It's official! Ink jet prints equal or surpass Eastman's dye transfer prints!

[slackercruster]

I have completed preliminary tests comparing the finest traditional Eastman Kodak dye transfer prints with machine made dye transfer prints aka as Ink Jet prints. The preliminary tests show the $1.50 ink jet print to equal or surpass the $250 Eastman's dye transfer print.

I will be sending in a full report within a few weeks. But it wont have dye stability tests done. Those will hopefully be in by Dec. (Although Ctein is not holding his breath, since he says my tests are worse than useless!) But for the rest of you that can see farther than your nose, you may find the report interesting.

If you are not familiar with Eastman's dye transfer process see my earlier post on it:

Remembering Dye Transfer Color Printing - PentaxForums.com

 

[unpopular]

Interesting. What methods did you use to run the tests?

 

[gsgary]

Bull**** they will last like a proper print

 

[slackercruster]

I will go over it all when I write the report. I will send in photos and scans when the report is done.

Basically I scanned a high grade dye transfer print. Processed the TIFF scan files into 3 exposures with LR. Then turned the 3 images into 1 fused image using Photomatrix. (Now, I'm not talking freaked out tone mapping. I am talking about *invisible*, very slight HDR that just adds depth.)

The resulting ink jet print from that 3 into 1 image equaled or surpassed the dye transfer print with IQ. Without the 3 into 1 process, the scan of the dye transfer lacked some depth and transparency. (This again confirms that you can get some HDR effects from 1 RAW image processed into 3 images with exposure adjustments in PP.)

I used a Canon 5 ink color printer from Walmart that cost $100. I used Epson 4 star pro paper. The printer was set to the highest quality printing setting.

The best way to have done this test would have been to spend a few thousand $$ for the finest Epson printers. Buy the Epson $700 ink packs and run a shootout with a hand rolled DT print made by Bob Pace and use the same the original chrome or negs he used for the DT to make the ink jet print. But I was not able to do that. So this was what I came up with and the results were astounding...even with a $100 Walmart 5 ink printer.

So, if my half ass setup will rival the finest Eastman DT prints I feel very comfortable making this announcement.

 

[slackercruster]

Bull**** they will last like a proper printe

Don't know what you mean?

But if your taling about ink jet archival dye stabilty...they are very good.

If you are talking about water resistance to running. Trad DT's can't compare to ink jet. They run all over hell.

Full report on the dyes to come out about Dec '12.

 

[unpopular]

Dye Transfers aren't terribly archival, iirc. Acid-fixed dyes are not lightfast, inkjet PIGMENT inks are probably the most archival color process available, coming close to silver if kept in a controlled setting.

I'm interested in exactly what you're measuring, is it archival quality or image quality. I have a hard time believing that ink jets are "as good as" dye transfer by any quantifiable measure (other than archival quality), perhaps a statistically proven qualitative study could be carried out, though you'd need a large sample.

 

[compur]

Basically I scanned a high grade dye transfer print. Processed the TIFF scan files into 3 exposures with LR. Then turned the 3 images into 1 fused image using Photomatrix. ... The resulting ink jet print from that 3 into 1 image equaled or surpassed the dye transfer print with IQ. Without the 3 into 1 process, the scan of the dye transfer lacked some depth and transparency.

Huh? Scanned a dye transfer print, then printed the scan on an ink jet and then compared the ink jet print with the original dye transfer? Huh?

 

[unpopular]

^^ yeah, I didn't read that right.

If so, this is a pretty worthless study, imo. There are ways to do this kind of thing, but this isn't it.

For a quantitative analysis, you would use a standard target and take microdensitometry measurements, comparing various qualities such as grain/dot size, grain/dot accutance, grain/dot pitch, grain/dot frequency, etc. I'm not really sure what that would prove as far as overall "quality" - all you could say is "the dye transfer had a higher grain pitch" or "the ink jet had more accutance".

For a qualitative analysis you could print the same image using inkjet and the another using dye transfer on identical paper stock, and poll a large sample of a technique-niave population asking them to determine which print looks better in a controlled setting. You'd want to use a range, probably scoring each print like 1-10.

Ideally you'd want to control for "image/process" bias, where one image for, whatever reason, just looks better in one process than the other. But that would get complicated fast.

One problem using something like Photomatix is that there is no way for peers to analyze what was done to the image, and if that processing has any effect. If you used something non-proprietary and open, such as enfuse or reinhart, we could look at the algorithm to determine if processing played a roll. ATM we'd just have to take your word for it.

