topic Re: GPS Accuracy in EOS DSLR & Mirrorless Cameras

GPS Accuracy

rcolman — Fri, 23 Jan 2015 16:59:53 GMT

GPS in a dSLR is critically important to me. I photograph archaeological sites and location is really important. In fact, I bought a 6D just because of the built-in GPS.

I am not wild, however, about the accuracy. I am usually outdoors with a clear view of the sky, but still see that points wander about when they should be at a single point.

I wonder if the GP-E2 is more accurate and REPEATABLE than the built-in 6D GPS?

Rick

Re: GPS Accuracy

Skirball — Fri, 23 Jan 2015 21:44:36 GMT

I don't know much about the GPS on the 6D. I own one, but never turned it on. But, just as a general comment - if accuracy is important to you, I'd really look into a separate GPS unit. I'd guess the one in the 6D (and 70D, 7D2) is more a marketing feature than anything else.

I have a GPS for my bike, and a watch for running, and the accuracy isn't terrible good in that either. I upload it to a website that then compares it to a map, and fits it to roads and trails as best it can. But before that process there are certainly outliers.

My phone is better, because it triangulates with the cell towers. But even then, sometimes Google maps thinks I'm driving 50 feet off the side of the freeway.

Again, I don't know too much about the specs of GPS, but I'm guessing that if you really need accuracy that you'll need more than a consumer GPS.

Re: GPS Accuracy

TCampbell — Fri, 23 Jan 2015 22:20:34 GMT

The GP-E2 is more accurate. It is more sensitive and a bit more accurate than the GPS on the 6D and is also better at getting a fix in poor signal areas where the 6D may not be able to get a fix. Also, it includes an electronic compass -- so in addition to recording latitude/longitude/altitude, it also reccords the magnetic direction the camera was facing when you took the shot (the 6D doesn't have a compass). Lastly, it's also weather-sealed.

As I look at the map of photos I've taken with this GPS and then compare to where I know I was standing, the position fix is very close (within a few meters - which is about as good as typical GPS can get.)

KEEP IN MIND, however, that you get to set the update interval. For battery saving purposes, you can configure the GPS to update every 1 sec, 5, 10, 15, 30, 1min, 2min, 5min intervals (you can save battery life by putting more time between intervals and also the GP-E2 has a "logging" mode and it can log longer with the less-frequent interval.) THIS MEANS that suppose you set the GPS to a 30-sec update interval (which is what I use) and you talk several feet away then take a photo just before the next update interval, the GPS will not know that you've moved. Both the 6D GPS and the GP-E2 GPS have those same update intervals.

I tend to arrive in a spot and then work with that spot for several minutes before moving on... so a 30 sec update interval is fine for me. If you're moving around a lot... increase the update-interval.

Re: GPS Accuracy

rcolman — Fri, 23 Jan 2015 22:58:27 GMT

Thank you very much for your comments,. They are appreciated. That said, however, there is a HUGE difference in between using GPS on a bike ride versus being on the side of a cliff, 30 miles down a dirt road in the middle of nowhere.

Built-in camera GPS is a great advantage, both in the field an during image and data analysis. For example, how much fun would it be to manually geotag a couple of thousand images after the fact, hoping that something has not gone wrong?

Sony SLT cameras were among the first high quality DSLR cameras to offer built-in GPS and I owned three different bodies. GPS quality was reasonable, perhaps better than the 6D, but I moved away from Sony to Canon for other reasons.

However, the trend is away from built-in GPS, which is very sad. It is of great value to even casual users - most of whom say that they don't need without even trying it.

Re: GPS Accuracy

rcolman — Fri, 23 Jan 2015 23:03:09 GMT

Tim, thank you. Do you know of any test data, informal or otherwise, comparing the units? I don want to spend another $300 unless I really need to do so ...

Re: GPS Accuracy

MHampton — Wed, 29 Apr 2015 15:38:27 GMT

I just really tried out the GPS on my 7DmII and when I checked the coordinates later, they were all over the place, even though I was shooting at a fixed location. All the photos were tagged within about a 3000ft area, but I was stationary with my tripod at a bird watering station.

Re: GPS Accuracy

hsbn — Wed, 29 Apr 2015 16:27:04 GMT

I have the 6D and the GPS is terrible. It jumps all over the place, even if you just step back and forward a bit. I always have to drag the coordinate to the right location in Lightroom afterward. I'm mostly taking landscapes so I just need to know general area. But for your purpose, I think it is not good.

