Google Maps GPS problems

Oh, there's absolutely a general Galaxy S GPS issue. The question here is did it get addressed since Verizon released their device months most everyone else. I haven't seen any problems on my end. Others are saying they are having problems. With the other Galaxy S devices, they were ALL experiencing the same issue. The Fascinate seems to have some working, and some that don't.

I just booted my DINC for the first time in a week or so. My SF GPS works fine, but compared to the DINC its just so sloooooow in aquiring and locking on. Its not in the same league. I hope they can fix it with a software update.
 
GPS lock-on is very slow on the Fascinate.
I don't use it much, but I needed it yesterday and it was borderline annoying waiting for the lock.
It's alright though, nothing is perfect.

[tapatalk on android]
 
I don't think they'll ever work the same. The Incredible uses a Qualcom chip w/ integrated GPS. Samsung uses a Broadcom GPS chip in the Galaxy S devices (in fact, Broadcom's newest). This chip is supposed to be very energy conservative, so perhaps we're seeing a bit of that going awry, with possible delayed wake-up times.

It seems that most GPS issues with other Galaxy S devices were problems relating to positional accuracy. From my quick research, the BCM4751 should be able to get down to 2m accuracy. I have seen that accuracy, consistently, with my device (given optimal conditions).

I think it's more likely that the accuracy issue has been addressed, but that in doing so, something else has cropped up where many people are seeing very long-lock on times.

Just a thought.
 
Firstly, I'm not suggesting you do this, but just reporting it as something that worked for another person.

Ripped out the battery, rebuilt my roaming profile (*228 -> 2). Now I'm sitting on 9 fixed out of 10 known satellites, 6 meters of accuracy. Snappy as hell to locate, too.

Seems like this might improve aGPS in some manner. I've done it on mine, but my GPS still works fine, as it did originally.
 
Firstly, I'm not suggesting you do this, but just reporting it as something that worked for another person.



Seems like this might improve aGPS in some manner. I've done it on mine, but my GPS still works fine, as it did originally.
I was wondering if it was an power conservation thing since I've read the Dinc is horrible with the battery and that the SF is much better in that regard. I don't get that GPS icon in the notification bar unless/until I open Maps/Navigator, even if GPS is 'ON'.

Could you dumb that procedure down for the fancy TV challenged? What is/How do you 'rebuild the roaming profile'? Mine always seems to work (even sitting in my living room) but it is often slow to lock (even outdoors).
 
I was wondering if it was an power conservation thing since I've read the Dinc is horrible with the battery and that the SF is much better in that regard. I don't get that GPS icon in the notification bar unless/until I open Maps/Navigator, even if GPS is 'ON'.

Could you dumb that procedure down for the fancy TV challenged? What is/How do you 'rebuild the roaming profile'? Mine always seems to work (even sitting in my living room) but it is often slow to lock (even outdoors).

It's about as dumbed down as it gets, but I can see the confusion.

Dial *228, choose option two. When it's done (~10 seconds), reboot the phone. Go outside, test GPS. :)
 
I already posted some of this above, but I will expand on it a little bit.

The first thing I did when I received my Fascinate, was open it up and enable all 3 GPS settings. I installed Google Maps, opened Nav, and went outside and watched as the phone connected to Verizon's data network and quickly grabbed my AGPS position. With AGPS out of the way, the GPS icon continued to blink as it searched for satellites. This is the part that took forever. Forever being ~8 minutes. When it locked, it locked well and I was able to navigate without any issues. I exited Google Maps and left all 3 GPS settings enabled. I went inside for a few hours, and then went back outside. Turned on Google Nav and again it found me via AGPS first followed by my satellite based position shortly thereafter.

The next day, I got in my car, opened up Nav, and typed in my destination address. Once again, it quickly found my AGPS coordinates. However, I waited ~5 minutes to get a satellite lock.

Tuesday, I flew up to Edmonton & took both my Droid X and Fascinate with me. I picked up my rental car, drove outside the garage and pulled to the the side of the road. I opened up Nav on both devices and typed in my destination address. The Droid X locked my satellite coordinates in ~1 minute. The Fascinate took ~8 minutes. I could have been half way to my destination in the time I sat there waiting for the Fascinate to lock. That said, it did lock, and the navigation continued to the hotel without any issues.

