[BigMackCam]

My wife has a 21 inch 4k iMac. We deem it unacceptable for final processing. Thats close to 200 dpi. We find that too small for adequate image evaluation. My old 27 inch iMac is better for 1:1 viewing at 2650x 1600 or closer to 110 dpi. My experience suggests that's about optimal for editing.

...

You are actually limited in this case, you by the number of pixels the screens, but by the number of pixels your eye can detect, which is estimate to be about 6MP while a 4k picture is 8MP. So in this case, it doesn't matter how many pixels you put up there, your eye doesn't have the capacity to differentiate if you are taking in the whole image.

Thanks, Norm. That makes sense.

So my next question... kind of alluded to in earlier posts, but not clearly enough:

Given that my eyes aren't really capable of evaluating individual pixels on a 15.6" 4K screen, if I increase reproduction ratio beyond 100% to, say, 200% or even higher, will that work? Or will the upscaling and use of multiple pixels to represent one real pixel cause problems / artefacts?

I'm quite comfortable with the fact that the 15.6" 4K screen - and indeed, probably, my existing laptop's 17" FHD screen - won't be ideal for purposeful evaluation of photos for serious use. I'm comfortable with the idea of investing in a larger screen for serious editing - but it won't be on my shopping list for some time. Until then, I have my existing 23" FHD external monitor which has been pretty decent for photo editing when I'm working from my home office...

[normhead]

Given that my eyes aren't really capable of evaluating individual pixels on a 15.6" 4K screen, if I increase reproduction ratio beyond 100% to, say, 200% or even higher, will that work? Or will the upscaling and use of multiple pixels to represent one real pixel cause problems / artefacts?

For us, it wasn't good enough, but we had a better screen to work on. We just didn't see it as clearly as on my old 27 inch iMac. But if you don't have an old 27" iMac, maybe 200% is the better option, I wouldn't try more than that though. I'd try and get my resolution set to an effective 100-150 dpi if I were attempting that.

A 17" 4k screen so what? maybe 14 inches across for 3840 pixels would be 225 dpi. I have no idea what values you'd get going to 2x, but it would seem like it should be close to the 100-150 range.

We just didn't find with our hardware it worked well for us using the 200 plus dpi machine. It was fine for most work, just for printing she like to bring her work over here. Less so now however than when we first got that computer. She seems to be getting used to it.

[BigMackCam]

Do yourself a favour and consider adding an external larger (than 15") screen.

It's definitely on my list, though I'd like to learn how to get the best out of the 15.6" built-in monitor too

4k TV sets are quite cheap these days and can be easily connected by HDMI or DP to laptops.

Is an inexpensive 4K TV likely to offer 100% of sRGB gamut?

[kaseki]

A useful site for learning about particular display capabilities and the testing thereof is TFT Central - LCD Monitor Information, Reviews, Guides and News.

For example, this review will show how well a particular monitor conforms to the sRGB and Adobe RGB mappings. (Also, I vaguely remember that there are very few displays that actually achieve 100% of either, and there is some reason for this that I have forgotten.) http://www.tftcentral.co.uk/reviews/asus_rog_swift_pg27uq.htm

---

Last edited by kaseki; 11-28-2018 at 06:23 PM.

[normhead]

A useful site for learning about particular display capabilities and the testing thereof is TFT Central - LCD Monitor Information, Reviews, Guides and News.

For example, this review will show how well a particular monitor conforms to the sRGB and Adobe RGB mappings. (Also, I vaguely remember that there are very few displays that actually achieve 100% of either, and there is some reason for this that I have forgotten.) Asus ROG Swift PG27UQ Review - TFT Central

My solution was to buy a Samsung monitor, and a Samsung TV, the monitor on my order iMac is also a Samsung monitor so colour values are consistent. That's all I need.

[BigMackCam]

A useful site for learning about particular display capabilities and the testing thereof is TFT Central - LCD Monitor Information, Reviews, Guides and News.

