GPS problem

[dkabat]

Just downloaded some images  GPS tagging reports Error - Failed to excute external command. 

Could someone please tell me how to fix this?

Dan Kabat

[peterkrogh]

Dan,
Are you on Vista?
I've been battling this problem as well. The same files work on XP and Mac, but kick that error on PC.  I have tried changing the programs (both IIP and EXIFtool) to run as administrators, but no luck.

Marc?
Peter

[dkabat]

No Peter - using XP SP3.  Have just downloaded program again and will reinstall to see if that works.

Thanks
Dan Kabat

[dkabat]

Did repair on program and tried to ingestion again.  Same error.  Only thing I can think I am doing differently, is that I am choosing to ingest one day's worth of pictures at a time rather than the whole card.

Dan Kabat

[dkabat]

Just tried to download full card, not partial - same problem.

Using 3.2.04

[MikeBaynes]

I've also been seeing this error message on XP SP3.  In my case the ingestion completed but the GPS data didn't make it into the resulting EXIF.  Marc has been debugging my example files.  See this thread:
http://thedambook.com/smf/index.php?topic=4100.0
   -Mike

[dkabat]

The only system differences I can see is Lightroom 2.4 and updating Second Copy, thses plus the Microsoft security downloads is I think the only changes since my last card download of a month ago when I had no problems.

Dan

[Marc Rochkind]

I'm out of town until Wednesday (July 1), and I'll try to see what I can come up with when I return.

The failure to execute the external command (ExifTool) is a completely different problem from the one I'm working on for Mike, which is a failure of ExifTool to insert the GPS data when it does run. In the past, on Vista, running as admin has always been a fix for the failure to execute problem. This is the first I've heard of it on XP.

--Marc

[ianw]

Marc,

Another GPS query - though not related to these issues.

I took some long exposure images from my car on Sunday night to get images of light-trails.  It's a bit hit and miss - and mostly miss!  After ingestion the log reported that only a small number of images had any GPS data added.  My GPS device is set to record positions every second and my images were all taken with 15 second exposures.  My speed was anything from stationary to about 50mph.

The images when stationary - test images at the start to check exposure and composition from strapping tripod to front seat! - all seem to have got GPS embedded, despite there being a level of 'drift' i.e. I was stationary but the GPS log indicates some movement.  Once I started driving then no GPS was embedded.  I can understand this in a way as which GPS log point do you embed out of the 15 created during the exposure?  The first, the one in the middle or the last?  When you may travel 300 meters during the exposure I can understand why it's not easy to add GPS data.  Unless it's a trade secret can you describe the logic you use to decide whether to embed data or not?

Ian

[Chris Bishop]

I have this problem also -XP SP3. I had actually given up on it. Marc is such a good helpful guy, my previous requests for help have all started out blaming the software, and ended being my fault I daren 't ask again, as I hadn't got into GPS tagging, but want to. I carry the GPS on all my outings anyway.
Chris Bishop

[peterkrogh]

Ian,
I think that there is no secret logic here.  I'm guessing that what is happening is that the tracklog as saved by the GPS is discontinuous, and at the point that the camera tags the time, the tracklog is empty.  I have had to merge tracklogs made by my garmin in order to fill in these points.  I have done this with the windows version of the garmin software.

Peter

[ianw]

Peter,

I've checked my GPS device and it was set to record points every 5 seconds, not every second.  I then checked the tracks in Garmin Mapsource.  There are some breaks in the GPX file but these are due to me driving through tunnels, where reception is understandably lost.  There is also a 'break' at midnight, where one GPX file is closed and another started for the new day.

So the tracks are contiguous.  It's just that during the 15 second exposure there are 3 track points to choose from.  It seems that ImageIngester can't decide which one to use.  I'm not sure why it worked when I was stationary at the begining of my trip as there were still 3 points to choose from.  Maybe these points were at the "same" location - allowing for some GPS drift.  The images taken when moving had points that could have been 100 meters or more apart.  It might be this sort of distance change that is making ImageIngester decide that it can't allocate a GPS position to an image.

It's not the end of the world as I can tag the images manually.  There will only be a couple to do, as most of the images will get binned.

Ian

[peterkrogh]

Ian,
When the tracks are discontinuous, try merging them into a single track and then run again. The intermediate points will be interpolated.
IIP is simply locating the EXIF time on a spread between known points.

So if you have one tracklog made in new york on sunday, and one made in San Francisco on Friday and merge the tracklogs, IIP will locate all the pictures made during the week along a vector between the two cities. All the Tuesday, Wednesday and Thrusday pictures would be spread along a straight line between the coasts.

If the tracklogs were not merged, then there would be no data for the Tuesday, Wednesday and Thursday photos, and IIP would not assign any positions.
Peter

[ianw]

Peter,

I'm not explaining myself very well.

My GPX file is continous enough for most of my images.  I'm at work at the moment so don't have the exact details in front of me so the following values are examples.  If I have continuous track points every 5 seconds at 23:31:05, 23:31:10, 23:31:15, 23:31:20, 23:31:25, 23:31:30 etc and take a picture at 23:31:12 with a 15 second exposure then which track point should it use?  Should it calculate a position based on the start of the exposure i.e. approximately half-way between the points at 23:31:10 and 23:31:15?  Or should it be based on the end position of the exposure i.e. approximately mid-way between the points at 23:31:25 and 23:31:30?  Alternatively should it be based on the position half-way through the exposure i.e. the position at 23:31:19.50?  The distance travelled during the exposure could be up to 300 metres so there's quite a position difference between these options.

I might try this with the latest version of ExifTool when I get home to see what that does.  This can now calculate GPS details when given a GPX file to look at.  Indeed if this works well enough it's functionality that Marc could remove from ImageIngester and rely on ExifTool to do.  IIP would still have to prompt for the correct GPS files, but would no longer have to do the actual calculations.

Ian

[peterkrogh]

IIP uses the EXIF time, so it's up to the camera to create the time at the start of end of the exposure.  I'm not sure this is even convered by the EXIF spec, so it may be done differently by different cameras.
Peter