Oh, I forgot to update test with the same heuristics as the main app
I might do that today
RUi
On Wed, Oct 15, 2008 at 01:33:06AM +0530, Vikas Saurabh wrote:
The test application is blurting out lots of output, following is a
snippet(from the middle of the output) which sort of showed
What release do you use?
How long since your last reboot?
Have you been using other applications that use accelerometers?
I'm using Om2008.9 and I'm noticing instability in the accelerometers,
so maybe some of your problems are derived from it.
Rui
On Wed, Oct 15, 2008 at 12:48:24AM +0530,
I am using 2008.8 with regular updates, so I should be pretty updated :)
hmmnot sure, but it might be some 10 hours. No other apps were using
accels.
What is meant by instability? I was seeing continuous output from 'test'
program (although with similar values). Is that the instability?
Yet another one - this time in Python. Should be easy to port, though.
Features:
- no dimming
- no stopping if backlight not lit
- badly adjusted timers
- reopening the file at every check
- able to set the right position even if device is nearly lying
- does not switch if on verge of two modes
On Mon, Oct 13, 2008 at 10:26:48PM +0200, Oscar Casamitjana wrote:
screenfp = fopen (/sys/class/backlight/pcf50633-bl/brightness,w);
Excellent idea, I'll implement it in my newRotate and make a new
release.
BTW, I still can't commit...
https://admin-trac.openmoko.org/trac/ticket/1563
--
On Tue, Oct 14, 2008 at 06:15:58PM +0100, Rui Miguel Silva Seabra wrote:
On Mon, Oct 13, 2008 at 10:26:48PM +0200, Oscar Casamitjana wrote:
screenfp = fopen (/sys/class/backlight/pcf50633-bl/brightness,w);
Excellent idea, I'll implement it in my newRotate and make a new
release.
BTW,
This seems pretty stable than the previous one. The brightness effect is
nice, but the event, apparently, is hitting even if the phone is moved a
little (i was trying to type a message).
The brightness thing started to feel too much pain, so I killed the rotate
instance (from a separated ssh
Hi,
Thank you very much for your input. Can you try to run a bit of the
'test' I include? I'm starting to suspect different accelerometers might
give different outputs and so the heuristics might not make sense
anymore, I hope that's not the case.
You can return the brightenss with
echo 63
The test application is blurting out lots of output, following is a
snippet(from the middle of the output) which sort of showed some difference
when I tapped the screen.
Types: a(2), b(2), c(2)
Codes: a(0), b(1), c(2)
Value: a(-18), b(-1152), c(-54)
vertical straight
Types: a(2), b(2), c(0)
Vikas Saurabh wrote:
On Wed, Oct 15, 2008 at 12:10 AM, Rui Miguel Silva Seabra
[EMAIL PROTECTED]
mailto:[EMAIL PROTECTED] wrote:
On Tue, Oct 14, 2008 at 06:15:58PM +0100, Rui Miguel Silva Seabra wrote:
On Mon, Oct 13, 2008 at 10:26:48PM +0200, Oscar Casamitjana wrote:
Ón7~˪{ó½:ó¶ÛmõÓ_4Ó
rotate
Description: Binary data
/* -*- tab-width:4; c-basic-offset:4 -*-
* rotate.c -- determine Freerunner orientation.
* Author -- Chris Ball [EMAIL PROTECTED]
*
* This file is free software; you can redistribute it and/or modify it
* under the terms of the GNU
Oscar Casamitjana wrote:
Hi all
here you have my rotate version.
This version is based in Chris Ball's rotate, but with some changes:
* The adquisition engine is encapsulated in a thread.
* Before rotating screen, it ensures two seconds for the new position
* Backlight turns black while
On Mon, Oct 13, 2008 at 2:26 PM, Oscar Casamitjana [EMAIL PROTECTED]
wrote:
Hi all
here you have my rotate version.
This version is based in Chris Ball's rotate, but with some changes:
* The adquisition engine is encapsulated in a thread.
* Before rotating screen, it ensures two seconds
Angus Ainslie wrote:
1) inverted the screen was improperly split left and right
2) rotate seemed to stop working completely. killed it and restarted it.
3) on returning to normal orientation the x dimension was right but
the y dimension remained from the previous rotation. leaving my
14 matches
Mail list logo