About 36 hours later I grab my car from valet, open up Nav on both the X and the Fascinate, and punched in Edmonton Airport. This time I did not sit and wait for either to grab the satellite coordinates. As I started to drive, both the Fascinate and X grabbed my AGPS. Seconds later the X was using Satellite based coordinates to Navigate me toward my destination. The Fascinate took almost 10 minutes. I couldn't believe how far in my drive I had gotten before it actually started working.

When judging how well - or if - the GPS works properly on this phone, you need to take several factors into consideration:

1) Is it able to grab your AGPS position quickly? I would say in my case, yes it can. This will work both indoors and outdoors - and is based off of information the Fascinate is getting from the network I am currently connected to.
2) Does AGPS accurately report your location? The answer to this question is going to vary wildly. Sometimes the calculation will be extremely accurate, other times you could be several blocks away. There is nothing exact about AGPS.
3) Are you able to get a satellite lock quickly? In my experience, you will never get a satellite lock indoors, because you need a clear line of sight to the sky. That is why you want to enable AGPS (for apps that need a general idea of where you are - like weather apps), and Satellites (for apps that need a greater level of accuracy - like Nav). If you currently have Navigation open (not maps - but the actual Navigation function), and your GPS icon is blinking, you DO NOT have a satellite lock, and turn by turn Nav is completely useless.

Now, if we have people reporting that they are getting all of the above to work flawlessly, no matter where they go, and after days have passed between use, then those of us with problems have a hardware vs. software issue, and it will never be addressed IMHO.

So, when people say theirs works perfectly, I'd be curious if you are making a distinction between AGPS and a satellite lock. Again, if you are in the Nav app, and your satellite icon is blinking, you do not have a satellite lock. The maps will say "searching for GPS on it".
 
Last edited:
Having worked with GPS for nearly 10 years there are a few things that people must understand.

1) The initial GPS lock (cold boot) will take up to 12 minutes to lock however some manufactures pre-populate generic almanac data in the factory that will take that time limit down to the 3-5 minute range. I don't believe that Samsung does this.

2) GPS doesn't work indoors (read that again if you need to). Those of you getting a "GPS fix" indoors aren't using GPS satellites. It is using triangulation of cell phone signals from the cell towers or wifi signals (aGPS). Based on this fact, if your house is only covered by one or two towers triangulation is NOT going to work and you are NOT going to get a lock - it's never going to happen!!! Those of you in metropolitan areas with lots of towers will get near pinpoint accuracy using nothing but the cell signals. Newer chips can also use wifi signals to locate you. For example, I've been using the same SSID for 8 years at my home. When the Google Street View car drove by my house they logged the SSID of my wireless router (remember there was a big controversy about this) and geotagged it. I can lock instantly in my basement with wifi on thanks to this.

3) So why doesn't the Fascinates GPS work as well as some others? - Well, if I knew that I'd probably be working for Samsung by now. That said I suspect that the Fascinate (and other Galaxy S phones) have a bug in how they use almanac data and/or how they use aGPS (cell and wife signals) to populate their almanac data.

Commercial GPS chips (without aGPS assistance) even with near perfect almanac data can still take close to 30 seconds to lock. The farther away you are from the last almanac on next power up the longer it's going to take to lock. Once you travel ~50 miles away it could take 4+ minutes to get a lock.

GPS devices with aGPS, such as our phones, can use aGPS (cell / wife data) to "seed" the almanac data to greatly increase lock time with the satellites. I suspect this seeding of the almanac data or use of that almanac data in obtaining a lock may be where the Fascinate has a bug.

Another key thing to understand is that in order to use cellular / wifi triangulation your device MUST know the GPS coordinates of the towers / routers. This information is obtained from a server over the 3G connection. I've noticed that some of the "GPS fixes" posted on the internet are changing the server that this data is obtained from (it appears the change is from Spirant to Google). Better / faster access to this aGPS data will obviously improve lock times. Also, more accurate aGPS data will also improve lock time. I don't know if the Fascinate is able to store any of the aGPS tower / wifi router location data locally or if it has to re-download it every time. If it's not able to keep a local store then this could also increase lock time as it has to re-populate the aGPS data.

In short there is a lot going on "behind the scenes" when using aGPS and there is a lot that can go wrong. Fortunately, in the case of the Fascinate it appears that it has a fully functional and very accurate GPS chip when just using straight GPS from the satellites. Unfortunately, when aGPS is thrown in (which is software based) things don't appear to work so well. My only concern would be if there is a bug in how the Fascinate (and other Galaxy S devices) use their almanac data it could be a lot more difficult to fix as it may be in the GPS chips firmware (which would still be a software bug) but flashing the firmware on the GPS chip via a OTA update (while possible) is anything but easy and Samsung doesn't want to have to replace thousands of phones due to bricking the GPS chip.

