Ad
Ad
Ad
Pages: [1] 2 »   Bottom of Page
Print
Author Topic: Eye One Photo Profiles with Current Epsons  (Read 11017 times)
ericstaud
Sr. Member
****
Offline Offline

Posts: 384


WWW
« on: December 11, 2006, 05:56:59 PM »
ReplyReply

Does anyone use the EyeOne Photo setup to create custom profiles for the Epson 2400, 3800, 4000, or 4800?  I am curious about the quality of the results, the need for profiling (whether there is any improvement over canned profiles), and whether the EyeOne setup is as good as services available on the web?

I just don't see many posts about people working with their own print profiling systems.
Logged
ericbullock
Jr. Member
**
Offline Offline

Posts: 67


WWW
« Reply #1 on: December 11, 2006, 06:34:09 PM »
ReplyReply

Quote
I just don't see many posts about people working with their own print profiling systems.
[a href=\"index.php?act=findpost&pid=89943\"][{POST_SNAPBACK}][/a]

Most likely due to the cost involved. I would say you are looking at a minimum of $1000 for a decent spectrophotometer based system. I will also say that any package that does not incluse a spectrophotometer is a waste of time (scanner based solutions, etc).

The quality of the iOne profiles is good, especially when compared to the more expensive solutions. If you like to tinker a lot (or do not want to rely on third party profiles or services), then the iOne is the tool for you.

Regards,

-eric-
Logged
jjlphoto
Sr. Member
****
Offline Offline

Posts: 467


« Reply #2 on: December 11, 2006, 10:16:26 PM »
ReplyReply

I printed some TC9.18RGB targets on my Epson 3800 today, and am letting them dry overnight. Tomorrow I pan to read them with my GMB i1 Pro and Match 3 software. I see there is a v3.6.1 updater available, so I'll install that as well. I will report back.
Logged

Thanks, John Luke

Member-ASMP
ericstaud
Sr. Member
****
Offline Offline

Posts: 384


WWW
« Reply #3 on: December 11, 2006, 11:23:22 PM »
ReplyReply

Thanks John and Eric,

I have used the Eye1 Photo with my R2400 in the past.  The Heavyweight Matte and Premium Luster profiles I created were nearly identical to the canned profiles.  I think the blacks and grays were slightly more neutral, but without a side by side comparison I would never have noticed.  This means I really used the Eye1 Photo to verify how close my particular R2400 was to Epsons standard, and found out mine is pretty close.  

I am not a big fine art printer.  The only 16x20's framed and hanging on walls that I have made were printed in a darkroom.   With the upcoming purchase of a 3800 I thought I would revisit the issue.  My understanding is that the "Pro" model printers are linearized at the factory.  So custom profiles would likely not be necessary or make very little difference versus the canned Epson profiles when using Epson Media.  It seems the real reason for custom profiling is for the use of third party papers.  Does this kind of thinking generally agree with how everyone else views the issue?

It seems to me that the Eye1 Photo at $1500.00 is the the same as the $250 Eye One + 13-20 custom paper profiles.  Not such a bad deal if you do very much printing or experimenting with papers.
Logged
digitaldog
Sr. Member
****
Online Online

Posts: 8107



WWW
« Reply #4 on: December 12, 2006, 09:22:40 AM »
ReplyReply

Quote
Does anyone use the EyeOne Photo setup to create custom profiles for the Epson 2400, 3800, 4000, or 4800?  I am curious about the quality of the results, the need for profiling (whether there is any improvement over canned profiles), and whether the EyeOne setup is as good as services available on the web?

I just don't see many posts about people working with their own print profiling systems.
[a href=\"index.php?act=findpost&pid=89943\"][{POST_SNAPBACK}][/a]

Try the canned profiles from the US first, they are excellent. If you have a lot of papers that are not supported, then it's time to considering building your own if you want to spend the time and money. The quality of the package you mention will do a fine job. You're paying the most for the Spectrophotometer and can upgrade to more robust profiling software in the future if you wish. You can also use the instrument to measure ambient light, strobe, and profile the display, a projector etc.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
jjlphoto
Sr. Member
****
Offline Offline

Posts: 467


« Reply #5 on: December 12, 2006, 09:45:21 AM »
ReplyReply

Quote
It seems the real reason for custom profiling is for the use of third party papers.  Does this kind of thinking generally agree with how everyone else views the issue?
[a href=\"index.php?act=findpost&pid=89990\"][{POST_SNAPBACK}][/a]
 

Many of the more popular third party paper vendors supply profiles on their website at N/C for popular printers (hint-Epson). I have read that some folks have tried the 4800 profiles on their 3800 with good results, but I have not tried that route yet. It souldn't be too long before those paper vendors get those 3800 profiles on their website as that really helps sell the paper.
Logged

