NOTICE: is no longer in use from 1st March 2024. Please use the OpenStreetMap Community Forum

I was checking consumer tests of cameras and I see that a compact camera ("superzoom" type) offers embedded GPS.

GPS may be useful for surveying and for WikiLovesMonuments photos, but the camera costs some 50 € more than equivalent ones. Are the coordinates found by such embedded systems precise enough to be used directly?

I saw the POV camera question but I'm not interested in video etc.

This question is marked "community wiki".

asked 21 Apr '15, 08:55

Nemo_bis's gravatar image

accept rate: 0%

edited 21 Apr '15, 18:34


This sounds more like an invitation to a discussion (and information that will go out of date rapidly) - maybe a mailing list might be a better place for it?

(21 Apr '15, 09:59) SomeoneElse ♦

I don't think the technology of GPS and cameras is evolving that rapidly. The topic is both too specific for a mailing list (where it would soon get lost in the archives) and too open-ended for a wiki page (where it would be hard to account for personal experiences). I have experience with the alternatives and I've chosen the Q&A site because I think it's the best place.

(21 Apr '15, 18:39) Nemo_bis

Given that this depends

  1. on the camera in question
  2. on your definition of "precise"

it is not possible to give an answer other than: given a good fix and clear skies the coordinates are likely to roughly describe where you took the photograph (not where the object that you photographed is).

permanent link

answered 21 Apr '15, 11:43

SimonPoole's gravatar image

SimonPoole ♦
accept rate: 18%

These cameras likely use a standard GPS chipset but the results will vary. My experience with one was that the accuracy was comparable to a handheld GPS when it got a fix. But the time to obtain a fix from power on would be up to 15 minutes. Also when taking a series of photos while walking around with the camera GPS enabled, the update rate was unacceptable; a photo might be stamped with the previous location even though there was a movement of several meters.

permanent link

answered 21 Apr '15, 12:17

Mike%20N's gravatar image

Mike N
accept rate: 17%

Thanks. Just to check, as "correct" position you were expecting the position where you shot the photo, not the position of the photographed object, right? Sometimes the coordinates are technically correct but don't make any sense because the object was very distant from the camera. Sorry if the question is silly; in my experience in Wikimedia Commons many insert wrong coordinates for this reason. :)

(21 Apr '15, 18:44) Nemo_bis

My experience with two (very similar) cameras is much the same as Mike N: Slow to first fix. Slow update rate. I also noted that they usually only see a relatively small number of satellies (never seen better than 5) - implying a poorer fix than a dedicated GPS that is able to process more satellite signals (my phone rarely sees fewer than 7, often 10 or 12, including Glonass).

permanent link

answered 21 Apr '15, 13:33

G3YAC's gravatar image

accept rate: 20%

Thanks. Do you think it's possible that cameras are designed in a way that makes it harder for them to "see" the sky? I mean, phones are often designed to have the antenna have good signal, while I imagine the GPS could be anywhere in the middle of the camera. I wonder if there's a way to check such stuff before purchase.

(21 Apr '15, 18:42) Nemo_bis

If you have a gps or a smart phone just set the cameras clock to match gps time and your gpx trace can be used to fix the position your jpegs exactly. JOSM and Garmin Basecamp can do this along with several other programs. See

permanent link

answered 21 Apr '15, 22:38

andy%20mackey's gravatar image

andy mackey
accept rate: 4%

It is 3 years after the post. Here is an update from my experience of the latest and greatest cameras. They are NOT accurate even if your speed is simply walking.

I tried out the Sony action cam x3000 and a go pro 7

permanent link

answered 25 Oct '18, 18:58

Mapping%20the%20world's gravatar image

Mapping the ...
accept rate: 0%

could you please try to define what you mean by "NOT accurate"? e.g. is it possible to capture a parking lot with several aisles with the cam's gps?

(25 Oct '18, 19:08) aseerel4c26 ♦

Hi Aseerel4c26, If you could help me with a question that I am about to post, then I can give you an accurate answer. But my instinct suggests that there would be no way to capture parking lot aisles with the accuracy.

My Latitude is at 43.0 degrees, longitude 85.67 I am trying to figure out how many feet do I move for every .00100 movement on latitude and how many feet do I move for every .00100 longitude. If I can get the answer to this question, then I can tell you how many feet I am off. (There is a website that gives me 99% of the information, but they cut off the decimals too early.)

As for my data points while mapping with the Go Pro 7 vs. a standing test of my Huawei Tablet, they are as follows.

Huawei Tablet                 GoPro7:                       
Long           Lat            Long
42.96277000  -85.6452141      42.962500     -85.64500
42.96275150  -85.6453750      42.962500     -85.6452778

As you can see, the Longitude of the GoPro was exactly the same for the data. (it repeated this for 100 meters, meaning I would have passed any drunk test perfectly, yet its highly unlikely I walked a perfect straight line)

(25 Oct '18, 22:56) Mapping the ...

I found a longitude/latitude calculator that uses DD.
For myself, .0001 latitude changes about 36 feet. I had a difference of .00027 difference of the most accurate reading and the Gopro 7 reading. It was close to 100 feet off while taking a reading every one second and walking at a speed of around 12 feet per second.

(26 Oct '18, 00:05) Mapping the ...

Thanks! Yes, that coordinate chopping is unusable for most mapping activities.

From 43.000, -85.670 to 43.001, -85.671 it is 111.32 m to the North and 81.41 m to the West or 137.91 m diagonally. See this JOSM screenshot. The default web editor iD seems not to be able to display coordinates, sorry. The alternative web editor Potlatch2 can show mouse pointer lat long, which helps, too.

Just post your question as a new question entry here on this site if you are interested in converters/tools.

(26 Oct '18, 07:40) aseerel4c26 ♦

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text]( "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:


question asked: 21 Apr '15, 08:55

question was seen: 6,886 times

last updated: 26 Oct '18, 07:40

NOTICE: is no longer in use from 1st March 2024. Please use the OpenStreetMap Community Forum