Forum: General Topics

Forums / General Topics / Spikes in altitude and speed

 

Subject:Spikes in altitude and speed 

ianmanton

0:11
Saturday
6-Feb-2010

Hi again all

I've just been out for a walk and decided to track the path again via TMJ. What's come back is rather bizarre, I would expect the odd spike here and there but there is constant fluctuation such as jumping from 99m alt to 483m and back to 119 in about 40 seconds! At one point I appear to have disappeared 268m below sea level lol! Also appear to have surpassed 40km an hour quite a few times peaking at 87mph - wish that was true!

My question therefore is, is this normal? I was in the open at all times with no overhead objects or trees etc so would have thought the signal would have been strong enough to avoid frequent "hiccups" per se...

Running version 0.7.3 of TMJ basic on blackberry 8900, enabled "include cellID/signal" recently - whether that would cause anything to go awry? Have made the track public so my rambling makes sense!

Cheers
Ian

http://www.trackmyjourney.co.uk/track/tqvzpD4tqCGCf
 

Stephen

18:06
Saturday
6-Feb-2010

Location:
Surrey, UK

Phone Model:
BB 9800 Torch
BlackBerry 8900
SE W910i
Nokia 5800

Hi Ian,

Crikey, that must be one of the 'noisiest' tracks I've seen!

I think what you're getting there simply shows the rubbishness of the internal GPS on the 8900! So it is normal, albeit very frustrating! I've had similar spikiness myself, and I now very rarely use the internal GPS, instead using a separate bluetooth device, which gives a far superior track.

You were probably a bit unlucky to get such a noisy track though - the GPS satellites are constantly moving across the sky, so there are times when fewer are visible, which makes it harder for the phone to get a good fix. Maybe this was the case last night. Do you find that the location accuracy varies at different times?

Were you hand-holding the phone, or was it in a pocket. I think I read somewhere that the GPS antenna is at the bottom of the 8900, beneath the space bar, so making sure that this has a clear view may help. Also, did you first turn TMJ on during the walk, or when stationary (and outside) at the start. I've found with other devices that they often get a better/more reliable fix if left stationary outside for a couple of minutes after switching on.

One thing you can do in TMJ to help reduce the spikiness when walking is set the Activity Profile/Max Track Speed to say 5mph, which will stop any trackpoints from being stored if the current speed is above this threshold. This can give quite a dramatic improvement in removing the spikes, although if the signal is excessively spiky (as in your case), it could result in parts of the track not being recorded at all. Worth experimenting with though. (Don't forgot to change profile if you then get in a car/bike etc - otherwise you'll end up with nothing being recorded!)

Cheers,
Stephen

(ps: Don't worry about the CellID settings - they won't make any difference to the GPS signal)


 

ianmanton

21:21
Monday
8-Feb-2010

Hi Stephen

It may well have been activated on the move, but was in an external pocket rather than buried deep within the recesses of my jacket - but agree this could have an influence (along with naff phone-ness)!

Meanwhile I've just had it on whilst flying - and notice when viewing the graph options on the website it appears to cap out at 6km altitude on the y-axis - is there a way to increase this?

Lastly - thinking about subscriptions - are there any plans to offer a lifetime licence lenght, or other promotional type things lole a half price renewal option or anything like that?

Cheers
Ian
 

Stephen

20:36
Tuesday
9-Feb-2010

Location:
Surrey, UK

Phone Model:
BB 9800 Torch
BlackBerry 8900
SE W910i
Nokia 5800

Hi Ian,

The altitude capping is very odd. Looking in the server logs there are a substantial number of trackpoints (from all users) with an altitude of precisely 6054.00m. Further investigation suggests that these have all come from BlackBerry phones, so I'm wondering if it is perhaps a limit imposed by the phone, rather than an issue in TMJ. I was about to suggest that next time you fly to check the BlackBerry's built-in GPS status (Options/Advanced Options/GPS), but looking at my own 8900 this doesn't actually show an altitude readout, only latitude and longitude :-(

One other thing you might try (to at least eliminate TMJ as the cause) is to enable GPS Logging and set the TMJ Menu/GPS/Log Options/Log Type to Raw GPX. This log file should then show whether the altitude is being provided correctly by the phone and then TMJ is doing something strange, or whether the phone itself is the issue.

(I have definitely correctly recorded a higher altitude using an external GPS device, but that was in 2007, so there may be another bug introduced since!)

With the subscriptions I'm not really able to offer a lifetime license, as the vector maps and TMJ website generate ongoing costs that I have to cover. However a reduced renewal price does sound reasonable, I'll bear this in mind!

Cheers,
Stephen

 

ChrisM

21:45
Wednesday
10-Feb-2010

Location:
Bedfordshire, UK

Phone Model:
Sony Ericsson W770i, Blackberry Playbook(?)

Hi Stephen (and all)

With regards max altitudes, if you like, have a look at my 'FlightHome' track. That shows points in excess of 10,000m recorded in May2009. That agrees roughly with the cruising speed that the captain said we would be at, so I don't think anything was being truncated there. I guess this suggests there was no lim it imposed by TMJ then...(?)

Regards,

Chris.
 
 

ianmanton

22:05
Friday
12-Feb-2010

Hi all

Incredibly frustratingly my return flight was today but I couldn't get GPS to find any satellites at all until I stepped out the airport back home! SImply refused to find anything in any application including the setup/advanced setup/gps main BB menu.

Rather annoyed as I can't figure out why it wasn't finding anything! Wondered if it was as I had mobile network off but it had no problem locking on on the way out nor when i tested that scenario later on to check.

Nonetheless you were right about the data being shown in the track file itself being capped at 6054 rather than the graph being the problem. Bizarre - guess we'll need someone on a plane with a blackberry to test further at some point!

Cheers
Ian
 

(You must be logged in to post a reply to this thread)