human eyes vs. film

[Filip Plesha]

If you could analise how much light does the human eye need to get an image

would it be possible to say to what film ASA speed does human eye compare?

and what would that speed be?

 

I know it's a bit difficult to do that because there is no exposure

time,it is instant.

But let's supose that we see in 24 fps like film cameras do.

so considering how deep the human eye can dig into shadows when

the "iris" of the eye is wide open,what would that film speed be?

Would it be like 1000ASA or more?

[Guest Ultra Definition]

I think that you need to qualify what F stop is the lens. Some kid posted here that the eye opens up to F2.8. Another problem is that the eye loses the ability to see color at low lighting level. It's B/W sensitivity is many times higher than color sensitivity, plus at low lighting levels it is more sensitive to higher color temperature light. You can increase the color temperature and gain 3 or 4 F stops in sensitivity.

[Filip Plesha]

Yes but there must be a way to determine this.

If you consider that your eyes see on moonlight,or even starlight

(they are probably wide open then) and if you imagine

that your eyes are a camera photographing at 24fps

you would be able to say an ASA speed equivalent.

 

From my photographical experience i would say that

you need about one minute exposure with a ASA 800 film

to get a picture of a subject iluminated by starlight under f-stop of 2.8

 

Now if you convert this to 1/48 sec exposure time for

a 24fps camera,you would need a 3 270 000 ASA film

which is only 10 stops faster than 800 ASA

 

So if human eye was a MP camera loaded with film shooting at 24

it would have this speed if you consider a starlight iluminated subject

to be the limit of the human eye.

Pretty amazing.

[Guest Ultra Definition]

I think that at 10 Lux the eye still sees color, although not as vividly as at higher levels. It's B/W light sensitivity may be 5 to 10 F stops higher, depending on the color temperature of the light source. But all these figures are just guesses.

[Phil Rhodes]

Hi,

 

I have read that the human eye is capable of detecting "a few" photons impacting a rod or cone site. It is extremely sensitive, although like most electronic imaging devices, it loses colour rendition and becomes soft and noisy in the dark.

 

The problem is that it's almost impossible to quantatively test the human visual system because your brain makes on the fly decisions and fills in your impression of sight with information from memory and other senses. Everybody's brain makes different decisions on this kind of thing.

 

Phil

[Filip Plesha]

Yes i agree. There is the effect of filling the gaps with imagination and memory.

[Rob Belics]

Let's see. The retina is made of rods and cones. The cones are responsible for color while rods for intensity. But rods are about ten times more sensitive than cones which is why images lose color in darkness. There are also about 20 times more rods than cones.

 

So there is a lot of processing going on in the eye itself since there are only about a million fibres that leave the eye going to the brain.

 

The focal length of the eye is from 12.5 to 16mm. The critical flicker frequency is around 60hz, meaning the eyes temporal response can detect flicker slower than that but can adapt as in the movies.

[Fredrik Backar FSF]

One also comes to think in terms of resolution.... It must be quite high I suppose?

I do however have some faint memory that there aren`t that many receptors in all...

 

Another thing. Is the eye sensitive to "invisible light" too. Ultra violet and so on. But the brain only processes the "visible" range???

[Filip Plesha]

I don't think the resolution is that high.

I think its maybe like a fine grain 65mm negative (not the print

or a copy)

 

You can test the resolution by putting images of higher and higher resolution in

front of you,after some time you wont see the difference.

[Phil Rhodes]

Hi,

 

> Is the eye sensitive to "invisible light" too. Ultra violet and so on.

 

Not as far as I know. The rods and cones just don't react to wavelengths outside the visual spectrum. Which, er, is why you can't see them.

 

Phil

[Guest Ultra Definition]

the resolution is quite high, but only in the cener, then things get blurry. Check for yourself. You see nearly 180 deg. horizontally, but the angle, where it's very sharp is narrow; there is also a lot of signal processing going on in the brain to vary that angle and I guess the wider it gets, the less resolution you get.

[Filip Plesha]

It is high,but it is not much higher than a medium format still

negative or 65mm film negative.And i mean that in the area

where your brain is activly scanning the image.

[Alvin Pingol]

Speaking of the eye vs. film, I'm actually quite glad they don't react the same to light.

 

Sure, the eye can see dozens of stops of overexposure, but this is because of temporary retina burn, e.g. looking into the sun for 1 second, you have to wait a few minutes for your eyes to return back to normal.

 

Or if you watch TV for a long time at night, you have a huge dark spot in the middle of your eye's image for a few minutes.

 

Those old tube CCDs were sort of like this, weren't they? ;)

[Guest Ultra Definition]

The old tube CCDs were not CCDs but pickup tubes or tube sensors - Vidicons,

Saticons, Newicons, etc.

 

The eye can't see dosens of stops of overexposure. It's total range is about 15 stops.

 

You can burn out CCD by pointing camera at the sun.

 

You can get dead CCD pixel(s) just by taking your camera in the plane and having it in the luggage and turned off all the time.

[Dominic Case]

Is the eye sensitive to "invisible light" too. Ultra violet and so on.

Depends on what you mean by "sensitive".

The rods and cones only detect "visible" light. That's why it's visible. However, the retina can be burned by UV just as skin can. Cheap sunglasses that block visible light but not UV are very dangerous as they cause the iris to open up, so admitting more UV rather than less.

