How do I decipher distance data in EXIF? For editing I use PSE. Exif data is found under File -> File info. A while ago I was looking through all of that, and under "Schema" found...

Morning Star Loc: West coast, North of the 49th N.

For editing I use PSE.
Exif data is found under File -> File info.
A while ago I was looking through all of that, and under "Schema" found "aux:ApproximateFocusDistance:4294967295/1" for a photo of a small forest fire next to Mount Robson.
For a photo of Mount Robson itself, the numbers were "20/100" and for one of my granddaughter on the couch, "320/100"

Could someone please explain how I use these numbers to figure out the appr distance from camera to subject? As they are written they make absolutely no sense at all, exept for the last one. Assuming metres, 320/100 would give my 3.2 metres (about 3 1/2 yards) - which is about right. But what about 20/100 for a mountain or that ten-digit number for the next mountain?

Rongnongno Loc: FL

For EXIF explanation you should resort to google...

Morning Star Loc: West coast, North of the 49th N.

Rongnongno wrote:

For EXIF explanation you should resort to google...

Hey Ron, you should know by now that I like to look up things and I also know that Google can be my friend. In this case though, it wasn't. I spent over an hour following links that sounded like they would have the answer to my question, none did. Even on the Adobe website I got nowhere.
Because the exif pop-up says "Powered by XMP" I looked up that as well, no joy.
So, I placed my question here, hoping that someone knows how to interpret these numbers that seem to make no sense...

Rongnongno Loc: FL

Let me check if I find something then.

Rongnongno Loc: FL

http://www.exiv2.org/tags-xmp-exif.html

now using a more pointed search...

http://www.awaresystems.be/imaging/tiff/tifftags/privateifd/exif/subjectdistancerange.html

I believe it is in there.

http://www.cipa.jp/std/documents/e/DC-008-2012_E.pdf

Depending on the 'short number' it seems that you can determine if the number is in Kilometers. Miles or nautical miles.

Which makes no sense unless dealing with microns!!!

Morning Star Loc: West coast, North of the 49th N.

Rongnongno wrote:

http://www.exiv2.org/tags-xmp-exif.html

now using a more pointed search...

http://www.awaresystems.be/imaging/tiff/tifftags/privateifd/exif/subjectdistancerange.html

The second link is no help at all. The first link claims "Subject Distance - Rational - XmpText - Internal - EXIF tag 37382, 0x9206. Distance to subject, in meters."
OK, that would make sense for the photo of my granddaughter on the couch, 320/100, would be 3.2 metres, and yes, that was about the distance I stood away from her.
But what about the other two numbers? 20/100 would then be 1/5 of a metre, and 4294967295/1 4.29 million km. Neither even approaches reality, there must be another explanation.
I'll keep on searching....

Edit:
Didn't see your third link till I posted this. That document says distance is in metres and gives 20/10 as an example. But that still doesn't explain the other numbers.

Morning Star Loc: West coast, North of the 49th N.

Morning Star wrote:

A while ago I was looking through all of that, and under "Schema" found "aux:ApproximateFocusDistance:4294967295/1" for a photo of a small forest fire next to Mount Robson.
For a photo of Mount Robson itself, the numbers were "20/100" and for one of my granddaughter on the couch, "320/100"

...snip...

Now here's a clue: On the 4/3 forum, I found a message quoting the exact same long number above! Googling that number does give results, but all dealing with subjects I know nothing about. I'll take it as some sort of error message, not a distance message. The photo it belongs to, is haze because of the smoke, may have something to do with it.

Rongnongno Loc: FL

Morning Star wrote:

Now here's a clue: On the 4/3 forum, I found a message quoting the exact same long number above! Googling that number does give results, but all dealing with subjects I know nothing about. I'll take it as some sort of error message, not a distance message. The photo it belongs to, is haze because of the smoke, may have something to do with it.

Still looking, I will post again ONLY if I find something of interest.

Apaflo Loc: Anchorage, Alaska

Morning Star wrote:

The second link is no help at all. The first link ... (show quote)

The type of values are defined by the Exif Version 2.2 standard. You can probably find it by searching on "JEITA CP-3451". Whatever, it says that Exif tag 37382 is a type "RATIONAL", which consist of two LONG integers (which means two unsigned 32 bit values). The value is the first LONG divided by the second LONG. Perhaps not particularly useful, but it does let you know that what you have been assuming is correct.

The second bit of information you need is that many lenses do not provide anything like precision distance information. Not even close. There are some lenses that have only half a dozen or so values, hence the granularity might be pretty great. (And a popular lens that has only 1 or 2 values for distances between 100 meters and infinity might well result in the exact same value being found on many many images.)

Your best bet might be to download the Exiftool package. It gives a figure in meters, and doesn't require that you do the math. But it isn't any more accurate than the lens providing the information.

Morning Star Loc: West coast, North of the 49th N.

Apaflo wrote:

The type of values are defined by the Exif Version... (show quote)

OK, I just pulled up a photo of my favourite chickadee on the bird feeder.
Distance says 585/100 - in metres: 5.85m. Whipped out my dressmaker's measuring tape, and measured 5.81m.
So, these "smaller" numbers then, would be centimetres divided by 100 equals metres.
The ridiculously large number, I will treat as some sort of error, or the camera simply telling me: Sorry, but I can't read that distance.
Leaves the 20/100 - Mount Robson certainly was further away than 20cm.
Maybe it should also be treated as some sort of error.
In any case, I already know more than I did last night ;-)
Rongnongno and Apaflo, thank you both for your input!!
If I ever find an easy to understand explanation, especially of the "strange" numbers, I'll post again.

Searcher Loc: Kent, England

The long number, 4294967295/1, converts to hexadecimal FFFFFFFF which is infinity in the Exif distance calculations.

For distances over 3 metres, just divide by the 2nd number (as you did), for distances under 3 metres you may get a completely false reading.

If you have macro mode switched on, the numbers may also change.

I read as much as I could from the links given, but stopped when I nodded off . . .

Searcher Loc: Kent, England

If you download (free) Picture Information Extractor from http://www.picmeta.com/products/picture-information-extractor.htm

an enormous amount of Exif data is available when the camera and lens supports it.

The program opens in the usual way with an inbuilt browser, click an image and the data shows up instantly. This includes focus distance which shows in metres.

The Exifdata tool is a more comprehensive but is complicated to use.

rehess Loc: South Bend, Indiana, USA

Apaflo wrote:

The type of values are defined by the Exif Version... (show quote)

I believe, but don't understand, why the lens is providing this information. Why isn't the body the source of distance data and focusing activity? Is this true of MILC as well as DSLR?

jefren Loc: Montgomery, Alabama

To provide a guess to Apaflo's question, I suppose that the camera knows whether the image is in focus, but knows where the lens is focused only if the lens communicates that information and the camera records it.

DWU2 Loc: Phoenix Arizona area

Searcher wrote:

The long number, 4294967295/1, converts to hexadecimal FFFFFFFF which is infinity in the Exif distance calculations.

For distances over 3 metres, just divide by the 2nd number (as you did), for distances under 3 metres you may get a completely false reading.

If you have macro mode switched on, the numbers may also change.

I read as much as I could from the links given, but stopped when I nodded off . . .

In addition to what Searcher has said, I would note that the number 4294967295 is 2^32-1, or 2 to the 32nd power minus 1. This is, I presume, the largest number the software could report. Hence, infinity.