For example, this review will show how well a particular monitor conforms to the sRGB and Adobe RGB mappings. (Also, I vaguely remember that there are very few displays that actually achieve 100% of either, and there is some reason for this that I have forgotten.) Asus ROG Swift PG27UQ Review - TFT Central

Thanks for the info.

Just to be clear for the purposes of the thread...

The 15.6" 4K screen on the mobile workstation I've ordered covers at least 100% AdobeRGB and therefore well over 100% sRGB (indeed, it's guaranteed to do so), as well as having a 600 nits brightness rating at max setting. So gamut isn't going to be a problem so long as I'm able to apply suitable profiles, or select emulations, for different use cases.

The resoltuion continues to concern me based on a number of responses, though. Aside from getting a larger external screen - which I acknowledge will be helpful (or even necessary) for quality editing results - I'm still trying to figure out if there's going to be a negative impact on photo editing using the built-in 15.6" screen due to its high resolution, and whether magnification beyond 100% (which I'm certain will be necessary) will work effectively or introduce problems...

[kaseki]

First, I'm pretty well convinced I can't answer your question. The answer seems to relate to the question: When Mint or other OS magnifies an image, what sort of interpolation does it perform for smoothing as pixels are mapped into groups of pixels?

Some insight might be derived from this blurb, slightly old but still relevant for treatment of resolution and magnification.

Confused about HiDPI and Retina display? ? Understanding pixel density in the age of 4K | EIZO

[beholder3]

It's definitely on my list, though I'd like to learn how to get the best out of the 15.6" built-in monitor too
Is an inexpensive 4K TV likely to offer 100% of sRGB gamut?

I fear the situation is complicated. Probably your tiny notebook screen is the absolute best option to do color editing. And it is probably the worst option for most other aspects.

Using internet testing sites will help you find out about color spaces of TVs and monitors. If you dont mind using translators, then the site PC Monitor Tests & Fernseher Tests - Prad.de - Ausführliche Monitor Tests & Fernseher Tests offers a lot of very thorough TV and monitor tests.

But: I do wonder how relevant the color space thing is for you. Where do you watch your images the most? If that is a device which is not covering the color space as well, you are not really gaining something in the end. Then you have the question on calibration. IMHO calibrating your main photo watching device is more important than color space alone. For printing a robust chain of calibration profiles (monitor and printer) is even more relevant.

You should also keep in mind that the mentioned 6 MPx human eye resolution are valid for the 43mm limited field of view only. That means they are valid if your face is exactly 15.6" away from your screen - and that sounds a little close. I would guess your eyes will be farther away thus dropping effective resolving power to 3-4 MPx.


I use a calibrated not-expensive 49" TV set (still using 125% upscaling in windows) and like it very much. While I do know that probably some color information gets lost, the joy of watching my images (both during edit and later) in that size and glory this outweighs the drawbacks - for my personal priorities.


I suggest you make up your mind about

a) the way/end-devices you watch finished photos mostly and

b) which aspects are your priority.



It is a very individual question in the end.

[TonyW]

Me neither, I consider yours to be the last word.

Steve

(...neither agitated nor annoyed...)

Thank you for your kind and generous offer. I was hoping for discussion if not acceptance. Still I will accept and here it is Word.

Tony

(...definitely not agitated or annoyed, just saddened to see myths perpetuated in spite of a large body of evidence to the contrary...)

[normhead]

Did you put something out there for discussion?

From your post.....
Clarkvision: Printer detail and ppi

First fo all Roger N. Clark presents no credentials. He appears to be just some blogger.

The whole first paragraph is based on the following line....
"One arc-minute corresponds to 0.003 inch at a distance of 10 inches."

The inverse then is calculated at 344 dpi, at 10 inches..
My comment. 10 inches is a ridiculous viewing distance.

The inverse of 0.003 = 344, but it takes at least two pixels to resolve something, so double this number and we get about 700 ppi as the resolution of the eye at one arc-minute.

