where there’s smoke there’s forum fire
Hello all,
I would like to pose a small experiment to someone that has a bad pixels in their video in a D7K.
Cap on, in Live View, does a Still photo have the same bad pixels in the same 'relative' position?
Cap on out of Live view, are the bad pixels gone?
My best,
Mike
Remember that Sony warned new Alpha A33/A55 overheat when recording video. There are rumors that Nikon D7000 and Sony A55 use the same sensor or at least a shared sensor design.
This hot pixel issue can be related to the sensor itself.
abetanco said:
Remember that Sony warned new Alpha A33/A55 overheat when recording video. There are rumors that Nikon D7000 and Sony A55 use the same sensor or at least a shared sensor design.This hot pixel issue can be related to the sensor itself.
Sony was brave enough to acknowledge the problem, what is Nikon waiting for?
abetanco said:
Remember that Sony warned new Alpha A33/A55 overheat when recording video. There are rumors that Nikon D7000 and Sony A55 use the same sensor or at least a shared sensor design.This hot pixel issue can be related to the sensor itself.
Wait. You're trying to tell me hot pixels are exacerbated by heat? And that said pixels reside on a sensor?
huh.
;)
To answer Mike's question, NO, I did not see hot pixels with or without live view when taking stills. NR was set to off, cap was on, etc. I was looking at JPGs on the computer, but also scanning NEFs in the camera back at maximum mag.
Interestingly, I came home and was set to do this experiment and took a little test vid with cap on to establish where the hot pixels were and didn't see any! The camera was cold, though, from being in the car all day. Sure enough, an hour later the hot pixels were back in video roughly where I remember them being when I first tested this a month ago. The stills with no hot pixels were taken after the camera had warmed up.
I wonder what other people are seeing.
My Serial starts with 3008XXX, and have zero problems thus far. Great performance, and huge upgrade from my d70s lol.
mcammer said:
To answer Mike's question, NO, I did not see hot pixels with or without live view when taking stills. NR was set to off, cap was on, etc. I was looking at JPGs on the computer, but also scanning NEFs in the camera back at maximum mag.
Hi mcammer,
I was barking up the wrong tree. Sorry.
The pixels from the video are visible in Live View and always in the same position in all resolutions relative to the LCD screen, which I had thought might be something, but of course, had I thought it out more would have realized that still images aren't the same resolution as video.
They don't get their imaging from the LCD panel, where the video might get their images.
My best,
Mike
Mike Gunter said:
They don't get their imaging from the LCD panel, where the video might get their images.
I might be misreading what you intended to convey, but I can assure you the LCD is downstream of the same headwaters that feed the video. The LCD is not midstream of any data-flow. It is an independent end-point.
Drab said:
I might be misreading what you intended to convey, but I can assure you the LCD is downstream of the same headwaters that feed the video. The LCD is not midstream of any data-flow. It is an independent end-point.
Hi Drab,
Headwaters meaning CMOS sensor, yes, but I think there's more afoot in video.
The defective pixels lie in the video processing seen in the LCD.
I don't buy the hot pixel in 'High Definition' sensor argument at all - with the exception that overexposure will show noise; the output in video isn't close to the number of pixels of the sensor. The content is processed and resampled downward.
I had wondered if the content was 'passed through' the LCD. Can one, using a switch, record the display data?
My best,
Mike
Mike Gunter said:
The defective pixels lie in the video processing seen in the LCD.
Remove the word "processing" and we're all on the same page.
Mike Gunter said:
I don't buy the hot pixel in 'High Definition' sensor argument at all - with the exception that overexposure will show noise; the output in video isn't close to the number of pixels of the sensor. The content is processed and resampled downward.
As others have mentioned, there is a very very low chance Nikon uses every line of the sensor in video mode. The Jello Effect typical of CMOS sensors is caused by slow line read speeds. Reading every line when only 1080 lines are needed is very unlikely. Forget for a moment that downsampling is computationally expensive, reading every line is almost impossible on current hardware. The buffer just isn't large enough. This is evidenced by the fact there is a slowdown at all when shooting high-speed continuous. Obviously the mechanical limitations of the shutter/mirror determine the maximum frame rate, but the fact one can fill the buffer and experience a framerate slowdown is proof, not evidence, that the RAM buffer is full. If Nikon was reading the entire sensor in video mode @ 23.976 FPS the RAM buffer would need to be orders of magnitude larger than it is. One could shoot 6 fps forever.
