Google
  Web www.gpspassion.com


GpsPasSion LIVE!
www.flickr.com
This is a Flickr badge showing public photos from GpsPasSion Live !. Make your own badge here.

www.NaviBlog.com



Versions

Links/Liens




Portal/Portail
Rechercher

- -

Polls/Sondages
Sondage
Pour vous guider sur la Route :
GPS Mobile (SEM)
GPS Intégré
Smartphone
Autre
Voter  -  Résultat des votes
Votes : 2166




Club GpsPasSion
Soutenez le site!

USA: (US$)
EUROPE: (€)
Guide Paypal


GpsPasSion Forums
Home | Profile | Register/Enregist. | Active Topics | Search/Recherche | FAQ
Username:
Password:
Save Password
Forgot your Password?

 All Forums
 English Forums - Hardware
 Garmin nüvi forums
 GPS acquisition nüvi 7xx / SIFv2 AGPS impact ?
 New Topic  Reply/Répondre
 Printer Friendly
Next Page
Author Previous Topic Topic Next Topic
Page: of 17

gpspassion

93884 Posts

Posted - 08 déc. 2007 :  16:00:15  Show Profile  Visit gpspassion's Homepage  Reply with Quote

Updated 20080208 : I'm happy to report that SiRFInstantFixII is really helping my 760 a lot once it's built-up a good set of synthetic ephemeris data. As expected it did take a couple of days of regular use for that to happen, but on two short outings on two consecutive days since, the 760 locked in as quickly as the SiRFinstantFixI enabled TomTom GO920T, even a bit quicker on the second day, which makes sense since the SIFv1 data was 3 days old and is not updated onboard.

Here are some numbers : (same "protocol" as in December, see below)
- TTGO920T with SIFI : 12/12 seconds
- TTGO720T no AGPS : -/55 seconds
- nüvi 760 with SIFII : 12/8 seconds

What a change compared to the non SIFv2 2/3 minute TTFF! While Garmin clearly did not come up with a very "GPS efficient" design for the 7xx series, SiRFInstantFixII largely makes up for it in terms of TTFF, a good thing Garmin chose SiRF and not MTK or their own "bravo" chipset as they are not (yet?) "Auto-AGPS" capable...

I looked at the logs I recorded for my TTFF testing and unfortunately the 920T logs are corrupted so i'll have to them again, but I'm seeing some accuracy issues on the 760, which likely indicates that the quality of the synthetic ephemeris data computed on board is not as good as the "server data" used for SIFv1 and naturally not as good as the live ephemeris data broadcast by the satellites. Now to see if this has an impact on the user experience or not :



Updated 20080131 : While not in the notes for the v2.6 update on the Garmin site, this update adds SiRFInstantFixII to speed up the fix, provided some conditions are met, read about the pros and cons of SiRFInstantFixII in this article in the "Technical forums".

In theory this update should solve the sensivity problems on the 7x0 series that many of us found and that are discussed in the following pages.

How long for SIFII to work ? : in theory 24 hours are needed for the data to be collected and for a full set (30 satellites) of ephemeris to be calculated.

SIFII can populate the ephemerirs gradually as you use it, 2 hours here, 3 hours there, or maybe there is a "silent" mode that populates while the unit is off, I think Centrality had that option for their own "Autonomy" that likely led the way for SiRFinstantFixII. Or maybe it needs to be powered andd be in "sleep mode" like what Anders described in this topic, I think he might have been running an early version of SIFII already as my previous testing with a "standard" firmware did not show his results ;-)

I have turned off my unplugged 760 and placed it next to my reradiating antenna and will wait 24 hours to see what happens.

Posted 12/2007 before SiRFinstantFix v2 became available with FW v2.6


Analyzing TTFF on the map

This has been discussed in the main 7xx user reviews topic but I'm starting a new topic as the TTFF (time to fix) is a real concern to me on the nüvi 760 series.

Granted the 7xx uses the high sensitiviy SiRFstarIII chipset, but you still need to feed gps signals to that chipset, something the large patch antenna of the 3xx and 6xx series that faces up and is at a good distance from the RF interference of the screen and other electonics did well. On the 2xx and 7xx it's a different story...

The "ephemeris" 28db-Hz constraint
Just a quick reminder that the "bottleneck" is not at the tracking level as the SiRFstarIII chipset can track down to 14dB-Hz (the lower the signal level the lower the accuracy though), but to download the live ephemeris data from the satellites, something that requires a strong signal level of 28dB-hz, regardless of how powerful the GPS hardware is. You can overcome this limitation with what I call Internet A-GPS that downloads a 50Kb file containing predictive ephemeris data valid for 7 days, but Garmin have chosen not to implement it so far.

