Eye Oscillations and Framerate

H

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Blue's News found an interesting article on oscillations and framerate that should spark a lively discussion this morning.

Humans can see frame rates greater than 24fps (although plenty of people will argue that they can’t on the internet). Once you’ve accepted that fact, the next question is why do movies at 48fps look “videoy”, and why do movies at 24fps look “dreamy” and “cinematic”. Why are games more realistic at 60Hz than 30Hz?
Click to expand...
 
Supposedly the upper limit found when testing the finest Jet pilots, was 72 FPS in a moving image, and 240+ in a non-moving.
For a non-moving test, they used a wheel with a cut out in it. The wheel material blocks a very powerful lightbulb, humans were able to distinguish the shape of the cutout (when the wheel spun that cutout into the user's line of sight) usually in the mid 200 FPS range, where in a moving image, like we see in a game, or movie, peak humans were unable to distinguish differences past 72 FPS (coincidentally the FPS max setting for the Source engine).

I guess the ultimate test would be to have a 60, 72, and 72+ display, side by side, in a double blind experiment to see if people could notice a difference. Hell this could be done with rigs on this forum.

The conditions would require the images to be presented as the same color, contrast and brightness at their respective framerates (so black and white would be ideal for this test as we are only measuring perception of motion). There are most likely other considerations, but other than that, place a slow moving item on the screen until a user can differentiate between the lowest and next lowest FPS screens, and increase it's speed, until we find the right speed at which the user can find determine the fastest FPS display. I believe, if we perform this test correctly, the numbers may reflect a human perceptional FPS cap of 72.

I know I will upset absolutely any person on this forum who bought themselves something nice, and is convinced they are superhuman!
 
Kalabalana said:
Supposedly the upper limit found when testing the finest Jet pilots, was 72 FPS in a moving image, and 240+ in a non-moving.
For a non-moving test, they used a wheel with a cut out in it. The wheel material blocks a very powerful lightbulb, humans were able to distinguish the shape of the cutout (when the wheel spun that cutout into the user's line of sight) usually in the mid 200 FPS range, where in a moving image, like we see in a game, or movie, peak humans were unable to distinguish differences past 72 FPS (coincidentally the FPS max setting for the Source engine).

I guess the ultimate test would be to have a 60, 72, and 72+ display, side by side, in a double blind experiment to see if people could notice a difference. Hell this could be done with rigs on this forum.

The conditions would require the images to be presented as the same color, contrast and brightness at their respective framerates (so black and white would be ideal for this test as we are only measuring perception of motion). There are most likely other considerations, but other than that, place a slow moving item on the screen until a user can differentiate between the lowest and next lowest FPS screens, and increase it's speed, until we find the right speed at which the user can find determine the fastest FPS display. I believe, if we perform this test correctly, the numbers may reflect a human perceptional FPS cap of 72.

I know I will upset absolutely any person on this forum who bought themselves something nice, and is convinced they are superhuman!
Click to expand...

I was with you until your snide little passive-aggressive remark at the end that shows you have no understanding at all of how consumer displays work. Not to mention when the viewer receives feedback about display content, the results you mention become invalid.
 
Kalabalana said:
Supposedly the upper limit found when testing the finest Jet pilots, was 72 FPS in a moving image, and 240+ in a non-moving.
For a non-moving test, they used a wheel with a cut out in it. The wheel material blocks a very powerful lightbulb, humans were able to distinguish the shape of the cutout (when the wheel spun that cutout into the user's line of sight) usually in the mid 200 FPS range, where in a moving image, like we see in a game, or movie, peak humans were unable to distinguish differences past 72 FPS (coincidentally the FPS max setting for the Source engine).
Click to expand...

Incorrect. It's 300.

Here's proof, just queried it in the console myself:
Code: 
"fps_max" = "0" ( def. "300" )
          notconnected
          - Frame rate limiter, cannot be set while connected to a server.
 
Nothing looks better at 24FPS. Ever. Period. That is the bigger, more BS, myth. Never in my life have I ever thought, "Gee, I'm glad this is in 24FPS."
 
60 Hz is nothing. NOTHING. Imagine walking or driving a car at 24 fps haha lol. What about race drivers and fighter jet pilots? Nah man, 24 is da shit! It's "filmic"

dandragonrage said:
Nothing looks better at 24FPS. Ever. Period. That is the bigger, more BS, myth. Never in my life have I ever thought, "Gee, I'm glad this is in 24FPS."
Click to expand...