The other argument is that in video mode the sensor's duty-cycle is significantly higher than in any normal still photo mode. As has been said these are "hot" pixels and not stuck pixels. Video equals at least a 50% sensor duty-cycle (24fps @ 1/48th of a second) if Nikon is able to fully power-off the sensor and amplifiers between frames. There is likely a power-up and power-down lag of up to 100ms for such a large die, so we'll call that a 55% duty cycle (.02083 seconds vs .02283 seconds of sensor power per frame).
In high-speed still shooting, let's assume a worst case. You're shooting in the dark, trying to sneak in one non-blurred shot. 1/20th of a second. That's 0.050 seconds of sensor activity per shot, .002 seconds added for power up and power down delays, 0.052 seconds of powered sensor * 6 = 0.312. A 31% sensor duty cycle. So to get the sensor as hot as recording video one would need to shoot high-speed continuous @ 6fps @ 1/10th of a seconds, forever and ever. But wait - you can't: The buffer fills. This is why hot pixels are exacerbated in video when compared to stills.
Mike Gunter said:
I had wondered if the content was 'passed through' the LCD. Can one, using a switch, record the display data?
The LCD in most consumer electronics is fed by its own framebuffer. No comprehensive tear-down of Nikon cameras exist, so I can't even comment on what chip they are using as the LCD controller, but there is no reason to expect them to have reinvented the wheel.
While one could conceivably fork the data path post-framebuffer, it would take a custom piece of hardware, gaining nothing while costing a pretty penny. The way dual-outputs are normally done (iPods, portable DVD players, etc) is to fork the video data before the LCD controller (which is where the framebuffer I mentioned is physically located.)
It gains nothing because the data in the LCD controller's framebuffer is 1 - uncompressed, 2 - aligned for the data bus to the LCD, and 3 - in the colorspace required of the display itself. Since the LCDs on consumer electronics are not "high gamut" displays this means the colorspace has been decimated, and there is not enough memory in said framebuffer to hold full colorspace. Said buffer expects aligned inputs and outputs, because latency is key, else one would get tearing and all sorts of other nasty visible artifacts on the display itself.
regards,
Saul
PS - I'll think some more about this and try to ponder a reason Nikon would have take an unconventional approach. It clearly IS possible, but for the reasons I mentioned above I see no reason to believe they did. Everything we see is explainable pretty robustly w/o exotic action on Nikon's part.
Does anyone know how wide spread this hot pixel problem is?
Is it only in the US, or is it world wide? Because I only hear about cameras being bought from Best Buy!
Has anyone experienced a problem with D7000 bright/hot/stuck pixels similar to mine? I've taken a series of JPEGs with different ISO settings and exposure times. Examining the photos in Lightroom 3, I see a few bright pixels starting at a shutter speed of 0.5 s at ISO=3200. The number of bright pixels increases with increasing ISO (probably no surprise there). It also increases with increasing exposure time, peaking at 0.8 s, at which point there are roughly 3 dozen bright pixels, most of them Red. I initially thought the frequency of these hot pixels might depend on EV. However, in exposures of 1 s or longer, all of the bad pixels vanish (!), and so there does not seem to be an obvious relationship to EV. I took all of the foregoing shots with noise reduction off, either with the lens cap on or in a fully darkened room. The bad/bright pixels are not the same as the saturated pixels that appear in long 30 s exposures taken at high ISO; they are entirely different pixels. Consistent with that, turning NR on does not have any noticeable effect in suppressing the bad/hot pixels.
In normal use of this camera, I can't see myself ever taking many 0.5-1 s exposures at ISO>2500. Most of my photos will be shot outdoors in bright light and so will lend themselves to typically much faster shutter speeds and much lower ISO values, where I see no evidence for hot/stuck pixels. However, 3 dozen bad/hot pixels does seem to be a far greater number of defects than other people have been reporting at high ISO. Would any of you experienced Nikon users keep this camera, or would you consider the problem serious enough that you'd suggest that I return it? I have to make a decision in the next week and would appreciate any advice.
Baba Ganoush said:
Has anyone experienced a problem with D7000 bright/hot/stuck pixels similar to mine? I've taken a series of JPEGs with different ISO settings and exposure times. Examining the photos in Lightroom 3, I see a few bright pixels starting at a shutter speed of 0.5 s at ISO=3200. The number of bright pixels increases with increasing ISO (probably no surprise there). It also increases with increasing exposure time, peaking at 0.8 s, at which point there are roughly 3 dozen bright pixels, most of them Red. I initially thought the frequency of these hot pixels might depend on EV. However, in exposures of 1 s or longer, all of the bad pixels vanish (!), and so there does not seem to be an obvious relationship to EV.
