Jaroslav Kysela wrote:
>On Tue, 19 Mar 2002, J Sloan wrote:
>
>>Agreed, the oss drivers should _at least_
>>be maintained as an alternative, e.g. for
>>those of us who want reliable sound with
>>*low latency*
>>
>><explanation>
>>I haven't checked lately, but not too long
>>ago the alsa drivers were found to be one
>>of the worst sources of latency in the kernel.
>></explanation>
>>
>
>You should really take care about your words. You've not written any
>technical reason to say these sentences.
>
Fair enough - I don't see any technical
reasons why alsa *couldn't* perform as
well, latency wise, as oss.
>We are not aware about any
>problems against low-latency.
>
Some folks on this list had been doing latency
profiling on their kernels during the past year
(sorry, I don't remember the exact dates) and
surprisingly the alsa driver was showing up
as one of the top sources of latency - IIRC one
possible explanation was "in-kernel mixing".
>Sure, OSS API emulation is only emulation,
>so there is additional layer which can be a bit slower than simplified
>native OSS drivers, but using ALSA API, we get really serious latencies
>even for multichannel hardware.
>
Thanks for the feedback, I'll have to do some
testing of my own and see what the situation
is at this point - I'd be very happy to learn that
my concerns are outdated and already resolved.
Joe
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Mar 23 2002 - 22:00:21 EST