the human eye 60fps thing

[carpediem]

On (almost) every forum when there is talk about fps, someone will say. "The human eye can't see more than X fps".

I remember an application that someone did long time ago that would display a cube, or something, rotating on the screen. With one side having customizable fps and the other being constant at 30 (I think).

The point with the app was to easily show you the difference between 2 different fps values.

No to my question. Does anyone know where to find this application? Been trying to find it but without any luck.

 

[T2k]

On (almost) every forum when there is talk about fps, someone will say. "The human eye can't see more than X fps".

That's actually BS: it's vary person by person.

BTW you can tell the difference between 60-70fps and 120, right?

 

[carpediem]

On (almost) every forum when there is talk about fps, someone will say. "The human eye can't see more than X fps".

That's actually BS: it's vary person by person.

BTW you can tell the difference between 60-70fps and 120, right?

Well, i can "feel" the difference between 125 and 333 in Quake3.

 

[Bouncing Zabaglione Bros.]

On (almost) every forum when there is talk about fps, someone will say. "The human eye can't see more than X fps".

That's actually BS: it's vary person by person.

BTW you can tell the difference between 60-70fps and 120, right?

Well, i can "feel" the difference between 125 and 333 in Quake3.

That's different - you're running a biofeedback loop, and you're detecting changes in the feedback. This is a much easier war for you to "feel" the difference, and involves a lot more than just what you are seeing.

When Doug Trumbell was doing research in the 80's on possible new film formats, he found that the upper limit on human perception was at about 81-87 frames per second (depending on the person). There are a lot of other things going on in the realm of perception (as opposed to just seeing) that means that around this level, people will be pretty well fooled by what they are seeing.

It's very important to note that what you see it not just what comes into your eyes - there's a lot of post processing that goes on in your head, and this has a great effect on what you are actually percieving.

There was a great demo where he would show what looked like a magician's stage act on his new film format. Everyone was fooled and though they were actually seeing a live act on stage, rather than a film artfully played back on what looked like a stage.

 

[AlNom]

If you show me 10 minutes of sustained 30fps, and then show me 10 minutes of sustained 60fps, I will not tell the difference (without careful examination).

Show me half the screen rendering at 30fps and the other half at 60fps, and I will most certainly tell the difference.

 

[Thowllly]

I just tested myself. Made a prog that displayed some rotating dots at full refresh rate, then I made a second version of it that ran at half the refresh rate (but still moved the dots at the same speed). I set the refresh rate to 160Hz (so one would run at 160fps and the other at 80fps). I selected randomly between the two versions without looking. After it started I looked at the screen and every time I was able to correctly identify the framerate within a second. (I did this lots of times and I got it right every time).

The difference is really obvious, but it was not the smoothness I noticed, rather it was that the particles that ran at 80fps had slightly blurred edges in the direction of movement. (edit: the 160fps particles on the other hand look really nice and sharp, but only as long as I was following them with my eyes ofcourse, if I didn't follow them, then they got 'motionblurred', as expected) I belive this is because the eyes moves continously with the dots, and if the framerate is lower than the refresh rate then the eye will move on, and then when the dot is redrawn a second time in the same position after the eye has moved on, it will look like there is a second dot lagging slightly behind the dot your trying to follow. These two dots then combine into one slightly blurred dot. So, if I had tested 160fps at 160hz against 80fps at 80hz, then I might not have been able to notice the difference. I tried to test at higher refreshrates, but my crappy monitor wouldn't go any higher than 160hz.... sorry for my rambling...

 

[carpediem]

On (almost) every forum when there is talk about fps, someone will say. "The human eye can't see more than X fps".

That's actually BS: it's vary person by person.

BTW you can tell the difference between 60-70fps and 120, right?

Well, i can "feel" the difference between 125 and 333 in Quake3.

That's different - you're running a biofeedback loop, and you're detecting changes in the feedback. This is a much easier war for you to "feel" the difference, and involves a lot more than just what you are seeing.

When Doug Trumbell was doing research in the 80's on possible new film formats, he found that the upper limit on human perception was at about 81-87 frames per second (depending on the person). There are a lot of other things going on in the realm of perception (as opposed to just seeing) that means that around this level, people will be pretty well fooled by what they are seeing.

It's very important to note that what you see it not just what comes into your eyes - there's a lot of post processing that goes on in your head, and this has a great effect on what you are actually percieving.

There was a great demo where he would show what looked like a magician's stage act on his new film format. Everyone was fooled and though they were actually seeing a live act on stage, rather than a film artfully played back on what looked like a stage.

Agreed. That's why i wrote that I can feel the difference. I just wanted to get a hold of that demo I saw long time ago to show people that there is a difference between 30-60 and 60-80ish, which I believe I can detect easily when switching. Some people seem to think that the brain is capped @ 60hz for visual input and I just wanted to be able to show them that demo

Seems like no one heard about it though.

 

[Nick[FM]]

As far as I know (and remember) the eye (or actually the brains) need 24fps in order to record it as a continuous movement. Anything above that makes it of course smoother to a certain point. All this is of course very personal, as some say 60fps "stutters" whereas others think 30fps looks as smooth as 100fps.

It's kind like the Hz of the monitors. Some say that they get headache from 60Hz and others can look at it without any problems. I can't stand 60Hz (get headache very quickly) and need at least 80Hz+ in order to look at a plain white image on screen. A friend of mine can't even look at 85Hz without getting his eyes sore. His "minimum limit" is 100Hz. It is all very much up to the person.

