Anyone have experince with persistance of vision effects?
I'm trying to make my Delux Flasher do really nifty stuff but I'm running into persistance of vision problems. If I run the blinking too slow, the flicker is visible, too fast and humans cannot see the effect.
Unfortunately, all I gots is 1 and 0 for outputs, and I gotta do nifty stuff with just that.
Hmmm.
-John C.
Persistance of Vision?
Moderators: Sparky, Moderators
-
Madman Lighting
- Posts: 1818
- Joined: Sat Mar 05, 2005 9:16 am
- Location: Serenity.
- Contact:
Persistance of Vision?
That Madman Who Lit Up Deep Space Nine
-
b5ranger99
- Posts: 551
- Joined: Mon Feb 23, 2004 1:11 pm
- Location: SC
- Contact:
John,
Can you define "too slow". IIRC anything over 15 cycles per second is going to have issues. If you are running slower than that it could be a duty cycle issue. If the LED is on over 50% of the time there will also be and issue.
Any details .e.g. the basic circuit (555 or PIC) would also be a help.
Joe
Can you define "too slow". IIRC anything over 15 cycles per second is going to have issues. If you are running slower than that it could be a duty cycle issue. If the LED is on over 50% of the time there will also be and issue.
Any details .e.g. the basic circuit (555 or PIC) would also be a help.
Joe
"When others do a foolish thing, you should tell them it is a foolish thing. They can still continue to do it, but at least the truth is where it needs to be."
Dukhat, Babylon 5
“Imagination is the only weapon in the war against reality.”
Lewis Carroll
Dukhat, Babylon 5
“Imagination is the only weapon in the war against reality.”
Lewis Carroll
-
Madman Lighting
- Posts: 1818
- Joined: Sat Mar 05, 2005 9:16 am
- Location: Serenity.
- Contact:
Hi Joe,
Its a programmable microcontroller and my minimum time resolution is 256microSeconds (ten to minus 6th). With a few clicks, I can reduce that to 64 uSec so resolution isnt an issue. With this rig, I can program a minium on or off time as short as 64uSec (with some jitter). My test rig is an array of yellow LEDs and yes, I know that they eye responds differently to different colors.
My problem is I dont really know where to lookup how the human eye responds to light pulses. What are the shortest on and off times the eye will notice as flicker? I achive variable brightness by pulse width modulation so I need to adjust the timing so they eye sees variable brightness as a smooth effect, with no flicker.
Its a programmable microcontroller and my minimum time resolution is 256microSeconds (ten to minus 6th). With a few clicks, I can reduce that to 64 uSec so resolution isnt an issue. With this rig, I can program a minium on or off time as short as 64uSec (with some jitter). My test rig is an array of yellow LEDs and yes, I know that they eye responds differently to different colors.
My problem is I dont really know where to lookup how the human eye responds to light pulses. What are the shortest on and off times the eye will notice as flicker? I achive variable brightness by pulse width modulation so I need to adjust the timing so they eye sees variable brightness as a smooth effect, with no flicker.
That Madman Who Lit Up Deep Space Nine
-
USS Atlantis
- Posts: 2388
- Joined: Sat Sep 08, 2007 6:44 pm
- Location: Galaxy 217, Orion Arm, Sol System, Sol III, 44° 53' N 93° 13' W (Local coordinate system)
- Contact:
Just as a spot-check
Standard TV runs 30 fps (frames per second)
Older Animated (Star Trek for instance) runs 24 fps
Most online GIFs run in the 10-12 fps range
Some studies show that the human eye can distinguish flicker at up to 100 fps
So, somewhere in there is the point at which the blink becomes 'steady on' to the human eye
Standard TV runs 30 fps (frames per second)
Older Animated (Star Trek for instance) runs 24 fps
Most online GIFs run in the 10-12 fps range
Some studies show that the human eye can distinguish flicker at up to 100 fps
So, somewhere in there is the point at which the blink becomes 'steady on' to the human eye
-
Chacal
- Posts: 3654
- Joined: Fri Jul 12, 2002 3:09 pm
- Location: Rio. Always unseasonably warm, even in the Winter, when we'll host the Summer Olympic Games of 2016
I didn't get the question. That's the essence of the effect: slower than "X", flicker. Faster than "X" smooth motion. The only 'in between' is at speeds in which some folk see blinky, while other folk see smoothy.
Sheer elegance in its simplicity.
Political unrest in dictatorships is rather like a round of rock-paper-scissors: The oposition goes on denouncing the regime on the papers, the regime censors the papers, rock-throwing ensues.
Political unrest in dictatorships is rather like a round of rock-paper-scissors: The oposition goes on denouncing the regime on the papers, the regime censors the papers, rock-throwing ensues.
-
b5ranger99
- Posts: 551
- Joined: Mon Feb 23, 2004 1:11 pm
- Location: SC
- Contact:
I think I understand the question.
I am working on a similar circuit myself.
If I understand correctly, you need to determine the frequency and the duty cycle. Using Kylwell's input as a standard if you set the frequency to 80 htz (72 + 10% approx.) as you reduce the duty cycle to give the impression of dimming the LED, when is it seen as dim vs start to look like it is flickering?
SCC-7107 USS Atlantis has a point that it is going to vary from person to person, but there must be a point that covers some large percentage of people.
I am going to default to the old engineering standard of X10 is safe. I would (will) start at 800 htz for a cycle time and 10% duty cycle minimum.
I just received my PIC programming station so it will be a couple weeks (real life
) before I can test out my theory. If you get a chance first let me know.
I am working on a similar circuit myself.
If I understand correctly, you need to determine the frequency and the duty cycle. Using Kylwell's input as a standard if you set the frequency to 80 htz (72 + 10% approx.) as you reduce the duty cycle to give the impression of dimming the LED, when is it seen as dim vs start to look like it is flickering?
SCC-7107 USS Atlantis has a point that it is going to vary from person to person, but there must be a point that covers some large percentage of people.
I am going to default to the old engineering standard of X10 is safe. I would (will) start at 800 htz for a cycle time and 10% duty cycle minimum.
I just received my PIC programming station so it will be a couple weeks (real life
) before I can test out my theory. If you get a chance first let me know."When others do a foolish thing, you should tell them it is a foolish thing. They can still continue to do it, but at least the truth is where it needs to be."
Dukhat, Babylon 5
“Imagination is the only weapon in the war against reality.”
Lewis Carroll
Dukhat, Babylon 5
“Imagination is the only weapon in the war against reality.”
Lewis Carroll
-
Madman Lighting
- Posts: 1818
- Joined: Sat Mar 05, 2005 9:16 am
- Location: Serenity.
- Contact:
Yes, you've got the idea. I need to compute frequency and duty cycle and write some code so that I get smooth transitions in brightness without the human eye seeing flicker.
I'm probably being a little sloppy right now in my math so I'll go back and double check.
Ugh! Right now I have to finish building a huge number of boards so I can restock the SSM store!
You will all have your new warp core effect kit (Delux Spinner) with this batch of kits. I have PERSONALLY HAND SOLDERED 200 LEDS to make those kits!
I'm probably being a little sloppy right now in my math so I'll go back and double check.
Ugh! Right now I have to finish building a huge number of boards so I can restock the SSM store!
You will all have your new warp core effect kit (Delux Spinner) with this batch of kits. I have PERSONALLY HAND SOLDERED 200 LEDS to make those kits!
That Madman Who Lit Up Deep Space Nine