Human visual acuity of 20/20 is being able to discern 1 arcmin in bright light. With trig, you get that, if you're holding the screen a foot away, that amounts to 0.000349 inches, which is 1432 dpi. If you're holding the device 2 ft away, it's 716 dpi. And against your face at 6 inches, you get 2865 dpi.
Could I impose for a citation for this? A quick check indicates that in the industrialized world, between 50% and 70% of individuals require corrective lenses.[1][2] That seems to indicate that the _average_ human does NOT have 20/20. It's possible in developing countries less than half the population does not need vision correction, but I'm not terribly convinced.
[1] National Centre for Social Research and University College London. Department of Epidemiology and Public Health, Health Survey for England, 2001. 2nd Edition. Colchester, Essex: UK Data Archive, June 2004. SN: 4628.
[2] Global Eyeglasses Market (2005). Eyeglasses MCP-2532: A global strategic business report. Global Industry Analyst, April 2008.
> “Normal” visual acuity for healthy eyes is one or two lines better than 20/20. In population samples the average acuity does not drop to the 20/20 level until age 60 or 70.
Thank you. I think I see where you're coming from.
That article has some interesting phrasing which I think is causing confusion. (For me at least.) It looks like they're using 'normal' to mean 'average in the range' some of the time and 'mean population visual acuity' other times. Mostly they're using the former. So 'normal vision' or 'average eyesight' refers to the range of 20/20 eyesight and not 'mean eyesight'. That doesn't explain that second sentence, though, which seems to go back to 'mean human visual acuity'.
At the very least, they could stand to revise the writing in that sentence. If they did mean to imply that the average person has 20/20 vision, I think they're going to have to provide a source. I did slightly more digging and I couldn't find any journal publications that list less than 60% of adults as having corrected vision. Either the remaining 40 have strongly superhuman vision or the number is off. I think there is at least one blind person (legally worse than 20/200) for every person with super human (20/10, the physical limit for humans) vision.
A simple numerical test based on that last point seems to illustrate the unlikeliness of a 20/20 average. To state again, 20/10 is the physical limit for human vision. Any higher and it's not physically possible to pack more photoreceptors into the retina or focus the pupil tighter. Bad (legally blind) vision is in the realm of 20/200. The National Federation for the Blind[1] shows there are roughly 6,700,000 blind Americans of all age groups. We will assume that they're exactly at the legal bound for blindness and not worse. To find how many superhumans we need to make the average 20/20, I think we can use (6700000200 + x10)/2 < 20 We need x to be less then -133999996. So it's simply impossible for the average vision to be 20/20.
It's still possible that number is correct and that the average American has 20/20, but at this point the majority of the scientific publications and formal surveys do not support the idea. If you can find additional sources (I'm so sorry to keep asking), please let me know.
No, the thing you are missing is that corrective lenses are a thing that exists. We're really quite good at taking someone with 20/100 vision and turning that into 20/20 or better.
Oh! Oooh! I see what you're saying. I thought you were asserting the mean _UNCORRECTED_ vision of an American Adult was 20/20. It's certainly much more plausible that the corrected vision is beyond 20/20.
I would recommend, for the sake of clarity, adding '20/20 corrected' or '20/20 rectified' to your original post. If I hear someone say, "I have 20/20 vision," my first thought is, "this person has perfect vision", not, "this person wears glasses."
I'll go back and add that clarification on my replies.
EDIT: It's too late to go back and edit my topmost comment. :(
Most of those people (IME) need the lenses to correct for short-sightedness, which if anything would make them better rather than worse at discerning phone pixels.
You use your phone at arms length? I am usually holding mine at like 6 inches to 1 foot from my face. I could very much tell the difference between the 318 ppi of the nexus 4 vs the 445 ppi of the nexus 5 with my usage patterns. I am skeptical that I would notice further improvements though.
> I am usually holding mine at like 6 inches to 1 foot from my face
Question: how do you type / swype with your device so close? I've just tried and my focus kept shifting from the back of my typing hand to the screen and back, very tiring.
