Reading mode: Light Dark

10d dead pixel question, HELP!!!!!!!!!!!!!!!!

Go to last post
S

sammmysams

Active member
Messages
73
Reaction score
0
Location
US
I did a dead pixel test with the cap on in a dark room...found one small red dot in the middle and another smaller blue next to it and nothing else...this is a brand new camera....what do the experts recommend? Is this enough to return it for??? Thanks
 
that depends...how often do you plan to photograph with the lens cap on? ;-)

I have more than two. I have one big one and about a half dozen with lens cap on for anything over 30 seconds. No big deal for me since I don't do long exposures. Two isn't bad depending on your exposure time. If you decide to exchange it, you might have more, you might have less. It's gambling.
                            • -- - - - - - - - - - - - SMoody
http://www.pbase.com/smoody
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
 
iso 100, large jpg, at both fast and long shutter speeds
Don Trust said:
sammmysams said:
I did a dead pixel test with the cap on in a dark room...found one
small red dot in the middle and another smaller blue next to it and
nothing else...this is a brand new camera....what do the experts
recommend? Is this enough to return it for??? Thanks
Click to expand...
Not enough info.

What ISO?
What shutter speed? (how many seconds/parts of seconds)
was it jpg or raw? if raw, check this out...
http://forums.dpreview.com/forums/read.asp?forum=1019&message=5014150
Click to expand...
 
I have one white pixel .. I noticed while examining an image from a shapening action I was testintig. (a real life image .. not a lens cap test)

After I noticed the pixel, I put the lens cap on to test to see if I got the same dead pixel all the time, and to see if there were any other dead pixels

I tested in manual mode f3.5 1/200 shutter large higest quality JPEG, so that I would see the effects for a typical exposure and file format I typcally use. The pixel is present in images from iso 100 - 3200 but not very notiable at iso 100-400.

To bring out the dead pixel so that I could see it while viewing the image at 1:3 (just to make sure that there was only one dead pixel) I used a laplacian filter. Here is the filter map:
0 1 0
1 -4 1
0 1 0
(scale 1 bias 0)

at iso 100 I had to apply it 3 times to be able to see the pixel at 1:3, at ISO 1600 It was clearly visible after 1 application of the filter

The filter will bring out any dead pixels, and amplyfy noise to create a nice star field.

So my question is -- is one dead pixel acceptable?

If most DSLR's have one dead pixel, I don't think it is worth it to trade it in. But if it is a rarity to have a dead pixel, I think I would consider trading in my camera.

when I make prints I can look for the dead pixel, and set it to the average of its surroundings, or the median of a 5x5 area around it .. that isn't really a problem.

PS: I also did a test of overexposed areas, and didn't find any black pixels.
 
  • I tested at large fimage with high compression, at ISO 100 the dead pixel is not there at all .. even after applying numerous laplacian filters. at iso 400 it was visible after 2 applications of the filter. This may be intersting to some users, so I fugured I would post it here.
I performed this test becase I went back and looked at the first images I took with my camera (april 8) and couldn't find the hot pixel, so I wondered if the hot pixel had been there all the time, whether it had appeared over time, or weather the compression setting I was using (because I only had a 128MB card) was removing the white pixel.

-ryan
Biff.ca said:
I have one white pixel .. I noticed while examining an image from a
shapening action I was testintig. (a real life image .. not a lens
cap test)

After I noticed the pixel, I put the lens cap on to test to see if
I got the same dead pixel all the time, and to see if there were
any other dead pixels

I tested in manual mode f3.5 1/200 shutter large higest quality
JPEG, so that I would see the effects for a typical exposure and
file format I typcally use. The pixel is present in images from
iso 100 - 3200 but not very notiable at iso 100-400.

