Subject: Re: [linux-audio-dev] Timestamping?
From: Steve Harris (S.W.Harris_AT_ecs.soton.ac.uk)
Date: Fri Jun 08 2001 - 14:32:07 EEST
On Thu, Jun 07, 2001 at 05:02:18PM -0700, Simon Per Soren Kagedal wrote:
> struct AudioTimeStamp
> {
> Float64 mSampleTime; // the absolute sample time
> UInt64 mHostTime; // the host's root timebase's
> // time
> Float64 mRateScalar; // the system rate scalar
> UInt64 mWordClockTime; // the word clock time
> SMPTETime mSMPTETime; // the SMPTE time
> UInt32 mFlags; // the flags indicate which
> // fields are valid
> };
Looks to me like they have one, global sync mechanism, wouldn't a cleaner
implementation be to have 0 or more sync plugins that have SMPTE etc. typed
outputs.
As Paul said, many of these sync signals are more complicated than an
integer. The "word clock time" one is particularly baffling.
- Steve
This archive was generated by hypermail 2b28 : Fri Jun 08 2001 - 15:08:49 EEST