Exif time stamp (windows) different from as seen in camera (D800e)

[debejyo]

<p>Hi Folks,<br /> I am reading the time stamp of an image on my camera, and comparing the time stamp in exif (windows right click properties). The time stamps differ by about 10s. The camera is obviously accurate (I verified that by clicking an image at an exact time). How does the camera display the correct time but my computer cannot? How can I get to the correct time on my computer?<br /> Thanks.<br>

Edit: I noticed that reading the exif in Photoshop (after saving as TIFF from NEF) shows the correct time in File>File Info. </p>

[Andrew Garrard]

<p>Hi Debejyo. Well, that's weird. Just checking: when you view in Windows, you <i>are</i> looking under the details, "date taken" field, which would come from EXIF? If you're looking at the file modified date, you'd probably get the time the file finished writing to the card, whereas the EXIF would be the time the shutter was released. I'd be perfectly prepared to believe a 10s difference, especially if there's something in the queue and you have a slowish memory card.<br />

<br />

Just a thought. Otherwise, I've not really noticed this problem, but I can't say I've looked for it.</p>

[debejyo]

<p>Andrew, take a look at the attached image. Left side is from photoshop and right side is from windows explorer.</p>

[debejyo]

<p>Ok, here is the photo.</p><div></div>

[Andrew Garrard]

<p>Thanks, Debejyo. So... I'm guessing the image was taken at 3:59:19am, as seen in the EXIF in your viewer. Windows may or may not be able to show this figure for a JPEG in the file details dialogue, but I doubt it knows about a NEF file. The file took, I guess, 11s to finish writing to card storage on the camera, which is the file modification date (3:59:30am). It occurs to me that this may also be the difference between the start of an exposure and the end of one, depending on what you were shooting. Checking my images, I don't really see a difference, but they were short exposures and fast cards.<br />

<br />

The file creation date on Windows is probably the point at which you copied the file from the card - hence a bit later, and why the creation time is later than the modification time (it was <i>modified</i> when originally written, but the creation is for the specific file location). And I'm guessing the creation date for your viewer is substantially later because it refers to the TIF conversion rather than the original file.<br />

<br />

I'd vaguely assert that this behaviour is logical, but it could also be a completely misleading explanation and nothing to do with what you're seeing! Still, I hope the (mildly) educated guess helps.</p>

[ShunCheung]

<p>First of all, I am merely speculating. I think the 10-second difference could be between the time the image was captured and the time the image file was completely written onto the memory card. The latter should be the creation time of the image file.</p>

<p>For example, if you are capturing sports or wildlife "machine gun" style, your DSLR could be buffering those images for several seconds as they are gradually written onto the memory card(s). When you have many images in the buffer, a delay of several seconds is actually common, especially if you are using some older, slow memory card.</p>

<p>Try capturing just one single image and check whether there is any time difference, or the difference is tiny. There may actually be several slightly different time stamps, but they may appear to be the same since in most cases, they are very close.</p>

[ShunCheung]

<p>Sorry Andrew, you and I are giving the same explanation.</p>

[debejyo]

<p>Thanks Andrew and Shun. That all makes sense. I was shooting on a continuous mode, so it was buffering. My exposures were short (1/60 or faster) and my slowest card was 95 Mbps. Anyway, I'll ignore what windows says. I'll just extract the exif information and rely on it. Now at least the truth is verified.<br /> If it is relevant to anyone reading this, I just discovered that the exif also contians a subsecond time, to the 1/100 ths of second, but it appears that D800e records it to the nearest 1/10 ths of a second.</p>

[thirteenthumbs]

<p>Check your computer clock against http://nist.time.gov/ . There is a setting to synchronize your computer with this service also. NIST = National Institute of Standards and Technology. <br>

I use it to set all my clocks and equipment manually if it does not have an auto sync feature. </p>

[pete_s.]

<p>There are software that can change the date and time of when the file was created to the the date and time in the exif info.</p>

<p>Exiftool is the ultimate in exif manipulation and extraction but it's hard to use for non-computer experts. </p>

 

[marc_rochkind]

<p>Just to help clarify the situation, most images have up to 6 date/times. Inside the image, there is a TIFF date/time, and two EXIF date/times, one for date/time original and one for date/time digitized. Then, the file that contains the image has date/time last accessed and date/time last modified. There might be a date/time of creation or date/time of last change to the file's properties (not contents), depending on the OS. I doubt that it's documented which one Windows Explorer is showing. So, as a few people above noted, it's not unusual for these date/times to differ by a bit.</p>

<p>Because files move around, get copied, etc., processing software should use the internal date/times, not anything reported by the file system.</p>

[heimbrandt]

<p>Exif Pilot is free and easy to use:<br>

http://download.cnet.com/Exif-Pilot/3000-2192_4-10779393.html</p>