To bring out the dead pixel so that I could see it while viewing
the image at 1:3 (just to make sure that there was only one dead
pixel) I used a laplacian filter. Here is the filter map:
0 1 0
1 -4 1
0 1 0
(scale 1 bias 0)

at iso 100 I had to apply it 3 times to be able to see the pixel at
1:3, at ISO 1600 It was clearly visible after 1 application of the
filter

The filter will bring out any dead pixels, and amplyfy noise to
create a nice star field.

So my question is -- is one dead pixel acceptable?

If most DSLR's have one dead pixel, I don't think it is worth it to
trade it in. But if it is a rarity to have a dead pixel, I think I
would consider trading in my camera.

when I make prints I can look for the dead pixel, and set it to the
average of its surroundings, or the median of a 5x5 area around it
.. that isn't really a problem.

PS: I also did a test of overexposed areas, and didn't find any
black pixels.
Click to expand...
 
I went back and looked thhrough many images and found that the dead pixel started showing up sometime between IMG_6169 and IMG_6243, 4 days after I received it.

My camera started at 4807 (continued numbering from my memory card)
so the pixel started after 1362 and before 1436 exposures.

-ryan
 
sammmysams said:
iso 100, large jpg, at both fast and long shutter speeds
Click to expand...
Well, don't know what 'fast' and 'long' mean to you, but at ISO 100 my opinion would be that you shouldn't see any hot pixel until about 1/2 sec or slower, and that's being very liberal.

Trouble is, as many have pointed out in other threads, if you trade it in, you may get one that's worse, or has some other problem.

Mine has 4 hot pixels (3 in a clump of red at the top middle) that show up at ISO 800 and above at about 1/60 or slower. At ISO 100, nothing until about 1/2 second. At ISO 100 @ 120 seconds (2 minutes) there are only 7. Since the camera is perfect otherwise, I am biding my time until I see them on normal pictures, or until near the end of the warranty when I'll send it to canon for adjustment.

Quote from my wife: "That little tiny dot is what you're worried about?!" Then she laughs all the way out of the room! :-)

You get no breaks in my house. ;-)
 
Don, i reshot the pic and at iso 100 and 1/2" and raw did not see the red pixel.....does this sound ok to you???
Don Trust said:
sammmysams said:
iso 100, large jpg, at both fast and long shutter speeds
Click to expand...
Well, don't know what 'fast' and 'long' mean to you, but at ISO 100
my opinion would be that you shouldn't see any hot pixel until
about 1/2 sec or slower, and that's being very liberal.

Trouble is, as many have pointed out in other threads, if you trade
it in, you may get one that's worse, or has some other problem.

Mine has 4 hot pixels (3 in a clump of red at the top middle) that
show up at ISO 800 and above at about 1/60 or slower. At ISO 100,
nothing until about 1/2 second. At ISO 100 @ 120 seconds (2
minutes) there are only 7. Since the camera is perfect otherwise,
I am biding my time until I see them on normal pictures, or until
near the end of the warranty when I'll send it to canon for
adjustment.

Quote from my wife: "That little tiny dot is what you're worried
about?!" Then she laughs all the way out of the room! :-)

You get no breaks in my house. ;-)
Click to expand...
 
Don shooting with raw and iso 100, i could not find any dead pixels at 1/2", 1", 2", or 4". I am wondering if jpeg (this is where i found dead pixels) is more prone to seeing dead pixels??? thanks for the help
sammmysams said:
Don Trust said:
sammmysams said:
iso 100, large jpg, at both fast and long shutter speeds
Click to expand...
Well, don't know what 'fast' and 'long' mean to you, but at ISO 100
my opinion would be that you shouldn't see any hot pixel until
about 1/2 sec or slower, and that's being very liberal.

Trouble is, as many have pointed out in other threads, if you trade
it in, you may get one that's worse, or has some other problem.

Mine has 4 hot pixels (3 in a clump of red at the top middle) that
show up at ISO 800 and above at about 1/60 or slower. At ISO 100,
nothing until about 1/2 second. At ISO 100 @ 120 seconds (2
minutes) there are only 7. Since the camera is perfect otherwise,
I am biding my time until I see them on normal pictures, or until
near the end of the warranty when I'll send it to canon for
adjustment.

