horizontal banding in noise
 

Can someone with an xh-a1 close their iris entirely (or put lens cap on) and set gain to 36db ... the noise on my unit is a grey and noisy, but if I leave the unit on like this for half an hour, the image gradually gets more and more "banded" and stroby .. as in, strobing horizontal noisy lines, like analogue interference I suppose. This increase in visible "banding" patterns seems to correspond to the bottom of the A1 getting warm .... As it heats up (and gets worse), it gradually becomes slightly apparent in lower and lower gain levels ...

Is this normal? Or should I return the unit?

Is this an issue that you can see with normal use of the camcorder, or just when shooting the inside of the lens cap at 36 dB gain for 30 minutes or more? 36 dB is a lot of gain, about 64x or 6-stops in film speak.

The noise from the CCD (as well as noise in other analog circuits) is in part thermally generated noise. As the CCD (and other analog circuits) warm up you can expect the thermal noise component to increase with temperature, and if part of the CCD is warmer (a thermal gradient across it), its noise level may be higher at one side than the other. I suppose the thermal gradient might cause other high gain image artifacts as a result of how the CCDs are read as well.

It is probably a good part of why 36 dB gain is considered useless by many people and not found on many camcorders.

I can't say whether or not it is normal for the XH-A1, or any other camcorder for that matter - I've never tried that test or heard of anyone else doing it. If the camcorder serves you well for your normal shooting, you can monitor progress and make your final decision before warranty runs out. But if it causes you to loose sleep, return it.

After 30mins+ of being powered on, the horizontal strobing interference-like noise clearly visible at 12db, and faintly visible at even -3db. This is all tested in low to lowish light conditions.

The 36db test is interesting as you can clearly see the "flat" greyish noise gradually get more and more "banded", until it's a big mess of strobing bands after ~30mins... it is an exaggerated version of what I'm seeing subtle hints of at lower gain levels... I've never seen these 'patterns' in the noise of other cams before (xl1, xl2, pd170, etc).

Is this on a monitor with live camcorder output or image recorded on tape?
HDV or SD mode?

DV and HDV, visible on camera plus through component out to 42" plasma .. NOT yet viewed on PC, as windows 2003 doesn't seem to play happy with the cam. Going to try capture on an XP box shortly...

Really I'd like someone to check their A1 to see if they have the same phenomenon, as if they don't then I'll try to get it replaced ...

But yeah, I'll try to get it captured asap .. currently in the middle of a 4-day edit/grade of a music vid, so a bit manic.

What presets, if any, are in effect?

none, AGC (autogain) off, no noise reduction or anything ...

crap. now my thirty-minute art film "lense cap in the dark at +36db" is totally ruined!

curse you canon!!!

I'd suspect that Don's explanation is pretty much the story and that there is nothing faulty with the camera. The user manual will even caution that after the camera has been on for a while, manual focus needs to be checked because warming of the lens and circuitry can affect focus slightly.

Cheers for the answers so far, it's clear that I'm going to have to capture a sample movie at -3 and +12 gain (after the cam has been powered on for a while), as the fluttering 'interference' lines are unlike anything I've seen on any other digital cam .. almost like RF interference. It seems very faint at -3, while at +12 or higher there's intense strobing of horizontal bands.

I was expecting either, a) people know exactly what I'm talking about, as it's really quite visible, or b) no one knows what I'm talking about, so the cam must be faulty, as it's really quite visible ;)

Sorry, but No, I will not run my camera at +36 gain for a half an hour with the lens cap on to see if I encounter the same garbage.

Kind of like asking a man to run his parked Ferrari's engine at 9K RPM for a half an hour to see if it blows as well.


What is your whole point in this exercise...
I mean, have you shot any actual footage with the camera?

Exactly, I think you have ruined it already since now you also see it at 12db, either that or your cable from the camera is near a power cord, that could introduced interference also, but I wonder what on earth made you do that in the first place? are you a UFO researcher?

I'm surprised that I've gotten such a negative response here, I don't see how I've explained this so colossally badly.

To re-iterate:-

- I'm getting a horizontal banding noise, like analogue interference
- It is not visible when I first turn the cam on
- It is worst after 30mins (turned on, not recording)
- Bottom of the cam gets warm to the touch
- It is very slightly visible at -3db, after being on for a while
- It looks like epileptic inducing strobing bands at 36db, once the cam has been on for a while
- There are strong visible banding noise at 12db, after the cam has been on for a while
- I've not seen it on the approx half a dozen pro cams I've owned.


