1d MarkIII - Determining shutter count

canonloader wrote in post #4808860
And a few years ago, a guy said he had found cold fusion, showed a glass of water with some bubbles in it and demanded everyone believe him. Did you? Did anyone? Of course not. They may have wanted to, free power would change the world, but they demanded proof, tell us how so we can duplicate it, they said. PPppfffftttt, and we all know the rest of the story.

All I asked for was a Mk3 RAW or jpg with the exif intact, so I could try a few things, prove or disprove it myself. Incredible.

Mitch, i will send you a raw file from a mkIII when i get home tonight. just PM me your email address.

Plenty of MKIII JPGs with intact EXIF scattered throughout, like my ISO registry thread.
https://photography-on-the.net/forum/showthre​ad.php?t=430454

Download this file and tell me how many actuations are on the body with CS2/CS3

http://www.jasonhollis​ter.com/tech/count.tif

Now open it with 1Dcount.

I know that this body has more than what shows in Photoshop. You cannot reliably use Photoshop to count your actuations. If you owned your Mk.III from new and are CERTAIN you have not shot more than 65535, then it's most likely accurate.

Buying used? Shot LOTS of 10fps frames and think youor over 50k? Then no, Photoshop is not going to help you get an accurate number.

Take the number PS gives you and add 65535. If you have shot over 65535 say by 12986 for a total of 78521, then the Exif will show 12986. It misses the 65535. Now if you have bought used how can you be certain that the "odometer" has only rolled over once? 1dCount will show the full 78521. This of course excludes Mk.III bodies. Currently, after 65535 you are screwed. It rolls over to zero with no way to know unless Canon says so.

Last time, you cannot reliably use Photoshop to count your actuations.

Enough?

CS2 says 16680, 1Dcount says there's nothing there, after I converted it to a jpg. It doesn't see a tif at all.

cosworth wrote in post #4807274
You're kidding me right?

1Dcount/Cancount counts past 65535. The numbers from Photoshop do not. any many have seen that under 65535 Photoshop is still incorrect. Never use Photoshop to determine shutter count. Photoshop also cannot reliably read any Mk.III files.

I 2nd this..

Can you please check if this shows a rough shutter count through CS2/CS3?
JPG with exif:
http://www.warrenwilli​ams.co.nz/2008/JAN_007​8.jpg
I'm guessing approx 25,000 shots. 7 month old body

Well, so far, none of these images have shown the same two info layout. I was trying to find it using a hex editor, but can't find the placeholders for the shutter counts on MkIII images at all. This last one is showing 24576/65536, but it's not for image count.

Yup, several of us have gone the hex editor route as well... including the author of the free 1D shutter counter posted on a different thread. But keep testing, it may be in that exif soup somewhere!

It would almost have to be, since Canon seems to be able to find it. There's a lot of data in these things though. Good stuff to do when I'm snowed in.

It doesn't have to be, there are other items stored in the camera that are not projected into the EXIF data. If you could reverse-engineer the communications to the camera, you may be able to figure out how to request things that are stored in the camera not part of the photos. There is no need for Canon to store shutter count and other maintenance related items in the EXIF. Better yet, if a Canon insider could get a copy of the technician's program they use to communicate with the cameras, then share it!

It is not possible to ascertain the shutter actuations on a MK3 body. Only Canon can pass on this info.

The 1D classic puts the count in the EXIF, after that model, it seems it got added to the hex data of the image itself and not recoverable with EXIF readers, which is also where the EXIF data is. Afterall, Canon wants a test image to tell you what it is, right?

It's also possible it's stored on one of the other chips inside that only Canon know how to access, along with in the hex data of the image.

There is a service by Canon where you send them a raw or jpg, and they tell you from the photo what your shutter count is? The only thing I have read is that Canon, when they have your camera, can tell you what your actuations is. If Canon can tell us from just a file, then it has to be in the EXIF somewhere, but I have never heard of this.

canonloader wrote in post #4814793
It's also possible it's stored on one of the other chips inside.

Let's hope your digging proves this theory wrong!

Would Canon be evil enough to store the shutter count across several, noncontiguous bytes?