Stats of the human eye

Discussion about everything. New games, 3d math, development tips...
Post Reply
3DModelerMan
Posts: 1691
Joined: Sun May 18, 2008 9:42 pm

Stats of the human eye

Post by 3DModelerMan »

I was thinking about the way the human eye works the other day, and I started wondering: if the eye was an LCD screen. What would it's effective resolution and refresh rate be? Eventually our TVs will reach that point and TVs won't need to be upgraded any more.
That would be illogical captain...

My first full game:
http://www.kongregate.com/games/3DModel ... tor#tipjar
cobra
Posts: 371
Joined: Fri Jan 23, 2009 2:56 am
Location: United States
Contact:

Re: Stats of the human eye

Post by cobra »

Well, the brain processes vision at approximately 60 hertz, and the resolution is infinite because there are no little sectors to represent a color. It's all just light being processed by our brains. Look up how the earliest cameras were designed and moved into the digital age and you'll understand how they had to figure out how to organize visual data.

Television won't reach that point any time soon. The data has to end somewhere, and infinite resolution couldn't really be possible because of this if nothing else. And since the human brain can only process visual data at ~60 hertz there's really no need to take it further than the higher resolutions we have today (some even being above 100 hertz) because at some point it won't make a difference anymore. I think anything above 60 hertz is to avoid eye strain even though it's not visibly apparent to us, right?
Josiah Hartzell
Image
CuteAlien
Admin
Posts: 9670
Joined: Mon Mar 06, 2006 2:25 pm
Location: Tübingen, Germany
Contact:

Re: Stats of the human eye

Post by CuteAlien »

As light is catched by photoreceptor cells in our retina, if you manage to control the amount of light reaching each one you should be able to create perfect visual illusions which can't be topped anymore. Wikipedia says it's 90-120 million rod cells and 4.5-6 million cone cells.
IRC: #irrlicht on irc.libera.chat
Code snippet repository: https://github.com/mzeilfelder/irr-playground-micha
Free racer made with Irrlicht: http://www.irrgheist.com/hcraftsource.htm
kazymjir
Posts: 727
Joined: Sat Feb 20, 2010 4:05 pm
Location: Munich, Bayern

Re: Stats of the human eye

Post by kazymjir »

I will not agree with you about resolution, Cobra.
The resolution don't have to be infinite.

First:
Science don't know what is the smallest object in our universe, and can't even proof that there is something like "the smallest object".
But we are talking about eye. Eye can't see everything in the universe. Eye sees only light.
So, there are little sectors to represent color. These little sectors are waves of photons which hit your eye, and they are representing color dependently of length of wave.

Second:
The same thing, we are talking about eye.
Can you notice single atom, molecule, or maybe something more bigger like viruses or bacterias without using microscope?
Why do you need infinite resolution if you can't notice such things?
You said, brain can process vision at about 60 hertz, but in our universe are lot of things that move in much higher frequencies.
But if brain can process vision only at about 60 hertz, "LCD screen" (which about 3DModelerMan mentioned) doesn't must have higher frequency.
Same is with resolution. If eye can't notice infinitely small object, or even things like bacterias, why this "LCD screen" must have infinite resolution?
3DModelerMan
Posts: 1691
Joined: Sun May 18, 2008 9:42 pm

Re: Stats of the human eye

Post by 3DModelerMan »

Doesn't the eye have an array of photoreceptors in the back? If that was the case then vision would be the brains interpretation of point cloud data right? So there's really no reason to go higher with refresh rate then.
That would be illogical captain...

My first full game:
http://www.kongregate.com/games/3DModel ... tor#tipjar
cobra
Posts: 371
Joined: Fri Jan 23, 2009 2:56 am
Location: United States
Contact:

Re: Stats of the human eye

Post by cobra »

kazymjir wrote:I will not agree with you about resolution, Cobra.
The resolution don't have to be infinite.

First:
Science don't know what is the smallest object in our universe, and can't even proof that there is something like "the smallest object".
But we are talking about eye. Eye can't see everything in the universe. Eye sees only light.
So, there are little sectors to represent color. These little sectors are waves of photons which hit your eye, and they are representing color dependently of length of wave.

Second:
The same thing, we are talking about eye.
Can you notice single atom, molecule, or maybe something more bigger like viruses or bacterias without using microscope?
Why do you need infinite resolution if you can't notice such things?
You said, brain can process vision at about 60 hertz, but in our universe are lot of things that move in much higher frequencies.
But if brain can process vision only at about 60 hertz, "LCD screen" (which about 3DModelerMan mentioned) doesn't must have higher frequency.
Same is with resolution. If eye can't notice infinitely small object, or even things like bacterias, why this "LCD screen" must have infinite resolution?
Hi. You misunderstood me. I said the screen couldn't have infinite resolution. And, as you said, I said that it doesn't need a higher frequency. And that's also untrue what you said, because a lot of LCD screens go far above 60 hertz.

Also, infinite resolution != an all-seeing eye, so what you say about microscopic objects doesn't make sense. Yeah, you can see more detail with what is large enough to see (and more sharply/defined), but it doesn't mean you see *everything* in existence.

The reason I brought all of that up is because he said this:
Eventually our TVs will reach that point and TVs won't need to be upgraded any more.