I had a friend argue with me about how at 48 fps he could see all the bad makeup and poor decors in those silly jackson movies. I couldn't believe it. He was angry at the technology which allowed him to see the film's shortcomings. Some people...
 
We're going to need at least 1000 fps for things to look realistic. The larger percent of your fov it takes, the higher FPS you're going to need..
 
Kalabalana said:
Supposedly the upper limit found when testing the finest Jet pilots, was 72 FPS in a moving image, and 240+ in a non-moving.
For a non-moving test, they used a wheel with a cut out in it. The wheel material blocks a very powerful lightbulb, humans were able to distinguish the shape of the cutout (when the wheel spun that cutout into the user's line of sight) usually in the mid 200 FPS range, where in a moving image, like we see in a game, or movie, peak humans were unable to distinguish differences past 72 FPS (coincidentally the FPS max setting for the Source engine).

I guess the ultimate test would be to have a 60, 72, and 72+ display, side by side, in a double blind experiment to see if people could notice a difference. Hell this could be done with rigs on this forum.

The conditions would require the images to be presented as the same color, contrast and brightness at their respective framerates (so black and white would be ideal for this test as we are only measuring perception of motion). There are most likely other considerations, but other than that, place a slow moving item on the screen until a user can differentiate between the lowest and next lowest FPS screens, and increase it's speed, until we find the right speed at which the user can find determine the fastest FPS display. I believe, if we perform this test correctly, the numbers may reflect a human perceptional FPS cap of 72.
Click to expand...
I wonder if its any coincidence that when CRT's were the norm, 60hz with predominantly white screens gave me a headache while 72hz and higher did not. I could see the flickering on the screen at 60hz but not at 72hz, 75hz, and 85hz settings. I don't have this issue with any other screen tech, just CRT's.
 
CaptNumbNutz said:
I wonder if its any coincidence that when CRT's were the norm, 60hz with predominantly white screens gave me a headache while 72hz and higher did not. I could see the flickering on the screen at 60hz but not at 72hz, 75hz, and 85hz settings. I don't have this issue with any other screen tech, just CRT's.
Click to expand...
You're comparing apples and oranges. A CRT doesn't have or need a backlight, the higher refresh rate requirement is due to their lightning fast pixel response times. It's the price of zero ghosting and blur in video.

As for image quality it's as much or more about interpolation than frame rates. Higher FPS is often accomplished by inserting interpolated frames between existing frames of a video, the more interpolated frames you add the more artificial it looks.
 
dgz said:
We're going to need at least 1000 fps for things to look realistic. The larger percent of your fov it takes, the higher FPS you're going to need..
Click to expand...

Nah, your fovea can only see about 10 degrees off center or so. So you basically can only see clearly in a cone shaped 20 degrees wide (think about that one the next time you get on the road!) the rest is blurry as all hell so framerates don't matter out there :) Although you do tend to have better ability to pick up movement out of the corner of your eye so perhaps it is better frame rate... just blurred to not get framerate drops :D
 
the 84hz per eye seems to jive with the Valve VR testing found that for sickness free VR 75 to 90 fps is needed
 
I really wish that all videos, games, etc. Would be at least at 75fps.

At the lame 24fps movie speed, I can see individual frames.

48fps might be better, but I have still not seen a movie where if I actually looked, I couldn't see the individual frames.

A jerky mess is what it is.
 
cyclone3d said:
I really wish that all videos, games, etc. Would be at least at 75fps.

At the lame 24fps movie speed, I can see individual frames.

48fps might be better, but I have still not seen a movie where if I actually looked, I couldn't see the individual frames.

A jerky mess is what it is.
Click to expand...

I can clearly see frames at 24fps in movies. It is so bad that I have to unfocus, or avert, my eyes during panning shots or things like car chases. The arc distance between frames for an object is huge on a movie screen. Huge. I can't understand how anyone defends it.

I have seen the first two Hobbit movies at 48Hz in IMAX 3D. It is glorious. Sure it can look odd (for a few seconds). Sure you might notice more detail (which could include makeup). In every way 48Hz is superior to 24Hz, despite 48Hz not being perfect.

