The marquee features of thee iPhone 14 Pro and iPhone 14 Pro Max are unquestionably upgraded cameras; a new always-on display; and a Dynamic Island’ that changes size and shape to accommodate various types of alerts, notifications, and interactions. For my circumstances and use case, I can roughly appraise these as having some potential utility, no utility, and some curiosity appeal.
However, digging deep in to the new iPhones' technical specifications, I found something that piqued my interest and which I thought could potentially offer me significant utility, and that is their support for “precision dual-frequency GPS.”
Apple's objective with the addition of dual-frequency GPS is a simple one, to improve the precision of GPS data on these new iPhones. For me, this has the potential to translate in to something meaningful when navigating the world around me with either my cane or guide dog - it could result in navigation apps such as Microsoft Soundscape, BlindSquare, and GoodMaps Explore delivering more accurate and reliable information about intersections, destinations, and what's nearby.
Below, I set out the results of some early and rudimentary tests. My intent wasn't to determine whether the iPhone 14 Pro offers more accurate GPS data than a single band iPhone. I take that as a given. The goal of my testing was to determine whether this would make a demonstrable and meaningful difference for my specific use case.
TLDR: Don't buy an iPhone 14 Pro or iPhone 14 Pro Max on the basis that you think that it will enable apps such as Microsoft Soundscape to be more accurate.
What is Dual-Frequency GPS and Why Should I Care
The turn-by-turn directions and information on what's nearby delivered by apps such as Microsoft Soundscape, BlindSquare, and GoodMaps Explore are a result of a GPS signal received from a satellite that the iPhone uses to pinpoint its location.
GPS operates on a certain frequency system. The range of frequencies over which the signal operates is called a frequency band and there are multiple frequency bands. Previous iPhone models supported only a single frequency band – called the L1 band.
Limitations of the L1 band are that it's relatively slow and things such as tall buildings, atmospheric conditions, or even dense foliage can interfere with its signal and also create multiple ‘reflective’ signals which result in errors and inaccuracies.
Apple's dual-frequency GPS implementation combines the L1 band with the L5 band. The L5 band has a lot of advantages over the L1 band, like higher power, higher accuracy, and the ability to detect unwanted reflected signals. On paper, the result is that the new iPhones are capable of GPS accuracies on a centimetre level, compared to up to 5 meters from a traditional single-frequency GPS.
In short, the dual-frequency GPS of the iPhone 14 Pro and iPhone 14 Pro Max means that they should offer faster and more accurate GPS data than previous models and the iPhone 14 and iPhone 14 Plus.
For detailed information on GPS L1, L2, and L5 Frequencies, here's an article that claims that it will tell you everything that you need to know.
Test Hardware, Software,, and Conditions
My test devices were an iPhone 14 Pro with support for dual-frequency GPS and an iPhone 14 which supports only the single (L1( frequency band. Both iPhones were running iOS 16.0.2.
Because the retrieval and initial processing of GPS data is handled by the iPhone's hardware and iOS, developer's don't need to update their apps to support dual-frequency GPS, they essentially benefit from it for free. This meant that I could use the current App Store versions of Microsoft Soundscape, BlindSquare, and GoodMaps Explore for my test and they should all be capable of demonstrating any change to GPS data accuracy.
Soundscape and BlindSquare use OpenStreetMap for map data and points of interest, whilst GoodMaps Explorer can be configured to use either Apple Maps or Foursquare for its external points of interest. For my test, I had GoodMaps Explorer configured to use Apple Maps. Arguably there is no reason to include both Soundscape and BlindSquare, as the fact that they use the same map data means that the results should be identical. I chose to include both at this point as there are things that I like and dislike about each in regard to their interface.
On my test day, there was light cloud. I was in a low-rise city in the south of England. I mention this, as the results are likely to vary depending upon weather and environmental conditions, so your own experience may not match my own.
Test #1: From the Couch
To get a quick and dirty indication of whether there was going to be any difference in GPS data accuracy, I figured that I didn't even need to leave the couch in my living room. I could simply fire up each of my test apps and see if the iPhone 14 and iPhone 14 Pro would agree or not on the distance to some nearby locations.
Both iPhones and all three apps were accurate and consistent in locating my current address.
However, when I use the apps to explore what was nearby, some differences in results appeared.
GoodMaps Explorer mostly reported street addresses and points of interest as being closer when using the dual-frequency GPS iPhone 14 Pro than it did with the iPhone 14. Within a radius of 100 metres, this difference ranged between 1 and 2 metres. Locations beyond this radius were typically reported as being 5 metres closer by the iPhone 14 Pro.
Things were less clearcut with Soundscape and BlindSquare, where the reported distances to points of interest and placed markers were typically the same on both iPhones. They disagreed on no more than 20% of locations. However, when they did disagree, the difference was as much as 10 metres. Interestingly, when there was a difference, the dual-frequency GPS iPhone 14 consistently reported the location to be further away than the iPhone 14, which was the opposite to my experience with GoodMaps Explorer.
My takeaway from test #1? That the 2 iPhones produced slightly different and seemingly conflicting information about the distance to nearby locations. This test didn't tell me which, if either, was more accurate. It didn't tell me if the difference would translate in to a meaningful difference in real-world use. I couldn't even be certain whether some differences were the result of OpenStreetMap versus Apple Maps rather than dual-frequency GPS versus single band GPS. Clearly more testing would be necessary.
Test #2: From the Yard
After determining from my couch that the 2 iPhones often produced differing results in regard to the distance of some nearby locations, I figured that the next step would be to find out whether this was still the case when outside of the house. This would bring me a little closer to the real-world conditions where I would want any improved accuracy, and would determine if my testing would proceed to a third, and final stage.
Being out in the open and well away from any buildings produced results that were very similar to those inside my home. So much so, that there's nothing of note to say about this second test. I can simply say “see comments from test #1.”
My takeaway? That I was surprised. Perhaps I shouldn't have been, as the nature of my property is such that it probably doesn't present too many issues for a GPS signal. This is likely a case of where your own experience might vary according to the type and construction of your property.
Test #3: From a Real-World Scenario
Having determined that the iPhone 14 Pro and the iPhone 14 could deliver different GPS data, I figured that I had only one option left - to grab my guide dog's harness and discover if one of the iPhones gave more accurate GPS data than the other in a real-world scenario, and -if it did - whether this would make a meaningful difference.
To allow me to perform a true side-by-side comparison of the performance of the two iPhones, I paired one with my Sony LinkBuds and the other with my AfterShokz Aeropex. This meant that I could have one phone talking directly in to my ears, the other talking to me through my cheekbones. No doubt an ‘interesting’ look, but a setup that proved to work well during my test.
I chose Microsoft Soundscape as my test app, as it's previously proved to be the best fit for my typical use case and to provide the best experience. The only difference in the configuration of Soundscape between the two iPhones was that one used a male voice, the other a female voice. This was so that I knew which iPhone was speaking to me during the test.
My test route was an approximately 2 mile loop around a low-rise residential area. It consisted of many intersections; public transport stops; points of interest; and a couple of markers that I've manually placed previously, so there would be plenty for Soundscape to tell me about along the way.
After all of the above setup, you might reasonably be expecting a positive payoff for the time that you have invested in getting this far in my post. Regretfully, and with some surprise, I can't give you one.
Within 200 metres of leaving home, I knew that a 2 mile walk wasn't going to be necessary to reach a conclusion. However, I persevered, just in case.
It didn't change things. The two iPhones and Soundscape delivered near identical callouts of locations. Some times the mail voice would speak first. Some times it would be the female voice first. Whichever it was, , the other voice was rarely more than a couple of steps behind. There were more occasions when the two voices were in perfect synchronisation than times when the callouts didn't at least partially speak over each other.
In short, there was no meaningful difference in the performance of Soundscape on the dual-frequency GPS iPhone 14 Pro compared to the single band iPhone 14.
Had there been a difference, my next step would have been to determine Soundscape on which iPhone was more accurate and by how much. With there being no discernible difference, this was now moot.
Test #4: From the Future
Following a request in the comments below, I performed some additional tests with BlindSquare after this article was originally posted.
Here are the key points:
- The “current GPS accuracy” quoted by BlindSquare for each iPhone was 5 metres. This was outdoors, on a clear day, and well away from any buildings or trees. With the dual-frequency iPhone 14 being capable of more precise GPS data than the single band iPhone 14, I wonder if the app itself is not calibrated to go below a 5 metre reading for this data. It's also possible that the ‘GPS signal-friendly’ conditions levelled the playing field somewhat in regard to this reading/calculation. This is all pure speculation on my part, so make of this what you wish.
- Using a custom marker created on the dual-frequency GPS iPhone 14 Pro, BlindSquare's distance callouts on the iPhone 14 would often jump by 2 or 3 metres as I approached the marker. On occasions the jump would be to a greater distance, even though I knew that I was moving closer. This was from a starting distance of 20 metres and walking at a slow and steady pace. In contrast, distance callouts on the iPhone 14 Pro were typically stable and appeared to accurately reflect my pace and the distance covered.
- When standing still at what the iPhone 14 Pro told me was 20 metres distant from my custom marker, the distance quoted by BlindSquare on the iPhone 14 was not stable. On occasions spontaneously jumping up or down by as much as 2 metres. These jumps were typically no more than 5 seconds apart.
- when approaching the custom marker at a slow walk, BlindSquare on the iPhone 14 did not give a distance lower than 3 meters. I had to stand still for nearly a minute before BlindSquare indicated that I was at the location of the marker.
- The 2 iPhones gave me different what3words results for the exact same position.
- Not surprising considering the above, the map coordinates were also different - x.39001,-x.12433 versus x.38998,-x.12435.
My takeaway? That I would prefer for those with a working knowledge of all-things GPS and BlindSquare to be the ones who attempt to pick the bones out of the above. However, it does seem reasonable to conclude that the dual-frequency GPS of the iPhone 14 Pro does deliver more stable and reliable GPS data than the iPhone 14.
My tests are clearly very far from scientific. Under laboratory conditions, I am certain that the iPhone 14 Pro does indeed deliver more precise GPS data than the iPhone 14. It's probable that this makes a meaningful difference in some locations, some atmospheric conditions, and even possibly when travelling in a vehicle.
However, for my typical use case, the dual-frequency GPS iPhone 14 Pro performed no better than the iPhone 14.
Colour me surprised. I truly approached this exercise thinking that I would notice at least some difference, and that I could use this to at least partially justify the purchase price of an iPhone 14 Pro.
As stated above, I expect other people's experiences to vary according to such things as location and atmospheric conditions. Accordingly, if you have either an iPhone 14 Pro or iPhone 14 Pro Max, I would love to hear your experience in the comments below. The Apple Watch Ultra has dual-frequency GPS too, so do please also share your experience if you have one of these. It would be particularly interesting, and likely very helpful, to hear from anybody who lives somewhere more challenging for GPS signals.
Hey David, Great research and post, thanks! I've been curious about this for sure.
I have two thoughts/questions and few answers:
Is it possible the maps are currently full of more L1 type data so more precise calculations are not available? Especially OpenStreetMaps I would not be surprised if a lot of the mapping was done with L1?
Of possible consideration for experimentation: when I first read about the dual frequency abilities the first thing i thought of was would this improve accuracy in self defined waypoints or points of interest? And would this allow for a more precise recall to the location I am looking for? When I previously tested this with Microsoft SoundScapes I think in the iPhone 12 Mini days I was not impressed with the lack of precision to find the point I wanted to return too. Yes, I was attempting to find a point kind of out in space without an immediate business address. I have the new Apple Watch Ultra
and I have done so far one test of the Waypoint feature and the answer is kind of yes and no. Compass says I've arrived even though I can be some distance away. But when I was checking the raw GPS data it looks very precise in that I saw changes in the longitude value just by taking a couple of steps and trying to make a new Waypoint, so it looks to me the Compass app is being possibly less precise than the data would allow it to be. Since Compass isn't really intended for the precision a cane travel user likes it isn't entirely surprising. Perhaps an enhancement to one of our more specialized apps to sync with the GPS data the Watch Ultra (or iPhone Pro) can provide would allow us to get a better idea.
Thanks for this post David, very interesting indeed.
I think I mentioned that my current thinking is that I’ll replace my 12 mini next year, most likely with a Pro model. This was one of the features I hoped would make that Pro option worthwhile. So this is not very encouraging.
I would though agree that it would be interesting to hear if it is any better at setting and finding markers that you set yourself. If not, then that’s pretty definitive.
Also, on a side note, I wonder does anyone know whatever happened with the European led Galileo system? It was meant to be more precise than the US owned GPS system. Or has it already been integrated and we simply continues to call it GPS?
This is off topic but...
How did you manage to buy both?
I understand O2 has a thing where you pay monthly for a phone, I'm on it, but did you really pay for 2 phones? That's going to be over £2000 to pay both off isn't it?
Great Work, but not using as intended.
Hi David, I would have never had the patients to do what you did to test all this and write it up, so I commend you. However, I still think the feature is valuable if used as they must have intended.
For one thing, only part of the L5 satellites are up and running according to what I read, and the apps you tested with do not show if you are accessing them on the L1 or L5 band, or how many of them you are reaching. Where this feature would help the most is driving downtown in big cities in your car where the GPS tends to get scrambled by tall buildings and sends you in circles. An example of this is in Seattle, where if you try to get an Uber from the bus station, the GPS thinks you are on top of the highway overpass instead of underneath it. So it keeps sending cars to the wrong place. I would still consider this to be real world navigatin that would help anyone, blind or sighted. If it can clear up that type of interfeerence for me, it would be welll worth the cost. Might be worth checking YouTube to see if anyone has done any testing in that type of scenario.
What a fantastic write-up! Thank you for taking the time to experiment and share with us. I find this interesting and definitely curious to know how these results would vary; I've found GPS to be pretty tough to work with at times in a city like Boston, for example.
Similar results for now
HI David. This was an awesome write-up! Thank you for taking the time to do this! I did not notice any sort of GPS improvements or digressions when testing this feature on a 14 Pro Max at my research center, I'm glad my results seem to line up with yours on one hand, but wish we would see improvements. I think we will in time, as all of the l5 stuff isn't fully ready yet. It's also interesting to note that you are in the UK and I'm in North America, but our results are similar. I think we may be here discussing it from a different perspective then, but we'll see what happens. This was a project I've been trying to figure out how to do for myself, but I don't have the ability to monitor 2 speech outputs at the same time. Again, thank you very much for this!
What about the accuracy of self created points?
Hi David, thanks for this very interesting post! If you still have the two Iphones; could you please do the following experiment as well and let us know the results? Create a "my places" point in Blindsquare with your Iphone 14 Pro (and set range to to 1 meter) and see how accurate the Iphone 14 Pro navigates to that point? To check the accuracy it is easiest to make that point at a spot you can check like your own front door or, a traffic light etc. And do the same with the Iphone 14. I Am not completely sure, but I think that, to get that accurate navigation, you need to create a point with a device that has the dual frequency feature aboard as well. And if you could do the same experiment in an indoor location, that would be great. I guess, creating an actual route with an app like Seeing Assist Move, would give the same kind of experiment. But just checking the accuracy of one, self created point, is a bit easier. Hope the Iphone 14 pro will give better results in this test..., because improved accuracy would be very nice...
Re: What about the accuracy of self created points?
I've now added a few BlindSquare experiences/findings to my post. Check out the “Test #4” section.
Greetings from Chicago. I want to mention that the test was conducted with open sky and not in a big city with tall buildings on each side of the street right next to the sidewalk like in Chicago or New York. It also was not tested on a residential street with an overhead tree canopy. It would seem likely that the results of the iPhone 14 Pro would be better in these environments.
Starting with the iPhone 12 series, the iPhone supports all four of the world’s satellite navigation systems. These are the Global Positioning System from the United States, GLONASS from Russia, Galileo from the European Union, and Beidou from the People’s Republic of China. Although GLONASS supports L5, it’s accuracy is only two or three meters. All Galileo and Beidou satellites support L5. Only a portion of GPS satellites support L5. Even if a device can receive data from all four systems, it must support dual frequency to have the accuracy within a centimeter. Given the world now has four satellite navigation systems, it makes one sound outdated and provincial to refer to software applications and services generically as GPS technology, unless speaking about the GPS system and standards specifically. The iPhone uses more than satellite navigation technology to deliver location data to iPhone users. Apple uses the position of cell towers in relation to the iPhone and the nearness of Wi-Fi hot spots to also locate an iPhone.
From a Developer
I developed an iOS GPS app a few years ago and at the time there was very little control over the hardware. You just received a signal from iOS as requested see https://developer.apple.com/documentation/corelocation/cllocationaccuracy
The dev has control over accuracy but it’s please give me accurate data if you can or best efforts by iOS and hardware.
The location has +- accuracy so I’d guess the location you get is the most accurate it’s found rather than some kind of average
The higher the accuracy the greater the power so devs are encouraged to use the least accuracy
Clearly a lot depends on the incoming signals and where you are.
Turn WiFi off when testing as it’s used to help locate you.