Thanks, John Luke

Member-ASMP
jjlphoto
Sr. Member
****
Offline Offline

Posts: 467


« Reply #6 on: December 12, 2006, 11:23:52 AM »
ReplyReply

OK-

Just made some profiles with GMB i1Pro and Match v3.6.1(latest). Targets made for Crane Museo Silver Rag, and InnoveFiba F-Gloss Brilliant White, and dried overnite, and I read them today. Just output some test prints, I need to let them dry a bit, but overall impressions are quite nice. Rich full gamut, nice blacks, the surface is one I have never seen before out of an inkjet printer. Not plasticy, not matte/velvety either. The Museo is warm, as predicted, the Innova is a clean white.
Logged

Thanks, John Luke

Member-ASMP
PetterStahre
Newbie
*
Offline Offline

Posts: 40



WWW
« Reply #7 on: December 12, 2006, 02:59:06 PM »
ReplyReply

Quote
Does anyone use the EyeOne Photo setup to create custom profiles for the Epson 2400, 3800, 4000, or 4800?  I am curious about the quality of the results, the need for profiling (whether there is any improvement over canned profiles), and whether the EyeOne setup is as good as services available on the web?

I just don't see many posts about people working with their own print profiling systems.
[a href=\"index.php?act=findpost&pid=89943\"][{POST_SNAPBACK}][/a]

I currently use the EyeOne Photo package for my R2400 and 7800. There are small, but in my mind significant, differences between standard profiles and the ones you make yourself. For paper from  third party vendors the difference can be even more significant. However, you have to be ready to invest a lot of time in learning/testing.

My main complaint about the EyeOne Photo package is the lack of averaging - reading the charts 2-4 times and averaging the measured values to get better data. I do this for "complicated papers" (often matte fine art papers) using Excel before I continue in the EyeOne software, and it works fine - just obviously not the most efficient workflow. I use averaging when I see banding in gradations from my test prints, and that usually solves it.

The other complaint is that I think (others might help filling in here since I'm also curious) that the maximum EyeOne Photo test chart containing 918 patches is not enough for all papers to make smooth transitions mainly in deep&rich shadows. Again, this concerns mainly matte fine art papers. Glossy ones are quite easy to profile. I've run into this problem in another recent thread (regarding profiling for ImagePrint) where I got help to solve one error (my software settings) but still can't get an enough smooth profile for a specific paper, using the EyeOne Photo package. I think I need larger test charts (more patches) to make an even smoother transition. And for that I need another software.

The alternative is the ProfileMaker 5 Photostudio Pro (from Gretag/X-rite, but there are of course other software) but it's an expensive option. It has averaging, larger test charts and more.

(Note: In Sweden there is a 50% discount if you order the PM5-package at the latest tomorrow (december 13). Last year in december the same discount were offered and it then seemed to be an international campaign. Maybe the same this year and something to investigare before you buy? I will upgrade to PM5 and see where that leads me.)

Cheers,
Petter
« Last Edit: December 12, 2006, 02:59:50 PM by PetterStahre » Logged
jjlphoto
Sr. Member
****
Offline Offline

Posts: 467


« Reply #8 on: December 12, 2006, 03:43:21 PM »
ReplyReply

Petter-

I did notice some banding on my test prints. How do you make a profile with averaging? Actually average out all the numerical data? Wow, sounds like a lot of work, and how do you make the resulting profile then with those new numbers? Anybody here make profiles using the "averaging method"? I find it hard to believe that is what inkjetart profiling service is doing.
Logged

Thanks, John Luke

Member-ASMP
PetterStahre
Newbie
*
Offline Offline

Posts: 40



WWW
« Reply #9 on: December 12, 2006, 04:20:36 PM »
ReplyReply

Quote
Petter-

I did notice some banding on my test prints. How do you make a profile with averaging? Actually average out all the numerical data? Wow, sounds like a lot of work, and how do you make the resulting profile then with those new numbers? Anybody here make profiles using the "averaging method"? I find it hard to believe that is what inkjetart profiling service is doing.
[a href=\"index.php?act=findpost&pid=90118\"][{POST_SNAPBACK}][/a]

The profiling service companies uses PM5 or similar software so they have the ability to use averaging. But whether they use it or not is another question. However, in some cases the difference is very obvious (when comparing profiles created using avering or not).

