Gps performance (sirf 3 vs qualcomm)

[Barkeeper1979]

Hello,

I have a question. I read that in the diamond and the touch pro htc doesnt use the sirfstar 3 anymore but the qualcomm gps receiver.

The sirf 3 is a very good receiver but i dont have any experince with qualcomm. I read a review on the internet that qualcomm's performance is less effective than the sirf 3.

(http://www.gpspassion.com/fr/articles.asp?id=175&page=6)

Does anyone have any experience with the new qualcomm in htc's diamond or touch pro?

[Black93300ZX]

Look in the Diamond section, there's plenty of talk of issues with the GPS. No one has the Touch Pro yet since it hasn't released, so no one here could answer your question.

[onesolo]

Look in the Diamond section, there's plenty of talk of issues with the GPS. No one has the Touch Pro yet since it hasn't released, so no one here could answer your question.
But can't you summarize it??

[pokevitek]

So, I had Sirf Star 3, now I have Diamond (Qualcomm navigation). For a car use, Qualcomm performs as well as SirfStar 3, for pedestrian use, it is unusable, due to Static navigation can not be swithced. Diamond least for 2 hrs and 20 minutes of navigation, due to small batery, but like I said, for pedestrian it is unuseable, and in car you can recharge it permanently, so, iGO and TomTom works well. COM4 baud 4600.

[Smaniac]

So, I had Sirf Star 3, now I have Diamond (Qualcomm navigation). For a car use, Qualcomm performs as well as SirfStar 3, for pedestrian use, it is unusable, due to Static navigation can not be swithced. Diamond least for 2 hrs and 20 minutes of navigation, due to small batery, but like I said, for pedestrian it is unuseable, and in car you can recharge it permanently, so, iGO and TomTom works well. COM4 baud 4600.

Is this a software or hardware limitation?

[foo]

for pedestrian use, it is unusable, due to Static navigation can not be swithced.
I have some difficulties to understand that. What is the point with "static navigation"?

[pokevitek]

First, someone thinks it is a software problem, someone thinks its hardware. Anyway, static navigation solves the probem with "jumping" GPS coordinates when staying on one place (in car, coordinates mey be +- 5 meters off), so, navigation will not actualise location in car if speed is under 3 km/s. That is it.

[foo]

Ah okay, I understand.
But this "static navigation" should be handled by the application not by the GPS driver. So the navigation software can and should decide by itself if there is some advanced map-matching applied to the coordinates.
Is this "static navigation" feature only an issue with TomTom or does this apply to GoogleMaps as well?
As far as I can remember my old TomTom 6 application had a pedestrian mode as well. I would assume this "static navigation" feature would be turned off in this case.
Does TomTom 7 have no pedestrian mode or is this really something that is handled by the GPS driver?

[mydexterid]

...
Is this "static navigation" feature only an issue with TomTom or does this apply to GoogleMaps as well?
...
Does TomTom 7 have no pedestrian mode or is this really something that is handled by the GPS driver?

I think you should ask this on the Diamond forum, because almost noone has this phone right now. But if you manage to get an answer, pls let me know, because I'm really interested in this question (I mean the 1st one).

[Nuri58]

Hello,

I have a question. I read that in the diamond and the touch pro htc doesnt use the sirfstar 3 anymore but the qualcomm gps receiver.

The sirf 3 is a very good receiver but i dont have any experince with qualcomm. I read a review on the internet that qualcomm's performance is less effective than the sirf 3.

(http://www.gpspassion.com/fr/articles.asp?id=175&page=6)

Does anyone have any experience with the new qualcomm in htc's diamond or touch pro?

I think you should ask this on the Diamond forum, because almost noone has this phone right now. But if you manage to get an answer, pls let me know, because I'm really interested in this question (I mean the 1st one).

