Hi! We see you’re using an ad-blocker. We’re fine with that and won’t stop you visiting the site.
But as we’re losing ad-revenue from this then why not make 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!
Joined: Mar 03, 2006 Posts: 7191 Location: Reading
Posted: Sat Apr 23, 2011 7:19 am Post subject: Garmin Proximity Alerts
I'm in the process of creating a program to write out .gpx files to override the Garmin default of 36 seconds as this setting pulls in cameras on adjacent roads and is therefore too broad.
I've cracked 99% of it and I have a working version.
My final problem is trying to understand how the POI Loader deals with proximity alerts.
Having read various articles I initially put 'feet' into the data field as that is the POI Loader setting I was using. This was clearly wrong as I was getting warned of cameras up to three quarters of a mile away!
I then read a few more articles and it seems that POI Loader uses Metres by default even though you have selected 'MPH and Feet'. This setting was more accurate but still not right as the alerts were now too close.
Next step was to use yards instead of metres which is better but still not right.
Example:
I've set my' Approach' figure to 15 seconds. Travelling at exactly 30 mph approaching a 30 Gatso, as soon as the alert triggers I've timed the seconds and it's 12, not 15. similarly at 50 mph the alert is triggering slightly too close.
My calculation is based on 15 seconds at 30 mph distance covered = 0.1250 of a mile = 220 yards or 201.168 metres.
= POI Loader has a mind of its own
I'll get there _________________ DashCam:
Viofo A119 V3
Joined: Sep 06, 2006 Posts: 1618 Location: East Hertfordshire
Posted: Sat Apr 23, 2011 9:44 am Post subject:
Just a thought - have you tested the Garmin 36-second default in the same way? Wondering if that's correct or a little bit out as well? _________________ David
(Navigon 70 Live, Nuvi 360)
Joined: May 04, 2008 Posts: 754 Location: Knaresborough, North Yorkshire.
Posted: Sat Apr 23, 2011 11:42 am Post subject:
Just a thought - have you tried the Ash10 CameraManager utility? This allows the user to set warning times (and overspeed warning times) for each type of camera.
And yes, the values produced by Ash10 in the gpx files do not exactly tally with those expected by simple maths - there is clearly something more to the calculation that I haven't grasped.
Joined: Dec 07, 2006 Posts: 564 Location: North Devon
Posted: Sat Apr 23, 2011 2:34 pm Post subject: Re: Garmin Proximity Alerts
Kremmen wrote:
My final problem is trying to understand how the POI Loader deals with proximity alerts.
I then read a few more articles and it seems that POI Loader uses Metres by default even though you have selected 'MPH and Feet'.
For some reason, Garmin used the weird '@speed' construct in GPX files and then needed to add a dialogue to find out what units it is in. The other entries in the GPX file are in metric units and the POILoader setting has no effect on their interpretation. (Why Garmin did this, is one of life's little mysteries )
and he wrote:
I've set my' Approach' figure to 15 seconds. Travelling at exactly 30 mph approaching a 30 Gatso, as soon as the alert triggers I've timed the seconds and it's 12, not 15. similarly at 50 mph the alert is triggering slightly too close.
Could this just be due to inaccuracies in the way you've measured exactly 30mph and 12 seconds? Speedo error and counting 'elephants' or 'mississippis' might do it
It might even be a firmware bug in your device...
Quote:
My calculation is based on 15 seconds at 30 mph distance covered = 0.1250 of a mile = 220 yards or 201.168 metres.
Looks good to me...
Quote:
POI Loader has a mind of its own
Correct! (and Garmin sometimes introduce undocumented changes between versions)
Last edited by PhilHornby on Sun Apr 24, 2011 7:08 pm; edited 1 time in total
Joined: Mar 03, 2006 Posts: 7191 Location: Reading
Posted: Sun Apr 24, 2011 6:35 am Post subject:
Thanks for the feedback
As far as speedo accuracy is concerned I used the Garmins own speed on a few occasions which differs from my cars digital speedo by circa 2mph at 50 so hopefully I can rule that one out.
