Best strategy for printing

My question concerns maintaining the quality of a photo for printing. I have two options: (1) design a document in InDesign, place TIFF files into the ID document, save as PDF, read PDF into Acrobat Pro and convert to TIFF for printing or (2) design the document in Photoshop with a set of templates, place TIFF files, and print. Obviouslly, ID is more efficient to design with because of its master pages, but where will I get the best quality print? Option 1 or option 2? I suspect Option 2 because there is less conversion going on.

thanks in advance.

My understanding is that InDesign is a colour-managed program, and so if you set up your document and CM options correctly then you should be fine. I think conversions are done at the output stage. However that is a big “if”, because I have struggled with ID CM settings where components of the document had different profiles, and particularly where I was sending the document in CMYK to an external printing service. This may reflect my very limited knowledge of ID, but nonetheless be prepared to invest some time in getting CM right. If you’re going to do this multiple times it’s a worthwhile investment.

Thank you, Brian for the insight. Three more things to note, which I should have included. First, these are grayscale images. Second, I am printing on an Epson 7880 printer with PiezoPro Inks. And, third, there is a very small amount of text on two of the documents pages.

Any other insights appreciated.

David

I was wondering about the relevance of this question to IJM products, so that now makes sense, David. The use of Piezo inks adds another level of complication. I’d need to think about this as I haven’t ever tried to do this. I may be out of my depth, and would prefer if someone else weighed in, e.g. @walkerblackwell.

The answer may depend on what your intended printing workflow is, including whether you intend to print with an ICC or not. If you intend to print with an ICC and you’re on a Mac, then I guess you could print direct to QTR from ID, as ID could do the profile conversion at the output stage. I have no idea whether ID can cope with greyscale ICCs - you may need an RGB version of the ICC.

If you’re not using an ICC to print, or you’re not on a MAC, then as you noted in option 1 you’ll need to output a PDF and use other s/w. I wouldn’t involve Acrobat. I’d open the PDF directly in Photoshop. I guess the question is what ICC to output the PDF with. TBH I’m not sure. Again, I don’t know how well ID works with greyscale ICCs so I suspect AdobeRGB, as the RGB version of Grey Gamma 2.2. Then you can save a TIFF and print with Print Tool (Mac) or QTRGui (Win).

At this point I start to wonder whether it’s not just simpler to go for option 2 and do it all in PS. This is what I do for my exhibition printing. I insert my digitised signature, image title, date and edition information under the image in PS and print. (This format is a little like what Brooks Jensen of Lenswork does, and also what traditional printmakers do.) I have an action recorded in PS that inserts and places the signature and text. The text is just dummy text that I manually edit. If your document structure is sufficiently simple then it may be easier to record a PS action to create your document from the photographic image.

If you print a grayscale calibration target through ID and calibrate to that it should work to then print design work through ID as long as the “colorspace” is the same (eg: Adobe RGB 1998 or Gray Gamma 2.2) for all your images. The target would be a validation of sorts.

IMO, most likely it will work fine but I have not tested it and don’t have time till the end of the week . .

best,
Walker

I do not have InDesign installed, so not sure of it’s capabilities. I am most comfortable in Photoshop.

An important thing to keep in mind is that you should always try to keep your greyscale images in 16bit, whether in RGB/16 de-saturated or Grey/16. This allows you much more headroom to do edits without introducing banding and clipping. If you start with an 8bit RGB image, for example from a phone, the first thing to do is bump it to 16bit. But before forcing to greyscale, see what you can do with the ‘Black and White’ function (in Photoshop). You can do some nice things with mixing the color channels into grayscale.

In Photoshop, some of the filters do not work in 16bit, but the main things for editing (curves, masks, sharpening, blurring) do.

David seems to be talking about the post-editing stage. 16 bit is certainly recommended for editing, but once that is finished then I don’t think that 8 bit hurts for layout and printing, despite some people insisting on a 16 bit printing pipeline.

InDesign’s main output format is PDF and you can’t really do 16 bit in PDF although some people have tried.

For grayscale it is required. Piezo will separate the tones so well that you can see banding in gradients that are 8bit because of the literal gray steps.

best,
Walker

My understanding is that the Windows printing pipeline is 8-bit, is that not right? Because if so, then I’ve never seen it, not even in the proof of piezo image.