Other "poor man's" tips would be to hold the 7xx outside your window with the back facing up before getting started in the morning or using an external antenna. In both cases it's a constraint that we shouldn't have to put with when the previous generation didn't require it and when the competition does so much better, especially TomTom with free QuickGPS (internet A-GPS) on all it's units (Globallocate Hammerhead on the ONE and SiRFstarIII on the GO).


GO920T, 760, C620T, 660, 250


Side by Side Testing
As usual you can't beat side by side testing to get a good picture of the "GPS issues" at hand, so I did several test runs with this process:
1. place various receivers side by side in my car
3. let them boot up in my underground garage where the signal level is too low (~16dB-Hz) to download ephemeris data
4. start the data logging
5. drive out.
6. time the TTFF
7. analyse logged data

20071208 - FIELD TEST #1
Just did a side by side warm test (4+ hours without using them) by driving out in the suburbs with the 920T, nuvi 760 (with SW 2.20 and GPS FW 3.00s) and Mio C620T ->
- 920T : < 10s (thanks to QuickGPS A-GPS)
- C620T : ~ 45 seconds (standard due to ephemeris download)
- nuvi 760 : 3 minutes and it might have lasted longer if I hadn't stopped at a light.

I would hate to start out in Manhattan with the 760 in the morning, I doubt the 760 would ever get a fix...

Since the hardware design won't change, maybe Garmin will bite the bullet and license SiRFinstantFix from SiRF as TomTom have done. I'm sure they could easily make the download available via WebUpdater.

20071209 - FIELD TEST #2
Back from my test drive, exact same route as yesterday and I'm happy to report that the TTFF on the 760 is somewhat improved, whether that is due to the v2.30 software update or a better satellite "environment" will have to be verified with more testing, will be doing two more test runs, one tomorrow morning and one tomorrow night.

Here are the results :
1. TomTom GO920T < 10 seconds (thanks to A-GPS)
2. Mio C620T = 50"
3. Nuvi 660 = 1'28"
4. Nuvi 200 = 2'00"
5. Nuvi 760 = 2'05" (with SW 2.30 and GPS FW 3.00s)

Similar results to jotne's (see bottom of page 2) and there is still a notable difference with the competition (excluding TomTom that has AGPS) and again this is a favorable environment in the suburbs with little blocage, but better than the 3+ minutes.

20071210 - FIELD TEST #3
Exact same route as previously. The 660 had a bit of a "hiccup today", the 920T (with a 3 day old ephemeris file now) still steals the show and the Mio C620T shows regular performance at <1+ minute.

Here are the results :
1. TomTom GO920T < 10 seconds (thanks to A-GPS)
2. Mio C620T = 55"
3. Nuvi 760 = 2'05"
4. Nuvi 660 = 2'15"

20071213 - FIELD TEST #4
Another test run, with screenshots of the logged tracks.

Here are the results :
1. TomTom GO920T < 10 seconds (thanks to A-GPS)
2. Route 66 Maxi (uNav chipset) = 1'04"
3. Mio C620T = 1'30"
4. Nuvi 760 = 2'40"

This is what the acquisiont process looks like on the map, with a bit of "wobbling" just after getting the fix, note that the uNav based Route 66 has some tracking problems :



Another consequence is the lack of accuracy resulting from the issues downloading enough ephemeris data. Only when I had to stop at the light 2 minutes later did the 760 have enough time to download more ephemeris data and plot a more accurate track :




Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?

Ads


danham

USA
7437 Posts

Posted - 08 déc. 2007 :  17:00:24  Show Profile  Visit danham's Homepage  Reply with Quote
Very interesting.

Are you saying that the field TTF times were measured while your vehicle was moving?

My 680 has taken as long as 3-5 minutes to lock while moving, but rarely takes more than about 60 seconds on a stationary cold start.

-dan

- Nüvi forum moderator -
Nüvi 760 in a '14 VW GTI & zumo 660 on a BMW F800 ST
Guide to working with pre-programmed routes: >> details <<
Language Guide / US Topo / 350 & 680 / MacBook & Intel iMac with OS X & Win XP / BaseCamp / Cape Cod, MA
Go to Top of Page

gpspassion

93884 Posts

Posted - 08 déc. 2007 :  17:09:46  Show Profile  Visit gpspassion's Homepage  Reply with Quote
Yes, driving out in the suburbs, in "warm fix" conditions, i.e. 4+ hours without using (them overnight actually). I do this test to replicate standard use of someone driving out in the morning.

@OrlandoRealtor - I checked and I have v2.20 so will update to v2.30 and check again but I can't see how it would make a different as far as TTFF goes, what really matters is the firmware of the GPS chip (a standalone device really - cpu+rf+software) and mine shipped with v3.00, i.e. SiRFIII v3.2.5 and if the signal level, the raw data isn't there, the firmware can only go so far anyway.

Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?
Go to Top of Page

Rgold

124 Posts

Posted - 08 déc. 2007 :  17:22:11  Show Profile  Reply with Quote
I tried 3 different Nuvi's 250 (Bravo), 250W (SiRF III) 650 (SiRF III) and Magellan 4040 (SiRF III). Magellan took the first place and slightly behind was the 250 with Bravo chip 25-35 Seconds from cold start. Now I'm using Nuvi 650 that locks in about 40-50 seconds. The Nuvi 250W was the worse, 1-3 Minutes. All this was measured while moving. I compared my 250W with a friend and his did better, 50 seconds to 1 min 20 seconds. This makes me wounder about Nuvi internal UHF receiver calibration.

Are they all coming calibrated proprerly from the factory?

NUVI650, DROID X CoPilot Live
Go to Top of Page

NoTrix

USA
182 Posts

Posted - 08 déc. 2007 :  17:25:27  Show Profile  Reply with Quote
Mine takes a full 1 minute "indoors" from a cold fix with 4 bars and continuous lock. About 30 seconds outdoors (5 bars). This is fine by me. It picks up better indoors than the HP Ipaq 310 I had for a while. By the way, that's with the update which was done before I used the device for routing.
Go to Top of Page

akennedy_gps

Canada
7 Posts

Posted - 08 déc. 2007 :  17:26:12  Show Profile  Reply with Quote
I have yet to roadtest my 770, but I attached an external antenna and placed it outside my window (I'm on the 14th floor). There is a clear view of about 180 degrees. I get a couple of bars within a few seconds, but it takes 5-6 minutes for them to turn solid. A few minutes more passed and I get a 3rd bar. That one locked on quicker and as soon as it did another 5 showed up (all very low power). This was from a cold start (more than 12 hours).

Once I had a position, with 4 good channels I was amazed at what I saw. I was sitting in my living room... moving between 4 to 27kmh. This didn't settle down at all. I watched for about 10 minutes. While watching the map I was moving all over the place... up to a block away. The accuracy was between 11-17 meters. This wandering continued even when I had up to 7 locked channels.

I've done this in the past with my iQue 3600 and never saw that. I'm going to try a test with both units later today.

I have sw ver 2.30

Edited by - akennedy_gps on 08 déc. 2007 17:28:44
Go to Top of Page

OrlandoRealtor

USA
300 Posts

Posted - 08 déc. 2007 :  17:34:25  Show Profile  Visit OrlandoRealtor's Homepage  Reply with Quote
Gpspassion, My TTFF on my NUVI 760 was NEVER as bad as yours is. Again when I updated to 2.3 it improved it immensely. Maybe they did have a Firmware update for the GPS chip that wasn't told to us. I had a 680 and now my 760 is way better on TTFF.
Go to Top of Page

danham

USA
7437 Posts

Posted - 08 déc. 2007 :  17:38:54  Show Profile  Visit danham's Homepage  Reply with Quote
GP:

Doesn't moving introduce lots of variables and wreck the validity of the test?

akennedy:

The "wandering" may actually be a sign of the newer receiver's increased sensitivity. And for a road navigator it has almost no effect on perceived accuracy as shown on the map.

-dan

- Nüvi forum moderator -
Nüvi 760 in a '14 VW GTI & zumo 660 on a BMW F800 ST
Guide to working with pre-programmed routes: >> details <<
Language Guide / US Topo / 350 & 680 / MacBook & Intel iMac with OS X & Win XP / BaseCamp / Cape Cod, MA
Go to Top of Page

gpspassion

93884 Posts

Posted - 08 déc. 2007 :  18:37:02  Show Profile  Visit gpspassion's Homepage  Reply with Quote
Yes, the wandering is due to high sensitivity, that's the price to pay, I had a whole write up about that in the 2005 chipset comparison (indoor testing page).

Clearly, driving introduces many different variables, but that's the point since I want to replicate real life usage, not get spec sheet data. Doing side by side testing is the only way to do away with "impressions" (works fine in my basement, car, home, etc...) by placing receivers in identical conditions, whatever they may be.

Based on the above feedback about the 250W and considering the 200 doesn't have any particular problems, I'm wondering if the 4.3" screen is not causing the problems here with too much RF interference and/or inadequate shielding.

Too bad there is no way of logging raw NMEA data to get a sense of the dB levels the chip has to work with.

