[Ti] sleep mode - Battery life

[David Remahl]

On fredag, jun 6, 2003, at 14:59 Europe/Stockholm, Michael Bigley wrote:

by me sweet medicine is still medicine and it isn't a candy ...
>>
>> sleep mode isn't the way to save the time (booting the system)
>>
>> sleep is for few minutes pause
>>
>> if Im going to stop the work for more just shut down
>
> but why??

I agree.

There is no reason why one should shut down the computer instead of 
putting it to sleep. It is very useful to be able to pull up the 'book 
and be ready to take notes or look up a web page within 2 or 3 seconds. 
No reason to shut down and have to wait 2 or 3 minutes for the computer 
to boot. The power consumption during the 2 or 3 minutes boot will make 
up for the power lost during sleep.

There are some services and hardware that don't like sleep (DigiTools, 
for example), but most handle it perfectly well. That includes network 
programs, such as personal file sharing and apache. Some programs will 
also react slightly strange when several hours of time pass within a 
few cycles. When the computer wakes up, system time is still what it 
was when it fell asleep. Then, after 10 seconds, it catches up. If for 
example a program is running an animation at the time, then it may be 
confused and jump to the last frame in an instant.

If you know you are not going to use the computer for a week or so, 
then you should probably shut it down, since the computer will use a 
few percent of the battery for each day in sleep mode.

> BSD and all Unix OS's by nature are designed to rarely be shut down. 
> Several of the "self-maintenance" operations, cron, log rotating, 
> locate, etc. are usually set to run in the off hours. They should only 
> be brought down for upgrades and maintenance.

Agreed. Note however that sleep mode will prevent self-maintenance from 
running just as a shut down computer will. I frequently leave my 'book 
running through the night, for that reason.

> I have traveled from KS to NY and back all while using only sleep mode.

> Early Microsoft OS's have brought about the "if there is a problem 
> reboot" and "reboot often to clear things up." Which is not the case 
> with UNIX, Linux, BSD, and OSX.

I rarely, if ever, shut down my computer for any other reason than a 
major system update. An uptime of more two weeks is the rule rather 
than the exception, and I never have any problems because of that. 
Occasionally restarting the Finder (logging out and back in again will 
take care of that), and not leaving Safari beta running 24/7 avoids 
mean memory leaks etc, which could otherwise bring the machine to its 
knees in inactive swap memory.

I find that it is actually easier to achieve a high uptime on a laptop, 
since it is possible to put it to sleep and it has automatic protection 
against power failure ;-).

> I don't know about earlier Mac OS's as I didn't like Mac's until OS X. 
> And yes I call it Oh-Ess-Ex because I came from the UNIX, Linux side. 
> Not the Mac "numbers".

The Mac OS 9 definitely was well served by an occasional reboot. If for 
no other reason than that it would hang spontaneously every so often 
;-). Lack of protected memory was the greatest reason. A crashing 
program could corrupt system memory and lead to undefined behaviour at 
a later point.

> Not that there is anything wrong with that.

Not at all :-).

/ Regards, David Remahl