Why? He was talking about what the human eye can perceive, then mixes in totally extraneous information seemingly related to the use of paired lines in test charts, but anyone who has used an old 72 dot matrix printer knows you don't need two lines, so doubling this number is inappropriate. In any case, if you move to a more reasonable viewing distance you need less resolution, not more. Think of a bill board.

The eye can do a little better, so perhaps 1000 ppi is the limit.

Perhaps? Really? That's evidence?
No it's not, a real scientist would have tested. Theory without empirical support is useless.

Laser printers used to be 300 dots per inch (dpi), but evolved to 600 and even 1200 dpi. Why? People could see ragged edges on letters on 300 dpi laser printers. At 600 dpi edges appear smoother. Some can tell the difference between 600 and 1200 dpi printers if the paper quality is high. This attests to the power of the human eye. Of course, this applies to high contrast subjects. But many images have high contrast components too.

Applying print standards where the 6 MP resolution of the human eye can be applied to small sections of a page from close distances is simply in my opinion an inappropriate use of data when discussing photography. Thanks to printer technology my images that off the camera at 100 lw/ph look sharp printed at 300 DPI. I've never detected jaggies in print, although I'm quite happy to defer to the printers assertion than it might when viewing text.

Many powerful images with impact often have incredible detail. If that detail translates to sharpness (called "tack sharp" or razor sharp") in a print, it can give an image that "WOW" factor when people look at it. Can modern inkjet printers deliver such detail?

The author provides no support for this statement. Where can I see one of these images, what exactly are we talking about? With absolutely no support for this assertion we simply must discard it as heresay. Many powerful images have practically no resolution. WHo took the world's most expensive photograph? - PentaxForums.com
Perhaps the author is suggesting eye popping resolution is one way to create a stunning image.... but that as well is speculation. The author without any evidence whatsoever implies, that more DPI can turn an ordinary image into an eye popping one. With all due respect, a few visual examples are needed to make this point.

Below are the images printed on HP premium plus glossy photo paper with the image set at 150 ppi, 300 ppi, and 600 ppi. Tests of resolution charts show the HP printer just resolves 600 lines per inch (lpi), thus 600 ppi prints might show an advantage. The prints were scanned at 1200 ppi and downsampled to 900 ppi for this presentation.

NO, your printer does not resolve 300 lines per inch at 600 DPI, it resolves 300 line pairs. If the printer is truly 600 DPI, and each line is one dot wide either alternating black and white, it will resolve 600 lines, or 300 line pairs. To me the argument that you need a two pixel line to produce one line were true, that would be more an argument for fewer bigger pixels than more smaller ones. What's the point of printing a one pixel line if it's too small to see? Making the dots bigger would do the same thing.

As I pointed out, I we have a 200 dpi plus monitor here and any one of those images if they looked like that on my 200 PPI screen at 1:1 viewing would be tossed. MY suspicion is more than 1:1 resolution has been employed to create this effect.

In dim lighting, such as a 60 watt light bulb at 3 feet (~1 meter) I can't tell any difference between the 300 and 600 ppi prints. But in good lighting (daytime near a window, or typical office lighting) the 600 ppi print is noticeably sharper.

OK this is not even evidence, this is an opinion. Has he actually done this? Evidence would be a double blind test side by side from a meter away. The problems with evidence based on human perception are well documented and at times hilarious. The human capacity to believe what they want to believe in the face of overwhelming evidence is a real phenomena.

The ppi resolution of images it prints can be no better than half the specified dpi of the printer. For example, if your printer spec says it is a 600 dpi printer, it can not resolve 600 lpi, but at most 300 lpi. Printers usually need several dots to reasonably impart the proper color to the image pixel. A search on the net should find more explanation of printer dpi versus image ppi, except the test here show that conventional 300 ppi maximum is not true with all printers.

