[Randomoneh]

As many of you know, if you keep increasing display pixel density, at some point there will be no real point to it because eyes won't be able to perceive those extra detail that resolution brings.


For many years, viewing distance has been recommended based on idea that apparent size of a pixel should be 1 arcminute. That translates to angular resolution of 60 pixels per degree.


But, if we want not be able to perceive higher quality, apparent size of a pixel has to be 0.3 arcminutes or less. That translates to angular resolution of 200 pixels per degree.



Full resolution here .


Of course, we can't just start recommending viewing distance based on angular resolution of 200 pixels per degree. Why? Because technology isn't there yet. We'd have to recommend either huge distances or small TVs.



Full resolution here .


So, my question is, now, at 2012, 4K is almost here, what angular resolution is good enough (not perfect) for recommendations of viewing distance? 100 plus?

 

[mtallent]

See the chart in my sig, I think it is pretty accurate. I set 7 feet from my 65 inch 1080P HDTV and I can just barely resolve a test signal of alternating black and white one pixel wide vertical lines. Further away and it just looks like a gray picture.


Mike T

 

[Randomoneh]



Quote:
Originally Posted by mtallent  /t/1418710/challenges-in-tv-viewing-distance-recommendations#post_22192167


See the chart in my sig, I think it is pretty accurate. I set 7 feet from my 65 inch 1080P HDTV and I can just barely resolve a test signal of alternating black and white one pixel wide vertical lines. Further away and it just looks like a gray picture.

Mike T

That chart is based on angular resolution of 60 pixels per degree. It may be good enough for you, however, it's not average.

 

[Otto Pylot]

What's wrong with just sitting at a distance that you feel comfortable with and you feel gives you the best resolution?

 

[Randomoneh]



Quote:
Originally Posted by Otto Pylot  /t/1418710/challenges-in-tv-viewing-distance-recommendations#post_22192661


What's wrong with just sitting at a distance that you feel comfortable with and you feel gives you the best resolution?

Nothing.


However, some things are wrong. For example, at distance x with xx inch TV viewer would benefit from 8k resolution. However, because he's using a chart / calculator that underestimates human vision by a great deal, he might buy 1080p thinking he wouldn't get any better image quality with 4k or 8k (provided content is there, of course) because chart told him he couldn't see the difference.


That's why I feel we should be true to ourselves and go with tested angular resolutions. We should at least be using 100 pixels per degree (apparent size of single pixel - 0.6 arcminutes), since most of quality rise happens to that point (see the chart).

 

[Otto Pylot]

Well I can't argue with the science but I think it's a silly argument to begin with. No two people have the same visual acuity, and while charts and stats are important and necessary in some situations, it ultimately comes down to what you like. It's the same argument as to whether you should calibrate with meters and software, software only, or just tweak to taste.

 

[Randomoneh]



Quote:
Originally Posted by Otto Pylot  /t/1418710/challenges-in-tv-viewing-distance-recommendations#post_22193312


Well I can't argue with the science but I think it's a silly argument to begin with. No two people have the same visual acuity, and while charts and stats are important and necessary in some situations, it ultimately comes down to what you like. It's the same argument as to whether you should calibrate with meters and software, software only, or just tweak to taste.

Try to think of it like this:


Human ear can register frequencies up to 20 kHz. You are out there buying headphones and you see one that can deliver 20 kHz and one that can deliver 40 kHz. There is no reason to spend money on latter if you won't be able to hear the difference. That's the case with displays too - people need to know where is the limit. Unfortunately, limit is almost always greatly underestimated.