I scroll/swipe left or right with my thumb, most of my use is reading comics or watching videos, video being where the resolution most matters doesn't really require much interaction at all.
As far as typing I hardly look at the screen then and ppi don't really matter then. If I am just checking my notifications it barely comes out of my pocket. But when we are in a discussion of ppi being useful, I assume we are talking about the uses where you are actually looking at the screen.
Edit: Just realized that when I type longer messages or emails I tend to hold it 2 handed landscape and hold it pretty close(around 1 foot away) and tap it out with my thumbs because I find it more accurate. So I guess that is how one would type with it that close.
I do not run flux, 6 inches seems accurate as if i stick out my thumbs they touch my glasses. Basically when I watch a video I want the screen to take up my entire field of view.
Is it even physically possible to notice the difference in that ppi at that range? I thought human eyes didn't have the capability.
I think everyone saying "yes I see a difference" should be doing some double blind validation. I've had people tell me they could tell the difference between 720p and 1080p at 20 feet. That is physically impossible for the human eye.
The human eye does not have a simple resolution like that, and yes it's VERY possible to notice PPI differences in the 300-600 dpi range, especially for certain forms of content.
At what distance though, without that your statement isn't helpful. We all know the human doesn't have resolution like a screen but if we ignore the optics of our eyes I think thats pointless. There are people saying they use their phones 6" away from their face. Ok, but that is not at all how anyone I know uses their phones at least distance wise.
Also 20/20 is not only not perfect vision, it's not even average [1]. So using that as the baseline is stupid, it's the low end not the high end.
Second the human eye does not have a fixed arc resolution, different specific detail scenarios have different degrees of accuracy. But if you're going to use this shortcut, you better go with a number on the upper end of things instead of the lower end, so more like .4 arcminute.
The net result is there is real, technical merit in going all the way up to 600 PPI assuming it's held 12 inches away. There's still arguably value to be had beyond that, but it starts getting more into just being a pointless spec war. Right now, though, we are very firmly well inside technical worth.
I guess thats the key, I never really have my phone closer than 12 inches from my face. I still think we'd be better off with better batteries than trying to pack more pixels into a window.
Most people don't watch their set from 20 feet away. And position is seldom static -- I tend to stand during football games, for instance, and move closer during intense plays or close-call replays. The same thing happens with smartphones, where people pull it in tight to look at fine detail, and a very high DPI allows you to do that with endless zooming/unzooming.
At arm's length, no. I don't know many people who stretch out their arms completely to use a cellphone or a tablet.
Usually people that I see (myself included, I have a Nexus 5 and a Nexus 7), hold our devices at about half an arm's length. At that distance, the pixels are discernible, especially if one has better than 20/20 vision.
It is easy for me to see jaggies on a 300 dpi device (the N7 and N10 are both in that ballpark, as was my old 720p phone). You can too -- go to a browser, and type "www" and then look at those letters. The jaggies are just what you'd expect from 300 dpi.
On the Nexus 5 (440 dpi), I can't see them anymore.
Resolutions above this seem unnecessary to me -- but for the N6, 1080p would mean 360 dpi, so that might be sub-perfect. 1440p will be perfect for sure.
> On the Nexus 5 (440 dpi), I can't see them anymore.
I can't see jaggedness on my iphone 4 at 'www' in the browser either, and that's at 326 PPI.
But fair enough. The low 400s range seems like a solid enough improvement. I just don't get the upper 400s like the Nexus 6's 493, and perhaps the 500 range next year. Obviously there's a bit of extra battery and a bit of extra processing hardware required for that, however small, I'd want to hang on to it beyond the low 400s PPI range.
Hmm, weird. I was going by my HTC One M7, which has a 1080p screen, but its ppi is 460ish, and the Nexus 6 is 493. I didn't realize an extra inch would require 1440p to have roughly the same resolution. I can't see jaggies anywhere at this resolution, though.
Also, don't be a stranger.