 

[Guden Oden]

Well, just FYI, the "church" (or great money-extorting scam, depending on how you look at it) of scientology teaches that the brain is capped at precisely 24fps. Of course, that number comes from the fact the "church"'s founder, L. Ron Hubbard, was a technological dunderhead and did not understand that movie cameras/projectors run at that speed for quite trivial reasons. He just thought that was the upper limit of the perception of the human mind and integrated that into his equally bogus teachings...

 

[carpediem]

]
It's kind like the Hz of the monitors. Some say that they get headache from 60Hz and others can look at it without any problems. I can't stand 60Hz (get headache very quickly) and need at least 80Hz+ in order to look at a plain white image on screen. A friend of mine can't even look at 85Hz without getting his eyes sore. His "minimum limit" is 100Hz. It is all very much up to the person.

Same here... I need at least 85hz to stop noticing a flicker. I don't notice any difference from 85 and up though.

 

[Bouncing Zabaglione Bros.]

Agreed. That's why i wrote that I can feel the difference. I just wanted to get a hold of that demo I saw long time ago to show people that there is a difference between 30-60 and 60-80ish, which I believe I can detect easily when switching. Some people seem to think that the brain is capped @ 60hz for visual input and I just wanted to be able to show them that demo

Seems like no one heard about it though.

IIRC, that demo is actually an old 3DFX demo - you can get it here

 

[AlNom]

] A friend of mine can't even look at 85Hz without getting his eyes sore. His "minimum limit" is 100Hz. It is all very much up to the person.

He doesn't wear glasses does he? or does he just have very good eyesight?

 

[Nick[FM]]

He doesn't wear glasses does he? or does he just have very good eyesight?

Actually he does wear glasses, but I have no idea what strength they are. I don't use glasses and I have a perfect eyesight (at least according to the doctors). Actually my left eye was diagnosed as "super sight" since I could see things with that eye only a handful of people can see.

 

[carpediem]

Agreed. That's why i wrote that I can feel the difference. I just wanted to get a hold of that demo I saw long time ago to show people that there is a difference between 30-60 and 60-80ish, which I believe I can detect easily when switching. Some people seem to think that the brain is capped @ 60hz for visual input and I just wanted to be able to show them that demo

Seems like no one heard about it though.

IIRC, that demo is actually an old 3DFX demo - you can get it here

I wish I could test it and find out, but my last 3dfx card (a righteous 3D voodoo 2), died a loooong time ago

 

[Guden Oden]

Lots of stuff matters wether a screen feels flickery or not, such as brightness of the screen, ambient brightness, if ambient light comes from fluorescent lamps, if you're tired, if you look at the screen out of the corner of your eye, and of course the characteristics of the screen itself...

I myself see some flicker at 75 but nothing at 85, but it's nothing that really bothers me.

 

[Nick[FM]]

Lots of stuff matters wether a screen feels flickery or not, such as brightness of the screen, ambient brightness, if ambient light comes from fluorescent lamps, if you're tired, if you look at the screen out of the corner of your eye, and of course the characteristics of the screen itself...

I myself see some flicker at 75 but nothing at 85, but it's nothing that really bothers me.

That's true, but for example we tested it with my friend on the same screen. Neither of us could stand 60-80Hz, whereas I was "ok" with 80Hz. He needed to turn up the Hz to 100Hz in order to look at the screen without feeling it in his eyes. We used a white fullscreen image to get the full effect of flickering. At least we thought that it worked well as a testimage.

 

[digitalwanderer]

Well, just FYI, the "church" (or great money-extorting scam, depending on how you look at it) of scientology teaches that the brain is capped at precisely 24fps. Of course, that number comes from the fact the "church"'s founder, L. Ron Hubbard, was a technological dunderhead and did not understand that movie cameras/projectors run at that speed for quite trivial reasons. He just thought that was the upper limit of the perception of the human mind and integrated that into his equally bogus teachings...

XENU!!!!!

Sorry, just gotta make sure.

 

[Basic]

Is this what you're looking for?

http://forums.gameon.co.uk/showthread.php?t=9989

 

[carpediem]

Is this what you're looking for?

http://forums.gameon.co.uk/showthread.php?t=9989

YES! That was it!

Thanks, oh so much!

 

[Nite_Hawk]

There are a lot of different things that are being talked about here.

Ok, first things first, read these:

http://webvision.med.utah.edu/temporal.html
http://www.ensc.sfu.ca/people/grad/brassard/personal/THESIS/node52.html

Basically, depending on the length of time and field of view, there is a threshold where our ability to precieve changes in brightness are limited. This might apply in several situations. The first is where you've got a CRT image, which is flashing a bright dot onto the screen, decaying at a certain rate, and then flashing again. Our ability to detect flicker will be dependent on how much of the screen is in our field of vision, they decay rate of the CRT (both how long each pixel is lit for, and how many times per second it is relit) and how bright each pixel becomes when it is relit.

Another thing we are talking about in here is detecting changes in the framerate, and it is a similar problem to the one above. High contrast changes will be more perceptable to the eye than low contrast changes, so while in one case 24fps might be fine (where the images are blured and thus result in more gradual contrast changes) while others might require much higher framerates (like computer generated video where there are abrupt high contrast changes).

For the framerate issue, another problem that ties in is that you are limited by your spatial resolution. Imagine for instance that you have a display that is 10 pixels wide, and you are running at 60fps. you want to move a black pixel on a white background from one end of the display to the other over a 1 second period of time. Unfortunately, due to the restriction on the spatial resolution, you are forced to an effective 10fps for a steady velocity across the screen. If you had 60 pixels instead of ten, you could move one pixel each frame, but with only 10 pixels you are limited to moving one pixel every 6 frames. In this case, motion blur would decrease the "choppyness" of the movement at the expense of bluring/ghosting like one experiences on an lcd.

Nite_Hawk