My experience with the Touch Diamond and iGO8 is that there is a time lag (when you stop the car the GPS continues moving 10 - 15 meters) and a lag consistently showing that my position is some 10 - 30 m behind the actual position. At least I have never seen iGO8 showing my position ahead of the actual. The latter becomes a problem when using smart or auto zoom when zoomed to the highest level. In that case it may look like i should turn at the next intersection and not the one I am in, and that is anoying. However, I believe its possible to set max zoom level also in the auto or smart mode somewhere, and that should improve matters (have not taken the time to do so).

However, the article you refer to does tell me that I'll have to live with some sort of issue, but by-and-large I find it is quite OK, especially if I zoom to a fixed level showing a little more than just 20 m road. I may add that iGO 8 in principle only runs on devices with SD support due to copy protection.

[robal]

I tested (on the road) both Diamond and Touch PRO.

Diamond:
Worse performance than SirfIII receivers I've had. No GPS lag. Still, the performance is acceptable.

Touch PRO:
Performance comparable to SirfIII. Got GPS lag though.


I'm sure that the put a better GPS antenna to Touch PRO, since it gets stronger signal, faster TTF and more reliable navigation.
I hope that GPS lag is software issue that will get fixed.

Cheers,

[Nuri58]

I tested (on the road) both Diamond and Touch PRO.

Diamond:
Worse performance than SirfIII receivers I've had. No GPS lag. Still, the performance is acceptable.

Touch PRO:
Performance comparable to SirfIII. Got GPS lag though.


I'm sure that the put a better GPS antenna to Touch PRO, since it gets stronger signal, faster TTF and more reliable navigation.
I hope that GPS lag is software issue that will get fixed.

Cheers,Sounds a bit wierd as TD and TPro is the same processor (and not up to the Sifr chip's performance - but that can be compensated by a well made software). Have you tried it using google map, that could be interesting. Or alternatively put it in pedestrian/off road mode and leave the road.

[danielherrero]

Lag on Raphael is driving me crazy......:confused:

[guitarguy]

Yeah, I'm seeing some pretty nasty lag myself... Just tested it out going from work to home last night, using iGo8. Seems when I stop my car, it keeps moving for a while, speed decreases quite slowly... Not quite as good as my Holux M-1200, but I'll keep testing and see...

Does anyone know/think that a software/radio/rom upgrade might fix this? Or is it just a hardware limitation?

[ukdj78]

I'm not getting hardly any lag using TomTom Navigator 6, seems perfect to me. I did set the baud rate to maximum, maybe that's why or maybe it's a software issue with IGo?

[guitarguy]

I'm not getting hardly any lag using TomTom Navigator 6, seems perfect to me. I did set the baud rate to maximum, maybe that's why or maybe it's a software issue with IGo?

Hmm, I'll try that tomorrow if I get a chance, although I have my doubts it'll make any difference. I was using the same setup on my Elfin, and although there was some very slight lag, nothing like what I see with the Raph....

But I'll try setting the baud rate to max and retry... I'll let y'all know...

[Nuri58]

I'm not getting hardly any lag using TomTom Navigator 6, seems perfect to me. I did set the baud rate to maximum, maybe that's why or maybe it's a software issue with IGo?
Most people report lag time regardless the navigationsoft. Maybe the iGO is a bit worse. I surely observe a speed dependend lag sometimes, but even while waiting in an intersection I sometimes observe that iGO informs up to 20 m to the intersection. Sometimes the turn indicator reads 0 m while the main screen surely shows that I am not yet at the intersection. Never I observerd that iGO showed a position ahead of the true position.

It is known that the TD processor is not as accurate as the Sifr chipset so I wonder if it's because the navi soft is averaging the position locking you to the road. Try to let it track your position in a mode not locking to the road (or use google map) and you'll see that the curser is moving even up to 20 m even the phone is left steady on the table.

[adolfotregosa]

i notice allot of gps lag with garmin mobile xt and little with tomtom 7 but it's there. Also i notice that the gps bars on tomtom or garmin do not maintain fixed, not the signal, that would be normal, but the satellite number. This is very strange when i have a few satellites with a very good signal and suddenly the gps software stops using it, why ??? If i was moving around i would accept that, but since i'm not moving why doesn't it maintain the same satellites ??

