Reason for GPS problems found!

[BlueStar88]

Andrew Bennett schrieb:

>  From reading between the lines on some of the e-mails from OpenMoko 
> staff, it seems that the interference with the GPS receiver was caused 
> (at least in part) by the SD reader being always-on, apparently because 
> the SD firmware is still incomplete or in beta, much like the rest of 
> the OpenMoko firmware.  The fix appears to be to enhance the SD reader 
> to only turn on when necessary.  From where I sit, that's a two-fer!  
> GPS gets fixed and the SD reader gets enhanced to a more complete 
> state.  I'm happy :)

Ahmm... to only turn the reader on when nescessary? Sorry, but that 
sounds like to be just a "work around" to me, not an enhancement to a 
more complete state. It sounds like selling a bug and it's work around 
as a feature we all have waited for..
If I got this right, it still remains to be a reception mess, if the 
reader will turned on on demand (for reading map-data for example). So 
where exactly is the enhancement? Not to have to pull out the SD card 
for using GPS/getting fix? Nice, but thats not enough.
If the GPS reception qualitiy drops for - lets say - 50%, while 
accessing map data on the SD-card, I risk to loose the connection to the 
satellites in case of having any bad (environmental) conditions. I'll 
loose precision at least. Thats very bad for driving in big cities with 
tons of obstacles around, where the pure navigation feature and it's 
maximum reliability is really needed for the driver. Permanent SD-card 
read access while reading map data is surely to expect for normal usage! 
It's not about just not beeing able to watch a movie from SD-card while 
GPS'ing, it's about the strong need for the GPS/SD hardware - in 
combination!
Sorry, but dynamically switching off the reader or clocking it down 
somehow is no real option, it is a serious failure in function for heavy 
routing application requirements.

To be honest, I personally expected a working mobile in the very basic 
hardware meaning, but with trillions of *software* issues of course. 
Almost all (very rough assumption) GTA02-people are having this issue 
now, even GTA01-users are coming out with confirmations, so why it was 
not detected (nor digged deeper) by personal full range use before mass 
production? I'm not whining, I'd just like to understand it to some degree.
Before someone suggests: I don't like the 'fallback' idea of using an 
external antenna, since it is very unhandy for bike and walk usage. 
Additionally the MMCX-jack will not like the permanent (often) mount and 
unmount procedures it'll cause.

My current thoughts are two sided. On one side I don't like to hurt the 
really great (!) project by returning the hardware. On the other side I 
fear, that this issue cannot be fixed to have the initially expected 
behavior.

I hope this bug gets busted *effectively*! :)

-- 

BlueStar88
________________________________________________________
PGPID: 0x36150C86
PGPFP: E9AE 667C 4A2E 3F46 9B69 9BB2 FC63 8933 3615 0C86

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.openmoko.org/pipermail/community/attachments/20080716/c15b2fcf/attachment.pgp