@OrlandoRealtor - can you check the GPS firmware, v2.90 or v3.00 ?
Go to Top of Page

OrlandoRealtor

USA
300 Posts

Posted - 08 déc. 2007 :  19:07:10  Show Profile  Visit OrlandoRealtor's Homepage  Reply with Quote
GpsPassion: GPS Firmware is 3.00s
Go to Top of Page

apersson850

Sweden
1274 Posts

Posted - 08 déc. 2007 :  21:09:22  Show Profile  Reply with Quote
quote:
Originally posted by gpspassion

@OrlandoRealtor - I checked and I have v2.20 so will update to v2.30 and check again but I can't see how it would make a different as far as TTFF goes, what really matters is the firmware of the GPS chip (a standalone device really - cpu+rf+software) and mine shipped with v3.00, i.e. SiRFIII v3.2.5 and if the signal level, the raw data isn't there, the firmware can only go so far anyway.

In this case you are wrong. I may speak prematurely here, so I'll not say any more than that you'll be surprised to see what clever firmware can do. Wait and see.

Anders
Go to Top of Page

OrlandoRealtor

USA
300 Posts

Posted - 08 déc. 2007 :  21:13:04  Show Profile  Visit OrlandoRealtor's Homepage  Reply with Quote
I agree Anders, I am amazed at what the update from 2.2 to 2.3 did for my 760. Garmin did not tell us the surprise that was in store.
Go to Top of Page

danham

USA
7437 Posts

Posted - 08 déc. 2007 :  22:43:33  Show Profile  Visit danham's Homepage  Reply with Quote
quote:
Originally posted by gpspassion

Clearly, driving introduces many different variables, but that's the point since I want to replicate real life usage, not get spec sheet data. Doing side by side testing is the only way to do away with "impressions" (works fine in my basement, car, home, etc...) by placing receivers in identical conditions, whatever they may be.



OK, I wasn't clear enough about "variables." While driving, you pass by tall buildings or under trees, for example. So if I drive to try to duplicate your test results, they will almost certainly be different.

I understand that you can get useful comparative data between units sitting side by side, but it would be "dangerous" to say that other users trying this will get similar TTFs.

-dan

- Nüvi forum moderator -
Nüvi 760 in a '14 VW GTI & zumo 660 on a BMW F800 ST
Guide to working with pre-programmed routes: >> details <<
Language Guide / US Topo / 350 & 680 / MacBook & Intel iMac with OS X & Win XP / BaseCamp / Cape Cod, MA
Go to Top of Page

akennedy_gps

Canada
7 Posts

Posted - 08 déc. 2007 :  23:07:07  Show Profile  Reply with Quote
OK, so I used my 770 for the first time outside today. The initial acquisition took about 10 minutes, but that was in a moving car. My friend didn't want to wait for it. There were some tall buildings around at first plus we were driving on the lower deck of a highway for a short time too. Hardly ideal conditions.

We stopped for lunch and about an hour later I switched it on and had acquisition in about 2 seconds! Awhile later, we parked in one parking lot, switched off the unit and then I turned it on again in a different lot about 400m away. This time it took about 10 seconds. Seems pretty good to me.

One thing that I did find a little confusing is that when I first got 3 locked channels, it still wouldn't give a position. Two were full power and the third was about half a bar. I couldn't get a location until I had four channels locked. With my iQue 3600, it would give you position as soon as it got 3. Do the Nuvi's need 4 or just 3 strong signals? I know you need 4 for elevation.

I'm using sw ver 2.30.
Go to Top of Page

jotne

Norway
1685 Posts

Posted - 08 déc. 2007 :  23:36:51  Show Profile  Reply with Quote
gpspassion

How many 760 have you tested?

I have had 660 for about one year and now own a 760 for one month.
Using my GPS every day, I have not noticed that my 760 takes longer time to fix.

Maybe you have a faulty unit?

Previously owned : GPS "eTrex Yellow" "GPSmap 76S" "GPSMap 60CS" "GPSMAP 60CSx" "Nüvi 660"
Unit I have now :"Nüvi 760 EU" "Nüvi 765" "Edge 305" "Colorado 300"
Go to Top of Page

bigej

USA
7 Posts

Posted - 09 déc. 2007 :  00:18:53  Show Profile  Reply with Quote
My 750 regularly takes close to 2 minutes if not longer to lockon without the car moving. The one time I left before it locked on it was close to 5 minutes and wasn't until I stopped at a long light.
gpspassion: in regard to your reference to the 760 having the Sirf chip: there have been two quotes from Garmin support staff lately, one stating that the 750 does not have the Sirf chip and one stating that all new Nuvi's have the Bravo chip. I don't know what is accurate?
Go to Top of Page
Page: of 17 Previous Topic Topic Next Topic  
Next Page
 New Topic  Reply/Répondre
 Printer Friendly
Jump To:
GpsPasSion Forums © 2002-2014_GpsPasSion/Manzanite Go To Top Of Page
This page was generated in 1,08 seconds. Powered By: Snitz Forums 2000 Version 3.4.05