While watching the snow fall tonight I was wondering why it appears in blurry streaks. Film is 24 frames per second, video is 30, and I try to keep PC games above 72, but what about real life and the mechanics of human eyes? Why do the snowflakes appear as streaks?
It’s 30 fps for the human eye pal. http://www.ping.be/powervr/fps_discus.htm
I hope this helps.
Snow flakes do not appear as streaks to my eyes.
There is a phenomenon called “persistence of vision” that has an effect similar to what you describe. In darkness, moving points of light may appear as streaks. Or upon closing your eyes, in daylight or night, there is an afterimage which fades out after several seconds. This is because the rods and cones at the back of your eyes continue to fire off signals through your optic nerve after the stimulation is removed.
I don’t know the potential firing rates of these nerves.
No, it’s not.
It depends on the individual.
30 is a safe limit for assuming adequate motion blur.
But, I’ve worked in computer animation for a long time.
And, with practice, I could pick out “jerks” due to less than optimal transformations between keyframes even in sequences played at 60 full frames per second.
Do the incandescent lights in NY subway stations still run at 30-40 Hz ? That was damned annoying !
Note that the rods and cones in the eye are not synchronized. So there isn’t a uniform frame to frame transition across the retina.
There is a framerate for humans but I don’t recall what it is. It takes time for the human mind to process inputs…just like a computer. As a result there are minimums in which we can’t perceive things. Fortunately our software is written such that these bumps get smoothed out and we don’t notice.
Of course, one might say the Universe itself has a framerate based on Planck Time (10[sup]-43[/sup] seconds) although it is debateable if you could characterize it as such.
Like Gyan9, I’ve worked in computer animation for some time. There is a surprisingly large variation from individual to individual in regards to their threshold for both ‘framerate’ and ‘refresh rate’. Some people will be perfectly happy with 12 fps in a computer game or animation. Others will think it still looks jerky at 30 fps. Even when considering one person, there are variables.
Generally speaking, the more motion blur in each individual frame, the lower the framerate equired to give a satisfactory illusion of smooth motion. Films have a framerate of 24 fps (in the USA at least), but most people would say they look smooth enough. A movie camera captures a relatively large amount of motion blur, which helps to ‘smooth out’ the motion when the film is played back.
Computer animation may have motion blur or not, depending on how it’s generated. I have done tests myself and found that the very same animation rendered with motion blur can look much smoother even at the same frame rate. Computer games don’t have motion blur (at least as far as I know), so a higher framerate is sometimes required to make them look good.
People are usually much less tolerant of large variations in brightness though. This is why the refresh rate of your computer monitor should generally be set to at least 75 Hz or so. Even 60 Hz is not a fast enough refresh-rate for many people: I can barely tolerate working at a monitor set to 60 Hz for even a few minutes. Even if a person does not conciously notice a flicker, it may cause eyestrain over long periods. To add even more variables, the eye is generally more sensitive to flicker in the peripheral vision. So, the bigger the monitor you are using (or the closer you sit to it), the higher the refresh rate needs to be.
To say that there is a frame-rate for the human eye isn’t technically correct. There is a shutter-speed, if you will, that is beyond our power to detect. While it varies by individual, it would translate to a fps of 25-40.
The reason it isn’t a true frame rate is that our eyes are taking in a constant analog stream of information. It is not blocked off into distinguishable chunks and streamed together, and we wouldn’t have the processing power to do that as quickly as we do with our vision. Your rods and cones are struck by even those photons that you don’t resolve.
[quote]
Computer games don’t have motion blur (at least as far as I know), so a higher framerate is sometimes required to make them look good. [/qutoe] They are starting to have motion blur to make the games look better.
This video card and almost all higher end gaming video cards now support motion blur.
http://www.hardwarecentral.com/hardwarecentral/previews/1760/5/
If I follow this discussion at all, I’m wondering how the shutter speed or frame rate or whatever we call these limits of our perception apply to a phenomenon I have observered: when a car passes me in an intersection, say, if I blink, or if I try to sweep my vision over the car, I can often get an image of the spokes of his hubcap - as if I’ve strobed it. Now, I know that effect and understand it when it occurs under fluorescent lamps - but in sunlight, it seems that I’M the one doing the strobing. But how does that connect with this discussion? Am I taking in the view for about 1/30 of a second, and all I get is one image, rather than a blur? This may be the answer I’ve been looking for. I’ll stay tuned.
If the human eye is 30 fps why do we have monitors set for 75? Is it to decrease the likelihood that the refresh of our eyes doesn’t coincide with a refresh of the monitor?
BTW, I thought this was going to be a theoretical debate on whether the universe was discrete or continuous.
Ya, KidCharlemagne so did I. There’s such a thing as a “Planck second” that is (I believe) the smallest measurable amount of time. No cite, cuz I’m lazy. It’s in that Stephen Hawking book I believe, though.
Now, that’s not to say that time may not be continuous. Maybe it is. I like the idea that the universe is a big Finite State Machine.
Not quite. Planck measures are the smallest which our current physics can handle meaningfully. We can certainly talk about smallest intervals of time than 10[sup]-43[/sup] seconds and there might be physical meaning to them, but we can’t use them in quantum theory and get meaningful results.
I spent the first year of graduate school learning about the first 100 milliseconds of visual perception. It is extremely complex. Computer animation is a very poor analogy for human perception and gives inaccurate insights (pun intended) to the OP question.
I’m with Waverly that the concept of “frame-rate” is an incorrect way to think about vision. A frame was first a mechanical method to photograph discrete shots. The eye is totally different. And I mean totally. The eye and related stuff is a freakish, beautiful, astounding, alien thing when you really get to know it (like my ex-girlfriend).
The visual effects that people notice (blurring, not blurring, strobed images) are likely due to the different levels of memory and types of eye-movement, some of which are counter-intuitive. The brain is far more weird than most people imagine.
Quantum physics? Planck? I don’t think this has much to do with the OP perceiving blurry snowflakes.
Think about this.
most people see 24fps as smooth motion, as in a movie
most people see flicker up to 70 Hz, but not at 200 Hz
a one-millisecond flash of light (1000Hz) is easily visible
So, when something I observed (spinning tires) is strobed to me, I’m only remembering one instant but I actually saw more?
The brain also takes shortcuts in an effort to get you a HUD as quickly as possible. That’s why films make us believe there is motion, that is why an image is filled in where the optical nerve attachment actually should leave a blind spot, and that is why many optical illusions give us the effects that they do.
My guess for the wheel spoke question is that a still image was what the brain thought made the most sense given the input, even if the split second of data did include some movement.
Cecil is highly skeptical of the claim that you can see strobe effects without a strobing light source. For instance, seeing hubcaps appear stationary or retrograde in ordinary sunlight. I am too. Persistence of vision would seem to contradict this, but I admit I know little about the physiology of vision.
How come the wheels of a moving car appear to rotate backward sometimes?
Some discussion thereon