View Full Version : flicker problem ...


Dean Harrington
December 19th, 2009, 08:14 AM
I have this problem that doesn't want to get resolved. Up to this point I've used everything I have to solve this problem. I've tried plug-ins: Foundry FurnaceCore de-flicker (nothing), all of the de-interlace plug-ins in various modes of sensitivity and adjustment, blur filters ... any suggestions? I have to get this resolved. check it out.

Robert Lane
December 19th, 2009, 09:12 AM
"Holy frequency shift, Batman!..."

Call me crazy but that looks like a "synchronicity" issue. Did you shoot this at 60i and was the room lit with flourescent lights? If so chances are this "flicker" is the relationship between the shutter-speed of the camera being very close to the frequency of those FLO lights.

If (and I'm guessing what the scenario was) that was the setup there's no filter or effect you can apply to remove this problem as it's "hard-coded" into the actual imagery. It's a similar problem to shooting a tube-type TV set and being able to see the raster line paint up and down or worse, freeze in-place based on the camera's shutter speed.

Or it could be an alien spacecraft scanning the room for lifeforms to replace Cartman... that's just a guess too.

Marty Welk
December 19th, 2009, 09:22 AM
i wonder what applying a temporal filter and using it set way to high would do?
after all it is noise in time, and strong enough temporal filtration will even blur frames themselves together, trying to hide noise that is not consistant from frame to frame.

Marty Welk
December 19th, 2009, 09:40 AM
yup that fixes it
wana see
Hotfile.com: One click file hosting (http://hotfile.com/dl/21170132/438c63e/flickerBGone.avi.html)
i hope your mac has a xvid codec , because i dont do quicktime here, so i had to convert it out of QT first, before i could play with it. this just like your video you posted is just a test thing, but it kinda shows that a strong temporal filtration/softening thing could wisk that away at a Cost , of the pic being a bit smoother and having more motion blurring.

Dean Harrington
December 19th, 2009, 04:58 PM
i wonder what applying a temporal filter and using it set way to high would do?
after all it is noise in time, and strong enough temporal filtration will even blur frames themselves together, trying to hide noise that is not consistant from frame to frame.

Got a name on that filter and any settings?

Dean Harrington
December 19th, 2009, 05:06 PM
i wonder what applying a temporal filter and using it set way to high would do?
after all it is noise in time, and strong enough temporal filtration will even blur frames themselves together, trying to hide noise that is not consistant from frame to frame.

yup that fixes it
wana see
Hotfile.com: One click file hosting (http://hotfile.com/dl/21170132/438c63e/flickerBGone.avi.html)
i hope your mac has a xvid codec , because i dont do quicktime here, so i had to convert it out of QT first, before i could play with it. this just like your video you posted is just a test thing, but it kinda shows that a strong temporal filtration/softening thing could wisk that away at a Cost , of the pic being a bit smoother and having more motion blurring.

At this point I'll try anything this side of destroying the image. I checked out your filter ... it works. Exactly which filter and what settings? Thanks

Dean Harrington
December 19th, 2009, 05:19 PM
"Holy frequency shift, Batman!..."

Call me crazy but that looks like a "synchronicity" issue. Did you shoot this at 60i and was the room lit with flourescent lights? If so chances are this "flicker" is the relationship between the shutter-speed of the camera being very close to the frequency of those FLO lights.

If (and I'm guessing what the scenario was) that was the setup there's no filter or effect you can apply to remove this problem as it's "hard-coded" into the actual imagery. It's a similar problem to shooting a tube-type TV set and being able to see the raster line paint up and down or worse, freeze in-place based on the camera's shutter speed.

Or it could be an alien spacecraft scanning the room for lifeforms to replace Cartman... that's just a guess too.

Shot in on the nanoFlash at 720-60p on an EX3 ... the room was lit with what appeared to be tungsten bulbs. Dan from NanoFlash went through the Nano menu with me and it appears that when PSF-progressive is turned on in the Nano and the camera is set-up as progressive there could be a problem ... I'm still not too clear on this point actually. I'll have to talk with Convergent Design again and really understand what this setting in the Nano does!
"Holy frequency shift, Batman!..." No sh_t Robin!