A little off topic: My basic workflow for "Excel-averaging": 1) read the chart at least 2-3 times but don't create the profile, just save the data. 2) import each file containing the measured data into Excel - using one sheet for each file/set of data*. 3) Duplicate one of the sheets into a 4th sheet .4) On the 4th sheet, let Excel calculate the average of the measurements from the three original sheets, placing each value in the corresponding cell on the 4th sheet, and limit the decimals to 4. 5) Export the 4th sheet into a text file (tab separated). 6) In the EyeOne software, create the profile using the values from your new file instead

(* This is the basic workflow - you might run into small problems when importing the data, at least if you're using "," instead of "." for decimals in your country&software as I do. Then you might have to use a work around in order for Excel not to change the values while importing (!). To check if this problem occurs you can simply open one of the data files outside Excel, in a plain text software (or Word), and check if the data seems to be identical to the data in your sheet.)

And while we're at it ... instead of measuring the same chart 3 times it's better in theory to print 3 charts and measure them once each and then do the averaging. (And then you should also walk backwards in circles around your printer while it prints the charts, 3 times or more, to transfer positive energy to your ink jet!  )

/ Petter
Logged
Dale Allyn
Full Member
***
Offline Offline

Posts: 225


WWW
« Reply #10 on: December 12, 2006, 04:36:32 PM »
ReplyReply

I spoke to Brian at ColorHQ.com this morning and he confirmed that ProfileMaker 5 is available through tomorrow at 50% discount here in the U.S. as well.

--
Dale
Logged

ericstaud
Sr. Member
****
Offline Offline

Posts: 384


WWW
« Reply #11 on: December 12, 2006, 07:54:33 PM »
ReplyReply

It sounds like PM5 is the better option for creating quality printer profiles.  It would be frustrating to spend $1500.00 on Eye1 Photo just to get Matte profiles with banding in the shadows, and limited profile editing options.

I have started another thread under MFDB:

http://luminous-landscape.com/forum/index....showtopic=13639

I think PM5 would be useful to create custom camera input profiles (for my P45) using the SG chart.  I would guess I could emulate Velvia vs Astia, create portrait profiles, or make more creative effects with input profiles.

In the past, I have really liked the Nikon Color Mode III setting to emulate Velvia, but it is only available throught the Nikon software.  With PM5, I might get the same velvia look through C1 pro.
Logged
PetterStahre
Newbie
*
Offline Offline

Posts: 40



WWW
« Reply #12 on: December 13, 2006, 09:35:46 AM »
ReplyReply

Quote
I spoke to Brian at ColorHQ.com this morning and he confirmed that ProfileMaker 5 is available through tomorrow at 50% discount here in the U.S. as well.

--
Dale
[a href=\"index.php?act=findpost&pid=90143\"][{POST_SNAPBACK}][/a]

I just found out it's the PM5 Photostudio Pro that has a discount of 50%, not PM5 Publish Plus which is required for averaging (and to be able to make your own charts). You can buy Photostudio Pro and upgrade the Measure Tool ($$ !) to get averaging at least, but if you would like to make your own test chart that's not enough. This information is from a salesperson in Sweden, maybe Publish Plus is on discount in other countries?
Logged
digitaldog
Sr. Member
****
Online Online

Posts: 8107



WWW
« Reply #13 on: December 13, 2006, 02:29:23 PM »
ReplyReply

Quote
My main complaint about the EyeOne Photo package is the lack of averaging - reading the charts 2-4 times and averaging the measured values to get better data.

What's the average deltaE for the differences here? I suspect you're wasting a lot of time doing this. The EyeOne will scan 100 samples per second. You're getting an average reading per patch (slower, you get more). On a device that's consistent over the sheet, averaging doesn't buy much of anything.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
PetterStahre
Newbie
*
Offline Offline

Posts: 40



WWW
« Reply #14 on: December 13, 2006, 03:22:12 PM »
ReplyReply

Quote
What's the average deltaE for the differences here? I suspect you're wasting a lot of time doing this. The EyeOne will scan 100 samples per second. You're getting an average reading per patch (slower, you get more). On a device that's consistent over the sheet, averaging doesn't buy much of anything.
[a href=\"index.php?act=findpost&pid=90333\"][{POST_SNAPBACK}][/a]

I base this only on practical experience. When profiling papers like the Hahnemuhle PhotoRag BW (for Photoshop) I noticed banding in some vivid colours. After doing 2 or more extra measurements and then averaging the data the banding dissapeared. I don't do this for glossy papers. And I thought PhotoRag to be a quite easy paper to profile, but all I can say is that I've got better profiles after averaging. The same has happened when profiling other paper like Hahnemuhle's William Turner and Fredrix Canvas.

