[Sphyrna-devel] Current status?

[Mike Montour]

On Sep 11, 2007, at 12:05 AM, pHilipp Zabel wrote:
>
> On 9/10/07, Dmirty Platonov <shadowjack at mail.ru> wrote:
>> Here is my thoughts about reg 0b
>> http://projects.linuxtogo.org/pipermail/sphyrna-devel/2007-July/ 
>> 000002.html
>> http://projects.linuxtogo.org/pipermail/sphyrna-devel/2007-July/ 
>> 000003.html
>
> Any success with approximating/interpolating that with a linear  
> function?

One of my tests was to make repeated measurements of the 0x11 value  
(the "CI" in the above messages) using the initial-acquisition register  
settings. With a constant 0x0b value the "CI" value would slowly change  
over time, presumably corresponding to the change in distance to the  
satellite. Increasing the 0x0b value by 1 unit would decrease the  
corresponding 0x11 value by approximately 0.2047.

To do this measurement I set the chip to measure the same PRN on  
several channels, varying only the 0x0b value between them. This let me  
isolate the changes due to the 0x0b value from the changes due to  
satellite movement + clock error.

  Unfortunately I was only able to get this to work with the "initial  
acquisition" registers; when I tried to go to the next stage (changing  
0x03 from 0x40000720 to 0xc05c1702, along with other register changes)  
the unit would lose the signal. Also, from looking at some of the  
traces there is not always a fixed ratio between the 0x0b value and the  
change in "CI".

> Personally, I'm just short on time, with way too many similar projects.
> That and the fact that I can't just hack on it for a few minutes when
> I feel like it because the battery is somehow always drained and the
> stupid thing doesn't run on USB power without it are the cause for my
> current silence.

For me, the biggest problem is that my apartment has a very poor view  
of the sky. To do any live testing I have to take my Neo + laptop out  
into the Big Blue Room. That, combined with a full-time job, limits the  
time that I have to spend on the project. The battery is not an issue  
because I leave my Neo plugged in to USB most of the time, and my unit  
does not have a problem if I leave it unplugged (and turned off)  
overnight.

> Whether or not the hammerhead will be in GTA02 does not matter to me  
> at all.

I plan to continue regardless of the GTA02 decision, but I would  
certainly be more motivated if my work was relevant to a mainstream  
product rather than to a dead-end prototype.