Ad
Ad
Ad
Pages: « 1 ... 4 5 [6] 7 »   Bottom of Page
Print
Author Topic: Imageprint - worth the $ ? - compare to QImage? other views please!  (Read 27287 times)
alfin
Jr. Member
**
Offline Offline

Posts: 77


« Reply #100 on: November 15, 2010, 05:27:00 AM »
ReplyReply

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.

ImagePrint uses bicubic and it does not apply any sharpening.
Logged

Lars Mollerstrom
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2885


« Reply #101 on: November 15, 2010, 06:21:07 AM »
ReplyReply

ImagePrint uses bicubic and it does not apply any sharpening.

Based on that information I can only conclude that Lightroom blurs the image. Not likely I think.


met vriendelijke groeten, Ernst Dinkla

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

Logged
Pandora
Newbie
*
Offline Offline

Posts: 6


« Reply #102 on: November 15, 2010, 07:07:54 AM »
ReplyReply

Mark Dubovoy's interesting article condemned me to more than a weeks worth of tests comparing IP and Epson prints.

For this evaluation I'm using IP6.1 with Ilford GFS on my R2400.  Having spoken to Colorbyte support I understand that the print engine for 6.1 on the R2400 is the same as that for IP 8.  Indeed they advised me that, from a purely colour management perspective, there would be no advantage upgrading to IP8 (for an R2400).

On the Epson side I'm using a custom made profile (i1 iSIS D50 with UV cut) with LR3.2 on a PC.  With IP - the RF3 profile on the lossless tif image exported from LR.  Both with perceptual intents.


Here are my observations.

I agree with with Andrew.  IP gives me a magenta shift in some blues.  In extreme cases it can be quite severe and prone to microbanding.

I agree with BradFunhouser. I prefer the skin tones IP produces. I find all Epson profiles ever so slightly yellow - clearly visible in skin tones.

I agree with much of what Mark Dubovoy says, particularly about the "slightly purer whites". But I very much disagree with him about IP's coarser dithering - I much prefer Epson's rendering - perhaps if I my prints were bigger, IP's dots wouldn't be so visible.  Mark says "more accurate blue tones".  I would say different blues sometimes blacker, particularly on Baryta papers, and sometimes more magenta.

With the Epson driver, one strength of my custom profile v the generic profile is that it improves shadow detail markedly.  Even so IP gives me still slightly better shadow detail.  This is only with perceptual intent.  For me, with this paper, printing with IP using a relative intent destroys shadow detail and greatly reduces Dmax.  When printing With a relative intent even the generic Ilford profile give much better Dmax and shadow detail than IP.

With the advent of IP7 Colorbyte produced a completely new set of profiles called DCM.  For colour work I believe these are backwardly compatible with previous versions - it is only in monochrome work that the DCM modifications are embedded in their software and require an upgrade.  In my experience the DCM profiles give yet more shadow detail. But, at this level, I am not sure this is always desireable or could not be achieved through manipulation in LR.

On another thread about Mark Dubovoy's article Eleanor Brown says she is now "getting stunning results from my epson drivers on both my 7900 and 9880 printers provided my files are properly prepared".  This, I think, is the clue.  Epson printing technolgy and profiling tools are continuously improving. I may be wrong but I feel that the recent updates for IP have been more layout driven than colour management driven and, unless IP ups the ante, the one will overtake the other. Perhaps, as BradFundhouser asks, maybe this point will arrive with X-Rite's new profiling offer. Unless, as I suspect, it has already happened.

From a print quality perspective only, is it worth the money?  If Colorbyte offered me a full refund I would probably not accept it.  But I hope and suspect I will not have to buy an upgrade for my next printer. Or, who knows, perhaps I will.  It can produce great images.  However, even though I own IP, I usually stick to LR and the Epson drivers.

And please forgive me if my R2400 is so now old hat that none of my comments are relevant.

Hope this helps,

"Joe User" (Peter).
Logged
alfin
Jr. Member
**
Offline Offline

Posts: 77


« Reply #103 on: November 15, 2010, 10:19:19 AM »
ReplyReply

Based on that information I can only conclude that Lightroom blurs the image. Not likely I think.

