Imageprint - worth the $ ? - compare to QImage? other views please!

[ErikKaffehr]

Hi,

Sorry, what is Measure Tool? I presume it belongs to some toolset?

Best regards
Erik

Printed then a 16 patch strip, which is built in Measure Tool.
The criteria is set like this:

Tomorrow I will do the same test with custom made profiles.
/Sven

[enduser]

Given that there is such a thing as male color blindness, I assume that all comments such as "a tendency for blues to shift magenta"  are based on hardware tests, not just eyesight.

[Ernst Dinkla]

Given that there is such a thing as male color blindness, I assume that all comments such as "a tendency for blues to shift magenta"  are based on hardware tests, not just eyesight.

Would hardware in the sense of two prints made with different profiles and viewed next to one another, under the same lamp and with one set of eyes be enough ?


met vriendelijke groeten, Ernst Dinkla

spectral plots of +100 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[Sven W]

Hi,

Sorry, what is Measure Tool? I presume it belongs to some toolset?

Best regards
Erik

It´s a module in the Profile Maker package. By X-Rite.
And will be replaced / added with the new i1 Profiler software, soon to be released.
/Sven

[Martin Kristiansen]

I had a Epson 4800 and took a chance and spent the money on IP. It was expensive but the combination of the work flow and what I considered to be improved quality made me pleased with the purchase. The acid test was when I upgraded to a Epson11880. The IP price was huge. I sold the 4800 with the RIP and purchased a new IP rip for the 11880. I guess that says something for me putting my money where my mouth is.

One of my issues not using the RIP was getting profiles from the various paper manufactures and they were all different. I would get a print just the way I like it on one paper then want to see it on another paper and the prints would be totally different. The problem is different manufacturers make profiles differently and with IP the profiles are all made in pretty much the same way.

As to the quality it contains a large subjective element but I like the quality way ahead of what I get from epson.

Ease of use and repeatability and an end to struggling with settings ended for me with IP. 

[alfin]

I really got confused the other day with all the discussions about Imageprint and people like Andrew claiming that the Epson print driver with canned profiles, print as good as or better than ImagePrint.

I haven’t used the Epson print driver for my Epson Stylus Pro 3800 for a long time and I started to think that something must have changed, new print driver software, new Epson profiles or just the fact that the printer manufacturers’ software now has catched up with the RIP suppliers. I just have to try for myself.

I made the prints using a modified Bill Atkinson test image you can find at Outback Photo:

http://www.jirvana.com/printer_tests/PrinterEvaluationImage_V002.zip

I have used this image many times before when I have compared papers and profiles and I’m familiar with it and I know what to look for.

The only Epson inkjet paper I have available is Epson Enhanced Matte, and since so many seems to agree that Epson profiles are good, I chose to use this paper. I first printed through Lightroom, using the canned Epson profile and then I printed through ImagePrint and compared both printouts under my Solux lamp. I got even more confused.

I printed the test image again on Innova White Gloss and Harman Gloss with its standard profiles and ImagePrint profiles and the results were the same. I also printed some of my own images with standard sharpening applied in LR, just to make sure it was correct what I saw.

The printouts from Imageprint is consistently sharper with more details, better micro contrast and however I try, I cannot find any obvious color shifts towards purple or red towards orange. The sky through the arch in the test image has a slight magenta shift, so don’t get confused by that.

This test has cost me some papers and ink, but to my eyes there is definitely not a chance in the world that Epson’s print driver with Epson canned profiles beats Imageprint. At least not for my printer model 3800.

Maybe the newest Epson printers have much improved drivers and profiles, I don’t know. I hope so, otherwise I really don’t understand why so many here are bashing a product that anyone taking the time to compare, easily can see for themselves is better.

[digitaldog]

The sky through the arch in the test image has a slight magenta shift, so don’t get confused by that.

Its not supposed to, its supposed to be blue, and if you see a magenta shift, well bingo, you’re seeing the shift (which in large part, is a Lab issue). That’s primarily why that image is in this test suite.

What even better test images for seeing the shift? There are these (at a price but well worth it):
http://www.roman16.com/en/

[Sven W]

Its not supposed to, its supposed to be blue, and if you see a magenta shift, well bingo, you’re seeing the shift

You must be kidding?
The sky in the arch is in ProPhotoRGB: 116, 108, 240. That is quite a magenta shift in blue in the FILE.

/Sven

[Ernst Dinkla]

Its not supposed to, its supposed to be blue, and if you see a magenta shift, well bingo, you’re seeing the shift (which in large part, is a Lab issue). That’s primarily why that image is in this test suite.

