Subject: Re: [linux-audio-dev] Re: OSC vs MIDI
From: Steve Harris (S.W.Harris_AT_ecs.soton.ac.uk)
Date: Sat Sep 04 2004 - 13:54:25 EEST
On Sat, Sep 04, 2004 at 01:44:28 +0200, David Olofson wrote:
> On Saturday 04 September 2004 00.16, Steve Harris wrote:
> [...]
> > > I *think* bluetooth should be able to handle the bandwith
> > > requirements of a mouse (125 Hz frame rate for the MX models,
> > > IIRC), but 15-20 ms latency spikes would be an issue that cannot
> > > be fully compensated for regardless of bandwidth. Jitter or delay
> > > (buffering to eliminate jitter) would cause problems with real
> > > time controllers, such as mice or musical controllers.
> >
> > Bluetooth can do realtime ADPCM audio, so a mouse should be no
> > problem. Maybe HID controllers use a different bit of the protocol?
>
> I dunno... Either way, ADPCM audio is just a matter of bandwidth, but
> latency seems to be the issue here. 15-20 ms (in addition to
> driver/OS polling latency + jitter and video buffering latency) is
> unacceptable for a gaming mouse, so I suspect that messages are just
> sent and handled ASAP, without timestamping. Just means you get
> terrible jitter instead, which is probably *worse* than the
> latency...
There cant be any latency jitter on the audio... I would guess that either
that part of the protocol is reserved for audio in some way, or the
logitech implementation is just rubbish.
- Steve
This archive was generated by hypermail 2b28 : Sat Sep 04 2004 - 13:58:52 EEST