Low Light Streaks - Is my camera faulty?

Ozpeter said:
...transmitter masts...
Click to expand...

Not a bad idea. At least it would explain why the streaks sometimes occur and sometimes not. But for me personally it becomes more and more clearer that this flaw exist in every gh1. Time will tell. But if so pana must fix it asap.
 
Nighthawk said:
I'm just now one of the lucky few that has 2 of these cams. I did a quick test with both cams in the same lighting, same settings, same kit lens. Yup, same banding issue. ...
Click to expand...

Hi Nighthawk: Especially since you seem to have 2 GH1s which exhibit this issue, and also a "location" where it seems to occur, please do the test with aluminum foil that I describe here:
http://www.dvxuser.com/V6/showpost.php?p=1679526&postcount=71

The purpose of this test is not to find a way we can all use the GH1 but to determine whether or not the issue you see (at least) is caused by external EMF interference.

Let us know what you find out. Thanks!

P.S.: You might wish to ignore John Caballaro's advise to put your foil-wrapped GH1s in the oven .... :2vrolijk_08:
 
What if it's a coating issue on the sensors glass? That might be why the issue deteriorates over heat up and down over time of use. They knock out these things so quick, what if the coating is not cured!

I say this only because; people report the problem disappearing after a awhile of use.

That's not possible with digital circuits, they work or they don't ( that's why the burn in time to see if it fails or passes with IC's).

It's these slow diminishing lines over time during hours of use reports we keep hearing; sound more like an analog action, which a coating would be akin too!

Most likely a stupid hypothesis.....
 
Last edited:
Since I don't have a GH1 for testing, I'm not sure if the following applies to its Creative/Movie Mode or not (the little icons in the user manual make my head hurt).

I just re-read this entire thread (that was fun!) and I don't believe anyone mentioned what Panasonic says on pages 88 & 89 in the user manual:

"... Noise may become visible when you take pictures in dark places. To avoid picture noise, we recommend setting [NOISE RED.] towards [+] or setting the items except [NOISE RED.] towards [-] to take pictures. ..."

I think this means turn Contrast, Sharpness and Saturation down, and turn Noise Reduction up.

If you set these before recording and they playback directly from the GH1 to a HDTV, does this reduce/remove the "noise/banding" issue we've been discussing?

(Playback on a HDTV may be more useful than watching on the GH1's LCD.)

Or, does adjusting these settings just make a mess of the GH1's video in other ways (gets too blurry, washed out, etc.)?

Or, do these features not seem to apply or have much affect on this issue?
 
Last edited:
Peter J. DeCrescenzo said:
Hi Nighthawk: Especially since you seem to have 2 GH1s which exhibit this issue, and also a "location" where it seems to occur, please do the test with aluminum foil that I describe here:
http://www.dvxuser.com/V6/showpost.php?p=1679526&postcount=71

The purpose of this test is not to find a way we can all use the GH1 but to determine whether or not the issue you see (at least) is caused by external EMF interference.

Let us know what you find out. Thanks!

P.S.: You might wish to ignore John Caballaro's advise to put your foil-wrapped GH1s in the oven .... :2vrolijk_08:
Click to expand...

Peter, I'll give that test you suggested a try ASAP but before trying I'll just mention that there must be something to that theory. The other night I was watching a DVD on my digital projector and brought out my cam to play with(wasn't a very good flick obviously). You can not imagine how things looked on the LCD then. I blamed it on the weird lighting it was dealing with considering the only light source was what was cast by the projector but, in hindsight, the amount of electrical gear hard at work in that room could easily be the culprit.

John's suggestion wouldn't apply to me as I mostly BBQ.
 
I was disappointed finding the vertical lines also in day time footage (1080/25)

It looks like it is more evident on more or less even surface color.

Johnnie




1247611800.jpg


1247611725.jpg


1247612065.jpg
 
Last edited:
PappasArts said:
What if it's a coating issue on the sensors glass? That might be why the issue deteriorates over heat up and down over time of use. They knock out these things so quick, what if the coating is not cured!

I say this only because; people report the problem disappearing after a awhile of use.

That's not possible with digital circuits, they work or they don't ( that's why the burn in time to see if it fails or passes with IC's).

It's these slow diminishing lines over time during hours of use reports we keep hearing; sound more like an analog action, which a coating would be akin too!

Most likely a stupid hypothesis.....
Click to expand...

...but I hope you're right. I've had the first cam for 2 weeks now and nothing's 'burnt' in yet. Still have the issue. I need to test more but it all seems to me like it's exposure levels i.e. something fine at 640 ISO when set to, say, 125 will of course not just be underexposed but will produce the banding to a certain degree and in conjunction with that a ghosting of images with the reds and areas of high contrasts with fairly slow camera movements. Like Peter I'm going through the threads and try a number of the suggested settings and see if I can get any improvement including Kholi's suggestion of keeping the camera recording for hours at a time in low light to see if I feel the 'burn'.
 
Wow, Opherba, even in daylight shots. I still haven't had a chance to get up and personal with any of my footage yet. I'm in pre-production for an upcoming shoot. I may just to hand off to my DP about these issues and hope he finds an answer.
 
Those daylight shots are very worrisome, I assume that is with the kit lens. A fast prime would help but it totally is a deal breaker. This is a serious issue for quality work as a B roll cam.

Aghhh 5D, GH1, 5D , GH1 driving me crazy.
 
OPHERBA said:
I was disappointed finding the vertical lines also in day time footage (1080/25)

It looks like it is more evident on more or less even surface color.

