Distance to next junction accuracy?

[Mr-B]

Does anyone else think that Navigator 4.1 overestimates the distance to the next junction? I think it's just about double!

[nej]

Do you mean when it says "exit in 200 yards", but the exit is already there?

I've noticed that on TomTom on pocket pc. The reason I put this down to is that rather than the exit being defined as the start of slip road, i.e. when you can just pull into the left lane, it is rather the bit where you cannot pull over any more, i.e (ascii-art time!)

| |
| | <- The exit is actually defined as here.
| \
| \
| \ <- You start to pull over here so expect the exit to be defined here.

That's just my thought anyway...

One place it's REALLY wrong is J8 of the M25 going clockwise. The slip-road is the best part of a mile and a half long (I've no idea why!) and if you listened to TomTom you'd be a mile past the exit before it even mentions turning off!

[Mr-B]

Kinda - but this happens with every type of junction, not just slip roads.

I have also seen a comment in a TT review which pretty much matches mine:

http://mypalmlife.com/index.php?p=589&more=1&page=4

"One very annoying and interesting blatant flaw is the GPS measurements. While it seems to match my Pioneer’s GPS for miles, it complete losses it at the yardage count. 100 yards is not 100 yards. None of the yards are yards. In the last day of testing, I drove slowly down one street to gauge it better and when it told me I was 20 yards from my turn, I stopped and looked. I was maybe 20 feet from the turn. Not even close to 20 yards."

I've reported it to TT, but unfortunately their support ppl only seem able to recommend swapping out of every single piece of hardware or else re-installation - which of course makes no difference at all.

[nej]

What GPS are you using? If it's anything confiugrable make sure it's using WGS84 as it's Datum to make sure it's properly accurate - I had that problem with my GPSSmart initially.

Also, bear in mind that theres always going to be a couple of seconds delay anyway. The GPS receiver gets the signals, calculates the position, creates the nmea string, transmits it, the software receives it, decodes it, calculates where you are on the map and has to act upon the instructions. It's quite a lot to do, and when you are moving at 30mph, by the time it's actually started speaking to you (it also takes a couple of seconds to say "in 200 yards turn left", don't forget) you are no longer where you were when the position was calculated!

And further to all that, GPS isn't centimeter accurate. It's more like 10 meters, although the "official" accuracy is even more than that.

[Mr-B]

It's the TomTom BT GPS receiver and I can't see any way to configure it via the supplied s/w.

I think Navigator auto-corrects the distance for motion since when I stop before a junction it often increases the junction distance a little (since it didn't know that it was going to stop). However I'll monitor this aspect more closely in future.
Also the announced distance is in sync with the displayed distance so there's no delay there.
However the distance does always seem to be overestimated by about double so I don't think it's anything to do with GPS accuracy, since it wouldn't matter at larger distances.

I guess I'll just have to find a graduated road to know for sure.

[Wobbly]

This may be a silly comment - but have you tried swopping to km/metres - is it any better - seems a bit simplistic I know, but may be worth a try...?

W.

[andrewh]

It is the same in km/metres as I use that. I really think that they have some oddity in where they position the reataive distance. If you drive up to a junction and the readout says 100m, it can be anything from 40 to 90. Of course driving instructions can be slightly delayed but the actual "Turn Right" instructiuon comes at the correct time.

Odd!

Andrew

[Wobbly]

I think this the same as one of the posters above said - the software uses predictive location based on estimated speed. I notice on my T3 that the speed is not the same as on the car's speedometer (in some ways I'd be surprised if it was), but this difference/lag may explain the phenomenon you describe as one's speed becomes less 'constant' towards a junction...?

W.

[nej]

Car speedometers are not really that accurate to be honest. Just because the needle says 40mph, it's probably 38mph or something. One thing I've noticed is that the disparity between the speedo and the gps gets more as you get quicker.

[Skippy]

[Mr-B]

I think this isn't about the accuracy of GPS per se, since at the junction the distance indicated is zero (or a small distance if not at the actual junction centre)!
It's also not about the accuracy of car speedometers (my speedo always overestimates slightly).

I think the real issue here is the ability of Navigator to accurately judge the approaching junction's distance. I think the algorithm must just need a tweak since it seems to consistently approximately double the distance from several hundred yards away right up to the actual junction. Location accuracy seems to be spot on in all other situations.

Of course I'm happy to accept that it may be that my sense of distance is completely off, but the fact that others have reported it makes me think otherwise.

[Snerkel]

Talking about speeds displayed by the GPS, in my not so scientific test over a measured mile (proper signed measured mile) at a constant 60MPH according to GPS, a tad over 65MPH according to speedo the measured mile showed the GPS to be near enough bang on (Etrex Legend).

This would suggest my speedo is reading almost 10% above actual speed, this makes sense as I believe speedo accuracy needs to legally be between -0% to +10% eg it can never read under speed.

We see you’re using an ad-blocker. We’re fine with that and won’t stop you visiting the site.
Have you considered making a donation towards website running costs?.
Or you could disable your ad-blocker for this site. We think you’ll find our adverts are not overbearing!