Quote from my wife: "That little tiny dot is what you're worried
about?!" Then she laughs all the way out of the room! :-)

You get no breaks in my house. ;-)
Click to expand...
Click to expand...
 
Minimal helf from this site...thanks...but i will provide something users might find helpful:

http://www.starzen.com/imaging/deadpixeltest.htm

I did some test shots in raw format at 100 iso abd 1/2, 1, 2, 4 seconds and then save to tiff and ran through this program......no dead pixels......yeah!!!!!

thanks to one and all for all the help!
sammmysams said:
I did a dead pixel test with the cap on in a dark room...found one
small red dot in the middle and another smaller blue next to it and
nothing else...this is a brand new camera....what do the experts
recommend? Is this enough to return it for??? Thanks
Click to expand...
 
sammmysams said:
Don shooting with raw and iso 100, i could not find any dead pixels
at 1/2", 1", 2", or 4". I am wondering if jpeg (this is where i
found dead pixels) is more prone to seeing dead pixels??? thanks
for the help
Click to expand...
I assume you mean hot pixels, since that is how this thread started?? As I understand it, there is a slighly better chance of hot pixels showing up in a jpg file if you have the sharpening setting tuned up a notch or two. Nature of the beast I suppose. In the test you made, it sounds like you're home free. Happy shooting.

Bottom line is,,,, if they don't show up in what are normal shots for you then don't fret it. I've found that with any kind of equipment (photo or otherwise) one can always find something wrong if you look hard enough. In the real world, if it works and doesn't normally cause problems, just enjoy using it.
 
With a six million pixel count, it's not unusual to have a couple hundred that show up under extreme conditions and about twenty or thirty that show up under the circumstances most of us test at. The technology used to produce a six million pixel chip just won't do any better than that. It's just that we never really see all the defects that are really there.
sammmysams said:
http://www.starzen.com/imaging/deadpixeltest.htm

I did some test shots in raw format at 100 iso abd 1/2, 1, 2, 4
seconds and then save to tiff and ran through this program......no
dead pixels......yeah!!!!!

thanks to one and all for all the help!
sammmysams said:
I did a dead pixel test with the cap on in a dark room...found one
small red dot in the middle and another smaller blue next to it and
nothing else...this is a brand new camera....what do the experts
recommend? Is this enough to return it for??? Thanks
Click to expand...
Click to expand...
 
A month ago, when I got my 10D, I had one hot red pixels surrounded by 4 less red pixels made up because of the translation of data from RAW to a real picture, even though it was only one red photosite that was bad. It didn't even show up until about ISO 800 and 1/20 seconds, but I considered this a defect. When I e-mailed Canon Canada about having it mapped out, because that is all I wanted to do as I didn't care to go to the trouble of returning it, they in fact told me that it was a defect and so I got an exchange. Luckily the stores in Canada seem to be better stocked with 10D's.

Anway, to say that to have 20 to 30 show up under extreme conditions is nonsense. Do you mean a 5 minute exposure or 1/60 at a high ISO that was described up above? If you can see it in a shot that you will regurarly take, its a defect. Think about it, if you regurarly shoot at night and need ISO 800 and shoot at only 1/30, will you be wanting to always have to take care of the pixel in photoshop? I think if anything under under 1/2 seconds shows up, it should be dealt with. Either exchange or send it to Canon to have it mapped out. But once you see the hot pixel in an everyday picture and not a lens cap shot, you'll never be able to forget about and it will drive you crazy knowing that its there.

Kiran
Ken N said:
With a six million pixel count, it's not unusual to have a couple
hundred that show up under extreme conditions and about twenty or
thirty that show up under the circumstances most of us test at.
The technology used to produce a six million pixel chip just won't
do any better than that. It's just that we never really see all
the defects that are really there.
sammmysams wrote:
Click to expand...
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top