Johnnie
Click to expand...

You need need to document this just as you have here with visual descriptions and get it to a service center... Panasonic needs to see it in their face before they'll do anything....

All it takes is proving it to them! Trust me, I have done this with Sony, Canon and Panasonic in the past. This is where the wheel that squeaks the loudest, gets the grease.....

.
 
Last edited:
My camera also has this problem. The banding, both vertical and horizontal becomes apparent around iso400 and can get pretty obnoxious by iso1600. I have found that if I overexpose (at least by the camera's metering system) the banding is dramatically less visible. At one stop overexposed, the banding is almost completely gone (still there but very faint at iso1600). If underexpose by 1 stop the banding is obnoxious and makes the video painful to watch. At the "proper" exposure, the banding is very noticeable if still faint.

I'm not a video expert, and all I have here to test with at the moment is iMovie, so I have no idea how the overexposed video would handle in post, but it looks usable in the sense I could show it to relatives and they would probably not notice any banding and would think the video looked wonderful. I just have to be careful of clipping the highlights a little earlier.

Is this consistent with other people's experiences?
 
holdingpattern said:
Is this consistent with other people's experiences?
Click to expand...

Yes. It's definitely most evident in underexposed areas, even if the frame as a whole is well exposed.

I've even seen it on reasonably bright footage (inside a well lit restaurant in daylight) with a 1.8 prime lens at 200 iso. It's most visible on flat underlit surface, but can also be seen on skin (if in shadow) which is really ugly.
 
looks like it's going to the service center

looks like it's going to the service center

Yeah - I just about had it. Got this thing since 1st July - and have been testing like crazy! ISO 100,200,400,640,800,1000,1600.... 720p, 1080p.... 25,30,40,50 frames - overexposed, underexposed, full manual, OIS on/off, iexposure on/off. Indoors, outdoors, manual focus, continuous focus, all the film modes, different lighting, different white balance, pixel refresh, sensor cleaning, shoot w/o lens on/off, color space s RGB/ Adobe RGB - got the same results as you guys. Today I tried the burn in thing - set the camera on tripod and let it run for 3 hours straight, underexposed in lowlight. Yesterday it was about 2 hours. I was tempted to buy a nokton 40mm f1.4 but common sense prevailed - what good would any great lens do if the streaking still appears?!
But I do note outdoors is great - in very challenging light it is fabulous. So this thing brings tears of joy AND tears of frustration!
Please anybody.... if you have any solution that works.... I have more vertical smear on the lefthand, and on occasion, consistent banding at the bottom of the screen. I also experience horizontal banding. Sometimes when the whole scene is evenly underexposed, the effect is like looking through a wire mesh - some sort of silkscreen.
Since the time I got it til now, it is not getting better - or at least my mind would like to think it is!
 
Camera is back to the shop.
Pictures describing the problem were send to Panasonic.

Will decide shortly if to try another one or to move on.

I don't believe in voodoo (let the camera run until the problem is gone)...

I am starting to think you North American guys are lucky not to have the camera in large quantities yet...


Johnnie
 
OPHERBA said:
Camera is back to the shop.
Pictures describing the problem were send to Panasonic.

Will decide shortly if to try another one or to move on.

I don't believe in voodoo (let the camera run until the problem is gone)...

I am starting to think you North American guys are lucky not to have the camera in large quantities yet...


Johnnie
Click to expand...


Excellent! Putting this repair service case in record is beneficial to everyone. Giving them photos to document this prevents them from refuting the issue exists, which they would take with, ( path of least Resistance ) if you don't stick it in their face. Make sure you keep track of all your interactions & discussions with whom who have correspondence with. This way others on DVXuser can build the case on top of yours when they send their cameras in, and confront them on the same issue. Good job.... Thanks!

Being proactive is the key to getting this resolved if it's a rectifiable issue..


.
 
From my searches the banding looks very much liked Fixed Pattern Noise (do a google search for more info on this). You can find images from the D3 and D700 that have a very similar appearance at high ISO. High iso for those cameras is 6400 though. The hypothesis is that the gain amplifiers for each row (column?) readout from the sensor need to be exactly calibrated to each other. If they are not some amplify a bit more and in an underexposed image the signal to noise ratio is lower so noise is exaggerated in the rows with higher amplification. Can a firmware update adjust the gain on the sensor line readouts? In the videos I take location, size of the banding is fairly consistent. Can anyone debunk this theory?
 
holdingpattern said:
From my searches the banding looks very much liked Fixed Pattern Noise (do a google search for more info on this). You can find images from the D3 and D700 that have a very similar appearance at high ISO. High iso for those cameras is 6400 though. The hypothesis is that the gain amplifiers for each row (column?) readout from the sensor need to be exactly calibrated to each other. If they are not some amplify a bit more and in an underexposed image the signal to noise ratio is lower so noise is exaggerated in the rows with higher amplification. Can a firmware update adjust the gain on the sensor line readouts? In the videos I take location, size of the banding is fairly consistent. Can anyone debunk this theory?
Click to expand...

Man, I'm open to anything right now. I've tried the black wrap suggested by Peter with little success but I'm not in the same lighting situation that I was before. I'll have to retry tonight. I'm trying to keep everything as consistent as I can. I'm also trying the suggested NR up with everything else down while recording for hours straight. This is where I'd kill for LIVE out right now. Would be much easier to judge in real time what's going on.
 
You must log in or register to reply here.
Back
Top