From: Daniel Gudlat <gudlat_at_web.de>
Date: Wed, 17 Dec 2003 11:08:26 +0100

Björn Stenberg wrote:
> If you have an "old style" recorder, try setting "Disk poweroff" to
> Yes and see if you notice a difference.

I did quite a bit of testing yesterday, and this brought RLOD frequency
down somewhat (somewhere near Rockbox 2.1 levels), but like Ingo Korb, I
still get them with some regularity.

My theory, which my and other's results of playing with the RLD test
build seem to support, is that what we call Red Led Dead, (I prefer
RLOD myself, in irreverent mockery of some other OS's characteristically
colored fatal error indicator) are actually two distinct but possibly
related phenomena: spontaneous RLOD, which happens when the unit just
sits undisturbed and RLODs "out of the blue" (what Olaf Richter is
getting), and movement induced RLOD, which is triggered by the unit
being shaken, stirred, vibrated or otherwise moved (what I and Johan
Vromans are seeing). Ingo, what are the circumstances of your RLODs
with the RLD test build?

I've never had spontaneous RLOD. In that, my experience closely mirrors
that of Johan Vromans. E.g. when my JBR is sitting on the passenger
seat, I never have RLODs, ever, likewise for when it just sits on a
table. OTOH, when I put it in the plastic hardshell compartment right
below the car stereo (where it is not cushioned from the bumps of the
ride), I regularly get RLODs and similar behaviour. In fact, the most
reliable way for me to get RLOD-like behaviour is to put the unit in the
hardshell compartment and drive on the autobahn for a while. Carrying it
on my belt rarely produces RLOD, however.

 From my experience and what I read here it seems to me like your test
build may cure spontaneous RLOD, but does nothing for movement induced
RLOD. So, if we can confirm this to be true, I'd vote for getting the
change into the daily builds to help those with spontaneous RLOD, and
then see about what if anything can be done about movement induced
RLOD...

Daniel Gudlat
Received on 2003-12-17