Re: Timing problem on PPC based Macs (PortAudio)
Re: Timing problem on PPC based Macs (PortAudio)
- Subject: Re: Timing problem on PPC based Macs (PortAudio)
- From: James Chandler Jr <email@hidden>
- Date: Fri, 20 Mar 2009 21:25:04 -0400
Hi Thomas
A bit more info, plus a question at the end for the gurus.
A hardware sync for your current application is probably unnecessary,
however the sync/speed issue is common enough that folks who work with
multiple audio/video machines often tie them together at a hardware
level, so that many machines share the same samplerate.
There were assorted sync solutions as equipment evolved. If a user has
reason to suspect that one of his audio gadgets has the best clock,
then his most-likely-accurate device may get chosen as the master.
Ordinary SPDIF cables may be the simplest hardware samplerate sync
(assuming your devices have SPDIF connections). SPDIF passes a word
clock signal along with audio, so an SPDIF cable connected from the
master device output to the slave device input can lock the slave
gadget's samplerate against the master samplerate.
It may not work with all SPDIF devices, because some audio devices
have built-in samplerate conversion, and supposedly some audio
interfaces can have different samplerate clocks on input versus output.
Dunno if an SPDIF cable would simple-sync two different mac computers'
internal audio hardware. Folks here probably know.
If necessary, you can buy high accuracy timebase generators. Some
audio-oriented sync devices primarily output word clock. Folks who do
audio+video often use black burst generators, which are accurate video
timebase generators. In the past it was not unknown for high-end audio-
only guys to use black burst generators for master sync, but I don't
know if that remains common practice.
Some high-accuracy dedicated timebase gadgets use precision crystals
enclosed in tiny thermostat-controlled ovens. Provided that the tiny
oven can keep the crystal at a constant temperature hotter than the
highest expected ambient environmental temp, the crystal always
operates at a constant temp and is more likely to operate at a fairly
repeatable frequency.
====
Here is a dumb question for the list:
How does a modern Mac's CPU timebase accuracy compare against a mid-
price digital audio device's samplerate accuracy?
Would there be reason to expect that a Mac Pro's UpTime or
mach_absolute_time nanosecond measurement would be more long-term-
accurate, less long-term-acurate, or about the same long-term accuracy
compared to a nice mid-price audio interface like a MOTU 2408mk3?
Am not requesting criticism of specific hardware. I'm just curious if
there are design features of mass-produced computer circuitry which is
more-likely-accurate than the samplerate of mid-price audio devices?
Am assuming that a Mac soundchip timebase may be 'looser' than more-
expensive audio interfaces, but is the hardware timer driving
mach_absolute_time especially likely to be 'really really accurate'?
Thanks!
James Chandler Jr.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden