NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/palmtops/2822

Text File | 1992-08-31 | 1.6 KB | 38 lines

Newsgroups: comp.sys.palmtops
Path: sparky!uunet!ulowell!iisat!mkseast!dale
From: [e-mail] (Dale Gass)
Subject: PC-3000: clock speeds and battery life
Organization: Mortice Kern Systems, Atlantic Canada Branch
Date: Tue, 01 Sep 1992 02:29:06 GMT
Message-ID: <[e-mail]>
Sender: [e-mail] (Dale Gass)
Lines: 27

I'm a little curious as to the actual effect that the clock speed has on
battery life on the PC-3000.

I presume that when the machine is idle (low-power mode on, and no keyboard
or processing activity) it is irrelevant what the clock speed is; the machine
is in some special idle state.

When there's keyboard activity or processing going on, a higher clock speed
will certainly chew up more battery life, but it will also get the work
done proportionately faster.  Now, I assume that there's some fixed cost
for *any* CPU activity (for auxilliary chips, etc.), regardless of the speed.
If this is true, then unless twice the processing speed burns up a fair bit
more than twice the battery life, it shouldn't really matter what speed you
use; in fact, the faster speeds could be better, since you'd have the CPU
out of idle mode for a much smaller amount of time.

I haven't seen any significant battery difference between 5Mhz and 10Mhz
operation, although it's hard to measure when your activities vary (sometimes
use the serial port, sometimes don't).

Anyhow, I've found the AC adaptor for a bit better price ($55 Cdn.), so I've
ordered one.  That should take some of the pain out of serial port usage.

-dale
--
 Dale Gass, Mortice Kern Systems, Atlantic Canada Branch
Business: [e-mail], Pleasure: [e-mail]