I have considered other existing apps but being a developer I want to see if I can do my own.
I have looked at the data of others and I suspect the same may occur as the proximity settings are roughly the same?
I've tried 2 different 'name' formats, one with the @ and one without and that doesn't make any difference as it looks like the 'proximity' setting holds the key.
Once I've cracked this item with a % add-on my next project is to try and remove some of the 'duplicates' where there is a camera on both sides, being as Garmin have omitted a directional facility from their Loader.
On the M25 for example the cameras are on adjoining gantries. This triggers 2 warnings. What I'm finding is that the 'alerts' seem to get stacked up. A second alert will not trigger until the first has finished. This seems to have the effect of sometimes producing the second warning when your right on top of the camera or even just past it.
An interesting project that may take a little time to perfect.
Joined: Dec 07, 2006 Posts: 564 Location: North Devon
Posted: Sun Apr 24, 2011 7:07 pm Post subject:
Kremmen wrote:
As far as speedo accuracy is concerned I used the Garmins own speed...
You could try using the Nüvi in 'simulation' mode, rather than trying to measure this for real. You have no control over the speed the simulation runs at, so you may have to adjust the speed of your alerts to be lower than reality for the purpose of your tests. (I just tried a Nüvi 250W, which decided to 'drive' @28mph, thus not triggering my 30mph alerts (which are set for 35mph anyway)).
The quick test I did on the 250W, revealed that the 15 sec warning I had asked for, was being honoured. (I use ASH10's Camera Manager).
and he wrote:
I've tried 2 different 'name' formats, one with the @ and one without and that doesn't make any difference as it looks like the 'proximity' setting holds the key.
I may have misunderstood what you're saying there, but the two types of alert are definitely different. When the "@" is specified, you have to be exceeding the specified speed for the alert to trigger. The contents of the GPX {proximity} field (or the 36 sec rule) determine when the alert is considered active. When "@speed" is not present, the alert is triggered purely by distance (measured along the road, or roads making up the route).
There are some caveats: different units behave slightly differently; different firmware versions sometimes change things and some units let you set things like 'continuous' alerts. You get strange effects if you slow down and speed up within the proximity distance - such as approaching red lights - the alerts tend to re-trigger repeatedly. If alerting POIs are not within some arbitrary (undocumented) distance from your route (75' ?), they don't alert.
Last edited by PhilHornby on Mon Apr 25, 2011 6:40 pm; edited 1 time in total
Joined: Mar 03, 2006 Posts: 7191 Location: Reading
Posted: Mon Apr 25, 2011 6:12 am Post subject:
Thanks for that.
I have indeed noted some strange behaviour. The M25 variable alerts were almost continuous when I had the @0 within the file. When I replaced this with #0 they were less active, which is what I wanted. I've since replaced the #0 with #70 and I'm just getting the single warning. I've also noticed the behaviour you mention when you slow down or speed up within the proximity. Also, I get a Redlight warning as expected. If I stop before the lights then as soon as I set off again I get another warning. I suspect this is by design as a further reminder.
Looking at Ash's file his proximities are more or less the same as mine. His are lower, so as you say, it looks like it's firmware related. As an example, Ash has '360' for 50mph whilst mine is calculating '366' so I would expect mine to be slightly longer.
I have some rework and testing to do but I'll get there in the end. At least I know the proximities are being read in after my 'feet' goof.
One thing I have noticed, even when I was using the neat .csv file on my 2200 is that I always get the approach warning, regardless of speed, unlike some reports of no warnings at all when under the limit. I wonder if Garmin have responded to criticism?
I have downloaded the latest 3.0 firmware.
Thanks for the simulation mode tip, never thought of that _________________ DashCam:
Viofo A119 V3
Joined: Jun 04, 2005 Posts: 19991 Location: West and Southwest London
Posted: Mon Apr 25, 2011 1:29 pm Post subject:
The speed data is recorded for "plain vanilla" Redlight cameras (as opposed to combined Redlight and speed), but it isn't necessarily used.
