Printing panos on roll paper

[msbc]

I have a Epson P800 and want to print some panorama images on roll paper. How do I work out the print dimensions for a given image? Example, I want to print an image 16" wide, on 17" roll paper - how can I calculate the print length and resolution (to determine if I need to up-rez)?

[Mark D Segal]

I have a Epson P800 and want to print some panorama images on roll paper. How do I work out the print dimensions for a given image? Example, I want to print an image 16" wide, on 17" roll paper - how can I calculate the print length and resolution (to determine if I need to up-rez)?

You should have a look at this thread as it contains useful suggestions, even though it focuses on borderless. Printing Borderless from Roll P800

You don't mention which application you are using for printing. The specific steps differ somewhat between Photoshop and Lightroom. Generally speaking, if the photo width is 16 inches and the paper width is 17 inches, you would set a one half inch border on either side of this width if you want it vertically centered. Also you would want to check that you are sending no less than 240 PPI to the printer. Much of this is covered in video tutorials free to paid-up members of this website. 

[Bart_van_der_Wolf]

I have a Epson P800 and want to print some panorama images on roll paper. How do I work out the print dimensions for a given image? Example, I want to print an image 16" wide, on 17" roll paper - how can I calculate the print length and resolution (to determine if I need to up-rez)?

Hi Mark,

16 inches at 360 PPI =   5760 pixels wide source image.
16 inches at 720 PPI = 11520 pixels wide source image.

So it depends on the original image size which settings to use for upsampling (always upsample not down-sample), and for subsequent output sharpening.

Then add equal sized margins for the additional printable area of the paperwidth in the print dialog.

A printing application like Qimage Ultimate (for Windows, but runs fine under Parallels or similar on a Mac) automatically takes care of all that (and uses superior resampling algorithms and halo-free output sharpening), and exactly reports the physical margins without having to make a single test print.

Cheers,
Bart

[Mark D Segal]

I should add a question to Reply #1: what are the pixel dimensions of the photo you are starting from? This will indicate how much resampling needs to be done to get to the final linear print dimensions you are seeking.

On use of software, If you are printing from Lightroom, it will automatically resample on the fly depending on the relationship between the original pixel dimensions of the file being printed and the PPI you instruct Lightroom to print at in the Print module. However, regardless that it resamples under the hood, it is good for you to know how much resampling is taking place.

To do that note the number of pixels on one dimension of the original file; divide the number of printed inches you are seeking into that number of pixels and this will give you the initial pixels per inch without resampling. Let us say you have 3000 pixels on one dimension and you want to print that dimension to 10 inches. This will give you PPI of 300. If you set 300 as the target PPI in Lightroom, there is no resampling and you will obtain a fine print from it. But let us say you want to print the same pixel dimension to 20 inches. In this case you only have 150 PPI original PPI to start with. If you wish to print that at 360 PPI, for example, LR would be performing a considerable amount of upsampling, more than doubling the PPI to get from your start conditions to your end conditions. While doable, the output quality may or may not meet your expectations. That is the manual approach to explain the logic. Lightroom can show you all this information automatically in the Print Module depending on how you arrange your view options settings. See martin Evening's Lightroom 6 book, pages 454, 455.

I don't know how helpful any of this is - best to provide us with as much information about your file and your software to be sure to give you the most practical advice for your situation. I also recommend reading-up the sources I've suggested here and above.

[BradSmith]

........ how can I calculate the print length and resolution (to determine if I need to up-rez)?

ALGEBRA!!

(sorry, I couldn't resist  )
Brad

[Mark D Segal]

You call that algebra?  . In my heyday dividing one number into another was called arithmetic. But I don't fault him for asking. This kind of stuff has been bedeviling newcomers to printing for years. Fortunately we now have excellent teaching materials on this website for clarifying these issues, which are easier for some than for others.

[BradSmith]

OK, now hopefully to be helpful....  for example, let's say that the pano as it exists now is 3000 pixels H  by 7000 pixels W.  Mr. msbc wants his pano to be 16" H. 

His first question is, "How can I calculate print length?"

Answer in Algebra
print W divided by image W (7000 pixels) = print H (16") divided by image H (3000 pixels)
Therefore, (remember that term in math books....)
Print W = 7000 times 16" divided by 3000 =  37.33 inches

To determine the resolution to send to the printer, multiply the 16" H and 37.33" W of your print by the desired printing resolution.  I use 360 ppi on my Epson. 
360 ppi x 16" = 5,760 pixels H    and    360ppi x 37.33" = 13,439 pixels W

[Mark D Segal]

Or perhaps more intuitively presented: if 3000 =  16 inches, (7000/3000)*16 = (2.33*16) = 37.33.

On your second part, what you are getting at is correct - the amount of upsampling that would be needed for a target printed (output) PPI; but resolution itself is usually defined in terms of PPI, so what you are deriving is total pixels per dimension that need to be generated for meeting the 360 PPI output resolution criterion. Whether calculated as total required pixels or output PPI, the extent of upsampling is the same.

[msbc]

Thanks to everyone for their input.

I'm using Lr as my DAM - raw converter is Capture ONE - but I prefer to use PrinTao for the final print. I was after general advise rather than for a specific print, so the algebra lesson is appropriate :-)

I think a suitable workflow for me would be:
1. Softproof in Lr
2. Export image to 16-bit TIF.
3. Check dimensions - use the algebra Luke ! If up-sizeng needed re-size prior to printing - I have ON1 10 so ON1 Resize would be a good option here.
4. Output sharpen - Nik or Topaz Detail.
5. Print at full image size with PrinTao.

Anything I've missed or could be improved?

[Mark D Segal]

Anything I've missed or could be improved?

