Quote:
Originally Posted by
nht 
He is. Power lines is the classic example that most folks are familiar with. The max observed human acuity is around 20/8 if I remember right and the average around 20/16. There are some sharp eyed folks out there.
Vernier acuity takes you down to around 8-10 arc seconds vs that 1 arc minute. This means a lot more in the real analog world I think than the digital one...the gigapixl camera can image to this level of detail but probably not much else...
Example from the Apple site:
http://images.apple.com/iphone/featu...5-20100607.jpg
In real life you can probably see all the power lines in that picture (all the way down that road). Even with a high end DLSR with good glass and huge 14MP sensor I bet you don't have that data in the raw imagery.
20/20 is considered to be a good normal eye. I'm seeing words like "perfect vision" being thrown about in this argument. But's that impossible. no one has perfect vision.
The power line example, and there are others, doesn't mean much when discussing displays. The absolute amount of light presented in a clear sky is far greater than what we see on our monitors, and especially in small mobile devices, where displays are less bright.
Contrast is everything. EVERYTHING!!!
Visual acuity simply is a meaningless concept without taking the contrast into effect. We have the same thing with telescopes. There is only one star that even the biggest current telescope can image, and that is our own. Every other star we see in a photograph is not being imaged. That is, it's not being imaged as a disk. What's happening is that so many photons are hitting that one sensing spot, that it images them. now, depending on the sensitivity of the sensor, that may be millions of photons, or just a handful, but it's enough to illuminate it, even though the photons aren't illuminating the entire diameter of the spot.
Our eyes have the same problem when looking at the night sky. we aren't imaging any stars as disks, just bright spots of light.
So the question is what that means.
I can set up an experiment in which a person is in a dark room. There is a sealed box in the room. A tiny pinhole is made in the side of the box facing the person, and inside, is a bulb. When we put the box so that the pinhole is further away than the eye can resolve we can still see the light coming from the pinhole IF it is bright enough!
But if we turn the brightness down on the bulb, at a certain point, it won't be seen. But if the box is brought closer, it will be seen again, until the brightness is again turned down. This can be repeated several times.
This is why when lenses are tested, or vision is tested, contrast is considered as well.
If we have an alternating pattern of white and black lines, where the white line is almost specular, and the black lines are total black, we will be able to see them, if we have a vision of 20/20 or about .6 arc minute.
But as we lower the contrast between the lines, our acuity drops. At a certain point, all we will see is a field of solid grey. Increase the width of the lines, and again we can see them at the lower contrast. Continue lowering the contrast, and again we just see grey.
This is true with color as well. We are more sensitive to green than to red or blue. It's very complex.
When someone comes on and says that we can see such and such, it's wrong. It depends on what we're looking at.
As far as photo's graphics and video goes, 300 ppi at 12" is considered to be about the finest detail that a normal, good eye can see. I totally agree, based on my decades in the photo industry, and my own bio studies.
Both what was written here, and Soneira's remarks are just wrong. They are taking the lab test results of human vision, which are intentionally very limited, as we just try to test ONE variable at a time, and imposing that onto the varied daily visual experiences we have.
I take particular exception to someone coming on and making specific remarks that have little to do with the real world situation. I'd hate to be in that class. It seems to be too limited.