Re: GPS Accuracy

jrhoffman75 — Wed, 29 Apr 2015 18:32:50 GMT

You can get a Hoya data logger from Amazonfor under $100. It's returnable. It comes with software that will geotag your photos by using the HOS and camera time stamps. Try it for accuracy and if it doesn't work you can return it.

A DSLR with GPS is a photo tool that can give you a close approximation of where the photo was taken. Sounds like you really want and accurate GPS tool to locate a site and grap a photo of it. Different focus😀

Re: GPS Accuracy

MHampton — Wed, 29 Apr 2015 19:45:20 GMT

I have other GPS devices to do it the manual way. I was hoping that if Canon was going to put a GPS locator in the camera that it would be accurate. If not, why bother. It will just make people complain. As my father always said, "if you're going to do something, do it right."

Re: GPS Accuracy

rcolman — Wed, 29 Apr 2015 20:01:31 GMT

Not that simple. Remember the engineering Rule of Thirds - you can usually have two out of three: accurate, cheap, or reliable. It would not surprise me that the built-in is fairly accurate, but not reliable. Even larger and more sophisticated hand-held GPS units sometimes have a function that will average multiple readings to get one that is close ...

My BIG question, which no one seems to be able to answer, is if I spend money and hassle with a larger GP-E2 unit, will it be more accurate and reliable? and how much.

Re: GPS Accuracy

MHampton — Wed, 29 Apr 2015 20:09:43 GMT

Or even, if you have the built in, as with the 7DmII, can you even use the external? They may not work together.

Re: GPS Accuracy

jrhoffman75 — Thu, 30 Apr 2015 02:18:33 GMT

Updating my posting: .http://www.learn.usa.canon.com/resources/articles/2012/eos6d_inside_gps.shtml

I stand by my earlier comment. Your primary need is GPS accuracy; you should be looking at a high end GPS and geotagging software. There's a reason professional GPS units cost more than a 6D.

Re: GPS Accuracy

TCampbell — Thu, 30 Apr 2015 01:30:20 GMT

I have the GP-E2... compared to what I am reading here about the 6D and 7D II... the GP-E2 is EXTREMELY ACCURATE.

If I take a photo while standing on a street corner... it not only gets me on the right intersection, it gets me on the correct corner of that intersection. Every single image I have checked is consistently where I recall standing when I took the shot (one might argue that I was actually standing a couple of feet to the left or right of that spot, but it's still a small margin of error to be within a few feet. It is not off by hundreds nor even dozens of feet.)

But a few notes are in order...

GPS satellites do drift in their orbit. Orbits are described by values we call "ephemeris" and these describe the shape of the orbit (orbits are always elliptical -- nothing technically has a "circular" orbit... it only seems close to circular). Also the angle of the orbit, the major axis, the ascending node, etc. etc. This is all very important becuase the unit has to calculus to determine where each satellite should be at any given moment in time. The problem is... the satellites drift a bit. The folks who mange these satellites update the orbital parameters (the data that the GPS needs to use to calculate the accurate position) and the GPS unit downloads the latest orbital parameters ... eventually.

When you first start a GPS, it's using your assumed "last known" position and "last known" orbital parameters for the satellites to try to get a fix. This fix will be "in the ballpark" -- but will not be accurate until the GPS has had time to get all the orbital ephemeris data updates from the satellite network. If you haven't powered on a GPS for a year... this can take a LONG time. I keep a hand-held GPS that I use on my boat and as it is an emergency back-up unit (only on-board in case the main unit gets fried when I'm a LONG way from shore) it hardly ever gets powered up. I have to power it up a couple of times per year just to test that it still works and let it get ephemeris updates. I've seen it take 20 minutes (no kidding) before it could finally get it's first position fix if it hasn't been used in too long. If you last used your GPS yesterday.. it's remarkably fast.

When you first get a new camera and go out shooting... power it up and leave it powered up with a clear view of the sky for a while... preferably at least 15 minutes. This should make sure the internal GPS gets the latest data and the position fixes should be fairly reliable IF the GPS has a relatively unobstructed view of the sky. If you are in a city with tall buildings or in a forest... the GPS signal is going to be weak and you may be position fixes that show you a block away from your true position (this is normal and it's just how GPS devices work.) Some better GPS antennas (like the GP-E2) have better and more sensitive antennas and will tend to get a reliable fix even in some of these marginal locations (like forests, etc.) where the signal is weak.