This conversation is about hot pixels in video. What you describe is the typical increase in noise levels as amplification (ISO) and exposure times increase.
They disappear "all of a sudden" because I'll bet you dollars to donuts you have long-exposure noise reduction turned on. For exposures of 1s or longer a dark frame (shutter closed shot) is taken and subtracted from your picture, removing the effect of systematic sensor-based noise.
ilhooq said:
So I think I've gone through all 15 pages and I see a few references to Nikon saying that (paraphrase) hot pixels are normal for video.I just returned mine to Amazon with one hugely bright white pixel at left center and three others forming a triangle a dead center of image. The one bright one (also visible in Live View) was enough for me to feel it was defective, the other three were just a joke, really. Even for non-pro video, the image is not usable.
I submitted an issue to Nikon USA via website (with my four other bodies, this has always been the best/easiest/fastest way to go), attached sample video... and they just completely ignored the issue. After four days, no one ever even picked up the ticket, so I sent it back and had to eat it for $50 for the shipping.
Extremely disappointed with the whole process, but my fault for wanting a half-price 5DMkII, I guess. Lost a lot of respect for Nikon with this.
I know that there are a few of you here who think that this could/should be a firmware issue, but has anyone had a camera serviced or repaired or otherwise handled by Nikon? Or is it point-of-sale only? Any NPS members gone through this yet?
I got a phone call on Monday, four days after I submitted info to Nikon that I was returning the camera, seven days after I reported it defective.
The guy on the phone was assiduously neutral on the topic, and offered no explanations, solutions, compensation, or mention of future engineering, firmware or fixes. No, I wasn't expecting any compensation, but fifty bucks on return shipping is _still_ fifty bucks for the privilege of beta-testing a bad body. I was, however, hoping for Nikon to articulate a coherent explanation here and also to offer a way to get a functional body into my hands.
In response, I got no commentary or assistance from Nikon Tech in any way, really, other than the call itself.
I'll be looking to rent Canon bodies for an upcoming DSLR test-shoot for a documentary feature-length project, with an eye to progressively dumping Nikon bodies and lenses in the next 12 months or so. I was in the middle of growing a business around Nikon equipment here, and it seems very clear to me that — between a lack of engineering/QC and a lack of tech/after-sales support — I have chosen the wrong company to rely upon.
Nikon is seriously burning its brand with this body, and is still very much not in front of the problem here.
From my point of view, spending payroll dollars to call up a customer and then offer absolutely nothing in the way of information or real support is not a long-term strategy in the early-21st century consumer/pro electronics business.
Hi Saul,
"Remove the word "processing" and we're all on the same page."
Unlike RAW, AVCHD is a highly processed file, clearly not through the LCD but before it and after it to output, but nonetheless compressed somewhere. Whatever energizes 'Live View' also switches on' Video'
Which leads...
"As others have mentioned, there is a very very low chance Nikon uses every line of the sensor in video mode. The Jello Effect typical of CMOS sensors is caused by slow line read speeds."
I'll leave the rest of your comments out. Good read and thanks.
But I do have some questions.
How would that happen optically? If only 1080 lines of the sensor (assuming that is the line group read for 1080) were read, how would one use any given focal length on a camera for the APS sensor?
My best,
Mike
Mike Gunter said:
Unlike RAW, AVCHD is a highly processed file, clearly not through the LCD but before it and after it to output, but nonetheless compressed somewhere. Whatever energizes 'Live View' also switches on' Video'
Live View IS video, without recording. I really don't see where you're going with this.
Mike Gunter said:
How would that happen optically? If only 1080 lines of the sensor (assuming that is the line group read for 1080) were read, how would one use any given focal length on a camera for the APS sensor?
Not sure I understand the question. Are you asking if skipping lines would change the effective crop factor?
The effective crop factor would only change if one were to use only the center 1920x1080 light wells. If one were to simply skip whole rows and whole columns, though, this would not be the case.
As I said, I can't prove this - but it is simple, makes sense, and fully explains what we see.
Hi Testing123,
You mentioned that all the pixels weren't sampled.
"As others have mentioned, there is a very very low chance Nikon uses every line of the sensor in video mode. The Jello Effect typical of CMOS sensors is caused by slow line read speeds. Reading every line when only 1080 lines are needed is very unlikely. Forget for a moment that downsampling is computationally expensive, reading every line is almost impossible on current hardware..."
What strikes me as odd about that, if that were the case, is that the you would only use only 1080x1920 lines of the APC sensor.
That was the question.
Maybe you didn't mean what you said; it happens to me all the time. ;-)
My best,
Mike
Mike Gunter said:
What strikes me as odd about that, if that were the case, is that the you would only use only 1080x1920 lines of the APC sensor.That was the question.