 

Actual visual sensitivity varies enormously: for one thing, the fovea or centre area of the eye contains mostly colour-sensitive cones, whereas the rods (more sensitive but colour blind) are in the peripheral areas of vision. Also the overall sensitivity of the eye can be varied by about 10,000:1. This is adaptation, and is affected by the iris, by sensitivity of the rods & cones, and other factors less well-understood. It's like being able to adjust from a film emulsion of EI 10at f/1.4 to EI 6400 at f/16 (roughly). Though I'm not sure if that range is correct in absolute terms.

 

At any of that range of settings, you can still see a full tonal range of at least 10,000 to 1.

[Bill Totolo]

My Spectra meter goes down to 0.1 footcandles. I can actually see pretty well under those circumstances once I adjust, depending on the color of the studio and objects around me.

 

So the eye must have phenominal low light sensitivity. I can only imagine how sensitive a cat's eyes are.

[Rob Belics]

A book I have says a moonless, cloudy night is .00001 candles per sq meter.

[Alvin Pingol]

>>The eye can't see dosens of stops of overexposure. It's total range is about 15

>>stops.

 

Is that so? I thought it would've been much more, for you could look into a dim lit room and "frosted glass" lightbulb at the same time, and be able to see ( a ) detail in the room, and ( b ) the texture of the frosted glass. The only time your eyes really blow out is if you look right into a bright lightbulb, i.e. you can't see the curly tungsten filament (unless you squint), or looking at the sun.

 

Also, and correct me if I'm wrong, but unlike film or CCD, if you keep pouring on more and more light onto a colored object, e.g. a red sweater, it won't eventually "burn out" to white. The red will just keep on getting brighter and brighter... right?

 

>>You can burn out CCD by pointing camera at the sun.

 

Yes, but as opposed to our eyes, pointing a CCD at the sun for 1 second won't do much, but looking at the sun for 1 second leaves a nasty "dark spot" on our images for about a minute.

 

Of course, too much sun time on either CCD or the eye is disasterous. ^_^

[Dominic Case]

you could look into a dim lit room and "frosted glass" lightbulb at the same time

 

That's probably not much more than 15 stops: but in any case there's some adaptation going on as you swing your eyes from the shadows to the bulb, - a bit like pulling the aprture as you pan the camera. Also, while you are looking at the bulb, your (more sensitive) peripheral vision is coping with the shadows.

 

The range of adaptation of the visual sytem to different light levels is more powerful than the range in any fixed situation.

[Mitch Gross]

Hey, I finally know what Dominic Case looks like! Maybe I should stick in my photo...

 

Any image in a film frame if exposed bright enough will go white no matter what the color. Eventually it is so bright that the information is lost. The same is true for the human eye, although it can be pretty painful to prove. At night, sit in total darkness for a long time so that your irises dilate fully and your brain is trying to pull as much info as possible from your rods & cones. Then switch on a bright light keeping your eyes wide open. Aside from a bit of pain as your pupils snap close quickly, you'll notice a bit of a white flash in your vision, quickly compensated for by the brain.

[John Pytlak RIP]

My SMPTE Paper "Shedding New Light on Darkroom Illumination" talks about human visual sensitivity, and has a few graphs of sensitivity data:

 

http://www.kodak.com/US/en/motion/support/...on/page01.blind

 

A dark-adapted eye is VERY sensitive, but not very discriminating of detail or color.

[Dominic Case]

A dark-adapted eye is VERY sensitive, but not very discriminating of detail or color.

 

Taking photos of the sky is quite interesting. Usually there are only one or two stars bright enough to show any colour at all to the naked eye (below the threshhold of the rods' sensitivity). A properly-exposed photo on colour stock shows the colours that are there as film has no equivalent of rod vision.

 

Anyone know why the BB tags don't work on my posts? I'm using IE5.

[Alvin Pingol]

That's probably not much more than 15 stops: but in any case there's some adaptation going on as you swing your eyes from the shadows to the bulb, - a bit like pulling the aprture as you pan the camera.

I was talking about looking at the bare bulb and the room at the same time (without "swinging" your eyes).

 

Also, while you are looking at the bulb, your (more sensitive) peripheral vision is coping with the shadows.

 

Ahh, that would explain it. B)

 

Oh, and I did a fun little experiment with my eyes last night. Without my glasses on and in total darkness, I sat next to the light switch and looked at the standby LED on my computer monitor. It was very blurry. When I turned on the light, as my iris "stopped down," you could see the LED becoming slightly more in-focus.

 

Very cool. :)

[Dominic Case]

When I turned on the light, as my iris "stopped down," you could see the LED becoming slightly more in-focus.

Do you need glasses?

 

In dim light, I can't read small print. My eyes only focus the print if it's too far away to resolve the letters :( . In brighter light, the iris stops down and increases the depth of field. It's that simple.

[Phil Rhodes]

Hi,

 

> When I turned on the light, as my iris "stopped down," you could see the LED

> becoming slightly more in-focus.

 

Are you sure that wasn't just a point-source lens flare issue? My eyes have a pronounced glow and streak flare, and I think most people's look the same given how artists draw bright lights. I have slight astigmatism in my left eye which produces an additional, irregular glow. These artifacts are more visible in high contrast situations, such as a point source like an LED in the dark.

 

Phil