What I've NOT said:-
- That I'm recording 36db for hours
- That I want anyone else to do the same
- That I want the camera to be usable at high gain

I've noticed a phenominon that affects all levels of gain, that is linked to how long the cam has been on, and that is 'extremely' visible at 36db, and becomes more subtle as you go down.

Since clearly no one knows what I'm on about..

looks like it's a fault.

Can you capture a small sample bit of it; say from 36 dB and 12 dB and post it were we can see it. Maybe attach a frame grab if that works.

What, if any, kind of lighting is leaning thorugh the lens cap? Incandescent, Flourescent, daylight? What shutter speed

Kris,

I set my A1 on manual with the gain at 36 dB. Of course I can see some noise, but not serious banding or strobing like you describe. After 30 minutes the noise is the same as when I started. At 18 dB I can barely see the noise and at 12 dB it is totally gone, even after being on for 30 minutes.

It is well know that the CCDs heat up with use and that the heat causes more noise, however, my camera did not get warm enough to feel or cause significantly more noise after 30 minutes. The bottom of my camera was not even warm to the touch on the bottom or anywhere else. I would suspect that there is a problem with your camera that is producing more heat than the CCDs alone and the extra heat is producing the noise that you are seeing.

What kind of battery are you using? Some batteries may produce more heat than others.

How was the camcorder supported? Anything unusual to limit air flow around the base?

Thanks so much Lloyd, that's exactly what I needed! The bottom of my unit 'does' get warm, and the banding noise is epileptic strobing at 18-36db (when the camera is fully heated up) .. note that if I have the camera on -3db for ~20 mins, then switch to 18+db, I get the banding strobing effect .. it's 'not' caused by running high gain for prolonged periods, it seems to be heat. The banding becomes visible all the way down to -3 (although it's faint).

I'll post footage for reference, just in case someone else has the same frustrating search for answers. I'm certain now that it's faulty. I was delayed in capturing footage as the cam won't communicate with windows 2003- I've now checked that the banding interference 'is' indeed recorded to tape.

Thanks for the help narrowing it down Don, but the cam was in normal conditions, hand-held or on a tripod, battery was the standard one that it came with ... This is in Scotland, so ambient temperatures are pretty low, including footage outdoors at night (with gain tested between -3 and +12).

Sorry to everyone if this wasn't explained clearly, but got there in the end.

"the cam won't communicate with windows 2003"

I'm not familiar with that version of Windows.

"I've noticed a phenominon that affects all levels of gain, that is linked to how long the cam has been on, and that is 'extremely' visible at 36db, and becomes more subtle as you go down."


Same with my camera, definite horizontal banding effect, that diminishes as gain is reduced. Thought it was a normal tradeoff but clearly makes everything above 6db unusable other than for "special effects". its visibe with the camera on power supply or battery, so don't think it has to do with heat, its there anyway the moment I switch on the camera.

Kurt, very interesting ... I'm currently arranging to return my cam, as Lloyd's message convinced me that it was a defect .. I figured that it's so obvious (at 12db+) that people must have noticed it, if it were normal ...

now I'm confused ;)

Agreed, it makes 12db unusable .. 12db on my XL2s was fine- bit noisy, but certainly didnt 'band'... I can handle the noise no problem, but the banding sticks out for me like a sore thumb

No surprise. Not all versions of Windows have equal driver support, and from what I've seen, server versions often may not have the same level AV driver support as desktop/workstation versions. All this is market driven no doubt.

Kris, affirmative, will do the same and bring it in, in 3 weeks , once back from a tour. Please post an update if your cam gets fixed in the meanwhile, very curious if Canon knows of the problem, there are probably a few more "noise bands" out there.

did you your the camera fixed?
 

Hi Kris,

was wondering whether you got your camera fixed in the meanwhile, respectively what Canon's comment was reagrding the banding issue?
I still haven't had time to send mine in.

thanks

Kurt

I shot a big, early fireworks show on June 30th. I used -3db and had the camera on for about 2 hours and 30 minutes. Towards the end, I could *easily* see the epileptic inducing strobing bands, which was very annoying. The noise did seem higher than usual as well.