Hope this helps clear up how GPS and aGPS works and lends some possible insight into what may be affecting the GPS operation of our phones.

Martian
 
Last edited:
Now, if we have people reporting that they are getting all of the above to work flawlessly, no matter where they go, and after days have passed between use, then those of us with problems have a hardware vs. software issue, and it will never be addressed IMHO.

So, when people say theirs works perfectly, I'd be curious if you are making a distinction between AGPS and a satellite lock. Again, if you are in the Nav app, and your satellite icon is blinking, you do not have a satellite lock. The maps will say "searching for GPS on it".

Good question. Since I've received my second unit, I've only tested the GPS once, on my roof, for it to get the almanac, lock on, etc.. It's been off since then (~3 days ago).

Tonight I'll walk outside and immediately turn on the GPS, GPSTest and MyTracks to record the walk. We'll see what happens.

Just to confirm, if you're testing straight GPS, WiFi should be off, and the only option checked of the three (in settings) should be "Standalone GPS Services" All other location services should be off (unchecked).
 
The slow lock or instability is what really bothers me. I do some activities that can take me into unknown territory on foot or skis. The behavior of the SF makes me a little nervous about it finding me if I really need it to. Most of the time it works OK but it is a tool I might need to get me out of being lost and , I kind of wonder.

Brandon, that is good info as to how the GPS is integrated into the cpu on the INC but the SF has an off-board setup (not sure these are the correct terms).
Maybe this is as good as it gets then?? That would be a concern for me.

Also, if anyone knows, is there an app that would let me shut down the radio to extend battery life and use the GPS / Nav section only? I was testing the SF before activating it, and the GPS worked as long as I stayed within the map I loaded via wifi.
 
In short there is a lot going on "behind the scenes" when using aGPS and there is a lot that can go wrong. Fortunately, in the case of the Fascinate it appears that it has a fully functional and very accurate GPS chip when just using straight GPS from the satellites. Unfortunately, when aGPS is thrown in (which is software based) things don't appear to work so well. My only concern would be if there is a bug in how the Fascinate (and other Galaxy S devices) use their almanac data it could be a lot more difficult to fix as it may be in the GPS chips firmware (which would still be a software bug) but flashing the firmware on the GPS chip via a OTA update (while possible) is anything but easy and Samsung doesn't want to have to replace thousands of phones due to bricking the GPS chip.

Thanks Martian. I've noticed that people having issues all seem to be using aGPS as well. I always use straight GPS, and I've never had a problem.

That being said, it has been documented that this new Broadcom chip (which replaced the BCM20751 on the older Galaxy S units, and coincidentally, why we don't have an FM radio), the BCM4751, is supposed to be super energy efficient. I'm still feeling like their might be some wake-up / idle timing issues associated with it at the driver level.
 
My only concern would be if there is a bug in how the Fascinate (and other Galaxy S devices) use their almanac data it could be a lot more difficult to fix as it may be in the GPS chips firmware (which would still be a software bug) but flashing the firmware on the GPS chip via a OTA update (while possible) is anything but easy and Samsung doesn't want to have to replace thousands of phones due to bricking the GPS chip.

Your entire explanation was a much more articulate/exact way of saying what I was trying to say, so thank you.

Regarding the specific quote above, this is a serious concern of mine too. I'm putting my faith into this list, in my hopes a solution may be found. GPS really is a big deal in my line of work. I absolutely love this phone overall, but the GPS is a nightmare when I need it.

Good question. Since I've received my second unit, I've only tested the GPS once, on my roof, for it to get the almanac, lock on, etc.. It's been off since then (~3 days ago).

What the hell were you doing on your roof? lol :D
 
The slow lock or instability is what really bothers me. I do some activities that can take me into unknown territory on foot or skis. The behavior of the SF makes me a little nervous about it finding me if I really need it to.
I'm confident it will always find you via satellites, so long as you have an open view of the sky, it just might take 5-12 minutes.

Brandon, that is good info as to how the GPS is integrated into the cpu on the INC but the SF has an off-board setup (not sure these are the correct terms).
Maybe this is as good as it gets then?? That would be a concern for me.
No need for concern. The only advantage of moving the GPS circuitry on-die or on-chip (not sure which it is with the incredible) is space savings and potentially a little (very little) power savings. GPS typically only updates once per second (ie. 1 Hz) so there is no speed increase. In other words, GPS is just as accurate (or inaccurate) if the chip is on-die, on chip, or off (ie. separate chip).