With luck, I'll get the chance to experience this again tomorrow if I can get out to see The Desolation of Smaug. Do yourself a favor and do whatever you can to see a good film in 48Hz 3D.
 
sfsuphysics said:
Nah, your fovea can only see about 10 degrees off center or so. So you basically can only see clearly in a cone shaped 20 degrees wide (think about that one the next time you get on the road!) the rest is blurry as all hell so framerates don't matter out there :) Although you do tend to have better ability to pick up movement out of the corner of your eye so perhaps it is better frame rate... just blurred to not get framerate drops :D
Click to expand...

i dont really care about this argument one way or another (because the real argument is far more complicated than anyone here is even close to discussing). but rods (outside the fovea) can see actually respond to much higher rates than the cones in the fovea.

it mentions it here:
http://www.cis.rit.edu/people/faculty/montag/vandplite/pages/chap_9/ch9p1.html
 
Alti said:
Incorrect. It's 300.

Here's proof, just queried it in the console myself:
Code: 
"fps_max" = "0" ( def. "300" )
          notconnected
          - Frame rate limiter, cannot be set while connected to a server.
Click to expand...
Even that's not true. That's the DEFAULT max framerate. I've messed with it up to 900 before doing a motion blur experiment.
 
Dunnlang said:
I can clearly see frames at 24fps in movies. It is so bad that I have to unfocus, or avert, my eyes during panning shots or things like car chases. The arc distance between frames for an object is huge on a movie screen. Huge. I can't understand how anyone defends it.

I have seen the first two Hobbit movies at 48Hz in IMAX 3D. It is glorious. Sure it can look odd (for a few seconds). Sure you might notice more detail (which could include makeup). In every way 48Hz is superior to 24Hz, despite 48Hz not being perfect.

With luck, I'll get the chance to experience this again tomorrow if I can get out to see The Desolation of Smaug. Do yourself a favor and do whatever you can to see a good film in 48Hz 3D.
Click to expand...
I saw the first Hobbit movie in both 24 and 48fps. First off, I do agree with you about panning scenes, 24fps feels too low for high motion content. That said, I generally preferred the 24fps compared to the 48fps. Reason being it just simply felt better. At 24fps, I would think "there's Gandalf, telling a story" and get into it. At 48fps I thought "There's Ian McKellen wearing a robe, giving a performance". In other words, 24fps felt like a movie, its own reality. 48fps felt like I was on the set watching actors perform. The explanation I've heard for this is at the lower framerate, your brain "fills in" more of the picture itself, which creates a different feel. That said, I've never found a GAME more enjoyable at 30fps than at 60fps.
 
JeffDC said:
You're comparing apples and oranges. A CRT doesn't have or need a backlight, the higher refresh rate requirement is due to their lightning fast pixel response times. It's the price of zero ghosting and blur in video.

As for image quality it's as much or more about interpolation than frame rates. Higher FPS is often accomplished by inserting interpolated frames between existing frames of a video, the more interpolated frames you add the more artificial it looks.
Click to expand...

Actually, that's not it. It's the scan lines, CRTs display an image, then wipe and redraw. LCDs display one image until they switch to the next. That's why CRT images look all flickery.

At one point I had a CRT that supported 100hz as a refresh rate and that was the only one that didn't make my eyes hurt after staring at it for a few hours. CRTs don't bother everyone but they used to drive me crazy. I'm so glad that LCDs don't do that, although image persistence can be an issue with LCDs. So we moved for a technology that has a problem with static images to one that has a problem with motion. Most newer LCDs are good enough now. I'm perfectly happy with the Dell U2312HMs I have now but I dread the problems I'm going to have trying to find a 4K screen that doesn't bother me.
 
Quix said:
At one point I had a CRT that supported 100hz as a refresh rate and that was the only one that didn't make my eyes hurt after staring at it for a few hours.
Click to expand...
Eye strain can result from a number of things, not just inadequate refresh rates. Yours is the first claim I've ever heard that 85Hz wasn't enough on a CRT to avoid it.
 
Quix said:
At one point I had a CRT that supported 100hz as a refresh rate and that was the only one that didn't make my eyes hurt after staring at it for a few hours.
Click to expand...

