My 2011 335xi's navigation system is really pathetic.
It frequently tries to take me to a correct address but leaves me a mile or more from the actual location.
Twice it has changed the town I entered (via bluetooth). I found the correct addresses in my address book, selected them, the BMW screen showed the correct towns, but when I selected "Start Guidance" the car changed the town --- once to the next town, once to a town 15 miles from where I was going.
Is this typical, or should I ask them to reinstall the software?
On earlier models the Nav used signals from the ABS sensors to work out the car's position and it would get confused if there was a problem with the connection.
Check whether you are using predictive text for data entry.
The car shows the correct position, but it has the wrong numbers for some streets.
The problem with the wrong towns was for data entered via the address book on my bluetooth phone. Initially, when finding the desired address in the list the town is correct, only after doing the final "start guidance" does the system change the town on me.
Can I do an engineer's reset?
I think I need to have the whole software system reinstalled.
Wrong numbers - you mean house address? These can be out a bit, especially where there are shops and the numbers are not shown.
If I enter Manchester, Wilmslow Road 700, then once I select it it changes the town to "M. Didsbury". This is not an error, but reflects that this address is in Didsbury and that Wilmslow Road runs through several localities (including some not in Manchester) in its several mile length.
This is techie speak for remove its fuse, wait several seconds and re-insert it. This will re-initialise the software and also force a [cold] first fix on GPS, which takes 30 minutes or so. Subsequently it will do warm fixes based on the known time and the fact the car rarely moves with the ingnition swictched off (but expect a wait after an overnight ferry trip).
Phones are quicker because they use the location of the nearest mast to get an approximate location first.
Unlikely, as it would probably crash if the program were corrupted.
Hmmm... There's no reason, other than a very poor-performing antenna, for a GPS fix to take 30 minutes. About 1 minute should suffice.
My company makes products that have GPS receivers in them, and typically they will pick up 7 or more satellites in the first couple minutes. (Only 3 are needed.)
GPS antennas in phones are not ideal, but normally adequate to get 8 satellites in a reasonable location.
Three suffice if you are on the surface (typically) of the sea, four otherwise for a full fix of x, y, z and t.
A first fix from cold (not knowing the approximate location or time) can take up to 30 minutes.
I had a work Blackberry, which seemed not to remember where it was or use the fairly accurate time. It could take up to ~30 minutes to get a fix, but then would be OK as long as the nav was on.
I mean it gets me to the correct street, but thinks I am at the correct number when I am actually a full mile away. (That was the worst case, but it is been 1/2 mile off a few times).
No, this is not that sort of case. Once it tried to take me to a town a full 15 miles away from the actual town, with several other towns inbetween.
Sounds like something to try.
But the GPS is not inaccurate, the navigation is. The screen does show my correct location, but the software often takes me to the wrong place.
It was a Blackberry, however using Clear Acquisition signals this is what you get.
The theoretical limit from a receiver doing a cold start is 12.5 minutes, however most basic receivers will wait for the start of the almanac, which might be up to 12.5 minutes later.
That is not my experience, with brand-new, never-booted-before products. Two minutes. Where is this theory coming from? You find the sattelites, and do the math, and no, the math doesn't take minutes.
" At a transmission rate of 50 bps, this gives 750 seconds to transmit an entire almanac message. " and if you have no idea where you are or what time it is then you need the entire almanac message.
Once you have that you can get a fix to within a few metres and 10nS.
I will guess that dizzy's "new" deveices do know the time wrt to UT to within less than a minute, because the clock will have been set at the factory or may even separately receive Rugby or similar.
Oh, no, they have an internal crystal-controlled clock that they set against the gps time when it's available and use as a base reference when it's not. Trimble first started doing that with a rubidium standard back in the late nineties but now everybody does it with a cheap crystal and an algorithm that calculates average drift and then corrects for it.
It's a trick but it's a pretty good one and it works well if you shut the thing off for an hour or so while you're in the supermarket. If you leave it shut off for six months it'll take a lot longer to sync up.
This stuff is really amazing, I never thought I'd see something like this in my lifetime. I remember watching navigators take celestial fixes through a little dome above the cockpit on transatlantic flights... we have come a long way...
I have a watch that does this - corrects itself every night AND recalculates the drift since the last sync. Cheap and FAR better than any fancy Swiss movement.
I did, but only because I was reading copies of "Navigation" during my lunch time while working at some laboratories in the early 80's. Perhaps the most quaint quote at the time was that with mass production one supplier (Texas Instruments) could see the price of a receiver falling to as little as $5,000 (~£10k today). As we know mobile phones with GPS are now less than £100 and a bare module about £10.
More than a "few". The math is not *that* huge, considering the compute power readily available.
Like one second, because the entire GPS subsystem is one little multi-chip module about 2 cm^2 in size that does all the work and reports the results. Dedicated logic indeed.
Hmm... This article supports the longer "cold" or "factory" time to first fix that you guys have been claiming (about 15 minutes).
formatting link
That doesn't match my industry experience, so I'm wondering if a faster but less accuarate postion may be reported before all of the Almanac data is acquired.
MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.