Always jumping around satellites instead of maintaining them and this is probably caused by the driver that decodes the gps/satellite information.

I really think that the problem comes from the driver that windows uses for the gps + the radio version.

I'm really confident that future ""HTC"" updates will solve the problem, not cooked roms having has base the actual htc rom(s)

Only time will tell , we will have to wait for the next htc update.

[Nuri58]

i notice allot of gps lag with garmin mobile xt and little with tomtom 7 but it's there. Also i notice that the gps bars on tomtom or garmin do not maintain fixed, not the signal, that would be normal, but the satellite number. This is very strange when i have a few satellites with a very good signal and suddenly the gps software stops using it, why ??? If i was moving around i would accept that, but since i'm not moving why doesn't it maintain the same satellites ??

Always jumping around satellites instead of maintaining them and this is probably caused by the driver that decodes the gps/satellite information.

I really think that the problem comes from the driver that windows uses for the gps + the radio version.

I'm really confident that future ""HTC"" updates will solve the problem, not cooked roms having has base the actual htc rom(s)
Only time will tell , we will have to wait for the next htc update.
Satellites are moving and the atmospharical conditions changing all the time affecting the signal you receive.
The chipset used by HTC is also not performing as well as the the Sirf III having a lower accuraccy, but that doesn't change the signal strength received of course.

[ekerbuddyeker]

Yeah, I'm seeing some pretty nasty lag myself... Just tested it out going from work to home last night, using iGo8. Seems when I stop my car, it keeps moving for a while, speed decreases quite slowly... Not quite as good as my Holux M-1200, but I'll keep testing and see...

Does anyone know/think that a software/radio/rom upgrade might fix this? Or is it just a hardware limitation?

its an IGO8 issue. i believe its programmed in so that if you lose sattelite it makes assumptions about where you are going (since you gave it the end point anyway, its fair to assume that you are continuing in that direction). What i wish they would do is put you 3 secs ahead of where it currently shows you, more often than not it would be an accurate assumption. I love igo8 but in a complicated intersection i often miss the correct turn.

[danielherrero]

Tired about the lag I ve done a little program that connects to GPSand analize GPGGA frames. This frame has a timestamp setted by the gps (not GPS time but the time when GPS generated that frame. Then when the program reads the frame simply compares the current time with that timestamp. This is a example result:
[0X2C46C0DA] - $GPGSV,4,1,15,02,,,,07,,,,08,25,052,,09,17,247,*7B
[0X2C46C0DA] - $GPGSV,4,2,15,10,47,175,,12,,,,15,57,314,,17,00,12 3,*4F
[0X2C46C0DA] - $GPGSV,4,3,15,18,09,319,,19,,,,21,10,298,,24,16,25 4,*4A
[0X2C46C0DA] - $GPGSV,4,4,15,26,69,350,,27,02,060,,28,51,082,*45
[0X2C46C0DA] - $GPGGA,084524.2,,,,,0,,,,,,,,*75
[0X2C46C0DA] - Processing GPGGA $GPGGA,084524.2,,,,,0,,,,,,,,*75

[0X2C46C0DA] - Date: 08:45:24:200 - 08:45:19:000 ->delay:5 seconds

Always 4 or 5 seconds delay (much delay¡¡). I am going to improve it to make a simple GUI where you can select Port and baud rate to make different tests with external/internal GPS/radio roms, etc.
Dani

[Nuri58]

its an IGO8 issue. i believe its programmed in so that if you lose sattelite it makes assumptions about where you are going (since you gave it the end point anyway, its fair to assume that you are continuing in that direction). What i wish they would do is put you 3 secs ahead of where it currently shows you, more often than not it would be an accurate assumption. I love igo8 but in a complicated intersection i often miss the correct turn.I don't think so. iGO does not anticipate for long. This you can verify in a tunnel - arter a few hundred meters it stops moving the curser despite your actions. My Blaupunkt has a gyro + receives speed signal so it continues with no problem. My Mio A701 did not suffer from this problem. So I am quite convinced the problem relates to the chip + iGO programming, the latter to a less extend.

