I noticed that a bright dot is showing up on my photos. It doesn't matter what lens I use and it's not my monitor. So this has to be a camera issue. Are these a dead pixel? And what can I do?
You can send your camera to service to have it (be glad there's only one new one) mapped out, get rid of it by using a LENR dark frame even for shorter exposures, or map it out using your raw converter if you shoot raw. If you shoot jpg the stuck pixel won't be so easily identified by post processing software.
Remediation depends partly on the camera body, make and model. Some bodies allow you to map these yourself, while other models require you to send the body in to a qualified repair and service shop to do the mapping.
Often this can also be removed in software, especially during RAW conversion.
Dead pixels are always off
Stuck pixels are always on (at least one color channel)
What raw converter do you use ? (odds are I can't help there, I don't use the common ones, but someone else probably could tell you how to do it).
I use Camera Raw 6.6. Interesting about the pixels .. dead and stuck. how would one know it's a dead pixel if it's always off? I'm confused on that one
Just read up on the warranty. Looks like my camera would go east or west of me. No place close to where I live.
I use Camera Raw 6.6. Interesting about the pixels .. dead and stuck. how would one know it's a dead pixel if it's always off? I'm confused on that one
Just read up on the warranty. Looks like my camera would go east or west of me. No place close to where I live.
Why does this happen and is it a normal issue?
It's normal. You've got > 21 million pixels and associated transistors on that sensor. The quality control that would be necessary to make sure every single one was perfect would make a sensor like that prohibitively expensive. I'd bet you that your sensor has more than one, but they were mapped out in the factory before boxing up your camera.
You usually only see the stuck pixel shooting dark fields. You'd see dead pixels shooting light areas, if they weren't mapped out, if the value wasn't integrated from surrounding pixels.
I am using DNG files. Can a DNG file be converted to a CR2 file? I have a card I've not downloaded yet. I can download those as CR2 files and see what happens.
What I'm understanding is .. if someone downloads their files as CR2's and Camera Raw takes care of the pixels ... then no one would know they even had a pixel problem. Am I correct in that assumption?
I'll read up on your links tonight Hubby home .. supper time.
Mary, yes it can feel like a "violation" when you first discover them. There are also loose pixels, lost and confused pixels, pixels in intensive care, "hot" pixels, "mild" pixels, pixels with ketchup, barbecued pixels, etc? Yes, I was surprised, too!!
But a handful of them are a normal part of electronics. As mentioned above, they show up in low tone areas, and also with longer exposures, high ISO. How much of a problem they are depends on where they are in the image. You might lose them if you crop. They are very easy to treat if they detract.
So, with only a handful of them, it is not really a camera fault, like if a knob fell off! So, lay that pistol down Ma!:D
CR2 files are the RAW files from your camera. You cannot go back to them from a DNG afaik.
Mary, yes it can feel like a "violation" when you first discover them. There are also loose pixels, lost and confused pixels, pixels in intensive care, "hot" pixels, "mild" pixels, pixels with ketchup, barbecued pixels, etc? Yes, I was surprised, too!!
But a handful of them are a normal part of electronics. As mentioned above, they show up in low tone areas, and also with longer exposures, high ISO. How much of a problem they are depends on where they are in the image. You might lose them if you crop. They are very easy to treat if they detract.
So, with only a handful of them, it is not really a camera fault, like if a knob fell off! So, lay that pistol down Ma!:D
CR2 files are the RAW files from your camera. You cannot go back to them from a DNG afaik.
Neil
BBQ'd pixels .. yummy
I still had my photos on my card so I downloaded them again. Only this time in CR2 files. I opened them in Camera Raw and then into photoshop. Pixel was still there. Yes I was hoping for some magic to happen and the pixel would of disappeared, but it was still there. Is there something I'm suppose to do within Camera Raw to make the bugger disappear?
Cloning .. I was cloning it out or else cropping it out today. I have one photo that looks like it had a glow to it like if you take a shot into the sun. Weird.
