AVS Forum banner
1 - 9 of 9 Posts

· Registered
Joined
·
48 Posts
Discussion Starter · #1 ·
I thought I'd share this formula - it's been a work in progress and may not be perfect (heck it may even have some serious flaws), but we can always refine the formula if need be (any feedback would be great).


The basic idea is to calculate the blurring/smoothness characteristics of a display, whether it be OLED, LCD or whatever else.


It takes in three objective parameters: Frame rate, response rate and duty cycle (the duty cycle is the proportion of time that a frame isn't black).


It outputs three subjective values: Flicker, Judder, and blur. These deserve some clarification.


* Flicker is what you get on CRTs (particularly 50Hz sets). Not many people notice it, but it can be very irritating for those who do.

* Blur is the typical phenomenon found in LCD sets.

* Judder is linked to the above two, but is more representative of jerky action. Some animations run at 12fps, and you can see the jerky switch between frames.


* Frame rate is measured in frames per second, and it applies to the video information also, not just the display's maximum potential.

* Response rate is measured in milliseconds.

* Duty cycle is measured from 0-1 (1 being 100%)


All of the output ratings are from 0 (worst) to 1 (best). So something like 0.5 would be subjectively 'semi - good/bad'.


Here are the formulas:


flicker = ( ((frameRate^4)/5000000) / (1+((frameRate^4)/5000000)) ) ^ (1-dutyCycle)

judder = ((frameRate^3)/50000) / (1+((frameRate^3)/50000))

blur = (500/ responseRate ^2 ) / (1+(500/ responseRate ^2 ) ) * ( ( (frameRate*(1-( dutyCycle /2))) ^4)/500000) / (1+( ( (frameRate*(1-( dutyCycle /2))) ^4)/500000))


(To avoid divide by zero errors for the "blur" statistic, avoid setting responseRate to zero. If you really want to, then just make it something like 0.000001).


Here are some results from the formulas. Remember they are not to be taken too seriously, since judging a score out of 10 (or 1 in this case) is a tricky subjective thing!

 

· Registered
Joined
·
48 Posts
Discussion Starter · #3 ·
Meh, I take it you're not interested in the upcoming OLED tech :p


These kind of stats are improved over current LCD/plasma sets, and may maximize after the recording industry standardizes on higher frame rates. Also, it's good in general to know how good a TV is going to be if blur, flicker, or judder bother you.
 

· Registered
Joined
·
22,124 Posts
how is "judder" (generally a result of 3:2 pulldown) related to the "flicker" (a function of screen refresh rate unique to crts) and/or "blur" (a function of sample and hold in lcd technology)?


as others have indicated, you might need to consider getting out more...



edit: i, for one, am interested in oled technology. but none of the things you are attempting to calculate appear to be germane. and if you are counting on a higher standardized frame rate, i think you'd best not be holding your breath.
 

· Registered
Joined
·
22,124 Posts

Quote:
Originally Posted by WaldorfSalad /forum/post/16832900


CliffsNotes?

someone with too much time on their hands trying to take their "subjective" observations and confuse them with "objective" results... and then attempting to draw conclusions from disparate data points...
 

· Registered
Joined
·
48 Posts
Discussion Starter · #8 ·

Quote:
how is "judder" (generally a result of 3:2 pulldown) related to the "flicker"

Oh right, I'm not speaking of that kind of 'judder' where one frame lasts longer than another, but rather simply where all the frames are the same time each - whether slow or fast. If each frame lasts exactly 1 second each (1 fps), that would score very low on the 'judder' scale. The 3:2 pulldown thing isn't taken into account at all. I suppose the formulas can be altered so that it is though, but I'm trying to leave the video source out of the equation, since I'm trying to rate the intrinsic display tech as priority (the video source can adapt if need be, well at least for future films).


With that said, try to fault the formulas and/or results (bearing in mind they're only approx, since everyone perceives somewhat differently anyway). As said before, any criticism welcomed.

Quote:
CliffsNotes?
 
1 - 9 of 9 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top