- Josiah
Josiah Hartzell
Image
hendu
Posts: 2600
Joined: Sat Dec 18, 2010 12:53 pm

Re: Stats of the human eye

Post by hendu »

The refresh rate varies by the person, and the area of the eye. The center is closer to 60Hz, the edges closer to 80-90Hz. This is why you can see CRT flicker much more easily from the sides.
ACE247
Posts: 704
Joined: Tue Mar 16, 2010 12:31 am

Re: Stats of the human eye

Post by ACE247 »

hendu wrote:The refresh rate varies by the person, and the area of the eye. The center is closer to 60Hz, the edges closer to 80-90Hz. This is why you can see CRT flicker much more easily from the sides.
Hey, wow! Tried it, its true!
Ps the way our eyes see is really determined by how our brains decode the info. Trust me you can see weird things if your brain doesn't do its job well.
devsh
Competition winner
Posts: 2057
Joined: Tue Dec 09, 2008 6:00 pm
Location: UK
Contact:

Re: Stats of the human eye

Post by devsh »

apple have actually gone to the effort of measuring eye resolution for their retina display. The retina display only works at >12 inches.
I would like to remind that your vision is projective and not orthogonal so the size of the eye-pixel will not be given in um or nm. It's steradians (solid angle), so at 20m away a display with 1mm pixels will surpass the resolution of your eye but at 10cm away you will be talking um sized pixels.
ACE247
Posts: 704
Joined: Tue Mar 16, 2010 12:31 am

Re: Stats of the human eye

Post by ACE247 »

Decided to expand on this old thread, hope someone reads it ;)

The human eye can see (maximum acuity) the difference between two lines .35mm apart at 1 meter, so maximum human visual acuity can be described as .00035 of the distance.

Really it's a relationship between size and distance, for instance if an object is 10cm^3 (that's 10cm wide x 10cm long x 10cm high) you would be able to distinguish it from the surrounding environment at a maximum distance of about 28571.42 meters away. That's 85714 feet or about 15 miles.

This is, of course extremely relative to the person we are talking about, the example above is someone with extremely good eyesight and taking into account that this is as seen with an ideal combination of both rods and cone receptors in the eye. Red or green objects would be more discernable in good lighting conditions(daytime) while green and blue would be more visible in poor lighting conditions. The rod receptors of the eye are not sensitive to red and most numerous (~120million) while the cones(colour vision) are only spread across a 0.3mm area in the ratio of 64% red receptors, 34% green and 2% blue. With blue being helped by the fact that blue light is of a lower wavelength and hence higher energy.
devsh
Competition winner
Posts: 2057
Joined: Tue Dec 09, 2008 6:00 pm
Location: UK
Contact:

Re: Stats of the human eye

Post by devsh »

dear friends we'd need a NEW way to display on screen

I think instead of having a beyond repair method of RGB display on LCDs etc where one R per one G per one B
we should actually START making diodes which are capable of displaying YUV directly???

this makes sense as the 120 million rods are most sensisitve to intensity (Y/luma)
and the 6 million cones actually give us UV/color

this is why movie formats get away with 4:2:2 or even 4:1:1 YUV where only the luma is 1080p but the UV are 540p or 220p

I think we should use a 20:1:1 YUV format for display with significantly great DPI (not resolution, because your 52" TV gives you a shitter pixel packing than a 15" FHD monitor)
REDDemon
Developer
Posts: 1044
Joined: Tue Aug 31, 2010 8:06 pm
Location: Genova (Italy)

Re: Stats of the human eye

Post by REDDemon »

This tecnquique is already used in game engines to reduce memory bandwith usage on the GPU for certain shaders.
Junior Irrlicht Developer.
Real value in social networks is not about "increasing" number of followers, but about getting in touch with Amazing people.
- by Me
Iyad
Posts: 140
Joined: Sat Mar 07, 2009 1:18 am
Location: Montreal, Canada

Re: Stats of the human eye

Post by Iyad »

cobra wrote:Well, the brain processes vision at approximately 60 hertz, and the resolution is infinite because there are no little sectors to represent a color.
Actually this is not true, the eye and brain does really not work like a computer, theres no refresh rate neither frameskip in eyes or brain. simply because the light is captured by chemicals (retinal and opsins) which change the conformation of these proteins and by a biochemical cascade, hyperpolarize membrane and send information to the brain, and the brain does not wait to receive these messages to interpret them. if some of the rods and cones membrane are still at the resting potential, no information will be sent to brain but you can have an image formed by the other rods and cones...

And the eye does have a resolution, it is determined by the angle between 2 ray of light that become undistinguishable.
Last edited by Iyad on Mon Feb 06, 2012 4:11 pm, edited 1 time in total.
#include <Iyad.h>
Mel
Competition winner
Posts: 2292
Joined: Wed May 07, 2008 11:40 am
Location: Granada, Spain

Re: Stats of the human eye

Post by Mel »

The fastest neural processes take as short as 0.0025 secs, so, our vision could reach 400hz, this doesn't happen because of the rest of the processing time.

Besides, the objects which fall on the center of the field of view of the eye are seen bigger than those which fall in the surroundings, and those seen out of the center, are also perceived as intensities (the color comes from the brain). But not only that. The eye doesn't see continuously, but diferentially. That is, we only perceive the diferences between two stimuli. If there was no diference (i.e. the eyes were completely stopped) we would see nothing, we see because we have 2 eyes, and they move randomly, very little, but enough to give the feeling that the vision is continuous, and threedimensional. The brain also builds the rest of the information we need based much more on the spatial memory than the temporal (we see better in places which are familiar to us, we form rapidly shapes from the images we see because the brain is used to that and so on, that is the base of the optic illusions). The human eye also sees a range of 2.000.000 colors.

The best a LCD TV could do is to increase the image rate so the motions were natural on every moment.
"There is nothing truly useless, it always serves as a bad example". Arthur A. Schmitt
Post Reply