What's the best way to compare papers?

[gonzalo]

Hello everyone,

I'd like to make a good selection of papers so I bought a few test packs and downloaded the Bill Atkinson test image. I was about to start printing when I thought whether I should soft proof the image for each paper or not. If I soft proof, I'll have better results and see the properties of the paper but it won't be the same image for all the different papers so it might not be a good comparison. On the other hand, if I just print it without prior soft proofing to see how the paper renders the image, I'll be able to see the differences among the papers but they might not be showing their full potencial, so the comparison isn't good either.

What do you think? How do you normally do it?

Thanks!

[Jager]

Well, I suppose it depends on what your goal is in comparing them.  If you're evaluating them to determine their individual characteristics and qualities, with an eye towards potentially choosing one or more for ongoing use, I'd individually optimize each via separate soft-proofs.  Especially if that's your normal workflow.

Welcome to the forum!

[Mark D Segal]

Hello everyone,

I'd like to make a good selection of papers so I bought a few test packs and downloaded the Bill Atkinson test image. I was about to start printing when I thought whether I should soft proof the image for each paper or not. If I soft proof, I'll have better results and see the properties of the paper but it won't be the same image for all the different papers so it might not be a good comparison. On the other hand, if I just print it without prior soft proofing to see how the paper renders the image, I'll be able to see the differences among the papers but they might not be showing their full potencial, so the comparison isn't good either.

What do you think? How do you normally do it?

Thanks!

Yes, you should soft-proof for each paper and profile combination you intend to work with, then actually make the prints and compare them. Be sure that the viewing conditions are consistent with the softproof conditions. There is a lot of educational material on this website taking you through all of that. The best comparison is just to look at them for the usual characteristics of print quality and see which you like best, then focus on perfecting your print preparation and printing using one or two papers to begin with - get to know them. You could also look up my various recent articles on this website, which are replete with different ways of comparing print quality based on some usual parameters. But in the final analysis, what you see on paper and what grabs you counts most.

[stockjock]

Well, I suppose it depends on what your goal is in comparing them.  If you're evaluating them to determine their individual characteristics and qualities, with an eye towards potentially choosing one or more for ongoing use, I'd individually optimize each via separate soft-proofs.  Especially if that's your normal workflow.

Welcome to the forum!

There are two different attitudes you can take to this.  If you are trying to find the paper that can produce the optimal print for your photographic style then you might want to soft proof them and optimize and evaluate the results.  And a perfectionist might even argue that you should custom profile each paper.  My attitude was a little different when I was going through this process.  I wanted to find the optimal paper for my style AND minimize my post processing time.  When I was printing on an Epson 3880 with Exhibition Fiber Paper I found I always had to tweak the soft proof to get the results I wanted.  With my current combination of the Canon iPF8400 and Canson Platine I rarely have to soft proof and tweak in order to get the results I want.  For me that is very valuable since it reduces my time post processing.  Your standards and style might be completely different of course. 

[GrahamBy]

then actually make the prints and compare them.

Amen.

Make prints of a photo you like, which you think represents your style or praxis or whatever, then pin them to a wall where you can see them regularly. Give yourself time to get over the "Wow! That's different, it must be better" and the "Yuck! That's different to what I'm used to" phases.

You might even end up liking several...

[Ernst Dinkla]

Hello everyone,

I'd like to make a good selection of papers so I bought a few test packs and downloaded the Bill Atkinson test image. I was about to start printing when I thought whether I should soft proof the image for each paper or not. If I soft proof, I'll have better results and see the properties of the paper but it won't be the same image for all the different papers so it might not be a good comparison. On the other hand, if I just print it without prior soft proofing to see how the paper renders the image, I'll be able to see the differences among the papers but they might not be showing their full potencial, so the comparison isn't good either.

What do you think? How do you normally do it?

Thanks!

The Bill Atkinson test image will pull the full potential from your paper choices + perceptual rendering with the proper printer profile in the application you print from and the media preset correctly set in the driver. I do not see what soft proofing adds to that, I doubt you can pull more from the selected paper by editing that BA image based on the soft proof and then make your print to compare with other paper qualities. You only introduce more variables if you go that way.  The best quality papers will show their quality fast. The color space assigned to your images, the gamut covered by your monitor and the viewing light not mentioned here. Go for the BA prints, judge them by sensible light and after that fill in the gaps in your workflow. Editing based on the soft proof might come in handy when papers with less quality limit the representation of the image, or the other way around.

Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2016 update, 700+ inkjet media white spectral plots

 

[Alan Goldhammer]

The Bill Atkinson test image will pull the full potential from your paper choices + perceptual rendering with the proper printer profile in the application you print from and the media preset correctly set in the driver. I do not see what soft proofing adds to that, I doubt you can pull more from the selected paper by editing that BA image based on the soft proof and then make your print to compare with other paper qualities. You only introduce more variables if you go that way.  The best quality papers will show their quality fast. The color space assigned to your images, the gamut covered by your monitor and the viewing light not mentioned here. Go for the BA prints, judge them by sensible light and after that fill in the gaps in your workflow. Editing based on the soft proof might come in handy when papers with less quality limit the representation of the image, or the other way around.

I agree with Ernst on this statement.  The Atkinson test image (or its closest relative the Outback test image) are designed to show up key issues with paper/ink combinations.  One needs to minimize any variables when printing so a proper comparison can be made.  The http://www.outbackphoto.com/printinginsights/pi049/essay.html discusses the important things to look for and were one to soft proof some of these would be altered.

Alan

[Mark D Segal]

The Bill Atkinson test image will pull the full potential from your paper choices + perceptual rendering with the proper printer profile in the application you print from and the media preset correctly set in the driver. I do not see what soft proofing adds to that, I doubt you can pull more from the selected paper by editing that BA image based on the soft proof and then make your print to compare with other paper qualities. You only introduce more variables if you go that way.  The best quality papers will show their quality fast. The color space assigned to your images, the gamut covered by your monitor and the viewing light not mentioned here. Go for the BA prints, judge them by sensible light and after that fill in the gaps in your workflow. Editing based on the soft proof might come in handy when papers with less quality limit the representation of the image, or the other way around.

Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2016 update, 700+ inkjet media white spectral plots

I fundamentally disagree with this - based on practical experience making thousands of prints on different media. Softproofing is a valuable aid to optimizing perceived image quality because without it you cannot properly predict the tonality of the printed image - the more so for MK papers than for PK papers, but true for both. Good printing is about reliable predictability between what you see on your display versus what comes out of the printer. Softproofing helps to mitigate the perceptual gaps between transmissive and reflected light.There is a reason why the photographers who use Lightroom and print to high photographic standards of excellence were very pleased when Adobe introduced softproofing to that application, and indeed why Adobe put the effort into developing it.

As for Rendering Intent, there is no valid recipe of one size fits all. Some photos look better with Perceptual RI, others with RelCol, and still others with Absolute or Saturation RI. There is a reason why Photoshop offers the four options and Lightroom two. RI is one variable amongst others to be used for optimizing the photograph and the softproof allows one to predict its impact.

[Alan Goldhammer]

Mark,

I think you are missing the key point here and that is how to initially compare papers not to maximize the quality of the print.  The BA or Outback test images have been designed to pick up flaws in printer/paper/ink combinations.  They are already for printing and it really doesn't matter whether you have a calibrated display or not - these test images will print fine without any further manipulation.  As Ernst points out one needs to pick the correct printer setting for the paper in question.  Any attempt at soft proofing is likely to compromise one or more of the selected images on the test print.  One wants to find the strengths and weaknesses of each paper in the absence of soft proofing.  I don't know how many sheets of each paper come in sample packs as it's been a while since I've purchased one.  If one has only 1-2 sheets of paper, printing out an unmanipulated test print of either the BA or Outback image set that will provide one with a good evaluation of the strengths and weakness of the paper.

Clearly, moving on from this point one would need to print out a number of different prints from one's own images using soft proofing to maximize image quality to decide whether the paper is worthy of a long term investment.

[GrahamBy]

Is the question to discover if certain papers in combination with the OP's printer have flagrant problems?

Or is it to discover which style of paper provides nuances more in line with Gonzalo's preferences?

[Mark D Segal]

Mark,

I think you are missing the key point here and that is how to initially compare papers not to maximize the quality of the print.  The BA or Outback test images have been designed to pick up flaws in printer/paper/ink combinations.  They are already for printing and it really doesn't matter whether you have a calibrated display or not - these test images will print fine without any further manipulation.  As Ernst points out one needs to pick the correct printer setting for the paper in question.  Any attempt at soft proofing is likely to compromise one or more of the selected images on the test print.  One wants to find the strengths and weaknesses of each paper in the absence of soft proofing.  I don't know how many sheets of each paper come in sample packs as it's been a while since I've purchased one.  If one has only 1-2 sheets of paper, printing out an unmanipulated test print of either the BA or Outback image set that will provide one with a good evaluation of the strengths and weakness of the paper.

Clearly, moving on from this point one would need to print out a number of different prints from one's own images using soft proofing to maximize image quality to decide whether the paper is worthy of a long term investment.

Hi Alan,

If hErnst is limiting his comment to the printing of test targets, I agree - soft-proofing is irrelevant. I thought he was making a more general statement. If I erred in that, I have no further disagreement.