I'm not sure if Garmin/Ash's program uses the speed number in the individual camera's name or not, but other brands just have all the Redlight cameras in a single file, so they would always have the same warning distance irrespective of the speed limit in force.
"Red/speed" cameras are split up into the individual speed-limit categories though.
There is the argument/opinion that while it is possible to miss a well-hidden camera through slight inattention, there's absolutely no excuse for missing a set of traffic lights at red, so we are lucky to have the ordinary red-light cameras in the database at all!
I have indeed noted some strange behaviour. The M25 variable alerts were almost continuous when I had the @0 within the file. When I replaced this with #0 they were less active, which is what I wanted.
I have a copy of Garmin's GPI_Viewer tool (obtained as part of the Content Toolkit and sadly, not for re-distribution). This makes it easier to see what makes it into the .GPI file, using various input parameters.
POILoader (or at least V2.6.0.0 that I'm using), is clever enough to know that "Name@0" is nonsense - it doesn't set an alert of any sort, based on a min. speed of zero. The same is not true of {proximity}0{/proximity} (even though it's kind of wacky, this sets a distance-based alert at exactly 0 metres (or feet ;-).
In the absence of GPI_Viewer, you can tell from the default audio if you're hearing a speed or distance-based alert. Distance alerts use "Passive Tone", which is a lower-pitched "Bong,Bong". Speed alerts use "Aggressive Tone", which is a higher-pitched "P-i-n-g".
Note: when experimenting with any of this, beware the use of 'magic' words (and numerals) in your POI file names (Gatso, Redlight,Speed etc). These will silently override all your settings! (Eye-of-toad, leg-of-newt programming: giving different results depending on the phase of the moon & state of the tide ;-)
and he wrote:
Looking at Ash's file his proximities are more or less the same as mine. ... As an example, Ash has '360' for 50mph whilst mine is calculating '366'.
EDIT: FWIW, I calculate that a setting of 366metres gives a delay of 16.374 secs and 360 gives 16.106 secs @50mph
(The uncertainty in the computed GPS fix is probably more than 6 metres anyway...)
I can tell you, that left to its own devices, when POILoader encounters @50 (in Imperial mode), it writes a proximity of 805m and a speed of 2235cm/sec to the .GPI file. (ie it defaults to 36.018 seconds).
Quote:
One thing I have noticed, ... is that I always get the approach warning, regardless of speed, unlike some reports of no warnings at all when under the limit. I wonder if Garmin have responded to criticism?
I've seen that effect on some unit or other in the past. Personally, I don't like it, because that's not what they document it as doing...
Joined: Mar 03, 2006 Posts: 7191 Location: Reading
Posted: Tue Apr 26, 2011 6:39 am Post subject:
Thanks
Your simulation suggestion was spot on. I did a few 'runs' across known cameras and in all cases where the simulation was at the prevailing speed limit the timing was 16 seconds. This was achieved under better circumstances as I had my eyes on the unit instead of the road and I used a proper stopwatch. Based on this I'm going to back off to Metres and re-test.
I'm happy with this and I can move forward and as I said I'm going to try and cut out duplication where the same camera type is on opposite carriageways thus avoiding the double warnings.
ps: does the .gpi interrogator give any clues to the Cyclops directional info ?
Slowly but surely, onward and upward........ _________________ DashCam:
Viofo A119 V3
Joined: Mar 03, 2006 Posts: 7191 Location: Reading
Posted: Thu Apr 28, 2011 3:51 pm Post subject:
Program output now in final testing and "the jobs a goodun"
Proximity alerts now behaving as expected.
Ability to remove most duplicate warnings caused by Garmin not allowing directional POI's via POI Loader.
Built in unzip feature.
Approach and/or Overspeed alerts.
[/list][/list] _________________ DashCam:
Viofo A119 V3
Posted: Today Post subject: Pocket GPS Advertising
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!
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
Or you could disable your ad-blocker for this site. We think you’ll find our adverts are not overbearing!
Hi! We see you’re using an ad-blocker. We’re fine with that and won’t stop you visiting the site.
But as we’re losing ad-revenue from this then why not make 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!