As you can see I was shooting at a high ISO, but I looked at other photos where I wasn't and it's still there. Bugs the heck out of me and I know it shouldn't.
So does one send the camera in for something like this or just let it be?
I read these two links. From what I understand is my Camera Raw is suppose to be taking care of this problem. I redownloaded my files to CR2, but it didn't take care of them when I opened them in CR and then into photoshop. That I don't understand.
The second link .. That explained a lot. Guess I just hold onto the camera and not send it in ... but .... What would you do Ziggy?
I read these two links. From what I understand is my Camera Raw is suppose to be taking care of this problem. I redownloaded my files to CR2, but it didn't take care of them when I opened them in CR and then into photoshop. That I don't understand.
The second link .. That explained a lot. Guess I just hold onto the camera and not send it in ... but .... What would you do Ziggy?
What do you mean by "redownloaded to CR2"? Did you mean that you redownloaded them from your camera as CR2, or did you convert them from native CR2 to something else and then back
Can you post a screen shot of a 100% zoom of the stuck pixels in the raw file before converting it to jpg or opening it in photoshop?
Electronic failure pixels aren't ISO or long exposure dependent.
What do you mean by "redownloaded to CR2"? Did you mean that you redownloaded them from your camera as CR2, or did you convert them from native CR2 to something else and then back
Can you post a screen shot of a 100% zoom of the stuck pixels in the raw file before converting it to jpg or opening it in photoshop?
Electronic failure pixels aren't ISO or long exposure dependent.
Mary m'lass, what you have captured there is a rare astronomical phenomenon - the opposite to a black hole, iow a white hole, aka a wormhole! The glow spilling onto the carpet is radiation from another dimension. I would not get too close! One of my dear old acquaintances, Alice, actually got taken down one of those, and she was never the same again. A very famous story, that one!
For Charlie's sake don't mess with that thing Mary, it might be the end of the world!
I still had my photos on my card so I downloaded them again. Only this time in CR2 files. I opened them in Camera Raw and then into photoshop. Pixel was still there. Yes I was hoping for some magic to happen and the pixel would of disappeared, but it was still there. Is there something I'm suppose to do within Camera Raw to make the bugger disappear?
Cloning .. I was cloning it out or else cropping it out today. I have one photo that looks like it had a glow to it like if you take a shot into the sun. Weird.
As you can see I was shooting at a high ISO, but I looked at other photos where I wasn't and it's still there. Bugs the heck out of me and I know it shouldn't.
So does one send the camera in for something like this or just let it be?
Using Live View for long periods can cause these to occur, as can very long time exposures. Then again, sometimes they occur (stuck and/or dead pixels) for no reason at all.
Since ACR is not correcting the pixel I suggest that it is probably a "hot" pixel, meaning that there is enough data coming from that pixel for ACR to think that it is a valid and working pixel.
Do let Canon service know the location of the pixel in the frame because their mapping may only automate the removal of stuck and dead pixels.
What do you mean by "redownloaded to CR2"? Did you mean that you redownloaded them from your camera as CR2, or did you convert them from native CR2 to something else and then back
Can you post a screen shot of a 100% zoom of the stuck pixels in the raw file before converting it to jpg or opening it in photoshop?
Electronic failure pixels aren't ISO or long exposure dependent.
What I did was download my card of photos again only I changed how they were converted .. CR2 instead of DNG. I just hadn't erased my card yet so I was able to do this. I did look back at some older photos .. weeks at most and the same issue is there. Since I'm slow at editing lately I didn't notice it till yesterday.
Here is a screen shot. Sorry .. it's probably not big enough for you to see what you may be looking for. I did increase the area bigger in photoshop. The second one scares me. There is a line too.
Mary m'lass, what you have captured there is a rare astronomical phenomenon - the opposite to a black hole, iow a white hole, aka a wormhole! The glow spilling onto the carpet is radiation from another dimension. I would not get too close! One of my dear old acquaintances, Alice, actually got taken down one of those, and she was never the same again. A very famous story, that one!