Nonsense. The author is talking about the paired lines on test charts. Again, back in the days of 72 PPI dot matrix printers, you could make a line visible with a one pixel wide line. The author is talking about a test chart, not an image. In an image where one part of an image may be say 10 dots wide and then become 11 then 12 then 13m , that one pixel makes a difference. In real life we rarely need test chart resolution.

I'm not clear on how this man carries on. He's already established that at 10 inches the human eye is capable of discerning 344 DPI. Now he's claiming we need 600 DPI to get 300. He keeps confusing resolving lines with resolution. The simple fact is, if you have more DPI than the human eye can discern because you've double the DPI, lets say your .003 arc is covering 4 pixels, your human receptor is now combing the information from 4 dots into one, and you've gained no more perceptual information, all you've done is reduce your ability to see that information. The limiting factor is the person, not the technology, so it doesn't matter how much you refine the technology, if the person can't make it out, then how does it add to the perception of sharpness?

On my HP1220C, which has 1200 dpi resolution, the closest lines are just resolved when the image is 1200 ppi, thus the printer resolves 600 lines per inch.

But at 10 inches we can only resolve 344 with our eyes, according to the author. My argument that modern printers and monitors produce much higher PPI values than is necessary for pleasurable image viewing. would seem to be supported. This is all about printer capability, and nothing about whether or not I actually like the print more done at higher res. In fact the only real evidence presented by the author suggests I do not.


This work is way to sloppy to be counted as evidence, and it's contradictory. It would seem to be a complete misrepresentation of the only real data in the article, that the human eye can discern 344 PPI at 10 inches. I suspect what the human eye can discern at 1 meter, at which distance the author claims to be able to see 600 ppi to be a lot less than that. The information I had prior to today was that a person could tell the difference between 300 and 600 DPI text at 8 inches, again, a ridiculous viewing distance for prints. Viewing at that distance, the viewer is not able to appreciate the compositional elements built into the print, only the technical clarity.

is that the discussion you wanted?
Do you have something better?

This is at best an anecdotal piece of speculation. At best, I personally believe the author is as fairly as possible relaying his perceptions, as altered by the his biases and pre-knowledge of which prints were which, perhaps he's on to something. But I have no way of knowing that. This is an article saying, I've done some messing around trying to illustrate something that I believe to be true, but it's not proof. IN fact, it's somewhat stunning in it's lack of any solid scientific work in 99% of the page. Nothing after the first line. The author is most notable for his ability to take two unrelated speculations and piece them into a point to support something he thinks is true, without supplying any empirical evidence to prove his speculation.

In this one the author strays from science in the first line, and just wanders further and further away into the realm of speculation complete with obvious misinterpretations about things that might (or might not) be of any relevance to a working photographer, with nothing but anecdotal speculation. This is not evidence this is speculation, maybe enlightened speculation, but speculation none the less.

The one thing I look for, a blind test with a group of at least 50 people, where one can demonstrate a significant preference for say 72, 144, 300 and 600 PPI prints of the same image, from say a meter away, where a preference for one of the prints exceeds 20% (which would indicate the results were not a product of random selection) by the standard deviation.

MY own observation based on my monitors (of which I have 4 ranging from 90 ppi to 220 ppi, is that from 20 inches, more than 160 dpi is wasted. And that is supported by the author's one scientific observation. That people can differentiate 344 DPI at 10 inches,. That suggests 160 DPI at 20 inches would be quite reasonable as a limit of human perception. At 1 meter (33 inches) my guess is 100 DPI would look sharp. Each human receptor taking in a .003 arc is going to take in more than one dot at 1 meter. probably a minimum of 4. Not only is the receptor going to combine 3 or four, that will decrease the resolution by 4x, meaning your 600 dpi is viewed as 150 DPI. Hope you're beginning to understand how ridiculous this argument is.

Equating line pairs, a way to measure resolution, to actual resolution is simply an unacceptable mistake. And anyone who's done the math knows, make a mistake in the first line, and the whole calculation is off, no matter how good the rest of it is. And in this case, the author just piles on error after error.

