I blew it in the previous post when I said Qimage doesn’t do color management. I’m not sure what I was thinking; I had forgotten that I had used the color management features in the previous generation of Qimage, and I couldn’t find them in the v2011.136 version that I am currently using. The reason I couldn’t find them was a program configuration issue. In the “job properties” screen at the lower right- corner of the Qimage main window, there are two pull-down menus for ICC profiles. The top one allows you to select the monitor profile, and defaults to the current operating-system-selected monitor profile. The second one allows you to pick the printer profile. However, in the default configuration of Qimage, a message is displayed to the effect that you should do color management in the printer driver, and it doesn’t show you any of your computer-stored ICC printer profiles. To change this, and allow Qimage to do color management, you need execute the following menu sequence: the Edit>Preferences>Color management (ICC). Once you do that, you check the box allowing Qimage to do the color management, and pick the printer profile you want from the drop-down list. There’s one little wrinkle: it looks like the access button for the drop-down list is grayed out; if you click on it anyway, the list appears.
I evaluate color management using a suite of TIFF images that I got from Charles Cramer. When I loaded the first one into Qimage, the colors were completely wrong on the monitor. All of the Cramer calibration images use CIE 1976 (L*, a*, b*) D50 as their color space. It looks to me like Qimage won’t handle Lab color, although it doesn’t complain when presented with a Lab image. Just to make sure, I printed the image using a profile I had created in Eye-One Publisher; it printed with the same wonky colors that I saw on the monitor. Converting the TIFF file to PSD in the Photoshop CS5.1 and reading the file back into Qimage produced colors that were in the same ballpark as the right colors, but they were pretty far off.
I don’t use Lab color in my current work, but in the past I have used it extensively. It wasn’t a great choice when Photoshop only supported eight bits per color plane. Because the Photoshop implementation did not provide for image-dependent scaling, they needed to pick minima and maxima for the chromaticity axes that made color space huge. That had the unpleasant side effect of causing posterization when quantized into 256 levels and then manipulated. Going to 16 bits per color plane eliminated that problem. I used Lab quite a bit for a while, and I still occasionally need to print those images. To do that using Qimage, one workflow would be: first sharpen the L* channel, then convert to printer color space, then let Qimage do the resampling. Another would be: convert to printer color space, and that to image resample and then sharpen. Neither is optimal for Lab image; sharpening should be performed after resampling, and it should usually be performed only on the L* channel.
I converted the calibration images to ProPhoto RGB using absolute colorimetric as the rendering intent. I printed them from both Photoshop (using ACE) and Qimage with perceptual rendering intent, and compared the results. They are visually identical, but not quite the same as the Lab image printed directly from Photoshop, showing that the fewer color conversions, the better.
So, I stand corrected. If you don’t use Lab color, Qimage does indeed do color management, and does it as well as the Adobe Color Engine. The last bullet in the preceeding Post should read:
- Export the image from Lightroom. Open Qimage, configure the printer driver, pick the resampling algorithm, and let Qimage send the file to the printer. Don’t forget to go back and delete the exported file.
I am not sure at all where y0u are coming from….and based on your response to my comment to your earlier post you seem to take any negative comment as “hostile”…oh, well.
I do not claim to be a color management expert…in anyway. However, I am not sure why anyone would leave and image in LAB color space. Most people convert to an RGB )or is offset press, maybe CMYK) space…..and, as I remember, PS does not provide a choice of rendering intent (I think it is always ‘relative”, by default, when changing color space).
If you somehow converted LAB using ‘absolute’, it is not surprising that it looked different than what PS printed directly from LAB….but interesting that PS and Qimage printed your ‘absolute’ conversion the same…..is that a negative to Qimage…??
As I said in my prior comment post, I think your main problems with Qimage are lack of study/knowledge….a.k.a user error.
As to your “process”, it show that you still do not understand what Qimage can do…
“…■Export the image from Lightroom. Open Qimage, configure the printer driver, pick the resampling algorithm, and let Qimage send the file to the printer. Don’t forget to go back and delete the exported file…”
Once you save set a print driver for a paper/papersize/icc group, Qimage automatically remembers that. If you want you can save setups, layouts, etc., etc, which will easily make subsequent printing extremely easy and almost fool proof…
Your paragraph starting with “I don’t use Lab color in my current work…” is a “fox” which I refuse to chase as it is not meaningful to the discussion.
You started your series with the results that neither PS or LR created great resizing and prints….I agree. Then, you praise ‘Perfect Resize’….which will require you to export a TIFF to PR and resize to print size and print….for each different size you want.
Do you realize that with Qimage, you can never resize the image? You do all your processing, adjustments, and sharpening so it looks good on the screen (not over sharpen, as you need to do in PS to get sharpening close for printing), then expost one TIFF/JPEG. This image can then be printed in Qimage at any size you want….and it is resized/sharpened automatically to the paper size you selected….on the fly, which no storage or need to later delete.
Obviously I do not feel that you have performed due diligence in your review. If you want to use PS, LR, PR, that’s fine….your choice. If you are going to do workflow comparisons, build fair comparisons. BTW….I think LR has the best workflow, but the printing quality compared to other options is lacking.
PS…no grammer is not something that concerns me, nor do I proofread for it….so you have another opportunity to make negative comments… 🙂
There’s a lot to deal with here.
I think the many advantages of Lab color as an editing space, and how that choice affects the color space in which images are stored deserves a post of its own. Stay tuned.
Photoshop and Qimage both provide the same choices for rendering intent upon color space conversion. Those choices are: Perceptual, Relative Colorimetric, Saturation, and Absolute Colorimetric. For some of the contents Photoshop allow additional options, such as black point compensation and whether or not to use diffusion dither. For each of the intents, Qimage allows the black point compensation option (I have no idea how you can do an absolute colorimetric conversion and perform black point compensation at the same time; as far as I know it is a color science tautology) but not the diffusion dither choice. Lightroom only allows Perceptual and Relative Colorimetric intents. I can’t post screen shots in this comment, so I’ll create a new post with the pictures.
I spent six years working for IBM as a color scientist, and I find the paragraph about absolute colorimetric conversion from lab to ProPhoto RGB incomprehensible. I’m not sure that this deals with what John is saying, since I can’t figure out exactly what John is saying but the general rule is that all conversions from one editing space to another should be performed using absolute colorimetric intent, assuming that the target color space has sufficient gamut to represent every color in the source image. Since there will be no gamut mapping, there is no reason to change state absolute colorimetric value of any color in the image. As I said in my post, the fact that Qimage and Photoshop made identical prints from the ProPhoto RGB image is an indication that the color engines in the two programs are substantially equivalent.
John makes the point that two images saves your printer driver settings. I do not dispute that point; Lightroom does the same, and I find it convenient in many circumstances. However, I rarely use the same driver settings for two batches of prints. First off, I have several printers, and the odds are that I am printing to a different one than the last time I used the program. Second, I use several different kinds of papers, and each one takes different driver settings and color profiles. Third, I print in many paper sizes, each with its driver setting.
John points out that Qimage deletes its temporary resized image. I don’t deny that. The image that I suggest deleting as part of the Qimage workflow is the one that got exported from Lightroom.