How does a camera stack up against human vision?
One of the most common questions people have when they start to learn about filmmaking is how it relates back to human physiology. We're a species that finds it easier to understand things through human analogies (our base 10 number system almost certainly has something to do with our hands having 10 fingers, for instance), and there are several areas of filmmaking that can be understood by thinking about anatomy. After all, a camera has a lens to see with and a microphone to hear with, just the way humans have eyes with lenses to see and ears to hear.
Unfortunately, the analogy isn't as perfect as we might hope. Exposure in a camera is driven by a combination of ISO, aperture and shutter speed. While our eyes quite obviously have apertures (the pupil) at the front, which adjust for exposure, is there an ISO and a shutter speed for the eye?
In a study published last week by Rockefeller University, there has been experimental proof that human beings are capable of seeing an single individual photon of light.
The answer isn't really so simple, because the human vision system is adaptive, and isn't just a matter of the eye receiving an image but also the brain interpreting what the eye sees. There is still research to be done into the limits of human vision and what is even possible for most people to see in the first place. In terms of shutter speed, this article by 100fps has a great breakdown of why it's such a hard question to answer, but the short version is that different parts of our eye perceive motion differently (it could be said that the edges of our vision have a lower refresh rate than the center), and even our bright-light and dark-light vision refresh differently.
Finding an answer to maximum ISO has been even harder, since it requires getting a sense of what is the absolute lower limit of light that can be perceived by a human. An additional complication is that we have two types of cells in our eyes: cones which give us color vision, and rods which are better in low-light situations. In a study published last week by Rockefeller University, there has been experimental proof that human beings are capable of seeing an single individual photon of light.
For reference, a 100W film light puts out something in the ballpark of 5e32 photons of light per second: that's 5 times 10 to the 32 power. Roughly 1e31 per 1/48th of a second. That's a lot of photons.
And we humans can see an individual photon of light. It's so tiny that you can't really give it an ISO equivalent. If you did, it would be in the billions, far outside what the system was intended to measure, or so high that it doesn't really have any conventional meaning.
Which is pretty cool. And it means camera makers have a long way to go to catch up.
I call shenanigans on this article, as low light cameras can easily see things that normal human vision is unable to see.
If you want a quick example, try reading a newspaper in a darkened movie theater when the main lights are off, unless you happen to be a superhero you won't be able to read a dang thing.
I haven't tried measuring what the "ISO" of my eyes are, but I know that in a moderately dark room things are much brighter on my camera's 7 inch monitor than they are with my normal vision.
I'm sure that with a little testing you could figure out roughly what the "ISO" of your eyes is by comparing what you see with your own eyes relative to what your camera "sees" on a monitor display.
My guess is that the "ISO" rating of our eyes would be pretty low compared to what modern cameras can "see".
Also human eyes are only f/2.1 in the dark : http://petapixel.com/2012/06/11/whats-the-f-number-of-the-human-eye/
July 26, 2016 at 5:04PM, Edited July 26, 5:07PM
agreed, my A7SII is way better than me
July 27, 2016 at 1:36AM
I don't even get banding in the blacks at night. My eyes are great! I'd be more interested in the compression rate.
July 27, 2016 at 6:36AM
June 24, 2017 at 7:09PM
I found a lot of interesting information from your post. and I am very grateful for your work. because this post can add my insight
July 4, 2021 at 9:14AM
I agree. My A7s can see better than me past about 300,000 ISO there abouts.
July 27, 2016 at 6:37AM
Our sight is too dynamic to be measured with ISO; each person has variables in the amount of rods cells. Besides, the eye is not the important part, is the brain where the image is constructed, and it is biased. We have a pretty good night vision for daytime mammals, but we see nothing in comparison to a cat. Bees, for instance, can see UV light, and use it to classify flowers; they also have a 'higher' fps, so if you could make a bee watch an analogical movie, it would see 35% of black screen, because of the space between the frames.
June 24, 2017 at 3:30AM
Not related to this post 100 percent per se, but I would suggest you all take a look, especially those who are starting out, at the following video by Tom Antos: https://youtu.be/Q7q94hWd4zA
Very helpful indeed. Thanks Tom. Cheers
July 26, 2016 at 9:16PM
Such a great video of Tom Antos and also Very useful information share here about F-number of the Human Eye. I am providing Nursing Assignment Help at Secure Assignment Help. I appreciate the way its written here. I will share this with my college people I am sure they will like it.