The test image is 3600x2700 pixels @ 360 ppi, so there is no need to resample neither in Lightroom nor in Imageprint for an A4 print. In Lightroom I set height to 19,05 cm (7,5Ē), deselect print resolution and output sharpening and I then assume Lightroom just sends the file to the printer driver. In ImagePrint there is no sharpening to choose and ImagePrint does not resample a file if itís at 360 ppi.
Logged

Lars Mollerstrom
madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #104 on: November 15, 2010, 11:16:50 AM »
ReplyReply

Correct, alfin, LR will neither resample nor output sharpen in the case you've described.

Also, regarding your earlier comments: I noticed that you did your tests on Enhanced Matte (a MK paper). I have found in my testing for the 3800 that Epson's supplied profiles for Photo Black (PK)-compatible papers are excellent, but the ones for their Matte Black (MK)-compatible papers less so.
Logged

Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2885


« Reply #105 on: November 15, 2010, 03:20:36 PM »
ReplyReply

In ImagePrint there is no sharpening to choose and ImagePrint does not resample a file if itís at 360 ppi.

I wrote "under the hood". There is an analogy with the Imacon scanners, you actually have to set negative values in the sharpening settings of the Imacon software to take out sharpening.  People comparing scanners are not always aware of that default value. That ImagePrint doesn't have an entry for sharpening doesn't have to mean there is no sharpening done.

I can imagine one thing that could make a difference in ImagePrint. Given the reported coarser screening and the here so far hardly mentioned 16 bit pipeline of ImagePrint, the translation to the cell size of the rasterisation can be more optimal. In a sense the resolution is reduced in favor of a better defined screening. I doubt that is the sole reason it looks sharper and I expect some print sharpening happens in that pipeline too. Whether you created equal conditions for Lightroom by taking out sharpening is still a question.

BTW, there is no way to set "original size" in Lightroom + driver to ensure that the input resolution actually matches the printer's native resolution = 360 PPI. That is what I do with Qimage when I print vector designs first rasterised in Photoshop to 360 or 300 PPI Tiffs.


met vriendelijke groeten, Ernst Dinkla

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


« Last Edit: November 15, 2010, 03:32:05 PM by Ernst Dinkla » Logged
alfin
Jr. Member
**
Offline Offline

Posts: 77


« Reply #106 on: November 15, 2010, 04:41:37 PM »
ReplyReply

I can imagine one thing that could make a difference in ImagePrint. Given the reported coarser screening and the here so far hardly mentioned 16 bit pipeline of ImagePrint, the translation to the cell size of the rasterisation can be more optimal.

I think you are getting closer now to the real reasons why ImagePrint looks better and sharper with more defined details.

I doubt that is the sole reason it looks sharper and I expect some print sharpening happens in that pipeline too. Whether you created equal conditions for Lightroom by taking out sharpening is still a question.

Maybe, but since neither of us know, itís just guesswork anyway. However, if itís all about sharpening (like Qimage), letís take Lightroom out of the equation also and print through Photoshop.

A print ready image, output sharpened with Photokit Sharpener and sent through Photoshop and the printer driver. The same image, without any output sharpening applied, sent through ImagePrint.
Are they equally sharp and detailed? Nope. ImagePrint still looks better with more micro contrast and details. Itís not about acutance!

BTW, there is no way to set "original size" in Lightroom + driver to ensure that the input resolution actually matches the printer's native resolution = 360 PPI. That is what I do with Qimage when I print vector designs first rasterised in Photoshop to 360 or 300 PPI Tiffs.

OK, you better discuss that with Eric Chan instead.
Logged

Lars Mollerstrom
David Watson
Sr. Member
****
Offline Offline

Posts: 396


WWW
« Reply #107 on: November 15, 2010, 04:59:05 PM »
ReplyReply

A big issue is whether you are printing from a Mac or a PC.  If a PC I cannot offer an opinion but if you are printing from a Mac then IP is a definite improvement on the native drivers.  Every Mac upgrade seems to bring its new printing problems. Whether it is the need to find and use drivers embedded in the OS or overcome contentions in the software when producing profile targets.  For my own purposes (landscape and architecture) IP solved all of these problems in one fell swoop.  I get very high quality and most importantly consistent results from both my 3880 and my 7900 printers - consistent between printers and from print to print -and I do not have to waste hours and hours generating new profiles  and setting up the OS/driver combinations.  IP bypasses all of the OS and application drives so there is no issue with colour management contention coming from the OS or the application.

In addition I do not use any Epson papers as I favour Hahnemuehle PhotoRag smooth and Canson Baryta Photographique so the Epson profiles are really of no use to me.  The IP profiles for these papers work very well indeed. 