For Charlie's sake don't mess with that thing Mary, it might be the end of the world!
:ivar
Neil
Neil there is a line now .. looks like that wormhole is leaving a trail rofl
Using Live View for long periods can cause these to occur, as can very long time exposures. Then again, sometimes they occur (stuck and/or dead pixels) for no reason at all.
Since ACR is not correcting the pixel I suggest that it is probably a "hot" pixel, meaning that there is enough data coming from that pixel for ACR to think that it is a valid and working pixel.
Do let Canon service know the location of the pixel in the frame because their mapping may only automate the removal of stuck and dead pixels.
It's going in. After seeing the line with the spot (this morning) I can see there is an issue with it.
I've never used live view yet. I will send a copy of one of the shots with it in and also explain to where it is.
Do I need to contact Canon first before sending it in? I've never done something like this before. I read that I do need to send a copy of my receipt to show it's still under warranty. Am I correct in my understanding of that?
Thank goodness I kept one of my 40D's .. gave one to my son and thought of selling or converting my other one to IR. I'm gonna miss my camera :cry
... Do I need to contact Canon first before sending it in? I've never done something like this before. I read that I do need to send a copy of my receipt to show it's still under warranty. Am I correct in my understanding of that?
What I did was download my card of photos again only I changed how they were converted .. CR2 instead of DNG. I just hadn't erased my card yet so I was able to do this. I did look back at some older photos .. weeks at most and the same issue is there. Since I'm slow at editing lately I didn't notice it till yesterday.
Here is a screen shot. Sorry .. it's probably not big enough for you to see what you may be looking for. I did increase the area bigger in photoshop. The second one scares me. There is a line too.
Since you are sending your camera in, it doesn't matter so much, but, could you explain again what you are doing when you say you only "changed how they were converted"? My understanding is that RAW programs map out the stuck/dead pixels based on their very distinctive appearance in the RAW file that reflects the demosaicing of the Bayer array.
If you convert the raw the file in any way, before trying to map out the pixels, the filtering algorithms can no longer find the stuck pixels. What I was hoping to see was a zoom into the pixel level from the RAW file (where you can see the individual pixels on the screen), I misspoke when I asked for a 100% crop, it should be more like a 400% crop, and not from photoshop version, but from the initial RAW view.
The crossing lines centered on the bright spot are something different, which makes me wonder if this is actually a single stuck pixel or not. A zoom in would see.
Since ACR is not correcting the pixel I suggest that it is probably a "hot" pixel, meaning that there is enough data coming from that pixel for ACR to think that it is a valid and working pixel.
I don't understand this Ziggy, no matter how much charge a stuck pixel is collecting, it's the pattern of the exposed pixel that the RAW converters are mapping out, more data in that one pixel, i.e. increased contrast between the stuck pixel and the surrounding pixels should make it easier for the raw programs to map it out. I think something else must be going on.
... I don't understand this Ziggy, no matter how much charge a stuck pixel is collecting, it's the pattern of the exposed pixel that the RAW converters are mapping out, more data in that one pixel, i.e. increased contrast between the stuck pixel and the surrounding pixels should make it easier for the raw programs to map it out. I think something else must be going on.
It's my understanding that the RAW converters which auto-detect either use either no current (dead pixel) or fully saturated (stuck pixel) "and" only those pixels which are isolated instances. In other words, clusters of pixels which are fully saturated won't trigger the re-mapping algorithm, for instance (and to my understanding).
Hot pixels are much more difficult to discern from some naturally occurring photographic content, so many, if not most, RAW converters don't truly handle hot pixels very well. (I realize that contradicts what Nasim said in his article, but it is my experience none the less.)
Since you are sending your camera in, it doesn't matter so much, but, could you explain again what you are doing when you say you only "changed how they were converted"? My understanding is that RAW programs map out the stuck/dead pixels based on their very distinctive appearance in the RAW file that reflects the demosaicing of the Bayer array.