 

[slackercruster]

Basically I scanned a high grade dye transfer print. Processed the TIFF scan files into 3 exposures with LR. Then turned the 3 images into 1 fused image using Photomatrix. ... The resulting ink jet print from that 3 into 1 image equaled or surpassed the dye transfer print with IQ. Without the 3 into 1 process, the scan of the dye transfer lacked some depth and transparency.

Huh? Scanned a dye transfer print, then printed the scan on an ink jet and then compared the ink jet print with the original dye transfer? Huh?

Yep. Without the original chromes / negs...that is how you do it.

 

[slackercruster]

^^ yeah, I didn't read that right.

This is a pretty worthless study, imo. There are ways to do this kind of thing, but this isn't it.

For a quantitative analysis, you would use a standard target and take microdensitometry measurements, comparing various qualities such as grain/dot size, grain/dot accutance, grain/dot pitch, grain/dot frequency, etc. I'm not really sure what that would prove as far as overall "quality" - all you could say is "the dye transfer had a higher grain pitch" or "the ink jet had more accutance".

For a qualitative analysis you could print the same image using inkjet and the another using dye transfer on identical paper stock, and poll a large sample of a technique-niave population asking them to determine which print looks better.

One problem using something like Photomatix is that there is no way for peers to analyze what was done to the image, and if that processing has any effect. If you used something non-proprietary and open, such as enfuse or reinhart, we could look at the algorithm to determine if processing played a roll. ATM we'd just have to take your word for it.

You are welcome to do the test to fit your needs. This is the test I am offering. Sure, it is jsut an informal test. Half ass done. But that is an important part of the tesdt as well. If the reults produce a DT rival, then a better process should yield even better results.

Either a printing medium can reproduce a print to equal a DT or not. That is the question. I could run the same scanned image through type R and it would look like crap. You use your eyes not a machine to decide, not science. Now, if your pixel peepng, then by all means take the test to the lab.

But thanks for the idea you have given me unpopular...

I will not post which print or scan is which. I challange any of you to tell which is the $1.50 ink jet and which is the $250 DT print. You wont be able to. Even if you saw them in person you could not tell unless you looked at the back of the prints or put them right up tom your nose and compared certain sharpness details. I've polled a few people on this in person...100% results show no one can tell.

Now, no scan can be as good as the original. But these scans, the way I processed them, come very close to 98% to 99% perfect copies.

 

[unpopular]

You can't expect a scan of a print to represent the print itself!

-----

Listen, I really appreciate this study, and I agree that a quantitative analysis isn't appropriate. But what you are saying, it's total garbage, and doesn't conclude anything aside from your impressions about dye transfer and inkjet. Why should I care about that?

 

[unpopular]

Also, even if you had the statistical data to back up your claim, all you have proven is that this particular DT print can be reproduced in inkjet.

 

[slackercruster]

OK, no problem. When I submit my test, you can skip wasting any time on it.

 

[unpopular]

don't be annoyed with me, I'm not the one who made the sweeping statement.

One important part of experimental design is knowing your limitations. This is more of a case study.

 

[slackercruster]

don't be annoyed with me, I'm not the one who made the sweeping statement.

One important part of experimental design is knowing your limitations. This is more of a case study.

That's OK. I don't mind defending it some. I wont argue about my limitations. I got plenty of them. My problem is that of time. I don't have time to fool around discussing or doing exhaustive tests. I just do what I can, put it out there and let people pick away at it if they like.

I also got ADD and get bored discussing things on and on. Which is about where I am at with this thread. And even with no ADD, I have a ton of things I am involved in and can only spare so much time to any one thing.

Anyway, when the tests come out, hopefully in the next week, you check it out and give some feedback on it if you like. I will invest more time in that thread to answer questions.

And keep a look out for my dye stabiltiy tests to come out hopefully in Dec. It will discuss the ink jet / DT dyes as well as many other color imaging and printing media. Nothing about science, just down and dirty tests with 6 months of sun exposure...with full photos to back up my sweeping claims!

PS:

The 'worse than useless' claim by Ctein is about my dye stability tests, not the claim made in this thread. The Yahoo dye transfer group seemed to have banned me, so I could not discuss this thread there. Seems Ctein is allowed to dish it out, but I am not allowed to answer back.