Re: Winmodem support, some performance tradeoff estimates

Thomas Sailer (sailer@ife.ee.ethz.ch)
Mon, 17 Aug 1998 13:19:46 +0200


Adam J. Richter wrote:

> background on an 8MHz MIPS R4000 processor. Let's say that this
> means the performance of using a soft modem is about 8 Pentium MHz,
> on the grounds that 1 32-bit Pentium MHz =~ 1 64-bit R4000 MHz in the

According to my own implementation experience, all you get for 8 Pentium
MHz is Bell 10x, i.e. 1200 bauds or maybe even 2400 bauds. I doubt
you'll want this :-)

NB1: for the code (which includes cycle counter measurement code) see
drivers/net/hamradio/soundmodem/*

NB2: If you design kernel space API well enough (which isn't
unfortunately
true for OSS) then almost all of the processing is a user space issue.
Also, you most likely will need floating point, as floating point is
much
faster than integer on some expecially intel processor models.

NB3: If it's so important for you to get winmodem support, make the docs
easily available, such as the ITU V series for a start. I'd likely be
willing
to contribute to such a project, but I'm too lazy to bother trying to
get these standards documents.

Tom

-
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