BOUNCE rtl@rtlinux.org: Approval required: Non-member submissionfrom [Adolf Ahmad MS <adolf@ratelindo.co.id>] (fwd)

[Der Herr Hofrat <der.herr@hofr.at>]

>From owner-rtl Mon May  7 20:24:45 2001
Received: from wbmail.ratelindo.co.id (rtiwb.ratelindo.co.id [202.155.4.226])
	by hq.fsmlabs.com (8.11.2/8.11.2) with ESMTP id f482OhJ18150
	for <rtl@fsmlabs.com>; Mon, 7 May 2001 20:24:43 -0600
Received: by WBMAIL with Internet Mail Service (5.5.2650.10)
	id <JR700N0N>; Tue, 8 May 2001 09:08:48 +0700
Message-ID: <41B6E7DDA9D5D3119B5700508B8B04E107536809@WBMAIL>
From: Adolf Ahmad MS <adolf@ratelindo.co.id>
To: rtl@fsmlabs.com
Subject: RE: [rtl] Maximum Sample Rate using RTLINUX
Date: Tue, 8 May 2001 09:08:40 +0700 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.10)
Content-Type: text/plain

Dear David Olofson,

Thank you for nice answer, of course this PPI 8225 card(using 8255 I/O chip
with Strobe Mode 1) and without Hardware FIFO/buffer. I only use RT_FIFO.
But if you have experience "What is the maximum Interrupt Frequencies of
RT_IRQ", tell me more.

Regards,


Adolfo

> -----Original Message-----
> From:	David Olofson [SMTP:david.olofson@reologica.se]
> Sent:	Friday, May 04, 2001 1:20 AM
> To:	rtl@fsmlabs.com
> Subject:	Re: [rtl] Maximum Sample Rate using RTLINUX
> 
> On Thursday 03 May 2001 05:49, Adolf Ahmad MS wrote:
> > Dear All,
> >
> > I need a help.....
> >
> > I have a problem during capture data stream with data rate(sampling) 782
> > kByte/s using PPI8255 Card(ISA Bus). To capture the data stream, i
> applied
> > IRQ triggering.
> >
> > But, some times many byte of data are lost. How many maximum transfer
> rate
> > on RTLINUX ?
> 
> No maxmum transfer rate, really, but there are hardware limitations on the
> 
> IRQ response times...
> 
> 
> > As info, there is no problem if applied this program to capture the data
> > stream with 100 kByte/S.
> 
> 100 kHz isn't bad for anything with a "normal" CPU in it! You're spending 
> most of the CPU cycles on context switching here, probably with an
> occasional 
> latency peak when the cache is exhausted... (That would be the several
> lost 
> bytes when you try to go beyond 100 kHz.)
> 
> 
> I *really* think you should consider using DMA, or at least some hardware 
> buffering... General purpose CPUs really aren't suitable for this kind of 
> interrupt frequencies.
> 
> Isn't there some hardware FIFO or ISA DMA support on that card? (I don't
> know 
> anything about the PPI8255...)
> 
> 
> //David
> 
> .- M A I A -------------------------------------------------.
> |      Multimedia Application Integration Architecture      |
> | A Free/Open Source Plugin API for Professional Multimedia |
> `----------------------> http://www.linuxaudiodev.com/maia -'
> .- David Olofson -------------------------------------------.
> | Audio Hacker - Open Source Advocate - Singer - Songwriter |
> `--------------------------------------> david@linuxdj.com -'
> 
> 
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/