Yes.

This may sound a bit "radical", but give it try: temporarily set aside that whole workflow, temporarily forget about all your software preferences and starting from the raw file do ALL of it in Lightroom, maximizing the capabilities of that application, which are extensive and can do everything you need to do very well, including the math. Print it, and look objectively at the results. Then go back to your multi-application workflow described in your post, use that and compare the result with the one straight from Lightroom. When you're done comparing, decide whether the complex approach adds decisive value relative to the easy one.

[dgberg]

+1

[msbc]

Mark,

The issue I see with that 'radical' approach is in regards to the output sharpening stage. Lightroom lacks any control in this area - only allowing you to select Low, Standard or High - with no visual feedback about which value is optimal. Printing 3 copies of each image to determine the best setting is significantly more effort and expense. A Lightroom only workflow may be simpler but does it give the optimal result? If I wanted simple vs quality I'd use Lightroom as my RAW converter rather than Capture ONE.

[Mark D Segal]

Start with Standard. For me it works just fine on 99.9% of my photographs. I've never ever made three prints to judge the best setting for output sharpening. The critical one to focus on is capture sharpening and you have complete control over that in LR. If you haven't tried what I'm suggesting you can't answer your question about how you know whether it is optimal. That is why I am suggesting you do the experiment I recommended. I'm not going to start a flame war between the Capture One crowd and the LR crowd, nor do I support pixel-peeping as the ultimate test of what works better - just do the tests, look at what comes out on paper and see for yourself - of course using the latest release of LR CC.

[BradSmith]

Mark's suggestion is excellent.  I think it is always good practice to TEST the simple method versus the intellectually "superior" method and look at the results. Let your eyes decide.
Brad

[Bart_van_der_Wolf]

Mark's suggestion is excellent.  I think it is always good practice to TEST the simple method versus the intellectually "superior" method and look at the results. Let your eyes decide.

Hi Brad,

I agree, but that does assume that the 'superior' method is not the result of prior testing.

However, with LR as the basis, it is a given that a lot can be gained (depending of image content), by using better Creative and Output sharpening (e.g. with 'Topaz Detail'). LR's upsampling is of decent quality, although that could also be improved a bit (or a lot, depending on image content) if one is a pixelpeeper (and those pixels do become bigger with output magnification).

Cheers,
Bart

[BobShaw]

I think a suitable workflow for me would be:
1. Softproof in Lr
2. Export image to 16-bit TIF.
3. Check dimensions - use the algebra Luke ! If up-sizeng needed re-size prior to printing - I have ON1 10 so ON1 Resize would be a good option here.
4. Output sharpen - Nik or Topaz Detail.
5. Print at full image size with PrinTao.

Anything I've missed or could be improved?

My workflow is delete 3 and 4. However I don't use PrinTao, I use Mirage Print.
I have never understood why people agonise printing in an image editing programme when there are printing programmes designed especially for that. You never have to worry about presets. Just select paper size, image size, profile and "print". Les Walkling got me onto Mirage Print at a week long printing workshop. He consults and tests for Epson, so if it's good enough for him then it's good enough for me.

I print usually big prints a metre wide. You can print basically any size that you get the paper into the printer and never have to calculate anything. Spend the time taking photos.

[msbc]

Bob,

I also have Mirage Print. I used it with my previous Epson 3800. When I updated to the P800 there was an additional licence required. Instead of upgrading I switched to PrinTao.

Anyway, wondering why you say delete steps 3 and 4? I don't recall Mirage offering any output sharpening options/control?

[Mark D Segal]

Hi Brad,

I agree, but that does assume that the 'superior' method is not the result of prior testing.

This suggestion assumes nothing about prior testing, because I have no idea when it was done, how it was done, or if it was done whether it deserves review with the most recent versions of the relevant software and methods, so that was the basis of the suggestion - no "a prioris". Sometimes it pays to be fearless and "re-iterative". :-)

[Mark D Segal]

I have never understood why people agonise printing in an image editing programme when there are printing programmes designed especially for that.

Well, let me explain it. Firstly, those printing programs generally require the files to be rendered pixel based formats (PSD, TIFF, etc) that can be large files. If one can print very well without creating that conversion and storage overlay, it may be advantageous. I know, conversions work well and "storage is cheap" so why worry, but in some respects "less is better". Secondly, printing from Lightroom is pretty flexible and easy, and with a fairly modest learning curve freely available one masters "presets" in the printer driver to the extent necessary - which for most people would be several at most. I agree that the bespoke printing applications have their advantages especially in respect of convenience, but from all I've seen, these days output quality between these various solutions is pretty much awash in terms of what you end-up seeing on paper, which I think is good news for those who both use them and don't use them. PrinTao by the way is a pretty clever piece of printing software - you may have seen my review of it on this website.

[Bart_van_der_Wolf]

Well, let me explain it. Firstly, those printing programs generally require the files to be rendered pixel based formats (PSD, TIFF, etc) that can be large files. If one can print very well without creating that conversion and storage overlay, it may be advantageous.

Hi Mark,

That's partly true, usually an 'intermediate' output file needs to be created for output, but it doesn't have to be stored indefinitely once printed, Just create a new one. Parametric editors can recreate the same output over and over again, when needed. A just in time (JIT) approach.

And if a program like Qimage is used, one only needs to output standard size (100% zoom) output files (even 8-bit/channel is possible), because Qimage produces better upsampling anyway and offers more flexible output sharpening. It even doesn't need to produce an upsampled output file for temporary storage, it upsamples+sharpens+colormanages (with optional dithering) and prints from memory and immediately sends it to the printer driver, block by block to avoid memory issues.

Cheers,
Bart