That blue in the arch is really on the edge. I have some test prints hanging here with that reference image. Starting from ProPhoto and sRGB, Ilford IGSPP11 printed on the Z3200 with Qimage Studio about a year ago. Perceptual rendering of the Ilford profile, a HP Color Center made profile, a HP APS made profile. The last shows slightly more magenta there with ProPhoto. All show it more in sRGB and the APS again most. I used another gradients target of Atkinson with the circles and it was clear that APS didn't handle the blue-purple-magenta as nice. It was a reason for me to avoid APS after that. But I'm sure that many wouldn't object or even see that difference.

met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[Ernst Dinkla]

The printouts from Imageprint is consistently sharper with more details, better micro contrast and however I try, I cannot find any obvious color shifts towards purple or red towards orange. The sky through the arch in the test image has a slight magenta shift, so don’t get confused by that.

The thread subject has Qimage as a reference. What kind of resampling and sharpening IP does under the hood compared to what  Lightroom does has to be considered too.


met vriendelijke groeten, Ernst Dinkla

spectral plots of +100 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[Bart_van_der_Wolf]

You must be kidding?
The sky in the arch is in ProPhotoRGB: 116, 108, 240. That is quite a magenta shift in blue in the FILE.

The original(?) Bill Atkinson's "Lab Test Image.tif", is in LAB colorspace, if I'm not mistaken. Could the conversion to ProPhotoRGB have introduced a shift to magenta? Your RGB values for Red and Green are significantly lower and overall a bit darker than the ones I get (121,114,246) in the secondary readout of the Lab file version. Conversions from Lab to RGB are prone to this magenta.

Cheers,
Bart

[GeoffM]

As has already been mentioned, obviously the canned IP profiles are built with a specific batch of paper and a specific ink set. I've seen some pretty substantial paper variance form time to time, but what about ink sets? Do they age, and does Epson (or any other manufacturer) make any claims about variance from batch to batch?

Guess I'm wondering if custom profiling, no matter which printing solution you use, is going to outweigh the mechanical advantages (ink laydown, dithering, etc.) that IP offers? (I know, I know, it's up to me to figure this out, but I'm hoping someone knows about the ink variance part of the equation.)

Thanks,

Geoff

[digitaldog]

The original(?) Bill Atkinson's "Lab Test Image.tif", is in LAB colorspace, if I'm not mistaken.

You are indeed correct and when you read that data in that space (and the sampling values play a role, I used 5x5) show ProPhoto at:137/138/255. If you also examine the blue areas of the yellow flower image, they too produce very similar ratio: 118/119/214 in ProPhoto. The Lab value ratio’s back this up too. And the soft proof in Lab, at least on this end on an NEC 3090 doesn’t appear at all magenta! Again, the reason Bill placed this image into the test suite is because its prone to shift magenta which is not desired.

[digitaldog]

That blue in the arch is really on the edge.

Yes it is, but its blue. The clouds and sky below, that’s a different story. Bill didn’t pop this in to the test suite with anything less than Bill’s keen eye for color and keen mind towards color management!

[BradFunkhouser]

I'm currently testing my new 9900 with ImagePrint 8.0 versus the Epson Driver and available profiles.  I've used ImagePrint since I had a pair of 7600s (one PhotoBlack, one MatteBlack) to a 9800 using the ImagePrint PhatteBlack solution.

Years ago I worked with Bill Atkinson's profiles on the 7600, as well as some custom profiles built by Chromix.  Inability to get good results on Canvas or Somerset Velvet with the 7600 driver led me to invest in ImagePrint.  It was a godsend at the time, as was the ImagePrint PhatteBlack offering with the 9800.

My general observation is that the ImagePrint color engine and profiles like to warm things up a bit versus other profiles (kind of like putting a camera in "portrait" mode).  Makes skin tones nice, but over the years I've regularly had to battle a magenta/red shift in the blues.

My main media are Epson Canvas Satin (formerly PremierArt Water Resistant Canvas for Epson), Somerset Velvet for Epson, Crane Museo Silver Rag, Epson Enhanced Matte, Epson Luster & Semimatte.

Now, with the 9900, I'm in a frustrating quandry.  ImagePrint 8.0 has a new color engine and of course a new set of stock profiles.  They've added a contrast boost that's absolutely killing me.  Even simulating old profiles and adjusting their shadow slider is barely getting me to acceptable looking test prints.  I use test images from Fuji, Andrew, Lindbloom, and a collection of my own images.  With ImagePrint 8.0 and the 9900, I'm battling the loss of shadow detail, which I've never had to do with ImagePrint before.