September 26, 2016 at 10:06PM
Maybe you can notice the one photon but you definitely can't see with it... I have to say my A7s sees a lot more than me in the dark...
July 27, 2016 at 8:22PM
Yes McCauley, I also tried that. Thanks, Charles for sharing informative information. Since I am nursing dissertation help service provider and would like to share with our UK teachers and appreciate such information.
September 26, 2016 at 2:04AM
One of the most common questions people have when they start to learn about filmmaking is how it relates back to human physiology. Check http://american-writers.org/ for getting an essay based on this topic.
July 27, 2016 at 10:19PM
Not mentioned (or realized) by anyone here is that night vision takes time to be at full capacity. Think about it. You turn the lights of in your room at night before bed; it seems almost black (assuming there aren't any bright lights outside the window). Then maybe 3-4 hours later you wake up to use the bathroom or get water and you can easily navigate the room and the whole house pretty much without needing to turn any lights on. A friend of mine who is an ophthalmologist (this is the real "medical" eye doctor, not an optometrist we are talking about) mentioned to me once that at a minimum it takes 30 minutes for our night vision to kick, and maybe even longer for full capacity. SO if you really want to compare your eyes to your fancy new high ISO camera, you need to be in dark for a while to make it valid.
But beyond ISO, it's really dynamic range that human eyes are so good at. It doesn't matter what camera: You simply can't have a camera roll from a dim interior and out into the bright sun the way we do every day. And of course color range too: it seems basically infinite.
Lastly- as the article says: we can train ourselves to see more and more subtle variations in color and tone. Any graphic designer or fashion designer could show you that.
July 28, 2016 at 3:21PM, Edited July 28, 3:22PM
I would argue that although it is evident that dynamic range is indeed better with our eyes, the exposure differential given by your example is just because our eyes are really good at exposing for what we are focused on. A camera could do so, but there is a lag between what the operator wants the exposure to be and the camera in terms of how fast the operator can adjust the camera.
June 27, 2017 at 8:03AM, Edited June 27, 8:03AM
It would be nice to know what ISO comes close to matching what a human sees in a dimly lit room or outside by a campfire, or sitting in a car with the dash lights on. Yeah, we can detect a single photon, but most of us want to know for example, if we're in a dim space and we can easily see the people and major scenery elements, what does that equate to in ISO as a starting point for filming?
July 28, 2016 at 9:14PM
Human eyes don't have ISO.
ISO is the number for measuring artificial sensor sensitivity.
October 5, 2016 at 5:53PM
It is a fantastic blog its really inspiring and full fill with knowledge
Invoice Books Printing London
March 19, 2017 at 12:48AM
Thanks a lot for the lovely article. I really appreciate the hard work you put in it.
March 19, 2017 at 11:42PM
I love all the posts, I really enjoyed, I would like more information about this, because it is very nice.Thanks for sharing.
Quick Step Master Installer East London
March 20, 2017 at 4:59AM
Thanks a lot for the lovely article. I really appreciate the hard work you put in it.
csgo smurf accounts
March 26, 2017 at 10:47PM
Very Nice. Really enjoyed reading your blog posts. Thanks for sharing useful information. keep sharing like this so i can keep myself update through this blog. jas hujan
March 27, 2017 at 2:03AM
The information you have posted is very useful. The sites you have referred was good. Thanks for sharing..
March 28, 2017 at 3:51AM
This was utterly fascinating! Thanks!
Wedding Videographer Eugene Oregon
July 1, 2017 at 8:44PM
The information you have posted is very useful. Thank you for nice and wonderful Information. Fire Watch
July 31, 2017 at 3:46AM
Great post! I am actually getting ready to across this information, is very helpful my friend. Also great blog here with all of the valuable information you have. Keep up the good work you are doing here. check out Redmi 4 Prime key features and Lenovo k6 Power Review
August 31, 2017 at 5:56AM
I found a considerable measure of interesting information from your post. Thank you so much for your work.
Cyberlink Powerdvd Crack
December 6, 2017 at 11:31AM
If you are looking for the stadiums when it comes to both the teams, Royal Challengers Bangalore play their match in their hometown. Usually, for playing the match, Bangalore team make their way to the M.chinnaswamy stadium.
RCB vs MI IPL Tickets
March 18, 2018 at 10:27PM
Peggo APK is very easy to see and download the videos from YouTube, but that downloaded will save in the YouTube account itself but not in the phone memory.
June 7, 2018 at 2:30PM