Reply
Thread Tools
Guest | Posts: n/a | Thanked: 0 times | Joined on
#91
Again. I repeat. Supl.google.com IS working here.
Dont know why some insists im faking it. I never do.
Might be a carrier/country whatever deal I have no idea, but it gives a good location in ~15 secs.
 

The Following 2 Users Say Thank You to For This Useful Post:
Posts: 650 | Thanked: 497 times | Joined on Oct 2008 @ Ghent, Belgium
#92
Again. I repeat. How fast you get a lock means nothing. When I set it to supl.google.com I get cr*p ACWP support, if I enable GPS I get a lock in <30 seconds.
It's *very* hard to reliably test what is going on, I'm using ACWP only as a measure because I can see clear differences between selected servers:
- set desired server
- reboot
- run acwp script I posted

Maybe I should wireshark the connection between n900 and supl.* to show what is going on.
__________________
Affordable mobile internet in Belgium: Try Mobile Vikings
2 GB, 1000 SMS and 15 euro of talk time for.... 15 euro
 

The Following 4 Users Say Thank You to petur For This Useful Post:
Posts: 958 | Thanked: 483 times | Joined on May 2010
#93
i actually dont really care if supl.google.com is working or not. what's more important to me is the end result i.e. accurate gps lock is achieved in a short time - which is what i am experiencing now. as for the agps cache - no such issue since my phone has a scheduled reboot task that runs everyday.
 

The Following User Says Thank You to droll For This Useful Post:
Estel's Avatar
Posts: 5,028 | Thanked: 8,613 times | Joined on Mar 2011
#94
Fine, but this thread isn't about your ends results (sorry ), but about *real* A-GPS. Not to mention that you *won't* get such good results *if* real A-GPS support would be needed to gain it (quickly) in the first place, but frankly, we don't care about you results. We care about A-GPS (in this thread, at least).
---

Now, about the real things:

Originally Posted by petur View Post
Maybe I should wireshark the connection between n900 and supl.* to show what is going on.
Very good idea. As said earlier, I remember someone respectable to mention, that even supl.nokia.com works *fully*, if A-GPS request is passed through a middle-man A-GPS server installed on N900 (the way, that local server "ask" supl.nokia.com, instead of N900 "asking" directly).

I can find the source quote for a life of mine, but it would be very interesting to RE it - after all, supl.nokia.com works for recently released Nokia devices. I bet, that they're filtering out requesting devices on purpose.

Knowing WTF is going on and what they use to identify, we could implement a way to mimic other device for our trusty N900 (and even implement possibility to do it from control panel, via CSSU, just like browser allow to change user agent).

/Estel
__________________
N900's aluminum backcover / body replacement
-
N900's HDMI-Out
-
Camera cover MOD
-
Measure battery's real capacity on-device
-
TrueCrypt 7.1 | ereswap | bnf
-
Hardware's mods research is costly. To support my work, please consider donating. Thank You!
 

The Following 6 Users Say Thank You to Estel For This Useful Post:
Posts: 31 | Thanked: 31 times | Joined on Jan 2013 @ USA
#95
Has someone packaged the proxy software at http://www.tajuma.com/supl/ somewhere?
 

The Following 5 Users Say Thank You to eccerr0r For This Useful Post:
Posts: 1,378 | Thanked: 1,604 times | Joined on Jun 2010 @ Göteborg, Sweden
#96
Originally Posted by eccerr0r View Post
Has someone packaged the proxy software at http://www.tajuma.com/supl/ somewhere?
I believe this is what Estel is referring to. Nice find...
 

The Following 2 Users Say Thank You to handaxe For This Useful Post:
woody14619's Avatar
Posts: 1,455 | Thanked: 3,309 times | Joined on Dec 2009 @ Rochester, NY
#97
Originally Posted by Estel View Post
Knowing WTF is going on and what they use to identify, we could implement a way to mimic other device for our trusty N900
It could also give us a way to simply push the data ourselves. More often than not when people use GPS they know roughly where they are (city, state, etc). Even if the AGPS guess is off by a good distance, it doesn't change the satellite constellation the GPS is looking for that much. If one could even set/enter rough default coordinates, as long as you're in the ball park (+/- a degree) it will cut the time needed for GPS to lock dramatically.

While the GPS code is closed (preventing us from directly injecting a guess at a starting point), this may be a good way to fake it. A local supl server that computes a rough guess and feeds it to the closed source that then injects the right data.

Would be a popular project if someone's willing to take it up.
__________________
Maemo Council Member: May 2012 - November 2012
Hildon Foundation founding member.
Hildon Foundation Board of Directors: March 2013 - Jan 15, 2014
 

The Following 6 Users Say Thank You to woody14619 For This Useful Post:
Guest | Posts: n/a | Thanked: 0 times | Joined on
#98
Originally Posted by woody14619 View Post
It could also give us a way to simply push the data ourselves. More often than not when people use GPS they know roughly where they are (city, state, etc). Even if the AGPS guess is off by a good distance, it doesn't change the satellite constellation the GPS is looking for that much. If one could even set/enter rough default coordinates, as long as you're in the ball park (+/- a degree) it will cut the time needed for GPS to lock dramatically.

While the GPS code is closed (preventing us from directly injecting a guess at a starting point), this may be a good way to fake it. A local supl server that computes a rough guess and feeds it to the closed source that then injects the right data.

Would be a popular project if someone's willing to take it up.
not quite that simple. The location servers return with a limited information, cached, making a near - but not complete set of location data. This data is cached. And the difference I believe. Is that the set of data is a bit late. Not real time. But, ofcourse we could cache similat response
 

The Following 3 Users Say Thank You to For This Useful Post:
Posts: 1,378 | Thanked: 1,604 times | Joined on Jun 2010 @ Göteborg, Sweden
#99
Originally Posted by eccerr0r View Post
Has someone packaged the proxy software at http://www.tajuma.com/supl/ somewhere?
So what is the consensus, would this intermediate proxy still work? Seems odd that supl.nokia allows the n9 et al but not the n900 yet would allow some arbitrary proxy supl. Or is it a certificate /ssl issue? Wireshark did not show ssl packets to supl.vodafone on my n900. It showed sslv3 for imap traffic.

But I am on one tenth knowledge here so someone smart with both an n9 and a n900 should sniff both supl'ing away...
 

The Following 3 Users Say Thank You to handaxe For This Useful Post:
Posts: 31 | Thanked: 31 times | Joined on Jan 2013 @ USA
#100
I was trying the supl proxy on my server and pointing my N900 to it. However results were inconclusive, I wasn't sure if my N900 was even happy with my supl server's certificates, making it tough to see what the real problem was...
I definitely did see some wireshark activity on port 7275 however, but didn't pursue it further after getting T-Mobile's server working... Granted a public server would be better as then it would also work on wifi...
 

The Following 2 Users Say Thank You to eccerr0r For This Useful Post:
Reply

Tags
a-gps, nokia n900


 
Forum Jump


All times are GMT. The time now is 03:44.