I fail to understand how anyone with the slightest bit of scientific training or training in statistics could take the linked article seriously for a half second after reading the first paragraph.

Hopefully this is enough to keep people from claiming to have all the correct answers based on this kind of "evidence" in the future.

[beholder3]

It is relatively trivial math to check the 6 MPx thesis.

You just need to take a simple office ruler with centimeter and millimeter markings and measure from which distance you can identify individual millimeter lines. Call it "one dot per millimeter" in one axis dimension.


If you can count the lines clearly from around 172cm viewing distance then you have a medical visual acuity (visus) of 1 (the human good standard).
That equals to resolving the 6 MPx.

[TonyW]

Thanks for the info.

Just to be clear for the purposes of the thread...

The 15.6" 4K screen on the mobile workstation I've ordered covers at least 100% AdobeRGB and therefore well over 100% sRGB (indeed, it's guaranteed to do so), as well as having a 600 nits brightness rating at max setting. So gamut isn't going to be a problem so long as I'm able to apply suitable profiles, or select emulations, for different use cases.

The resoltuion continues to concern me based on a number of responses, though. Aside from getting a larger external screen - which I acknowledge will be helpful (or even necessary) for quality editing results - I'm still trying to figure out if there's going to be a negative impact on photo editing using the built-in 15.6" screen due to its high resolution, and whether magnification beyond 100% (which I'm certain will be necessary) will work effectively or introduce problems...

Having used only briefly colleagues 4K monitors and printing from them I stand by my opinion as stated here. A larger monitor is always a nice thing to have and you will be perhaps a little limited at times by screen real estate but only time and your experience will tell if you can put up with a smaller section view of your ROI

The reason why 100% view may not cut it for your editing particularly noise and sharpening is that you will be looking now at images that were once say 3x magnified on your old monitor with your new 4k 100% zoom will not have the same magnification, so you will probably be uncomfortable and need to see images larger. Until you get your new monitor there is some guess work involved in what you will prefer to do

I think that these points may be considered factual and objective as opposed to subjective (but no less relevant):

Monitor resolution referring to a system whose quality how detail will be perceived by a viewer = Pixel format, screen size and viewing distance

Monitor resolution is impacted by the number of pixels in both horizontal and vertical

4K resolution is 3840 x 2160

4K resolution cannot be seen in isolation as affecting the quality as it also needs to be accounted for in the actual screen real estate and viewing distance – pixel density

The amount of pixels overall does not matter as much as the amount of pixels in a given area.

It must follow then that a monitor that is 30” wide will have a potential lower quality than a 12” wide monitor when viewed from the same distance. The identical quantity of 4K pixels spread over a wider area.

You may think or require different ways to measure/quantify ‘resolution’
If we express our monitor resolution as pixels per inch (PPI) then comparison of monitor resolution between different size monitors will always mean that the smaller screen has the higher resolution. Like the confusion that occurs with people using DPI when they should have said PPI (very common and no sin but..) it is probably as well to mention specific units of measurement

And It takes a lot more DPI to produce 1 PPI 🙄

[stevebrot]

First fo all Roger N. Clark presents no credentials. He appears to be just some blogger.

He has an extensive and impressive resume' and should know what he is talking about. That is no guaranty of course

ClarkVision.com: about R. N. Clark

My only complaint is that his Web site is sooo Web 1.0 and also hard on my eyes.


Steve

[BrianR]

He has an extensive and impressive resume' and should know what he is talking about. That is no guaranty of course

ClarkVision.com: about R. N. Clark

His photographic skills were also put into question, so his galleries are also worth a link:

Clarkvision.com Index of Photo Galleries

[dsmithhfx]

At first I found it a distraction, but got used to it very quickly. I just don't see it now. The important thing is your colour managed aps will be displaying correctly. You use firefox so follow these instructions to set colour management correctly: How to configure Firefox color management

Thanks for the link. I did not know you could configure browser color management.