Marty Welk
December 19th, 2009, 06:07 PM
Got a name on that filter and any settings?

its just the temporal smoother that is in virtual dub stock
surely Macs have some sort of temporal noise cleanup filter in one of the programs somewhere?
it might be called a "noise filter" things that would cleanup "Moving grain" on a pic to get rid of low-light gain noise. it analises 2 or 3 frames and tosses out stuff that is moving for no reason ??
i donno, i have never thought of it for this kind of noise untill now.

i checked to be sure (the web sample i did is so compressed) that it wasnt due to conversions or pre-compression filtering, by viewing the live output of virtual dub, so it must be the filter itself. i also tried different settings , it wasnt untill the filter was at 4 or 5 before it completly dissapeared that noise.

Marty Welk
December 19th, 2009, 07:38 PM
Shot in on the nanoFlash at 720-60p on an EX3 ... the room was lit with what appeared to be tungsten bulbs. Dan from NanoFlash went through the Nano menu with me and it appears that when PSF-progressive is turned on in the Nano and the camera is set-up as progressive there could be a problem ... I'm still not too clear on this point actually. I'll have to talk with Convergent Design again and really understand what this setting in the Nano does!
"Holy frequency shift, Batman!..." No sh_t Robin!

$2800 and you get THIS? umm, no thanks.
this type of noise doesnt look like anything to do with codecs, resolutions or scan rates, progressive vs interlace, or Digital signal transportation or digital recording.
(unless of course it was the lighting , any arc lighting , florescent mercury or sodium vapor Cfls)

The below Assume that it was incandescent lighting, as stated.

what it looks more like (but i donno) is an analog noise, or a EM injection noise into a analog.
power induction hitting the camera or analog wiring.
A "grounding" problem from the power. AKA Ground Loop, which is a badly named term for having a power or sine wave differential passing through things wired at different lengths from a AC power source. (gee now i know why they call it something simple)
other possibilities , someone using a cell phone, which will induce large RF waves into any (usually only analog) signals.

i must know more info, getting $^#%& like this into my video would simply destroy my brain cells :-) for days after too.

were you connected to AC power anywhere, with any aspect of your units?
were you connected TO boards or devices that are connected to AC, sound board, computer, switcher?
did you use Radio transmitters neer the camera or neer the nano , like wireless headsets?
Were any wireless mic Transmitters (not recievers) close to the camera or nano?
Were people allowed to use cell phones? was the noise constant or only during a period of time?
does the Nano use any A-D conversions internally?
was this also on the recording you did direct to the camera , or only the nano?

Dean Harrington
December 19th, 2009, 10:14 PM
$2800 and you get THIS? umm, no thanks.
this type of noise doesnt look like anything to do with codecs, resolutions or scan rates, progressive vs interlace, or Digital signal transportation or digital recording.
(unless of course it was the lighting , any arc lighting , florescent mercury or sodium vapor Cfls)

The below Assume that it was incandescent lighting, as stated.

what it looks more like (but i donno) is an analog noise, or a EM injection noise into a analog.
power induction hitting the camera or analog wiring.
A "grounding" problem from the power. AKA Ground Loop, which is a badly named term for having a power or sine wave differential passing through things wired at different lengths from a AC power source. (gee now i know why they call it something simple)
other possibilities , someone using a cell phone, which will induce large RF waves into any (usually only analog) signals.

i must know more info, getting $^#%& like this into my video would simply destroy my brain cells :-) for days after too.

were you connected to AC power anywhere, with any aspect of your units?
were you connected TO boards or devices that are connected to AC, sound board, computer, switcher?
did you use Radio transmitters neer the camera or neer the nano , like wireless headsets?
Were any wireless mic Transmitters (not recievers) close to the camera or nano?
Were people allowed to use cell phones? was the noise constant or only during a period of time?
does the Nano use any A-D conversions internally?
was this also on the recording you did direct to the camera , or only the nano?

My brain cells are disappearing quickly and yes, all of the above apply. Hooked up AC but near a sound board with translators close by. Humm ... interesting problem. I will continue to search for the right solution. Thanks
Just tried neat plug-in for noise reduction ... no change.

Dean Harrington
December 20th, 2009, 09:38 AM
After applying a large number of plug-ins I have finally founds something that is acceptable. I applied the 'Stop motion blur' plug-in in FCP with the following settings:
Time lapse - 1
Steps - 1
Initial opacity (52) flexible quality
Decay opacity - 0
Blend operator - add

Anyone else running into this problem may find this plug-in handy for more than just stop motion.
all the best

Steve Crow
November 27th, 2010, 08:50 AM
Thanks Dude, you saved my bacon!