Well yes. Of course your eyes are going to hurt, because your eye muscles are getting strained. You're supposed to take a 5 minute break every hour.
 
JeffDC said:
Eye strain can result from a number of things, not just inadequate refresh rates. Yours is the first claim I've ever heard that 85Hz wasn't enough on a CRT to avoid it.
Click to expand...

It depends on the CRT itself - the way the screens are made with various types of phosphor, they can fade out faster or keep an image longer for less flicker (but more blur).

I've had lots of monitors that were relatively decent at 85Hz, but on which 100Hz+ was a noticeable improvement. Higher-end stuff like a Mitsubishi Diamond Pro 2070SB and a few other similar (older) models to that one
 
I'm the biggest critic when it comes to framerate. I can't stand anything less than 60fps. But I would much rather have 120hz while gaming on my PC.

BUT when it comes to movies. I just saw Hobbit 3d 48 Hz its sooo weird, I didn't like it!! Well at least half of it. I did not like human characters as they looked fake, almost like a soap opera on TV. I just couldn't get in to the movie as much. It definitely took away from the fantasy of the movie. On the other hand the CGI looked absolutely stunning in 48fps 3D!! Weird!
 
Just because you're used to movies in 24FPS doesn't mean you're right when you complain about 48FPS. Framerate has absolutely no effect on realism or anything else. You are just used to garbage framerates.
 
dandragonrage said:
Just because you're used to movies in 24FPS doesn't mean you're right when you complain about 48FPS. Framerate has absolutely no effect on realism or anything else. You are just used to garbage framerates.
Click to expand...
I'd have to agree with this. At first it looked weird to me for about 20 minutes. After that I really appreciated the higher framerate in The Hobbit. It made the action scenes much easier to follow and appreciate.
 
I just know if I have problems with my new BenQ 144hz Z series going to be pissed used it for two days but the calibration was a pain.
 
dandragonrage said:
Just because you're used to movies in 24FPS doesn't mean you're right when you complain about 48FPS.
Click to expand...
So if you get motion sick, your body is just being wrong!
Framerate has absolutely no effect on realism or anything else.
Click to expand...
NO effect on realism huh? So you're saying 24fps is just as real as 48?
 
tetris42 said:
So if you get motion sick, your body is just being wrong!
NO effect on realism huh? So you're saying 24fps is just as real as 48?
Click to expand...

Actually, yes, motion sickness is indeed the fault of your own body.

I should say that higher FPS does not negatively affect realism, but I would also say that 24FPS should not be that big a problem for it, either.

I love how people are making up BS like higher framerates revealing flaws in the film and poor makeup and stuff. Framerate cannot do that. Resolution potentially could, but there was little to no resistance to going to 1080p. Not sure why all the idiots decide to speak up when we improve framerate.
 
dandragonrage said:
Actually, yes, motion sickness is indeed the fault of your own body.

I should say that higher FPS does not negatively affect realism, but I would also say that 24FPS should not be that big a problem for it, either.

I love how people are making up BS like higher framerates revealing flaws in the film and poor makeup and stuff. Framerate cannot do that. Resolution potentially could, but there was little to no resistance to going to 1080p. Not sure why all the idiots decide to speak up when we improve framerate.
Click to expand...
Well you realize that whether it's "wrong" or not, that doesn't change someone getting motion sick and having a valid complaint about the experience?

As for realism, I agree. 48fps absolutely felt more real to me, but again, 24fps felt like a movie, 48 felt like I was watching a live performance.

As for exposed makeup and flaws, I didn't see anything specific like that when I saw The Hobbit, although one possible explanation is that there is a reduction of motion blur at 60fps, so there could actually be more details being exposed on a subtle level because the picture is simply sharper while in motion. It's just a guess though, again, I didn't notice additional details myself.
 
tetris42 said:
Well you realize that whether it's "wrong" or not, that doesn't change someone getting motion sick and having a valid complaint about the experience?
Click to expand...

No matter what you do, there are some people who will complain about motion sickness. Be it 24FPS, 48FPS, moving in a car, moving in a train, moving in a plane, or standing completely still with their eyes closed, there are people with bodies so screwed up that they feel motion sickness. It sucks for them and if it were up to me, it would never occur, but regardless, their own bodies are at fault.
 
You must log in or register to reply here.
Back
Top