Subject: Re: [linux-audio-dev] matrix model vs. memory
From: Paul Davis (pbd_AT_Op.Net)
Date: Sat May 19 2001 - 21:36:21 EEST
>> >What seems easiest to me is a LADSPA like plugin to manipulate the gain.
>> >By making it an independent element on the graph it allows the normal
>> >graph sorting to optimize the buffer usage. It is also the most flexible.
>>
>> Karl, please see my reply to Abramo. The question is whether the most
>> fundamental method of transferring data from one port to another
>> includes a gain coefficient. If it doesn't, then precisely how gain is
>> implemented is irrelevant.
>>
>
>Right - I am suggesting that a gain coefficient should not be included -
>it just seems too complex and not nearly as effifient as the alternatives.
Well, I don't think it is complex. As I mentioned, there is already an
either implicit or explicit matrix of all the port connections. You
can either assume that there is an implicit gain of 1.0 when data is
moved from port to port, or you can include an explicit gain.
Its also much more efficient, since otherwise, gain operations have to
consist of:
src *= gain;
dst = src;
(with these two operations not occuring directly after one another,
since they occur in different parts of the system)
rather than:
dst = src * gain;
which occurs in one place.
of course, such a model does not stop there being other gain controls
present - i'm not suggesting that this be the only way gain is
controlled. i am just wondering whether making the gain between two
ports be explicit is a good idea, and it seems to me that it is.
--p
This archive was generated by hypermail 2b28 : Sat May 19 2001 - 22:08:19 EEST