[Mark Lindquist]

Hi Alan,

If hErnst is limiting his comment to the printing of test targets, I agree - soft-proofing is irrelevant. I thought he was making a more general statement. If I erred in that, I have no further disagreement.

Actually, I think both Ernst and Mark are both right.  The way I would do it would be to make a simple print and a soft-proofed one.  Your instincts about how you want to work, and your sense of work-flow should guide you.  If you are inclined to make a soft-proofed image, then make it.  If you wonder how it would have been in a simple print, do that too.  Testing is all about comparisons.  Not enough paper?  Make half sheets or test strips.

-Mark

[Mark D Segal]

Actually, I think both Ernst and Mark are both right.  The way I would do it would be to make a simple print and a soft-proofed one.  Your instincts about how you want to work, and your sense of work-flow should guide you.  If you are inclined to make a soft-proofed image, then make it.  If you wonder how it would have been in a simple print, do that too.  Testing is all about comparisons.  Not enough paper?  Make half sheets or test strips.

-Mark

Mark L.,

When you print, you are automatically always making a "softproofed" print. All the softproof does is show you *on your monitor* what is likely to come out of the printer, assuming your profile supports this reverse flow of information. You can use the softproof view of the printer target on the monitor to compare it with what comes out of the printer. If the two look similar, it means that your colour management set-up is OK. The advantage of using the target is that its colours and gradients (and their values) are well-known, so you can see in a jiffy whether the testing and comparing is leading you in the right directions. May I assume this is what you meant?

[Ernst Dinkla]

Hi Alan,

If hErnst is limiting his comment to the printing of test targets, I agree - soft-proofing is irrelevant. I thought he was making a more general statement. If I erred in that, I have no further disagreement.

It was not a general statement on soft-proofing/editing/printing that I made. In fact I described how I test/compare new papers to the papers I already use. In general I am not a big fan of soft proofing either but in this case I see it as a fundamental mistake to add it to the paper testing process. IMHO :-) It is like editing printer profiles to make them better, if needed it usually means the profile creation was bad in the first place. Bill Atkinson made terrific targets, don't fool with them is my opinion. If the printed target is bad it is either an issue with the paper quality or the color management choices (including bad printer profiles), not the target.

Edit, I see that the term soft proofing here is either used as a passive step in the process or a step that includes target image editing to the soft proof. I find the last fundamentally wrong and am afraid that the first does not offer enough information to guide a user to optimal paper qualities or best qualities. I really wonder why, when one aims to get the real thing in sight for the first time, virtual views are considered on that path. I am not even mentioning the texture, gloss, tactile aspects that add to that experience. For once start with printed proofs, learn the difference between monitor images and prints on that paper and then in practice judge soft proofing for that paper.

BTW, for the OP, you might waste ink, time and paper if you go blind into paper purchasing + testing. Several brands have the same or similar qualities with different labels. You could at least compare spectral plots available in SpectrumViz and guess which papers may print similar. Spectral plots are not telling 100% what the print may look like but I am sure it can make your testing process shorter.


Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2016 update, 700+ inkjet media white spectral plots

[Mark D Segal]

It was not a general statement on soft-proofing/editing/printing that I made. In fact I described how I test/compare new papers to the papers I already use. In general I am not a big fan of soft proofing either but in this case I see it as a fundamental mistake to add it to the paper testing process. IMHO :-) It is like editing printer profiles to make them better, if needed it usually means the profile creation was bad in the first place. Bill Atkinson made terrific targets, don't fool with them is my opinion. If the printed target is bad it is either an issue with the paper quality or the color management choices (including bad printer profiles), not the target.


Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2016 update, 700+ inkjet media white spectral plots

Hi Ernst, I think we are agreeing to agree. :-). But let me just suggest this: The nice thing about a target like Bill's is that those images and what they "should" look like hang out in one's mind. I can print one of these things and see almost instinctively by now whether or not the result is correct. And of course when we print it, we just print it - it's a baked cake - there is no role for softproofing at that point, because one is not going to adjust it. It is a set of test parameters. All that said, especially for those less experienced with these things,  softproofing comes in handy as an aid to assessing quality of the printed output  when printing on a matte paper for which the perceived (and measured) maximum black is far less than that of most monitors and even the paper white may be noticeably less "white" depending on the paper. Especially in those conditions, what comes out of the printer cannot be assuredly judged for adequacy unless you do compare it with its softproof view on the monitor. With a professional printer and wide gamut, good DMax paper such as Canson BP, IGFS, Epson Legacy PK papers, etc. viewing the softproof to compare with the print is less important. Of course if the printed output appearance deviates substantially from its intended softproof equivalent on the display, this is not an invitation to play with the target; it is a signal to look for a colour management or printer performance issue somwhere along the line.