But the Epson 9900 Driver with their supplied profiles presents a whole different set of problems.  Enhanced Matte and Luster have washed out skin tones, Canvas Satin has yellow cast, overall too light, and a significant red/magenta shift in blues compared to the enhanced matte and luster.  They don't provide a profile for their own Somerset Velvet.  It's a mess from my point of view.

For consistency across media, ImagePrint, even with the contrast problems, is still looking much better than the available profiles for the Epson Driver.

But here's what I don't know...  say I buy the new X-Rite i1/PhotoPro, build and edit my own profiles, or have Andrew or Chromix (or Chen, InkJetArt, BookSmart, etc.) build profiles for me, or get existing profiles from the Chromix or BookSmart libraries, will I be able to get the consistency across media and overall quality that I get with ImagePrint?  What tool would I use to do subtle edits to these profiles?  Would Colorburst Pro or EFI Colorproof XF be better for me?

Working on figuring out the answers.

[eronald]

Those of us with high-end spectros, color expertise and good eyes, can consistently get "their color" with the native drivers, if necessary by adjusting ink densities for new papers and making 5000-patch profiles. However I am willing to bet that Joe User with a Mac cannot get a decent profiled print out of Photoshop anymore, so I can understand that his frustration will lead him to Imageprint, or another RIP, which earn money by being stable and providing decent profiles.

Telling Joe User that it is "his fault" when prints come out too dark or too light also doesn't exactly help. Apart from RIPS, the whole industry model seems to be built around pointing a finger at someone else -Apple? Epson? Adobe? Xrite? and saying that it's their fault, and in the last instance getting an "expert" on payroll to explain to the user he is an idiot.

Edmund

I had a Epson 4800 and took a chance and spent the money on IP. It was expensive but the combination of the work flow and what I considered to be improved quality made me pleased with the purchase. The acid test was when I upgraded to a Epson11880. The IP price was huge. I sold the 4800 with the RIP and purchased a new IP rip for the 11880. I guess that says something for me putting my money where my mouth is.

One of my issues not using the RIP was getting profiles from the various paper manufactures and they were all different. I would get a print just the way I like it on one paper then want to see it on another paper and the prints would be totally different. The problem is different manufacturers make profiles differently and with IP the profiles are all made in pretty much the same way.

As to the quality it contains a large subjective element but I like the quality way ahead of what I get from epson.

Ease of use and repeatability and an end to struggling with settings ended for me with IP.  

[digitaldog]

I am willing to bet that Joe User with a Mac cannot get a decent profiled print out of Photoshop anymore, so I can understand that his frustration will lead him to Imageprint, or another RIP, which earn money by being stable and providing decent profiles.

The large number of users of just Epson and Canon printers who do not use either custom profiles or a 3rd party RIP would probably serve to prove you wrong.

Telling Joe User that it is "his fault" when prints come out too dark or too light also doesn't exactly help.

If they didn’t calibrate their displays properly, used the wrong driver, media settings or profiles, then yes, you can indeed blame them, it is their fault.

[eronald]

The large number of users of just Epson and Canon printers who do not use either custom profiles or a 3rd party RIP would probably serve to prove you wrong.

Quite a few other of these use the Photoshop.Epson "Printer Manages Colors" pipeline which works beautifully for Joe User, but doesn't quite count as profiled printing. Among the guys I met at Photokina, quite a few agreed they couldn't make printing work anymore.

If they didn’t calibrate their displays properly, used the wrong driver, media settings or profiles, then yes, you can indeed blame them, it is their fault.

You and I each have $20K in color equipment. You've written a book on the stuff, I've designed measurement hardware and written software that drives it. In other words, when we look at a setting, we know what it does or at least we think we know what it means. And in fact it is our job to get things right. When things don't work we have the patience to troubleshoot, and alternative components to substitute in the workflow. .

So far I've met just one guy who seems to be able to follow instructions which he does NOT understand. He seems to be a 75 year old pharmacist, and I guess people used to trust him with their lives.

I would suggest that Jane User would profit from having systems that self-calibrate because experience indicates that present color management won't work for her.

Edmund

[digitaldog]

Quite a few other of these use the Photoshop.Epson "Printer Manages Colors" pipeline which works beautifully for Joe User, but doesn't quite count as profiled printing.

Indeed, making your argument even less salient.

[eronald]

Indeed, making your argument even less salient.

I get a lot of emails every day offering me help to make my arguments more salient 

Edmund