That being said, it has been documented that this new Broadcom chip (which replaced the BCM20751 on the older Galaxy S units, and coincidentally, why we don't have an FM radio), the BCM4751, is supposed to be super energy efficient. I'm still feeling like their might be some wake-up / idle timing issues associated with it at the driver level.
Very good information here. This is the first I've heard what chip the Fascinate is using. Hopefully I can find a data sheet! Yes, it is entirely possible there is a bug with wake-up / idle but that is true for any device isn't it. I just encountered a bug in a updated firmware for a different chip that resulted in the chip occasionally failing to send a proper initialization string on power-up or reset. (gotta love intermittent problems) Had to work around it in software.

Regarding the specific quote above, this is a serious concern of mine too. I'm putting my faith into this list, in my hopes a solution may be found. GPS really is a big deal in my line of work. I absolutely love this phone overall, but the GPS is a nightmare when I need it.
I wouldn't worry just yet. Folks over at XDA have obtained a leaked update from Samsung for some of the other Galaxy S phones and manage to pull out some GPS related fixes that according to reports have "solved" the "GPS issues". My guess is the fixes are known, simple, and straight forward for the most part. Pushing them to millions of phones all over the world over various cellular networks is not so simple and requires lots of testing at both Samsung and each of the carriers, which is why fixes are so slow in coming.

Martian
 
Having worked with GPS for nearly 10 years there are a few things that people must understand.

<snip for succinctness>

Martian

Just a few questions... so the tower based triangulation is accurate down to the sub-3m range? Basing my experience on phones with fully functional aGPS I haven't seen this to ever be the case. The useful data gathered from the tower is really just the location of the tower and the almanac data. The phone can download that, and if it happens to be in view of multiple towers (and no satellites) it will strike a rough balance between them to better estimate your location. AFAIK there isn't sufficient technology (namely atomic-grade clocks in the towers and a terrestrial band receiver in the GPS chip) to allow for true distance based triangulation using just the towers.

Is that not the case?
 
You are correct in that you are not going to get 3m resolution from Cell triangulation but you can often do a lot better than "I'm somewhere in range of this tower". In some cases as good as 50m resolution but in other cases (rural) it could be an area of several kilometers.

You are also correct in that the cell towers do not have "GPS grade" clocks, but they do have pretty accurate clocks (you need pretty accurate timing when you transmit and receive at 850 MHz 1900 MHz). Therefore Time Of Arrival based triangulation can actually be fairly accurate, not sure on the exact number.

Also, you can get a fairly decent idea of your distance from the tower based on Receive Signal Level (RSL) alone (the further away you are from the tower the lower your Receive Signal Level - Given a know transmit power - which is available from the FCC, and the known receive level you can calculate distance based on signal attenuation - inverse square law). Having an approximate distance from tower of 3+ towers should be able get you within a few blocks radius, add in Time Of Arrival based triangulation and things start looking pretty good.

I don't have specific experience with Cellular based triangulation but I do know you can approximate the distance between a radio transmitter and receiver with surprising accuracy based on signal level.

Hope this information is helpful (or at least not boring)

Martian
 
Not to mention, the tower can usually provide pretty detailed information regarding itself besides location. For example, knowing if that tower is using a directional antenna versus omnidirectional can help quite a bit.
 
Now that I'm back in Seattle, I tried another test today. This time I went into settings and disabled both VZW location services and Google location services, and only used Satellites. This time it took about 4 minutes to lock. I'm not sure if it was a fluke that it was better, but it did go faster this time.

That said, Google NAV continued to say that it was searching for GPS, even though the icon had stopped blinking. I had to exit NAV and come back in again. In this case, it's probably better to use that Quick GPS program to grab the satellites, and then open NAV once it's done so.

There's something buggy going on with Google NAV/Maps and the GPS.
 
Now that I'm back in Seattle, I tried another test today. This time I went into settings and disabled both VZW location services and Google location services, and only used Satellites. This time it took about 4 minutes to lock. I'm not sure if it was a fluke that it was better, but it did go faster this time.

I don't think it was a fluke. I've noticed similar behavior, with lock coming more quickly only using the GPS services.
 

Latest posts

Trending Posts

Members online

Forum statistics

Threads
957,453
Messages
6,973,109
Members
3,163,815
Latest member
oCh33sYo