Maybe you didn't mean what you said; it happens to me all the time. ;-)
Nah - I mean that. What I don't understand is why you find that hard to believe? I don't claim to know the answers, but I think my model well explains observable evidence, but would love to have holes poked in it.
I think it is extremely likely they are only reading ~1080 lines. How many pixels of width they are using is a different matter, one I haven't figured out a way to test.
I don't believe they are reading only the MIDDLE 1080 lines, as that would affect the crop factor. But I believe they are only reading ~1080 lines total. If not I must believe Nikon now has two sensors from two sources which BOTH have CMOS line read speeds nearly perfect.
regards
-Saul
Drab said:
This conversation is about hot pixels in video. What you describe is the typical increase in noise levels as amplification (ISO) and exposure times increase.They disappear "all of a sudden" because I'll bet you dollars to donuts you have long-exposure noise reduction turned on. For exposures of 1s or longer a dark frame (shutter closed shot) is taken and subtracted from your picture, removing the effect of systematic sensor-based noise.
This thread began as a discussion about HOT PIXELS in general, which may show up most prominently in video, but also appear in still photos. Your concern might be about the video issue. I am interested in (and am asking more experienced people for help concerning) the more general problem of defects in the Nikon sensors. They are both legitimate issues for discussion.
You have lost your bet: NR was definitely turned off. I was very methodical and made sure of that, checking my NR menu selections both before and after taking the shots. My initial assumption was the same as yours, namely, that a dark frame was taken for the longer exposures, but unless there are some hidden NR parameters in the menus that I mayhave overlooked, I can assure you that NR was turned off.
PS: If you were right about the bright pixels I see in the short exposures being just normal, saturated pixels they should also have shown up in the much longer 30 s exposures I took with NR turned off. However, they do not show up in those 30 s exposures. They are an entirely different set of pixels. That is what is so baffling.
Does anyone know what is likely to happen to a camera with an allegedly defective sensor that is returned to the seller? Is it likely to be sent back to Nikon and refitted with a new sensor? Or is it more likely to be just repackaged and restocked? If the latter, won't there be quite a number of D7000s with blemished sensors floating around in the market for a long time to come? One reason I hesitate to return the camera I have is that any replacement I get for it might be even worse, perhaps a camera that someone else has already rejected and returned.
Baba Ganoush said:
Does anyone know what is likely to happen to a camera with an allegedly defective sensor that is returned to the seller? Is it likely to be sent back to Nikon and refitted with a new sensor? Or is it more likely to be just repackaged and restocked? If the latter, won't there be quite a number of D7000s with blemished sensors floating around in the market for a long time to come? One reason I hesitate to return the camera I have is that any replacement I get for it might be even worse, perhaps a camera that someone else has already rejected and returned.
Nope, thats why they sell refurbs(in the USA), consumer protection laws and companies reputation. I purchased a D300s instead of the D7K. I'm happy now that I have it, but was worried I was buying 4 YO technology, not the latest and greatest. Hope that the Nikon says something soon, bet the retailers would like to sell lots of D7K's and not see them being returned.
Baba Ganoush said:
You have lost your bet: NR was definitely turned off. I was very methodical and made sure of that, checking my NR menu selections both before and after taking the shots. My initial assumption was the same as yours, namely, that a dark frame was taken for the longer exposures, but unless there are some hidden NR parameters in the menus that I mayhave overlooked, I can assure you that NR was turned off.
The crow's out of the freezer, but I won't put it in the oven yet...
Both forms of NR are turned off? You only explicitly mention one...
Baba Ganoush said:You have lost your bet: NR was definitely turned off. I was very methodical and made sure of that, checking my NR menu selections both before and after taking the shots. My initial assumption was the same as yours, namely, that a dark frame was taken for the longer exposures, but unless there are some hidden NR parameters in the menus that I mayhave overlooked, I can assure you that NR was turned off.
PS: If you were right about the bright pixels I see in the short exposures being just normal, saturated pixels they should also have shown up in the much longer 30 s exposures I took with NR turned off. However, they do not show up in those 30 s exposures. They are an entirely different set of pixels. That is what is so baffling.
It is well known that Nikon applies some noise reduction on long exposures (I remember it was 0.4 secs, not 1 sec, but I maybe wrong), not "normal" noise reduction, but even before written RAW data. Thom Hogan, Iliah Borg, Marianne Oelund... and some other Nikon-gurus talked about that in a thread on dpreview forums time ago.
Perhaps that is the reason.
Saludos!
You must log in to post.
