From owner-freebsd-hackers Fri May 21 19:42: 6 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from Genesis.Denninger.Net (kdhome-2.pr.mcs.net [205.164.6.10]) by hub.freebsd.org (Postfix) with ESMTP id CEFF814F44 for ; Fri, 21 May 1999 19:42:03 -0700 (PDT) (envelope-from karl@Genesis.Denninger.Net) Received: (from karl@localhost) by Genesis.Denninger.Net (8.9.3/8.8.2) id VAA16665 for hackers@freebsd.org; Fri, 21 May 1999 21:42:03 -0500 (CDT) Message-ID: <19990521214202.A16648@Denninger.Net> Date: Fri, 21 May 1999 21:42:02 -0500 From: Karl Denninger To: hackers@freebsd.org Subject: New one - mp3 "clicks" during playback Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i Organization: Karl's Sushi and Packet Smashers X-Die-Spammers: Spammers will be LARTed and the remains fed to my cat Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi folks, I've gotten my Ensoniq card installed and the driver configured, and it works - sorta. mpg123 has been the recommended "playback" mode for MPG files. Well and fine. Except for one small problem: If the system starts taking interrupts, I get "clicks", roughly consonant with the interrupts, in the output. They're solidly correlated with disk activity. Now I could see if the system was busy for a long period of time that it could "get behind" whatever buffering is in the sound board, and you'd be screwed (due to data starvation). But this looks to be something different - it looks like there is NO buffering on the card at all, and that even the interrupt traffic causes stream disruption in the output. This can't be right - or is it? If it is, then any I/O during playing a sound file of some kind is going to lead to unacceptable interruptions and poor sound quality. Mpg123 seems to consume about 12% of the CPU when its running - I have a Pentium Pro 200 in this box. "vmstat" shows a 0.12 system load during a run, so I should be well within the boundaries. "mpg123's" buffering option (-b) doesn't affect the display of this problem in any appreciable way. Is this my hardware? This is a rather old Soundscape board, but I can't imagine they've changed that much. If the answer is to use a different card (one that has some buffering internally!) that's ok with me - what should I install instead? I DID note that the Ensoniq driver seems to want the Soundscape board on IRQ 9. That's a rather problematic interrupt to use in my experience in the past - is this part of the issue (the disks are on Adaptec PCI controllers)? Ideas? As it stands the decoder is useless in any environment where the system might actually have to do something while it's running. -- -- Karl Denninger (karl@denninger.net) Web: fathers.denninger.net I ain't even *authorized* to speak for anyone other than myself, so give up now on trying to associate my words with any particular organization. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message