If you convert the raw the file in any way, before trying to map out the pixels, the filtering algorithms can no longer find the stuck pixels. What I was hoping to see was a zoom into the pixel level from the RAW file (where you can see the individual pixels on the screen), I misspoke when I asked for a 100% crop, it should be more like a 400% crop, and not from photoshop version, but from the initial RAW view.
The crossing lines centered on the bright spot are something different, which makes me wonder if this is actually a single stuck pixel or not. A zoom in would see.]
I've attached a photo to show you what I do. In the circled box you'll see I can download my photos as a DNG file in the advanced section by clicking on the box. I usually have that clicked on. This time when I downloaded my photos I unchecked the box and then they will download in CR2 files. It's an either or kind of thing.
You mentioned mapping out .. when is that done and how is that done in Camera Raw? After they are downloaded to a file I've assigned I then open the RAW file into Camera Raw. I'm assuming it's in Camera Raw the mapping is done, but that is where I get lost.
Here is a 400% in Camera Raw. Screen captured and then increases by 200% for cropping.
Here is a 400% in Camera Raw. Screen captured and then increases by 200% for cropping.
Yeah, that's not a stuck pixel. A stuck pixel is just that, a single pixel. when you zoom in far enough in the raw file to see the individual pixels as a square board, like a chess board, you can see the cross pattern that is generated from the demosaiccing of the bayer array. And the flare wouldn't be there for a stuck pixel. That looks like damage to the sensor. Almost like laser damage, has your camera been around any laser pointers or light shows?
Hot pixels are much more difficult to discern from some naturally occurring photographic content, so many, if not most, RAW converters don't truly handle hot pixels very well. (I realize that contradicts what Nasim said in his article, but it is my experience none the less.)
So here, you mean electrical sensor noise from long exposure or an overheated sensor? Corel/Bibbles raw noise algorithm works decently.
Looking at dogdots latest post - this isn't normal electrical sensor noise. This is a damaged sensor.
Yeah, that's not a stuck pixel. A stuck pixel is just that, a single pixel. when you zoom in far enough in the raw file to see the individual pixels as a square board, like a chess board, you can see the cross pattern that is generated from the demosaiccing of the bayer array. And the flare wouldn't be there for a stuck pixel. That looks like damage to the sensor. Almost like laser damage, has your camera been around any laser pointers or light shows?
No my camera hasn't been around any laser pointer nor light shows. Just taking everyday shots .. nothing fantastic either :cry
Here are what 'stuck' pixels look like. The difference between the red and the green one is because of the Bayer filter.
This is SOOC in RAW with all noise reduction turned off and no enchancement of any kind. I just noticed that you had brightness and contrast turned up on that latest image you posted, but I don't think that increasing the brightness like that could turn this defined stuck-on pixel pattern into the hot spot that you have.
Comments
You can send your camera to service to have it (be glad there's only one new one) mapped out, get rid of it by using a LENR dark frame even for shorter exposures, or map it out using your raw converter if you shoot raw. If you shoot jpg the stuck pixel won't be so easily identified by post processing software.
Remediation depends partly on the camera body, make and model. Some bodies allow you to map these yourself, while other models require you to send the body in to a qualified repair and service shop to do the mapping.
Often this can also be removed in software, especially during RAW conversion.
So, which make and model is this camera body?
(Eric types faster than I type.)
Moderator of the Cameras and Accessories forums
It's on my 5DMKii and it's still under warranty. Something like this would be covered under warranty I would assume.
I do shoot in RAW, but I've not a clue how to do mapping or if I even can. All this is new to me.
Thanks to you both
www.Dogdotsphotography.com
It's just semantics.
Dead pixels are always off
Stuck pixels are always on (at least one color channel)
What raw converter do you use ? (odds are I can't help there, I don't use the common ones, but someone else probably could tell you how to do it).
I use Camera Raw 6.6. Interesting about the pixels .. dead and stuck. how would one know it's a dead pixel if it's always off? I'm confused on that one
Just read up on the warranty. Looks like my camera would go east or west of me. No place close to where I live.
