All Forum Comments
Unlocker wrote:
Have to strongly disagree there. A dead pixel is a manufacturing defect pure and simple. The Sale of Goods act 1974 covers this. Just because a certain amount of dead pixels are deemed OK by a manufacturer does not mean they should be accepted by you.
I can see that working with a display but with a DSLR how are you going to know if they replaced the sensor or just applied pixel mapping?Have to strongly disagree there. A dead pixel is a manufacturing defect pure and simple. The Sale of Goods act 1974 covers this. Just because a certain amount of dead pixels are deemed OK by a manufacturer does not mean they should be accepted by you.
Glyn
Ratcatcher wrote:
When you say over 4 months would it of still been within the warranty period.
Not confirmed by me but I would think that a single dead pixel on a k20D is not a warranty issue. With things like display monitors and TVs there is usually a minimum number of dead pixels that have to be exceeded before a warranty claim is accepted. When you say over 4 months would it of still been within the warranty period.
In the case of the K20D the pixel mapping function is provided to correct this fault and once applied it appears to be impossible to detect. With earlier Pentax DSLRs I believe that that warranty returns with dead pixels were addressed with pixel mapping at a service level.
Glyn
It took over 4 months and over 2500 shots before I spotted a dead pixel on my K20D and I then went back and checked the first shots taken with the camera and found it there too. Once I had seen the dead pixel it got really annoying. I could see it on most of my shots. However pixel mapping sorted it out fine.
I suspect that it is being spread to ajacent pixels by the jpeg compression but once found a dead pixel is easy to spot at 100%. Here is a 100% crop of the dead pixel, albeit in a dark part of a photo:
[IMG]http://i172.photobucket.com/albums/w20/GlynM/Misc/IMGP2574-dead-pixel-100_crop.jpg[/IMG]
Certainly in a jpeg a dead pixel has a characteristic shape of a cross. I found another on my K-7 which looked the same.
Glyn
I suspect that it is being spread to ajacent pixels by the jpeg compression but once found a dead pixel is easy to spot at 100%. Here is a 100% crop of the dead pixel, albeit in a dark part of a photo:
[IMG]http://i172.photobucket.com/albums/w20/GlynM/Misc/IMGP2574-dead-pixel-100_crop.jpg[/IMG]
Certainly in a jpeg a dead pixel has a characteristic shape of a cross. I found another on my K-7 which looked the same.
Glyn
lemmy wrote:
A bit of a mess really! ...
It is indeed and explains why the Pentax digital camera utility is labelling my Sigma 18-125mm as a Pentax F 28-80mm F3.5-4.5.A bit of a mess really! ...
Glyn
Comment by GlynM posted on ExifTool Tip at 01/09/2009 - 11:40
lemmy wrote:
That's a brillant and clear exposition - thank you.
Many thanks That's a brillant and clear exposition - thank you.
Looking at the Phil Harvey's web page for exif tool he does present the list of codes used for lens types and the text name he uses in the look-up table. His list for Pentax cameras can be found here
I think it is interesting to see that the last entry is:
'8 255' = Sigma 70-200mm F2.8 EX DG Macro HSM II
255, being a full 8 bits, looks suspicious and I wonder if it is also being used as a catch-all for some unknown Sigma values in a similar way to '3 255'.
I hope this helps,
Glyn
Comment by GlynM posted on ExifTool Tip at 01/09/2009 - 10:07
lemmy wrote:
... Does anyone know where the exif information comes from if the same lens is described differently in different programs? The lens must transmit some ID to the camera for the camera to know what it is at all.

If you look at the tag-id you will see that all the lens information is in a single tag, 0207. If you click on "Binary Data" at the top of the tag you will see what is contained in the tag. There is no text in here and I suspect the lens name is actually a numerical value and the application has to have a look-up table to convert these into the text that you see. I think many of the updates to these tools are to correct or update these look-up tables with new lenses and can be the cause of differences between applications and application versions.... Does anyone know where the exif information comes from if the same lens is described differently in different programs? The lens must transmit some ID to the camera for the camera to know what it is at all.

So in summary, the binary data is what is reported by the camera and everything below is interpretation by the application and is subject to the accuracy and updates of the application.
Glyn
Edited to add summary
Comment by GlynM posted on ExifTool Tip at 31/08/2009 - 22:25
I have been using Phil Harvey's exiftool for a number of years, certaintaly from before the stand-alone windows version when you had to compile the pearl source files yourself.
You do not have to be a command line guru to use it. I have put it on my desk top renamed as:
exiftool(-k -a -u -g1 -w txt).exe
I then just drag and drop photos onto it and it creates a text file with the same name and in the same place as the original file.
While I now use PhotoMe as my first choice because of its GUI I think these tools all apply some interpretation so a 2nd opinion is sometimes useful.
The text files from exiftool also create a record of the exif data at particular time and interpretation that I still often find useful to keep.
Glyn
You do not have to be a command line guru to use it. I have put it on my desk top renamed as:
exiftool(-k -a -u -g1 -w txt).exe
I then just drag and drop photos onto it and it creates a text file with the same name and in the same place as the original file.
While I now use PhotoMe as my first choice because of its GUI I think these tools all apply some interpretation so a 2nd opinion is sometimes useful.
The text files from exiftool also create a record of the exif data at particular time and interpretation that I still often find useful to keep.
Glyn
Comment by GlynM posted on ExifTool Tip at 25/08/2009 - 21:25
I have had my DA55-300 for over a year and have not had any issues with zoom creep.
My experience is that locking switches on lenses only engage when the lens is at its shortest length so are ok for transporting it around but don't help at all to keep the lens at any other length when it is being used.
The 2 worst lenses that I have for creep are a rather old, and well used, FA28-200 and a DA18-250.
Glyn
My experience is that locking switches on lenses only engage when the lens is at its shortest length so are ok for transporting it around but don't help at all to keep the lens at any other length when it is being used.
The 2 worst lenses that I have for creep are a rather old, and well used, FA28-200 and a DA18-250.
Glyn
Comment by GlynM posted on Zoom creep at 06/08/2009 - 20:30
Tubbymorton wrote:
...
It suddenly worked!
...
Don't worry it is so easy to get caught out with these complex devices. You are definitely not alone and this forum is here to help ...
It suddenly worked!
...
On the K-7 the ok button is used to toggle between using the 4-way buttons to control the AF selection or for selecting the settings that used to be found under a separate function button on the K10 & K20.
Initially I found it a bit confusing and often could not workout why I could not move the AF point, and a couple times why some of the function settings seemed to have changed, all because I had forgotten to press the ok button. I soon learnt to check for the presence of the little 4 way icon in the view finder when I had enabled AF selection and if necessary press the ok button the make it appear.
Fortunately the new firmware lets you set the default function and I found it very useful to make it default to AF sel.
Glyn
Comment by GlynM posted on Sigma lens incompatibility with K7 at 06/08/2009 - 16:55


Don't do this unless you are prepared to be disappointed but I have found the best way to check for hot pixels is to turn off slow speed shutter noise reduction (custom 17 on the K20D or 18 on the K-7), set a low fixed ISO, say 100, (to avoid high ISO noise reduction), set the camera to manual focus and take a couple of shots with the lens cap fitted on the lens. Inspect the result at 100% magnification or better. If you find anything then check if it is in both shots and then also check the same point on a recent normal shot.
But please don't blame me if you don't like what you find
Glyn
Edit - Just realised that you can't turn off slow shutter noise reduction on a K20D, only set it to auto or on, so this test only works best on a K-7 although I can see at least one hot pixel on my K20D doing this test with slow shutter noise reduction set to auto - Apologies, Glyn