[Nuri58]

Tired about the lag I ve done a little program that connects to GPSand analize GPGGA frames. This frame has a timestamp setted by the gps (not GPS time but the time when GPS generated that frame. Then when the program reads the frame simply compares the current time with that timestamp. This is a example result:
[0X2C46C0DA] - $GPGSV,4,1,15,02,,,,07,,,,08,25,052,,09,17,247,*7B
[0X2C46C0DA] - $GPGSV,4,2,15,10,47,175,,12,,,,15,57,314,,17,00,12 3,*4F
[0X2C46C0DA] - $GPGSV,4,3,15,18,09,319,,19,,,,21,10,298,,24,16,25 4,*4A
[0X2C46C0DA] - $GPGSV,4,4,15,26,69,350,,27,02,060,,28,51,082,*45
[0X2C46C0DA] - $GPGGA,084524.2,,,,,0,,,,,,,,*75
[0X2C46C0DA] - Processing GPGGA $GPGGA,084524.2,,,,,0,,,,,,,,*75

[0X2C46C0DA] - Date: 08:45:24:200 - 08:45:19:000 ->delay:5 seconds

Always 4 or 5 seconds delay (much delay¡¡). I am going to improve it to make a simple GUI where you can select Port and baud rate to make different tests with external/internal GPS/radio roms, etc.
DaniSounds interesting. Keep it up and let's know your findings Better yet let the manufactures know the problem.

[adolfotregosa]

Satellites are moving and the atmospharical conditions changing all the time affecting the signal you receive.
The chipset used by HTC is also not performing as well as the the Sirf III having a lower accuraccy, but that doesn't change the signal strength received of course.

of course but they do not change position 3 in 3 seconds and the same goes for the atmospheric conditions ;)

radio xxxxxx.31 is much better, so it just proves that the problem is being looked into by htc but we have to wait for newer updates

[Nuri58]

of course but they do not change position 3 in 3 seconds and the same goes for the atmospheric conditions ;)

radio xxxxxx.31 is much better, so it just proves that the problem is being looked into by htc but we have to wait for newer updatesI agree to the change in position, but the atmospheric conditions changes in seconds - just look at your TV signal how it can alternate in seconds from perfect to non existent and back to perfect. So with the extremly week signals (contrarely to TV signals) it will change quickly. Try to monitor the signal strenghts (eg GPS tuner) and see for yourself the variation. And since this is common to all GPS systems I don't think that it's a HTC SW update only.

[knight14th]

Tired about the lag I ve done a little program that connects to GPSand analize GPGGA frames. This frame has a timestamp setted by the gps (not GPS time but the time when GPS generated that frame. Then when the program reads the frame simply compares the current time with that timestamp. This is a example result:
[0X2C46C0DA] - $GPGSV,4,1,15,02,,,,07,,,,08,25,052,,09,17,247,*7B
[0X2C46C0DA] - $GPGSV,4,2,15,10,47,175,,12,,,,15,57,314,,17,00,12 3,*4F
[0X2C46C0DA] - $GPGSV,4,3,15,18,09,319,,19,,,,21,10,298,,24,16,25 4,*4A
[0X2C46C0DA] - $GPGSV,4,4,15,26,69,350,,27,02,060,,28,51,082,*45
[0X2C46C0DA] - $GPGGA,084524.2,,,,,0,,,,,,,,*75
[0X2C46C0DA] - Processing GPGGA $GPGGA,084524.2,,,,,0,,,,,,,,*75

[0X2C46C0DA] - Date: 08:45:24:200 - 08:45:19:000 ->delay:5 seconds

Always 4 or 5 seconds delay (much delay¡¡). I am going to improve it to make a simple GUI where you can select Port and baud rate to make different tests with external/internal GPS/radio roms, etc.
Dani

Could you send me this program so that I can test it on my SirfIII Eten X500, just to have some comparative values.