Why does this happen and is it a normal issue?
www.Dogdotsphotography.com
Me?
It's normal. You've got > 21 million pixels and associated transistors on that sensor. The quality control that would be necessary to make sure every single one was perfect would make a sensor like that prohibitively expensive. I'd bet you that your sensor has more than one, but they were mapped out in the factory before boxing up your camera.
You usually only see the stuck pixel shooting dark fields. You'd see dead pixels shooting light areas, if they weren't mapped out, if the value wasn't integrated from surrounding pixels.
Sorry, I thought you occasionally signed as "Eric". I must have you confused with someone else.
Regardless, you can type faster than I can type.
Moderator of the Cameras and Accessories forums
http://forums.adobe.com/message/4493772
Are you using DNG files by any chance?
More information about visible erroneous pixels:
http://mansurovs.com/dead-vs-stuck-vs-hot-pixels
Moderator of the Cameras and Accessories forums
I am using DNG files. Can a DNG file be converted to a CR2 file? I have a card I've not downloaded yet. I can download those as CR2 files and see what happens.
What I'm understanding is .. if someone downloads their files as CR2's and Camera Raw takes care of the pixels ... then no one would know they even had a pixel problem. Am I correct in that assumption?
I'll read up on your links tonight
www.Dogdotsphotography.com
But a handful of them are a normal part of electronics. As mentioned above, they show up in low tone areas, and also with longer exposures, high ISO. How much of a problem they are depends on where they are in the image. You might lose them if you crop. They are very easy to treat if they detract.
So, with only a handful of them, it is not really a camera fault, like if a knob fell off!
CR2 files are the RAW files from your camera. You cannot go back to them from a DNG afaik.
Neil
http://www.behance.net/brosepix
BBQ'd pixels .. yummy
I still had my photos on my card so I downloaded them again. Only this time in CR2 files. I opened them in Camera Raw and then into photoshop. Pixel was still there. Yes I was hoping for some magic to happen and the pixel would of disappeared, but it was still there. Is there something I'm suppose to do within Camera Raw to make the bugger disappear?
Cloning .. I was cloning it out or else cropping it out today. I have one photo that looks like it had a glow to it like if you take a shot into the sun. Weird.
As you can see I was shooting at a high ISO, but I looked at other photos where I wasn't and it's still there. Bugs the heck out of me and I know it shouldn't.
So does one send the camera in for something like this or just let it be?
www.Dogdotsphotography.com
I read these two links. From what I understand is my Camera Raw is suppose to be taking care of this problem. I redownloaded my files to CR2, but it didn't take care of them when I opened them in CR and then into photoshop. That I don't understand.
The second link .. That explained a lot. Guess I just hold onto the camera and not send it in ... but .... What would you do Ziggy?
www.Dogdotsphotography.com
That's ok. People always think I'm a guy on forums.
What do you mean by "redownloaded to CR2"? Did you mean that you redownloaded them from your camera as CR2, or did you convert them from native CR2 to something else and then back
Can you post a screen shot of a 100% zoom of the stuck pixels in the raw file before converting it to jpg or opening it in photoshop?
Electronic failure pixels aren't ISO or long exposure dependent.
they seem to me to become more unsightly under certain conditions, such as those
Neil
http://www.behance.net/brosepix
For Charlie's sake don't mess with that thing Mary, it might be the end of the world!
:ivar
Neil
http://www.behance.net/brosepix
I suggest sending the camera in for remapping.
Using Live View for long periods can cause these to occur, as can very long time exposures. Then again, sometimes they occur (stuck and/or dead pixels) for no reason at all.
Since ACR is not correcting the pixel I suggest that it is probably a "hot" pixel, meaning that there is enough data coming from that pixel for ACR to think that it is a valid and working pixel.
Do let Canon service know the location of the pixel in the frame because their mapping may only automate the removal of stuck and dead pixels.