Yes it is expensive to buy but it is a one-off cost and compared to the aggregate cost of quality papers and inks, not to mention the high cost of digital MF and lenses, I think it is affordable.


This was my earlier post on this subject and the thread reporting yet another set of printing problems with the latest 10.6.5 update just reinforces my view.

I have run the update and what do you know?  Imageprint runs with no problems at all on my Epson printers.  Good value I would say so!
Logged

David Watson ARPS
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2885


« Reply #108 on: November 16, 2010, 05:03:10 AM »
ReplyReply

I think you are getting closer now to the real reasons why ImagePrint looks better and sharper with more defined details.


When I check the eyes and lashes of the child on this page I see detail contrast and noise in the IP one. The old discussion about sharpness in fast and slow films comes to mind, the difference between chromogene dye clouds and B&W silver grain. I don't think the ImagePrint version is better here. There is not more definition. In halftone terms it is a bigger dot/ rougher screen and within that dot you then can define the dot sizes better based on the same image data input. It depends on the quality of the inkjet paper, the image content and the viewer's eyesight whether that is a good strategy.
http://www.luminous-landscape.com/reviews/software/imageprint8.shtml



met vriendelijke groeten, Ernst Dinkla

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

Logged
Robboo
Newbie
*
Offline Offline

Posts: 2


« Reply #109 on: November 16, 2010, 11:52:37 AM »
ReplyReply

Hi All,

I'm looking for a version of ImagePrint V6.1.... that includes the driver for the Epson 2400. Nu G4 that housed the software died and I'm trying to load it onto my G5. Every copy that I find does not include the Epson 2400 driver.

I thought I'd try to find it before having to upgrade to V8.

Thanks in advance!
Logged
Kirk Gittings
Sr. Member
****
Offline Offline

Posts: 1551


WWW
« Reply #110 on: November 16, 2010, 08:37:43 PM »
ReplyReply

When I upgrade printers, every couple of years, I try some alternatives to IP, but always come back to it. I am still on IP 7. What does IP 8 offer? I primarily print b&w.
Logged

Thanks,
Kirk

Kirk Gittings
Architecture and Landscape Photography
WWW.GITTINGSPHOTO.COM

LIGHT+SPACE+STRUCTURE (blog)
Sven W
Sr. Member
****
Offline Offline

Posts: 514


« Reply #111 on: November 17, 2010, 02:10:57 PM »
ReplyReply

Kirk
It' still the same color engine, so technically it's no difference to print via v7 or v8.
The support for newer print models in v8, does, of course, give you those printers advantages over older ones.
What is added in v8 is more of frames, text in images, background color, a better Spoolface, and so on. Take look here
And if you are on Mac, the latest build v7 also runs under SnowLeopard.


/Sven
Logged

Stockholm, Sweden
narikin
Sr. Member
****
Offline Offline

Posts: 862


« Reply #112 on: December 20, 2010, 10:19:56 AM »
ReplyReply

I have a 3880, and a 9800 and an 11880.  The total cost for these with IP8 is over $4500.  Thats insane, at least when you are not running a print business.

With the 11880 I use Bill Atkinsons profiles which are great. and I mean GREAT for the papers they cover. Thank You Bill.
I agree that generic profiles are a lot better than before, plus there is much less variation in the modern printers anyway. For example, the profile for Epson Exhibition Fiber on 3880 or 11880 is a remarkably good profile, I think it was made by the PixelGenius people for Epson, so not surprised.

I will one day test IP's EEF on the 3880 vs Epsons EEF profile, and see if it betters that, (doubtful, but will be impressed if it does)

Remember IP profiles are Generic too - they are not for your specific machine, unless you pay extra for that service.

Logged
Sven W
Sr. Member
****
Offline Offline

Posts: 514


« Reply #113 on: December 20, 2010, 04:15:43 PM »
ReplyReply

A weak argument, but you actually get a discount for add-on IP licenses.

/S
(11880, 9900, 7880, 4800 and soon 4900. All with ImagePrint)
Logged

Stockholm, Sweden
Tim Gray
Sr. Member
****
Offline Offline

Posts: 2002



WWW
« Reply #114 on: December 20, 2010, 04:41:24 PM »
ReplyReply

My use of a RIP is somewhat specialized - I use Quadtone Rip to adjust the individual ink densities to create a negative with the appropriate contrast range for a platinum print.  Unfortunately, based on a query to ImagePrint technical support, they do not support this kind of flexibility.
Logged
narikin
Sr. Member
****
Offline Offline

Posts: 862


« Reply #115 on: December 20, 2010, 04:50:11 PM »
ReplyReply

A weak argument, but you actually get a discount for add-on IP licenses.


