Hacker News Reader

I thought the same, but I was pretty surprised how bad my phone's GPS performance was next to cliffs. Especially altitude was off significantly.

When thinking of turn-by-turn navigation, keep in mind that car navigation will snap your position to roads. For a better comparison, try e.g. Google Maps in pedestrian or bicycle navigation mode. At least in my experience, the perceived "accuracy" drops significantly.

If you have a USA phone, bear in mind most are sold with glonass and beidou and support disabled.

Buy a phone or GPS from elsewhere in the world and you'll get to use all the networks and get much better accuracy.

EU phone, but yeah, good to point out GNSS isn't just GPS anymore. I also agree that the accuracy is insanely good most of the time; I'm just objecting the "accuracy test" using car turn-by-turn navigation :-)

Do you have a source on that? My understanding is that most consumer cellphone GNSS systems today are multi-constellation out of the box. If there is a satellite up there, the phone will use it, no matter which flag is painted on it.

Additionally, phone GPS chipsets are often built into the baseband modem and performance might be limited compared to state of the art (and just price, packaging limitations for antennas, assumption that a phone always uses A-GPS etc).

The dual frequency chipsets used in modern high-end fitness watches & drones have pretty spectacular performance even in incredibly built up areas with high tower blocks or under tree cover (they usually have a barometer as well to help with the altitude problem).

> keep in mind that car navigation will snap your position to roads.

Right. I chose driving as an example because someone navigating a city on foot could conceivably do it with low-precision signals like wifi. They can stop and wait for a better signal, or carefully study the map comparing it to street signs; a fix that puts them on the right block and that only updates once or twice per block is adequate, if not good.

Whereas turn-by-turn navigation is able to tell the driver "take the next right" 3 seconds before the turn - which shows the signal is being tracked precisely and regularly.

> For a better comparison, try e.g. Google Maps in pedestrian or bicycle navigation mode.

If you're into sports, take a look on Strava at people doing things like big city marathons: https://www.strava.com/activities/8078299417 https://www.strava.com/activities/51910242 where of course the positions aren't snapped to roads.

You'll see minor hiccups from time to time - but despite navigating between many tall buildings, you won't see errors of hundreds of miles.

GPS units can also include INS or INS like logic systems to prevent those types of jumping around. It can be difficult to tell from an application output just how processed your "dot" is.

It's actually fairly easy, for things like running GPSes.

You know the track isn't snapped to a map. And you know it's not built into a car, so no odometer or rate gyro.

Then just look at the trace as the sportsperson goes around a corner. Does the trace overshoot, then backtrack? That means the kalman filter is heavily smoothing with momentum. Does it produce a gently rounded corner that goes through a building? That means there's a moving average.

However if the traces display sharp corners when appropriate - then the signal is not being excessively smoothed.

(And I can assure you, even if a GPS watch has enough motion sensing to count steps, the swinging of arms and bouncing of feet mean wrist acceleration is noisy enough you've got no chance of detecting a runner turning a corner)

Right, I was mostly thinking about robustness against transient jumps due to transient multipath reflection as you navigate around an urban environment. If you know that your strapped to a walker/runner, you know that you can't travel 100m in ~3-4 seconds.

If you don't have other sensor data, like cell tower and Wifi, then GPS still has a pretty poor performance. Another problem of GPS is altitude. GPS is not really designed to give a good altitude reading, especially when you are in the mountains where only get satellites readings from over your head and don't have readings from the satellites on the horizon.

But in the mountains, altitude is also a very important information. I was up at basecamp in Nepal (south side) and Tibet (east and north side). The basecamp in Nepal is in a very tight spot. Bouncing signals is there a problem. Even radio is a problem.

BTW: I commute daily with train, here in Germany. I play Pokemon and I have problems getting a GPS fix with the latest iPhone when I'm not directly at the window.

This used to be true many years ago, but today, even cheap phone receivers use a very high number of satellites (so geometry is usually more favorable) and usually also contain a barometer, which can provide additional disambiguation via sensor fusion.

> especially when you are in the mountains where only get satellites readings from over your head

That might be true in a very steep ravine, but it's more of a concern for reaching communications satellites (of which there are much fewer or sometimes only one, depending on the system). Today, not getting a GNSS fix due to terrain is increasingly unlikely for the reasons mentioned above.

Also, in a hiking/mountaineering context, it's probably fair to assume that any GPS user will be located on the surface of the earth, so a precise lat/long fix is really all you need; you can determine the height using elevation models, which are pretty accurate these days.

> I commute daily with train, here in Germany

Some German trains were notorious for having bad cell signal due to metal-coated windows (applied for reflecting solar radiation). Try the same thing with an (open) car window and I bet you'll get excellent signal.

Some German trains were notorious for having bad cell signal due to metal-coated windows (applied for reflecting solar radiation).

It's the first thing I thought of, and it's why we had ceramic tinting put on our vehicle windows instead of conventional tinting (which in many cases, AIUI, has radio-blocking metal in it).

But in the mountains, altitude is also a very important information.

But wouldn't one carry a barometric altimeter in those cases? I mean, my iPhone and Apple Watch both have barometric altimeters, and those are consumer devices not designed to get you up a 7000m mountain.

Now, granted, that doesn't help the device that's "sewn into a jacket", but as a sibling commenter mentions, in a search elevation models could be used. Or if they can stick an altimeter in my watch, maybe there's one in this mysterious, unnamed device.

The basecamp in Nepal is in a very tight spot. Bouncing signals is there a problem.

I wonder how much of a difference the newer dual-frequency GPS receivers make. It makes a difference on my Apple Watch Ultra versus an older Garmin when running around in the woods, but I'm no mountaineer.

Most anyone who climbs Everest knows their approximate altitude based on certain landmarks (waypoints) that are very well known. The most obvious of which are the camps (1 through 4) but there are other landmarks that are well known.

If you're going off the main route however, then it's a bit of a different story.

Knowing your altitude on the main route is fairly easy. Those routes are entirely marked (roped) and have obvious geological features that mountaineers have memorized. The four camps en-route are also at the same altitude every year.

Getting outside information about your altitude certainly is important if you're doing a non-standard route or if you stray from the fixed ropes somehow.

> it's a proper satellite fix, with the precision to give turn-by-turn navigation instructions in good time

That’s not entirely true. Phones do sensor fusion when doing turn by turn navigation. Accelerometer, gyro, and barometer are especially taken into account, but also compass and trilateration via WiFi and cellular sources as well.

Dropped in just to say, no one understands GPS, including 99% of hacks and tech people... Go buy a $5 dongle and stream the feed, its just lat/long repeating maybe sometimes a altitude... but its still very weak and very unreliable... with mobile phones pulling wool to make you all believe its turn by turn. Ok bye.

> Dropped in just to say, no one understands GPS, including 99% of hacks and tech people... Go buy a $5 dongle and stream the feed, its just lat/long repeating maybe sometimes a altitude...

So you’re also one of the 99%! That “feed” isn’t the GPS signal, it’s the output of your cheap device. ;)

There are folks here however who are intimately familiar with gold codes, legendre sequences, Jaffe-Rechtin phase locked loops, RHCP signals, URA vs URE, etc.

Thanks Karen! Was suggesting people learn these concepts through the use of a usb dongle, but turns out you're the dongle.

Google does something pretty clever specifically for urban canyons too: https://blog.research.google/2019/02/using-global-localizati...

In my (limited, I don't often use my Pixel) experience, it does work much better than my iPhone's GPS, which often wants me to use the camera (in Apple and Google Maps) for urban canyon navigation using visual sensor fusion.

Your phone is using other signals, like maps of WiFi ssids to help locate in gps denied areas.