Cheers,

Mark

[Mark Lindquist]

Mark DS, yes.

Ernst, it's a good thing Jeff Schewe isn't around, LOL.

It is refreshing to hear your opinions and the encouragement to learn the craft of printing without relying on soft proofing.

For both of you, Mark and Ernst, simple is as simple does.

When an individual is learning, they must be free to make their mistakes to more fully understand.

Think of the combined years of knowledge you both have.

Cheers!

Mark

[gonzalo]

Thanks everyone for the replies. I think there is an interesting discussion going on.

First of all I must say that I work in a University and this selection is for the lab there so the students will be the ones choosing the paper the will print on. So it´s not just about about my personal style or preferences.

That said, I agree with Mark that, in general, softproofing helps you to visualize the result and therefore to maximise the print quality. At least I think I get better results that way. But my question about softproofing or not was aimed just for this paper comparison. In that case, according to most of you, it seems that softproofing is not a good idea, right?

[Mark Lindquist]

Thanks everyone for the replies. I think there is an interesting discussion going on.

First of all I must say that I work in a University and this selection is for the lab there so the students will be the ones choosing the paper the will print on. So it´s not just about about my personal style or preferences.

That said, I agree with Mark that, in general, softproofing helps you to visualize the result and therefore to maximise the print quality. At least I think I get better results that way. But my question about softproofing or not was aimed just for this paper comparison. In that case, according to most of you, it seems that softproofing is not a good idea, right?

In the case you site, absolutely not.  It will be confusing enough.  As Ernst has said - make the prints, make the comparisons.  Simple is, as simple does.

Mark L

[Pete Berry]

Hello everyone,

I'd like to make a good selection of papers so I bought a few test packs and downloaded the Bill Atkinson test image. I was about to start printing when I thought whether I should soft proof the image for each paper or not. If I soft proof, I'll have better results and see the properties of the paper but it won't be the same image for all the different papers so it might not be a good comparison. On the other hand, if I just print it without prior soft proofing to see how the paper renders the image, I'll be able to see the differences among the papers but they might not be showing their full potencial, so the comparison isn't good either.

What do you think? How do you normally do it?

Thanks!

I did this a few years back in the early days of the Baryta Revolution, comparing four barytas with two mid-grade RR's. All inked identically, printed through the iPF5000 driver with the "Special #5" media type - the max ink density setting for PK papers then. I was struck much, much more by the similarities rather than the differences I expected to see. What differences I saw seemed more related to the base paper tint rather than any color or gradation shifts. I did the same with MK papers, but with them there was a clear progression in IQ correlating with paper cost.

Here's a link to V700 scans of the six papers with identical settings, with the white point adjusted so that base tint differences are seen:

https://www.flickr.com/photos/8075035@N03/albums/72157624569134778

Pete

[Mark D Segal]

Gonzalo, let me rewind to your opening post, which for reminders said this:

I'd like to make a good selection of papers so I bought a few test packs and downloaded the Bill Atkinson test image. I was about to start printing when I thought whether I should soft proof the image for each paper or not. If I soft proof, I'll have better results and see the properties of the paper but it won't be the same image for all the different papers so it might not be a good comparison. On the other hand, if I just print it without prior soft proofing to see how the paper renders the image, I'll be able to see the differences among the papers but they might not be showing their full potencial, so the comparison isn't good either.

In the final analysis, it all depends on what your objective is. Perhaps we haven't focused enough on that. If your focus is mainly to see what each paper produces using a standard test image with no manipulation, the soft-proofing is indeed irrelevant. The printer/paper/profile will do what they do and you compare results. That certainly gives you indicative comparisons, but it may not give you all you want to know about full potential from a real-world photographic print workflow, because the photos will all need editing (a test target does not) and the kind of editing you do would likely vary depending on the paper/profile being used.   

So, if your objective is to see how you can optimize the print for each paper, then I wouldn't recommend using that target in the first place. As there are so many kinds of images, there is no "one size fits all". So in this case you would select a number of photographs having the characteristics that are most relevant to you, and print the same photo using no more than two or three different papers, each custom-optimized with soft-proofing to see which gives you the most pleasing outcomes. In a way, I did just this in my review of the Epson Legacy papers, so if you go there, you will see how editing plays differently for different papers/profiles and different kinds of photographs - and where soft-proofing is so useful. Once you have a small pile of prints of say at least half a dozen different kind of photographs each printed optimally on several papers, you will gain an appreciation of what paper you like best for which kind of photograph.