close but the image quality and screens are no match for 250lpi hybid screening.

Like the previous reply i dont know many operators that will charge bw rate on the igen for a colour job that contains bw. If you want bw we would run that on one of our mono digital printers and they have their own charge/click rate for quoting or run mono on the igen. The iGen would be quoted at colour. It does have the different click charges for bw/colour but if the job was mixed throughout it would all be run as colour, otherwise the collating time and cost would go up significantly. If your book was split simply it could be done, ie pages 1-32 colour, 33 onwards bw.

That's identical to how we price up our Digital work for our Docucolor 5000.