$4500 cost for software is a weak argument ?  - what software to you have that costs anything like that?

and more importantly what software do you have that allows you only to use it on one of your devices:
does Adobe charge per camera that you wish to use for Lightroom, or for Photoshop?
does Capture One charge you for processing Canon 5d2 files, then charge you again for profiles to process 1Ds files? and again for profiles for D60 files? and again for S95?!

No - you buy the software and you can use it on your devices, end of story.
It's a silly joke that you, an individual photographer have to pay for each and every machine you want to use IP on.
Logged
Sven W
Sr. Member
****
Offline Offline

Posts: 514


« Reply #116 on: December 21, 2010, 07:25:06 AM »
ReplyReply

You misunderstood. I meant that my argument was weak, it was more like an encouragement.

But for the rest of your reply I put in this:
Adobe charges you per computer. So does Cap1.
Silly jokes?

There are many software out there with very strict license policies. Often bound to an encrypted dongle.
Why? Because these small companies don't have the "muscles" as the Big Ones, and have to fight to protect their product and patents.
So there's no doubt that these companies can't have the prices you are used to. The developments and inventions from this sort of manufactures are very often progressive and a big help for the rest of the community. I don't mind supporting them.

And I can agree, that most of the "individual" photographer rarely need a sophisticated RIP these days.
But for me, running a print studio, it's definitely a must, and where IP's strength is unbeatable.
/Sven
Logged

Stockholm, Sweden
narikin
Sr. Member
****
Offline Offline

Posts: 862


« Reply #117 on: December 21, 2010, 07:43:16 AM »
ReplyReply

You misunderstood. I meant that my argument was weak, it was more like an encouragement.

But for the rest of your reply I put in this:
Adobe charges you per computer. So does Cap1.
/Sven

Aha - I get it Sven - yes agreed.

although I would respectfully point out that Adobe and Phaes One do not charge per computer, they allow a user to have at least two operational machines per licence code, and you can call Adobe and get a third activation most of the time.

yes for a print lab like yours, its maybe a necessity, for an individual who is just looking for better final quality, then its all about the profiles (Epsons dither pattern is much better in 2009/10 than it was in 2005) and as discussed above Generic profiles are often very reliable now, and the difference to IP ones hard to justify, if it exists at all.
Logged
tvalleau
Newbie
*
Offline Offline

Posts: 29


« Reply #118 on: December 21, 2010, 06:23:04 PM »
ReplyReply

I've read thru this thread (and understood it) after coming here to see if I could find an explanation for the fact that the IP profiles are so _small._

Greeting to this august group, and happy holidays.

I just got IP8 a few days ago, and am testing.  I've been making my own profiles using a (don't laugh) ColorMunki. I've used those, and of course the paper mfgr's profiles, along with Epson's (for my 3800.) I'm not knocked out by the results, but I'm still testing...

However, I did notice this: compare the IP profiles with others (my own; mfgr's, Epson's) and every one of them is 20% (or so) smaller. (Using ColorSync Utility on the Mac & holding one for comparison with the other.)

I certainly didn't expect that; at least not according to their literature, and what I'd read.  Any comments? Anyone care to explain that 1) either this is as screwy as I think it is, or 2) I'm an ignorant newb, and it isn't important because...?

TIA

Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9218



WWW
« Reply #119 on: December 21, 2010, 07:06:47 PM »
ReplyReply

I've been making my own profiles using a (don't laugh) ColorMunki.

Iím not laughing, it makes surprising good profiles!

Quote
However, I did notice this: compare the IP profiles with others (my own; mfgr's, Epson's) and every one of them is 20% (or so) smaller.

As my wife likes to say, size ainít everything <g>. There are private tags and metadata in profiles that can make the size larger. The measured data is often stored in X-Rite profiles which of course, makes them larger in size. Test the profiles using them with images and printing said images, donít read much into the actual size of the profiles.
Logged

Andrew Rodney
Author ďColor Management for PhotographersĒ
http://digitaldog.net/
Pages: « 1 ... 4 5 [6] 7 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad