Re: Winmodem support, some performance tradeoff estimates

Adam J. Richter (adam@yggdrasil.com)
Mon, 17 Aug 1998 09:12:56 -0700

Thomas Sailer wrote:
>Of course you slow down the system, because there will be a task that
>consumes
>25% of the CPU. But then again "most" users won't notice a decrease in
>interactive response :-) (you can simulate that if you want, just write
>a SCHED_FIFO
>task that consumes 25% of the CPU).

It is misleading to say that soft modem support would consume
"25% of the CPU" without saying what kind of CPU. You should say
that it would consume approximately some amount of MIPS. For example,
if your code consumes 25% of the CPU on a 100MHz Pentium, then it
consumes about 3.7% of the CPU on a 450MHz Pentium 2.

Anyhow, according to linker@ml.org, at least the Rockwell based Winmodems
have significantly more hardware than a truely soft modem, so it should
not be anywhere near the CPU load of a truely soft modem.

Adam J. Richter __ ______________ 4880 Stevens Creek Blvd, Suite 205
adam@yggdrasil.com \ / San Jose, California 95129-1034
+1 408 261-6630 | g g d r a s i l United States of America
fax +1 408 261-6631 "Free Software For The Rest Of Us."

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html