So yes, the GP-E2 is a better GPS. It also includes a magnetic flux-gate compass so in addition to adding a position fix to the image data, it also adds the direction you were facing when you took the shot and it also has a data-logger so it can keep a "bread crumbs" path of where you've been (I never use that feature, but it's in there.)

Re: GPS Accuracy

RobertTheFat — Thu, 30 Apr 2015 11:16:02 GMT

Very interesting article, Tim.

Has anyone ever thought of having GPS units in close proximity talk to each other about where they think they are? It might help newcomers and stragglers achieve a better fix. One of those cases where two (or more) heads are better than one.

Re: GPS Accuracy

BK_Shin — Sat, 02 May 2015 19:23:26 GMT

I have GP-E2 for my canon. I think geotaging is important because it gives to the picture a lot of information.

But briefly spenking, GP-E2 is not that accruate that you thought.

GP-E2 is still small gadget to record its geometic information. In some area to need exact location info, those gedget can get over 400 satellites in space. But GP-E2 as small gps couldn't get that much.

So, you alreadly have 6D which has inner gps. Even if you buy GP-E2, you can't expect remarkable differences between them.

Thank you.

Re: GPS Accuracy

rcolman — Sat, 02 May 2015 19:40:33 GMT

Can you verify that the external gps records camera direction within the image exit datA? If so, in degrees or ? Super important to me. Tnx.

Re: GPS Accuracy

TCampbell — Sat, 02 May 2015 19:42:19 GMT

@BK_Shin wrote:
I have GP-E2 for my canon. I think geotaging is important because it gives to the picture a lot of information.
But briefly spenking, GP-E2 is not that accruate that you thought.

I'm not sure what you consider "accurate" but as I browse through my photo history of shots taken with the GPS on the camera... the overhelming majority of these shots are tagged to within just a few feet (maybe 5'). Just occasionally I will find a shot that has my position info off by perhaps 10-12', but that's (a) unusualy and (b) I only have the GPS acquiring a position fix update once every 30 seconds and it's entirely possible that I happened to catch shots where I walked a few feet within that 30 seconds.

If you want a GPS to measure the movement of tectonic plates (need to measure fractions of an inch) then no.. it's not that accurate. But if you want a GPS for geotagging photos to recall your location when a shot was taken... it's amazingly accurate.

I can post a few examples with EXIF data and you pull up the locations on a map if you'd like. I have numerous images in which a Google satellite view will let you see what I was shooting, you can look at the subject in the photo, and you can tell that it pretty much pegs me spot on where I was standing at the time.

I have not used the built-in GPS on a 6D or 7D II, so I cannot compare. But I have numerous GPS units for cars, bicycles, boats, and cameras, and this unit is one of the more accurate GPS units I've used. I think it would be unrealistic to expect a higher degree of accuracy than the GP-E2 can provide without getting into commercial/industrial grade GPS units which wouldn't be in this price range.

Re: GPS Accuracy

jrhoffman75 — Sat, 02 May 2015 19:46:36 GMT

Yes. See page 20 of user manual.

Re: GPS Accuracy

rcolman — Sat, 02 May 2015 19:55:40 GMT

Don't have a unit but thinking about It.

Re: GPS Accuracy

TCampbell — Sat, 02 May 2015 19:59:51 GMT

@rcolman wrote:
Can you verify that the external gps records camera direction within the image exit datA? If so, in degrees or ? Super important to me. Tnx.

The GP-E2 does have an internal flux-gate compass and records the direction you were facing in degrees ("magnetic" bearing, not "true" bearing. To convert to "true" heading you would have to know the magnetic declination for your position (which is easy to look up.)

Electronic compasses generally are not particularly accurate -- they'll give a bearing which will be within a handful of degrees of correct bearing, but you can find a fixed point in the distance, check the bearing (the newer camera have a GPS-info screen so you can see it update in real-time), swing the camera off target, then swing back on target and notice the bearing info is not identical to your previous reading. It wont be off too much... just a few degrees. I generally regard it as being "close enough". Also, being magnetic, it can be affected by local magnetic interference.