The fact that the profiles sometimes gets better puzzles me but in the end I judge with my eyes. (And I don't like the time consuming process of measuring so I wouldn't do it if I felt I didn't need to.)
Logged
David White
Sr. Member
****
Offline Offline

Posts: 272



WWW
« Reply #15 on: December 13, 2006, 04:10:28 PM »
ReplyReply

I think that it would be informative to run another spreadsheet showing the deltas between the readings you are getting so that you would have an objective vs a subjective measurement.  If you have any software that can calculate delta-e values between measurement files that would probably be very helpful also.  As Andrew said, you are already getting multiple readings per patch that are averaged, so I'm not sure why you are seeing a visual difference unless there is something else going on here that is affecting the measurements.
Logged

David White
Stephen Best
Guest
« Reply #16 on: December 13, 2006, 05:06:19 PM »
ReplyReply

Quote
The other complaint is that I think (others might help filling in here since I'm also curious) that the maximum EyeOne Photo test chart containing 918 patches is not enough for all papers to make smooth transitions mainly in deep&rich shadows.
[{POST_SNAPBACK}][/a]

The current version of Eye-One Match will read targets larger that 918 patches but it's a bit of a fiddle.

Download Bill Atkinson's Eye-One Pro Targets from here:

   [a href=\"http://homepage.mac.com/billatkinson/FileSharing2.html]http://homepage.mac.com/billatkinson/FileSharing2.html[/url]

Pick one (I use the 1728 target), rename the reference file to "i1 Easy RGB 1.2.txt" and replace the file with the same name in the Eye-One Reference Files->Printer folder. Rename and save the original if you want. You can then read and generate profiles with Match based on the larger target. However, I prefer to use PM5 MeasureTool (a free download), measure the target three times and pick the best reading. You can use the Compare function to work out how close they are to each other and whether you've got a bad reading or not. When using MeasureTool, make sure you turn on the Spectral option to record spectral data so Match can detect OBAs and filter these accordingly. The main attraction of using the Eye-One package (3.6 and later) is the "LOGO Colorful" tables for Perceptual rendering.
Logged
PetterStahre
Newbie
*
Offline Offline

Posts: 40



WWW
« Reply #17 on: December 13, 2006, 05:59:52 PM »
ReplyReply

Quote
I think that it would be informative to run another spreadsheet showing the deltas between the readings you are getting so that you would have an objective vs a subjective measurement.  If you have any software that can calculate delta-e values between measurement files that would probably be very helpful also.  As Andrew said, you are already getting multiple readings per patch that are averaged, so I'm not sure why you are seeing a visual difference unless there is something else going on here that is affecting the measurements.
[{POST_SNAPBACK}][/a]

I don't know how to calculate the delta E, but for anyone who is interested in doing the math I've uploaded two Excelsheets to:

[a href=\"http://www.matbilder.se/test/glossy.zip]http://www.matbilder.se/test/glossy.zip[/url]
http://www.matbilder.se/test/canvas.zip

The glossy.xls contains 3 data sets/readings (3 sheets + the sheet for averaging). It's measured on a glossy no-name-paper similar to Epson Photo Paper Glossy.

The canvas.xls contains 5 data sets and also another routine for averaging which only uses the 3 best values for averaging (meaning the highest and lowest value from the 5 data sets for each cell is not used). (This is overkill, yes.) Measured on one of Fredrix Canvas.

Happy computing!

// Petter
Logged
Stephen Best
Guest
« Reply #18 on: December 13, 2006, 06:13:57 PM »
ReplyReply

Quote
I don't know how to calculate the delta E ...
[{POST_SNAPBACK}][/a]

Equations here (under Math):

   [a href=\"http://www.brucelindbloom.com/]http://www.brucelindbloom.com/[/url]

Bruce's free contributions to the colour science community are without peer.

BTW, if you manage to come up with Excel formulae to convert the spectral data to Lab, compute Delta Es and weight the values on a patch basis I'd be very interested! My Excel skills are nil.
« Last Edit: December 13, 2006, 06:43:57 PM by Stephen Best » Logged
jjlphoto
Sr. Member
****
Offline Offline

Posts: 467


« Reply #19 on: December 13, 2006, 11:20:44 PM »
ReplyReply

Quote
......I prefer to use PM5 MeasureTool (a free download)..........

.........The main attraction of using the Eye-One package (3.6 and later) is the "LOGO Colorful" tables for Perceptual rendering.
[a href=\"index.php?act=findpost&pid=90366\"][{POST_SNAPBACK}][/a]

A free download? It's usually over $500. What is the Logo Colorful table? I have the i1 Photo Package, and just updated to Match v3.6.1.
Logged

Thanks, John Luke

Member-ASMP
Pages: [1] 2 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad