GPS quality on the N95

July 12th, 2007  |  Published in Development, Mobile, Nokia, Nseries, S60  |  10 Comments

Earlier this week, Nokia released another firmware update for the N95. One key feature included in this release is the support for Assisted GPS (A-GPS) on the device. Assisted GPS works by combining the normal GPS satellite functionality with your cellular network connection. The end result is that you supposedly get faster, more accurate readings, while also using up less battery life.

But how accurate is GPS on the N95? Does this new firmware really make a difference? I’ve started doing some very unscientific experiments to find the answer and wanted to share some results.

The tests were done using Nokia’s Sports Tracker program with my N95. Sports Tracker works by recording your GPS position over time and using that information to display details about your speed, location, and distance travelled. One of the very cool features of this program is that it allows you to export your running routes and GPS data to a number of different formats including Google Earth.

I had a few workouts saved in Sports Tracker before the recent firmware update, and last night I went on the same run, but this time with the new firmware update. How did they compare?

N95 GPS comparison

The GPS route captured using the latest firmware definitely stays closer to the path than previously, however, the route also seems to zig zag a bit more than before. This can be critical if you’re using Sports Tracker to track distances as it can lead to inaccurate data.

For instance, I did a speed walk from my house to my friend’s house and I did the walk in 24 minutes. According to the map, the distance between the homes is 1.75 miles (2.82km). However, when I walked there using Sports Tracker, it said the distance was 2.18 miles (3.5km). This seemed like a huge difference, but when you compare the results on a map you can see why.

N95 GPS comparison

It seems that over time, the small zig zags in the GPS readings add up to big differences in distance, and also the average speed. According to Sports Tracker, I did the walk to my friend’s house at a pace of 5.4 mph (8.7 km/h), when in reality the pace was closer to 4.4 mph (7.1 km/h). So if you’re using the program to train or to accurately predict a workout, you should be cautious when examining the results.

What have other N95 owners experienced with the accuracy of their GPS data?

Responses

  1. Jon Wayne says:

    July 13th, 2007 at 11:16 am (#)

    N95 is a rubbish phone, it’s just an n80 in a fancy frock, that’s all. It’s better to have tomtom as a gps or a seperate reciever as the one built in is always messed up as you have just clearly proved.

  2. mapperz says:

    July 16th, 2007 at 9:17 am (#)

    Yes the lock is faster with A-GPS but the actual tracking is more ‘zig and zag’
    as your post shows.

    Prefer the option of A-GPS for quick fix and normal GPS for Sports tracker.

    How do you get a access to the nokia servers in rural location (on a mountain?)

  3. ziels says:

    July 18th, 2007 at 3:51 am (#)

    I have also updated my firmware to v12.0.013 but my tracking is also zig zaggy. Mine is not as apparent but it shows that I am going off the tracks.

  4. LIEW says:

    July 20th, 2007 at 10:51 pm (#)

    I FIND IT LONG TIME ALGO

  5. Rob says:

    September 30th, 2007 at 5:38 am (#)

    I’m really impressed with the phone, for what it is…a first of it’s kind. I’m sure Nokia will only continue to improve and build on what they’ve achieved with the N95. It’s been so stable since I upgraded the firmware to v12.

  6. Dude says:

    November 25th, 2007 at 4:15 pm (#)

    The N95 is a fun toy with a lot of stuff and a great screen, but everything else is substandard quality. Even with the latest firmware (v12). A-GPS is not accurate enough for real applications (better use a bluetooth GPS if you don’t want to miss turning instructions while driving), the camera is ok but slow and lags so no action pictures can be taken, the battery doesn’t even last through the day and multitasking is impossible because there isn’t enough memory and it KILLS APPS WITHOUT WARNING!

    And the feel is not what it should be, feels like cheap plastic and the dual slider feels loose, the playback buttons are wasted space as well.

    Oh, and did I mention the 3.5″ headphone jack? Now who’s the smart ass who decided to put it in the SIDE of the device? Now I would SWAP the usb and phono plugs on the board – because IF you want a plug in it while listening to music – at least I’d love to put it in a pocket. But can you? No.

    Bad design. I’m sorry. Nokia, you can do better! Please do it, quickly. I wasted my money on the N95 but never use it, but I still hope to get a better phone from Nokia!

  7. Bludja says:

    December 19th, 2007 at 6:34 am (#)

    I did a side by side test of 2 N95s recording tracks and didn’t see any of the zig zag. Post about it is here http://bludja.blogspot.com/2007/12/n95-gps-comparison.html

  8. localhost says:

    January 3rd, 2008 at 1:49 pm (#)

    I’m having some difficulties with the sportstracker as well (http://sportstracker.nokia.com/nts/user/profile.do?u=localhost) . I probably will have a run tomorrow again and I’m gonna bring the GPS filtering down to medium and the ‘average speed’ to once every 5 seconds. I’ve tried the sportstracker while driving to work and on higher speeds, the problems seems less.

  9. Testing the N82 with Sports Tracker (Beta) | geo2web.com says:

    January 20th, 2008 at 1:38 pm (#)

    [...] in the tracking – due a high filter correction method as default? here is an interesting article http://wubbahed.com/2007/07/12/gps-quality-on-the-n95/ Highlights the GPS being ‘hyper-active’ (note this was from an N95 with firmware [...]

  10. brotsje says:

    January 31st, 2008 at 3:28 pm (#)

    I am using sports tracker too, but with a Nokia N73 phone and an external GPS mouse (adapt marathon). I have the same zig zag in my route, so it’s not a N95 specific problem. The routes are 5 till 10 procent longer than the actual route is. Different types of filtering makes no differerence

Leave a Response