Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Apr 2003 16:08:56 +0200
From:      Bernd Walter <ticso@cicely9.cicely.de>
To:        Terry Lambert <tlambert2@mindspring.com>
Cc:        ticso@cicely.de
Subject:   Re: realtime problem
Message-ID:  <20030410140855.GP96083@cicely9.cicely.de>
In-Reply-To: <3E94AD3C.B17D2751@mindspring.com>
References:  <20030409114957.GN83126@cicely9.cicely.de> <20030409160038.B658@beagle.fokus.fraunhofer.de> <3E94AD3C.B17D2751@mindspring.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Apr 09, 2003 at 04:31:08PM -0700, Terry Lambert wrote:
> Harti Brandt wrote:
> I still recommoned shoving HZ up by over a factor of 10, and the
> quantum down by a factor of at least 5 less than the precision
> required.  That will (almost) guaranteed that you are always in a
> loose ballpark around your 1mS deadline, which should be less than
> 500uS.  You may lose out once in a while, but it shouldn't bee too
> often (I assume the thing's not going to be exposed to the Internet
> ;^)).

Well it depends.
I have two options:
- start from scratch
- start with GPL software
In the later case I have to expose...
But I believe that I will publish it even in the first case.
It's not going to be sold, so I don't have to make any garanties.
It's just like burning CDRs - it usually works, but you always know
that it may fail if you are out of luck.

-- 
B.Walter                   BWCT                http://www.bwct.de
ticso@bwct.de                                  info@bwct.de



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030410140855.GP96083>