Hello from Houston, Texas. Phil sent me an email but I still have some questions, hopefully there is something that I missed.
My camera: Nikon D200 with a GP-1 (GPS) unit. EXIFTool 11.77
I am taking photos of satellites as they go by overhead and need an accurate time of shutter opening (exposure 15 seconds) to the tenth of a second. So I got the Nikon which sets time by GPS - but the GPS time stamp is only to the second! Grrrr.
[File:System] FileModifyDate : 2019:12:02 19:18:28-06:00
[File:System] FileAccessDate : 2019:12:03 10:19:18-06:00
[File:System] FileInodeChangeDate : 2019:12:02 20:35:10-06:00
[EXIF:IFD0] ModifyDate : 2019:12:02 19:18:27
[EXIF:ExifIFD] DateTimeOriginal : 2019:12:02 19:18:27
[EXIF:ExifIFD] CreateDate : 2019:12:02 19:18:27
[EXIF:ExifIFD] SubSecTime : 16
[EXIF:ExifIFD] SubSecTimeOriginal : 16
[EXIF:ExifIFD] SubSecTimeDigitized : 16
[EXIF:GPS] GPSTimeStamp : 01:17:56
[EXIF:GPS] GPSDateStamp : 2019:12:03
[Composite] SubSecCreateDate : 2019:12:02 19:18:27.16
[Composite] SubSecDateTimeOriginal : 2019:12:02 19:18:27.16
[Composite] SubSecModifyDate : 2019:12:02 19:18:27.16
[Composite] GPSDateTime : 2019:12:03 01:17:56Z
I was hoping that the camera would get the internal time from the GPS and would record and display time to the tenth of a second - but that does not appear to be the case. I don't see a way that I could interpolate from the internal camera time.
Is it possible to set the camera time from my computer or from some other source? I have the Nikon D200 manual and it doesn't say anything about that - though the GP-1 can output time to a computer via another port.
Have I missed something?
https://sites.google.com/site/opticalsatellitetracking/
Thanks.
These questions are more appropriate for a Nikon forum I think.
But generally, GPS receivers update the position once per second, and store in integer timestamp corresponding to the position at the start of the second. So even if your camera time is more accurate, I bet your GPS fix is accurate only to the nearest second. It is very unlikely that the GP-1 interpolates between fixes. (But I could be wrong -- I'm not a Nikon expert.)
An alternative would be to log all fixes (eg. in a gpx file) and use some utility (eg. ExifTool) to interpolate the fix based on the camera timestamp. This requires calibrating the camera clock carefully, but is fairly easy to do.
- Phil
I'd forget the gps.
Assuming the time digitized is at the end of the 15 second exposure, looking at the time stamps, datetimeoriginal is the same.
Take a photograph of a clock showing 10th of seconds. You can the compare the time in the picture to the time in the metadata and get a "fiddle factor" for subsequent use.
Take your long exposure of the satellite. I suspect that (Datetimeoriginal + subsectimeoriginal - shutterspeed - fiddlefactor) will give you the closest you'll get to shutter button press.
Unfortuately, looking at a long exposure on a D80, the shutter speed appears to be only accurate to the second. The D200 is probably the same.
I have asked on the Digital Photography forum and they said it was more appropriate on the EXIFTool forum!
Let me look into calibrating the camera clock.
Its hard to find people who really understand how things work.
Charles
Quote from: Phil Harvey on December 03, 2019, 12:09:05 PM
These questions are more appropriate for a Nikon forum I think.
But generally, GPS receivers update the position once per second, and store in integer timestamp corresponding to the position at the start of the second. So even if your camera time is more accurate, I bet your GPS fix is accurate only to the nearest second. It is very unlikely that the GP-1 interpolates between fixes. (But I could be wrong -- I'm not a Nikon expert.)
An alternative would be to log all fixes (eg. in a gpx file) and use some utility (eg. ExifTool) to interpolate the fix based on the camera timestamp. This requires calibrating the camera clock carefully, but is fairly easy to do.
- Phil
This is a good idea, let me try to get a good fiddle factor.
It would seem that a camera that "can" control shutter speed to down to 1/1000 of a second would have a very good control over that shutter - could tell within a tenth of a second when it was released. Of course it is not reasonable to expect that even Nikon would have that accurate a shutter unless they were tuned or calibrated somehow, many of these cameras are made. I would be happy to get the shutter open (which is the start of the trail on the image) within a tenth of a second.
Charles
Quote from: Alan Clifford on December 03, 2019, 01:35:43 PM
I'd forget the gps.
Assuming the time digitized is at the end of the 15 second exposure, looking at the time stamps, datetimeoriginal is the same.
Take a photograph of a clock showing 10th of seconds. You can the compare the time in the picture to the time in the metadata and get a "fiddle factor" for subsequent use.
Take your long exposure of the satellite. I suspect that (Datetimeoriginal + subsectimeoriginal - shutterspeed - fiddlefactor) will give you the closest you'll get to shutter button press.
Unfortunately, looking at a long exposure on a D80, the shutter speed appears to be only accurate to the second. The D200 is probably the same.
Phil - can I get a contact for a Nikon expert? Do you have a person who gives you Nikon information?
No, sorry. I don't have any contacts like that.
- Phil
Good luck with Nikon; my experience with them has always been horrible.
You could try one of the forums at dpreview, https://www.dpreview.com/forums/1039 for example.
Alan -
Thanks! I have posted several notes on DP Review and have gotten some good help from people. It appears to be hard to talk to Nikon directly.
From a satellite tracking site - apparently you can get drivers for a laptop (etc) that will directly control the camera. I am gonna try that and see if my MacBook Pro can control the camera - and record shutter release times a bit more accurately.
Charles
Quote from: Alan Clifford on December 05, 2019, 10:40:28 AM
Good luck with Nikon; my experience with them has always been horrible.
You could try one of the forums at dpreview, https://www.dpreview.com/forums/1039 for example.