From mboxrd@z Thu Jan 1 00:00:00 1970 From: Clemens Ladisch Subject: Re: [PATCH 1/2] amidi: ignore not only Active Sensing but all System Real-Time messages Date: Mon, 11 Apr 2016 16:13:08 +0200 Message-ID: <570BB0F4.2020504@ladisch.de> References: <570AA980.3070802@ladisch.de> <570BA6B1.8060101@ladisch.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from dehamd003.servertools24.de (dehamd003.servertools24.de [31.47.254.18]) by alsa0.perex.cz (Postfix) with ESMTP id 645EF2604D6 for ; Mon, 11 Apr 2016 16:13:13 +0200 (CEST) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: alsa-devel-bounces@alsa-project.org Sender: alsa-devel-bounces@alsa-project.org To: Ricard Wanderlof Cc: Takashi Iwai , alsa-devel@alsa-project.org, Martin Tarenskeen List-Id: alsa-devel@alsa-project.org Ricard Wanderlof wrote: > On Mon, 11 Apr 2016, Clemens Ladisch wrote: >> Ricard Wanderlof wrote: >>> I would argue that it would be better to have both options, somehow. I.e. >>> active sensing is mostly a nuisance, but I can imagine occasions >>> (analyzing the output from a sequencer perhaps) where one would want to >>> keep the rest of the real time messages. >> >> I can imagine this too, and even more complex filters. >> >> But the amidi tool is designed to be simple, and works on the lowest >> level, with raw MIDI bytes. This makes it appropriate to handle SysEx >> stuff and to debug low-level hardware problems, but when you care about >> the semantics of the messages, you should use a higher-level tool, such >> as aseqdump. >> >> Filtering out clock messages serves an actual need. But I am not >> willing to add complexity for a problem that is, at the moment, nothing >> but a figment of our imaginations. > > My argument is really that currently amidi has an option for filtering out > active sensing, which has probably been added as it serves a useful > purpose in its own right. The patch replaces that filter with one that > filters out everything from F8 (clock) and up I consider this change a bug fix; I just forgot about clock messages when I originally impemented the filter. > I think it would be better to keep the existing option, and add a new > one, rather than to essentially change the meaning of the existing > option. That would be a new feature. Regards, Clemens