A few observations from a relative "Noobie"
Hi folks,
I have read the thread from beginning to end a few times and have run some tests of my own. I thought I would share my observations and comments.
It is a huge pity that the the clash of Titans between AngelDeath and Jaiy ended badly because I feel that Jaiy had some extremely valid points even if the were presented rather harshly. For instance, his comments about the NTP servers led me to do some tests of my own.
What I found is that if you run a "ping" test from the command window some of the servers fail to return the ping. Others ping pack promptly. I would heartily suggest to my fellow forum pals that before using a time server in their script that they run a ping test to see if the server they wish to use is valid.
For example: "ping hk.pool.ntp.org" times out every time so it is useless to use that in a script.
but.... "ping asia.pool.ntp.org" returns a successful ping.
See attached image.
I tested AngelDeath's Asia script from post#1 of this thread and, while it worked, I found the results to be excellent at times and then at other times I could not lock on to any satellites at all, even using a hot start with GPS Status. At times, the only way I could get any signal was to clear the A-GPS data and reload it.
So I decided to test Jaiy's method and ran a modified version of his script. The result is that while lock times are a few seconds slower, the acquisition of satellites appears to be more consistent. This may be due to only using an NTP server that actually responds to a ping test. Perhaps a non-responding server "hangs" the acquisition process? Hmmmm....
I would therefore strongly suggest you test this yourselves and modify your own scripts to make sure you are using a server that is up and running. You can find the servers for your area at
http://www.pool.ntp.org/en/
The gps.conf I am currently running looks like this:
NTP_SERVER=asia.pool.ntp.org
INTERMEDIATE_POS=1
ACCURACY_THRES=5000
CURRENT_CARRIER=common
DEFAULT_USER_PLANE=TRUE
REPORT_POSITION_USE_SUPL_REFLOC=1
I use ADB to push the file, clear the cache and Dalvik cache and then reboot. While the lock times are not staggeringly fast they are very acceptable and the results seem to be fairly stable and consistent.
It took me a while to get ADB to work before I figured out that USB Debugging has to be turned on.
As I said, I'm still a Noob!
I will be pleased to hear if anyone else has used this method and any comments/crticisms so we can all learn and hopefully move forward with faster GPS lock times. Perhaps this data can be merged with AngelDeaths work to produce a killer conf file. I hope so!
Cheers!
Phone Samsung Galaxy SII I9100
Android 4.0.4 / Rooted
Kernel Siyah v5.0.1
P.S. For the prurists: yes I know it's a config file, not a script!