LCDproc development and user support list

Text archives Help

[lcdproc] Backlight Shutdown Bug V0.3.3 (And [ANNOUNCE] LCDproc v0.3.4 coming soon :)

Chronological Thread 
  • From: choadster AT (William W. Ferrell)
  • Subject: [lcdproc] Backlight Shutdown Bug V0.3.3 (And [ANNOUNCE] LCDproc v0.3.4 coming soon :)
  • Date: Tue, 23 Jun 1998 14:33:29 GMT

On Tue, 23 Jun 1998 07:57:30 -0500, the chunks you blew at me arranged
themselves thusly:

>On Tue, Jun 23, 1998 at 11:28:34AM +0100, Chris Smith wrote:
>> > Hmmm.... Correct me if I'm wrong but this is a feature, not a bug =
>> > I didn't take a look at the 0.3.3 code, but in 0.3.2 there was a =
>> that
>> > told the backlight
>> > to turn off when the load is below 0.1 or something like this....
>> Okay. But it doesn't do it if you just type lcdproc and don't specify
>> any modes or contrast setting and it doesn't do it if you're not=20
>> displaying the Load meter (that latter I'll forgive it for as it =
>> that you're not interested in the load anbd doesn't check it anyway).
>It works like this. It is a feature of the XLoad mode. If you don't
>have the XLoad mode, it won't happen. The XLoad mode turns the
>backlight off if the xload screen is completely blank. That means
>that all the displayed load bars must be below a certain level (0.05?)
>and not just the current load.
>I just tested it and it does still do it if you do 'lcdproc' by
>itself. Perhaps when you were trying it, it was generating some short=20
>lived load that was keeping the light on. You have to wait until that=20
>blip moves all the way accross and off the screen to the left.

Erv's got it down just about right. The XLoad screen code currently =
the backlight status variables. LCDproc's outer loop actually makes the
function calls to the LCD driver to mess w/the backlight, but only the =
screen actually makes use of it.

This _will_ change in v0.4 ... all the backlight stuff will be =
(i.e. disable blinking, disable _any_ change in backlight status, turn =
backlight when loadavg drops below x, blink it when loadavg goes above y,
etc.). Also, since v0.4 will have a client/server, and since multiple =
could potentially be displaying their stats on screen, the server part =
actually drives the display and listens for stats from clients) will NOT
provide any functionality for backlight control past the actual blinking =
and the on/off code. The client(s) will control the backlight blinking, =
will (probably) be required per the LCDproc protocol to check the loadavg=
_every_ frame (not just when showing an Xload screen) and always give the
server instructions on what to do with the backlight.

Since each client will have seperate data structures, if LCDproc Server
switches to a screen from another client which _isn't_ blinking, it will =
course stop the backlight blinking stuff.

In fact I'm starting to think it won't make too much sense to use the
backlight much (aside from, um, turning and leaving it on :) when =
clients have connected to the server. Perhaps just the server could use =
it for
stuff ... I dunno *shrug* I'll figure something out.

One possibility is only allowing the local client (i.e. the one running =
on the
same machine as the server :) to control the backlight. That might make =
sense. Thoughts, people?

ALSO: ANNOUNCE! Um, bugs (heh-heh not related to the above conversation =
found in v0.3.3 have called for an updated version. This time we'll even =
to include all the files you'll need to compile! I think a new feature =
have squeezed into v0.3.4 but I don't remember if it did or not, or what =
it is
... Scott & I will release it hopefully sometime tonight :)

Please reply to
choadster AT
Reply-to field
set accordingly.

William W. Ferrell is not an employee of Hewlett-Packard.
The opinions expressed in this message are not those of
Hewlett-Packard, its employees, stockholders, or coffee-makers.
I speak only for myself.

To unsubscribe from this list send a blank message to
lcdproc-unsubscribe AT

Archive powered by MHonArc 2.6.18.

Top of page