GPS sometimes is half a world off

What Anders said, basically. Keep in mind that GPS uses a modification of the "great circle" method to locate a receiver. One satellite will give you a circle of possible solutions - constant distance from the satellite. Two satellites will give you two possible solutions (where the great circles intersect) while three satellites will give a fairly unambiguous position coordinate. Four satellites are generally required by GPS receivers to give a result because the clocks need to be synchronized as well in order to calculate both position and velocity.

If the GPS receiver displays a calculated position based on only two satellites (and it's really up to the software to ensure it's done correctly) you'd get two possible locations, but due to timing and position of the satellites, it shouldn't be a 180 degree error - the satellites aren't high enough to give solutions for 1/2 of the globe with only two signals. Seems to me there is something else in the software is mis-calculating the solution.

You forgot that navigation is a 3 dimensional thing. Thus using signals from only two satellites will give you a possible position along the intersection of two spheres. That's a circle by itself. Add a third satellite, and you get your position at two different intersections of three spherical surfaces. Add a fourth, and you get your position at a singular intersection of four spherical surfaces, but only if timing is perfect and there's no signal distortion. Otherwise your position will be inside a pyramid with spherical sides (or even worse to imagine, at the minimum combined distance from the outsides of an inverted imaginary spherical pyramid). Setting the clock correctly in the receiver is equivalent to minimizing the volume of this pyramid. That's how a simple GPS receiver can approach atomic clock precision by just listening to a few signals and applying some math to them.

No wonder sometimes these receivers think they are just... elsewhere!

Happened again yesterday when I was at a National Wildlife Refuge…completely out of doors the entire time.

Of course, but I think most people limit their position to some location on the Earth's surface. Commercial GPS receivers don't do a very good job of showing elevation either - the error is about 10x worse than that of the lat/long measurement. My point was that you're not going to get a 180 degree error in lat/long just because of an incomplete data set. It must be an error in the software.

Your phone will also be using the cellular networks and wifi to get a base idea of where it is, and then it can use the gps solely to key in on your exact position.

What I meant is that the position is calculated in three dimensions, then projected on the earth's surface using a certain map datum and geoid model. There's for example a difference from if the position is projected on the surface along a line towards the center of the earth, or if the line is perpendicular to the earth's surface. The earth isn't spherical, or both would have been the same.

But nevertheless, this is not the reason for suddenly ending up on the other side of the globe. That's some software issue.

pknight wrote in post #17546206
I'm sure it is normal. It just isn't good enough.

I worked last year with a 7D Mark 1 with the external GPS unit on the hotshoe. The camera itself was in dire need of an update (thank you Canon) but the GPS unit worked fine. Now we have a wonderful camera in the Mark 2 but the GPS is too unreliable and there's no option to fit an external one. At least 20-40% of the time it doesn't register readings. I'm on a helicopter with positioning set to very second and I rely on this information for my work. Resetting the camera so the Menu allows attachment of the external unit in lieu of the onboard unit would be welcomed as part of a firmware update.
Canon please solve this glitch on an otherwise excellent camera.

Are you sure you can't use an external GPS? As far as I know, the 6D uses the internal GPS when it has nothing else, but changes to an external if there is one. Note that I don't have any, so I don't know for sure.
But I do know that if I connect two GPS units to the 7D or the 1DX, they just use one of them without any issues. You don't need to set anything.

apersson850 wrote in post #17553037
Are you sure you can't use an external GPS?

Thank you for your input. However, naturally I've contacted Canon Europe over this and they say not. One reason is that the cable is the wrong one (I'm using the Canon unit) and the other is that there's no preparation for it in the Menus.
I do get the feeling I've been pushed away too quickly, the Canon Europe setup is designed to send you back to the same place i.e. in my case Canon Spain and they've not been helpful at all. Does anybody here know a direct route to Canon Japan? I've tried, written on their Japan site (in English) and clicked on what I thought were the right buttons but nothing from that either.

This is obviously different between models. The user's manual for the 7D Mark II specifically states that the GP-E2 can't be used on the camera, but for the 6D it instead says that an external GPS can be used for better GPS performance (read more optimal antenna).

The 1DX also works fine with two GPS units connected at the same time. It just uses one of them.

apersson850 wrote in post #17554527
... but for the 6D it instead says that an external GPS can be used for better GPS performance (read more optimal antenna).

The 1DX also works fine with two GPS units connected at the same time. It just uses one of them.

Thank you. This only goes to justify my request as being perfectly feasible. Now, who do I contact at Canon that will listen ...

m3photo wrote in post #17554751
Thank you. This only goes to justify my request as being perfectly feasible. Now, who do I contact at Canon that will listen ...

Probably nobody, because your request probably isn't feasible. If the camera were designed to be capable of using an external GPS, that capability would be built into the camera, just as in the 6D. That is, Canon seems to have no philosophical objection to the use of external GPS units. I suspect that there is a reason, related to the design of the 7DII, that makes it impractical to use an external GPS. If my suspicion is accurate there would be nobody who will listen who could make any difference.

For me, I am saving my 7DII battery power and getting much more accurate GPS data by using my free cell phone GPS app.