Sensor sizes 35mm format

[Longwatcher]

I finally decided to do some math just out of curiosity to see how many MP would fit on the head of a pin. oops, I mean fit on a 35mm sized full frame sensor.

The results:
- 1DsMkII .0072mm 16.7MP
- 20D .0064mm if FF 21.1MP
- Theoretical 4um FF sensor .0004mm 54Mp
(this is the size I have been told we can still get full ISO range out of eventually)
- Pro1 .0027mm if FF 120MP
- physics limit for visual light .00075mm 1536MP

Multiply by 12 for BMP file size given current dynamic range,
by 10 for 16bit TIF by 2 for 8bit TIF
by 1 for RAW
by .1 for jpeg level 8 compression

A Foven style sensor would have at least three times that value.

All the above should be taken with the knowledge that as pixel size decreases (MP increases) the noise will increase. Also the processing power and storage capacity needed will be significantly greater.

Just providing the results of my curiosity,

[Moppie]

Has anyone ever worked out how many MP's the human eye has?
Assuming one cell equals one pixel?

I would love to see a digital camera that used the output from the optic nerve to record the image :)

[Scottes]

Human eyes have more ability to resolve in the center. Hawks have even more. So the MP would lessen further from the center.

I wanna know the ISO of an Owl's eye, and how noisy it is.

[PacAce]

Has anyone ever worked out how many MP's the human eye has?
Assuming one cell equals one pixel?

I would love to see a digital camera that used the output from the optic nerve to record the image :)

The image out of that digital camera connected to the optic nerves of a human wouldn't look too good. It would like something like this:

http://www.tanseikai.com/pacace/images/eye-image.jpg

[timmyquest]

The image out of that digital camera connected to the optic nerves of a human wouldn't look too good. It would like something like this:

http://www.tanseikai.com/pacace/images/eye-image.jpg


Thats pretty neat.

This is kind of something i've been thining about in my head and i think it's waht makes photography so hard.

Your eye may see one thing, but really you have to use your mind to determain if what your eyes are telling you is going to make a good image.

[Moppie]

............It would like something like this...........

Oh Iv studied the human eye, and have a good idea of how the distrabution of cones and rods effects our perception of colour, light and movement.
A capture at say 1/2000th of a second might render and image like the one above, but imagine the ablity to control things like depth of field, and the massive wide angle when the eye is tracked around the field of view, and the data used to build up an image over say 1/100th of a second.
More than fast enough to capture even fast motion, and track your eye around to complete the picture in focus, with colour :)

[Canuck]

I forget where this was mentioned, but the human receives about 5.7TB of info from the 5 senses every second! I guess that it a little over 1.1 TB per sense average. However, if you are loking at say this pic in real life:
http://images.fotopic.net/?id=9615099
then there is little chance you will be using the sense of touch, so that means more info is gained thru sight than touch at that instant, I'd imagine. Enjoy the pic!

[Moppie]

Im guess Im going to have to buy some bigger CF cards.........

[robertwgross]

I forget where this was mentioned, but the human receives about 5.7TB of info from the 5 senses every second! I guess that it a little over 1.1 TB per sense average.

Wow. Is that in RAW, or JPEG?

---Bob Gross---

[PacAce]

I forget where this was mentioned, but the human receives about 5.7TB of info from the 5 senses every second! I guess that it a little over 1.1 TB per sense average. However, if you are loking at say this pic in real life:


But isn't that just the data transfer rate? An analogy: A DVD movie can send out a stream of info at a rate of about 7 mbps (if I recall correctly). But the video image in a frame is only 720 x 480 pixels. Anybody know how many rods and cones the human eye has?

[Scottes]

"...The rods are more numerous, some 120 million, and are more sensitive than the cones. However, they are not sensitive to color. The 6 to 7 million cones..."

From: http://hyperphysics.phy-astr.gsu.edu/hbase/vision/rodcone.html

[Scottes]

Heh. That's pretty interesting stuff.

[metoyou]

This thread is fun!
Keep posting please...I'm enjoying reading.

[Ronald S. Jr.]

LOL....they've probably forgotten what the info was...nearly 3 years ago.

[BillsBayou]

- Theoretical 4um FF sensor .0004mm 54Mp
(this is the size I have been told we can still get full ISO range out of eventually)
- Pro1 .0027mm if FF 120MP


I need to go see what we're discussing here. How do you get more MP out of .0027mm at FF than you do at .0004mm at FF?

[mkuriger]

wow talk about waking the dead!

[StewartR]

I finally decided to do some math just out of curiosity to see how many MP would fit on the head of a pin. oops, I mean fit on a 35mm sized full frame sensor.

The results:
- 1DsMkII .0072mm 16.7MP
- 20D .0064mm if FF 21.1MP
Lo and behold, the newly announced full frame 1Ds Mk III has ... 21.1MP.
Spooky.

[PacAce]

I need to go see what we're discussing here. How do you get more MP out of .0027mm at FF than you do at .0004mm at FF?

With everything else being equal, you can't. The difference between the two is the actual sensor site size. You can get more pin heads in a given area than you can nail heads although, if you made then both occupy the same amount of space around them, you would get the same number of pins and nails in the same given area. :)

But looking at those figures again, I'm thinking the .0004 is a typo and should be .004. Just a guess, though.