[danielherrero]

Could you send me this program so that I can test it on my SirfIII Eten X500, just to have some comparative values.
Unfortunately I was wrong. The time isnt timestamp but GPS time so It isnt valid for our goal:(

[knight14th]

You can sync the phone's clock with GPS-time using iGo or some other tools.

[danielherrero]

Yes. Then, if no lag the difference should be zero....Tomorrow I will post the program:D

[adolfotregosa]

guys i still think that the problem is the microsoft gps API related!!

download the htcgpstool.exe attached, run it, choose q7500 and open port, go to options and choose satellites, let it get a 3d fix ! when you have a fix (green bars) close the program and repeat the process several times and you will notice that the fix is under 5 seconds.

NOW, try that on any gps program. rarely it will fix in under 5 seconds, it looks like the api always forces a cold start and for me it's the windows API, something has to be wrong because the fix problem does not happen with that tool.

Smt is causing a lag/fix problem and i do not believe its hardware related or htc would have to recall many many touch pros :S

Final, on the htc tool , you can choose mft api and the program will shutdown with a error when you try to open the port.

Well, anyone that makes the test has the same opinion ?? or another theory ?

[danielherrero]

Hi. Here is the promised program.
Unzip it and put the exe in your device and run it.
Usage:
1) Go to a place with where gps is fix is possible.
2) Select port and rate (by default internal gps config, COM4 and 9600 bps).
3) Press Open
4) When the GPS time is available then you will see the delay (in seconds) between the timestamp and current time. Yesterday with and old radio rom I got 5 seconds.
5) The lower line shows bytes read and current real speed transfer. Its simply Bytes read / Test time. So as you can see 4800 bps is more than enought to read the stream correctly.
6) pause button. It stops reading the stream. If you pulse "Continue" button you will see a big delay as the program processes old frames. Finally delay shoulg get the previous value.
7) Close. Close the connection.
8) Exit. Close the program.

In my HTC Touch Pro with the new .31 radio I get 0 seconds (delay). With my kaiser I get 1 seconds.
Hope it helps us the find a solution for the gps lag.
Dani

[Nuri58]

In my HTC Touch Pro with the new .31 radio I get 0 seconds (delay). With my kaiser I get 1 seconds.
Hope it helps us the find a solution for the gps lag.
DaniI get 0 seconds as well (after a pause it shows a number for a split second then 0 seconds again.

[sanderd]

I have a 0sec delay too

ROM 1.90.401 NLD
Radio: 1.02.25.19
Protocol: 52.33.25.17U

I cannot check the iGo/TomTom lag (unless it happens with bicycle too ;)) atm though.
Also, Google Maps gets a fix (7 sats) inside the house after about a minute. Now google knows where I live :mad::p

SanderD

[Nuri58]

Well as long you bike on the roads i trust you could.

[Rabangus]

Today I tested GPS on TT7 on my Raphael side by side with my wife's Kaiser. The version of TT7, maps, poi's and favourites etc were all identical, and both units were running WM6.1.

Both had a bit of a lag, but the Raphael consistently had less lag than the Kaiser when approaching junctions etc.

The Kaiser GPS does get a lock much faster though - that's something I think I'm gonna miss from my old Kaiser.

Rowan

[golympio]

I was able to test the GPS yesterday as I have IGO 8 on my CarPC and IGO 8 on my Touch Pro: there was no noticeable difference between the two of them, and the voice indications were nearly synchronized (maybe 0.2s difference at 60 miles/h, sometimes Touch Pro first, sometimes CarPC first)

[danielherrero]

In my casi It depends of the moment. Sometimes no lag. Sometimes much lag......

[peter7]

Thanks for all your kind advice - it is now workig for me!