Moderator of the Cameras and Accessories forums
What I did was download my card of photos again only I changed how they were converted .. CR2 instead of DNG. I just hadn't erased my card yet so I was able to do this. I did look back at some older photos .. weeks at most and the same issue is there. Since I'm slow at editing lately I didn't notice it till yesterday.
Here is a screen shot. Sorry .. it's probably not big enough for you to see what you may be looking for. I did increase the area bigger in photoshop. The second one scares me. There is a line too.
www.Dogdotsphotography.com
Neil there is a line now .. looks like that wormhole is leaving a trail
www.Dogdotsphotography.com
It's going in. After seeing the line with the spot (this morning) I can see there is an issue with it.
I've never used live view yet. I will send a copy of one of the shots with it in and also explain to where it is.
Do I need to contact Canon first before sending it in? I've never done something like this before. I read that I do need to send a copy of my receipt to show it's still under warranty. Am I correct in my understanding of that?
Thank goodness I kept one of my 40D's .. gave one to my son and thought of selling or converting my other one to IR. I'm gonna miss my camera :cry
www.Dogdotsphotography.com
Go to this page and "Request a Repair":
http://usa.canon.com/cusa/consumer/standard_display/consumer_repair
You can choose which service site from that same page.
Moderator of the Cameras and Accessories forums
Thanks Ziggy
www.Dogdotsphotography.com
Since you are sending your camera in, it doesn't matter so much, but, could you explain again what you are doing when you say you only "changed how they were converted"? My understanding is that RAW programs map out the stuck/dead pixels based on their very distinctive appearance in the RAW file that reflects the demosaicing of the Bayer array.
If you convert the raw the file in any way, before trying to map out the pixels, the filtering algorithms can no longer find the stuck pixels. What I was hoping to see was a zoom into the pixel level from the RAW file (where you can see the individual pixels on the screen), I misspoke when I asked for a 100% crop, it should be more like a 400% crop, and not from photoshop version, but from the initial RAW view.
The crossing lines centered on the bright spot are something different, which makes me wonder if this is actually a single stuck pixel or not. A zoom in would see.
I don't understand this Ziggy, no matter how much charge a stuck pixel is collecting, it's the pattern of the exposed pixel that the RAW converters are mapping out, more data in that one pixel, i.e. increased contrast between the stuck pixel and the surrounding pixels should make it easier for the raw programs to map it out. I think something else must be going on.
It's my understanding that the RAW converters which auto-detect either use either no current (dead pixel) or fully saturated (stuck pixel) "and" only those pixels which are isolated instances. In other words, clusters of pixels which are fully saturated won't trigger the re-mapping algorithm, for instance (and to my understanding).
Hot pixels are much more difficult to discern from some naturally occurring photographic content, so many, if not most, RAW converters don't truly handle hot pixels very well. (I realize that contradicts what Nasim said in his article, but it is my experience none the less.)
Moderator of the Cameras and Accessories forums
www.Dogdotsphotography.com
Yeah, that's not a stuck pixel. A stuck pixel is just that, a single pixel. when you zoom in far enough in the raw file to see the individual pixels as a square board, like a chess board, you can see the cross pattern that is generated from the demosaiccing of the bayer array. And the flare wouldn't be there for a stuck pixel. That looks like damage to the sensor. Almost like laser damage, has your camera been around any laser pointers or light shows?
So here, you mean electrical sensor noise from long exposure or an overheated sensor? Corel/Bibbles raw noise algorithm works decently.
Looking at dogdots latest post - this isn't normal electrical sensor noise. This is a damaged sensor.
No my camera hasn't been around any laser pointer nor light shows. Just taking everyday shots .. nothing fantastic either :cry
www.Dogdotsphotography.com
This is SOOC in RAW with all noise reduction turned off and no enchancement of any kind. I just noticed that you had brightness and contrast turned up on that latest image you posted, but I don't think that increasing the brightness like that could turn this defined stuck-on pixel pattern into the hot spot that you have.