Proof of piezo doesn’t have a gradients.

Windows has ben 16 capable for a long time.

best,
Walker

So much insight! I think I am coming down to option 2 or something close to it. In a recent conversation with another photographer who has done a substantial amount of self-publishing of books, broadsides, and portfolios, she recommended doing the initial designing in ID where you can quickly make changes as you work to a final layout (in this case for a portfolio with multiple images on some pages along with images in the 2:3 ratio and 65:24 ratio) using low res images, exporting the ID file to pdf, converting the pdf to to tiff in Acrobat Pro, importing the pages into PS, then taking high res tiff files and placing them in place of the the lower res images. In the final step, taking the “PS TIFF” files and bringing them into Print-Tool to print with the PiezoPro inks and the QTR RIP. Many steps, but she believes she sees a quality difference using this approach.

Thoughts

You can open PDFs directly in Photoshop or Illustrator to replace the photos with higher-res and bitdepth versions.
I hate to suggest reading this thread with its snippy feel, but some useful info between the people-being-jerks-on-the-internet parts: https://forums.adobe.com/thread/2318909
If you do not have text overlayed on images, you can just drop the better versions of photos over the imported PDF, even if it opens flattened, flatten the final and save as TIFF for QT-Rip.

thanks, I will check into this!

I’d be interested in hearing why this other photographer thinks there’s a quality benefit to using Acrobat in between ID and PS, rather than just opening the PDF directly in PS. I also wonder what she would think of eburden’s suggestion to use that import script, which sounds like a neat approach if you’re on a Mac and able to use applescript. I’m not, and most of my ID documents have text on the images anyway.

David mentioned Print Tool, which indicates that he is on a Mac. Despite that I’d like to persist with this 8 vs 16 bit on Win thing, because there’s a risk that I might learn something. I said that the Windows printing pipeline is 8-bit because that seemed to be the received wisdom. You read it often and so I just accepted it. (I’m a Windows 7 holdout until the very EoL next January, and W10 may be different.) So I did a little research and found that an XPS print path was introduced in Vista and at that time extended back into XP, so it does seem to be the case that the capability to print 16 bit in Win has been around a long time. Something learnt already.

But as I read it, that isn’t a guarantee of 16 bit printing - the application and driver have to support it by using that particular print path. Canon has released XPS drivers (although there are reports of 16 bit causing CM issues). However I can’t see any evidence that Epson has. All the reports of their 16 bit drivers seem to be for OS X.

For the topic at hand, the question is whether QTR on Win supports 16 bit. The most recent set of release notes mentioning this issue was 2.7.0.0, which says:

!! Mac Users take NOTE !!

Major Upgrades to the Mac version of QTR Software:
8-bit and 16-bit image data supported – from app to driver.

The Windows implementation does not pass the data through
the print system like the Mac version does. So 16-bit data
changes noted above to not apply to the Windows version. However
QTR continues to use all 16-bits of a TIFF data file for
creating a print.

I don’t understand this - do you Walker? It reads to me like QTR on Win is not really 16 bit. If that’s the case then I’ve not seen any artifacts in gradients. Yes, you’re right, the PoP image doesn’t have any - I was thinking of other B&W test images that contain gradients.

I read the QTR text you quote as being clear that it does fully support 16bit TIFFs on WinOS.

Until I read this article, published a few weeks ago, it had been the case that to see a 10bit gradient on Photoshop you needed to have Quadro cards. I have always used those, and also use displays that handle 10 bit and 12 bit color. However…


Now, whether you are seeing the full quality of the gradients, especially in grayscale, your file remains high quality. And going from Photoshop saved TIFFs to QTR means that any OS issues are bypassed. And life is good.
The ability to display a proper deep gradient is critical to editing, of course.

There are two distinct issues here - what you see on the screen and what gets to the paper when you print. My comment was only about what gets to the paper if I print via QTRGui on Windows. That quoted text suggests some differences between OS X and Win in regard to the 16 bit thing, and I’d like to better understand what they are. I asked Roy to clarify, but didn’t get a response.