Re: [LAU] hardware - Intel CPUs

On Thursday 17 April 2014 18:18:05 david did opine:

> On 04/17/2014 12:13 AM, Gene Heskett wrote:
> > On Thursday 17 April 2014 06:09:56 david did opine:
> >> On 04/16/2014 02:54 AM, Gene Heskett wrote:
> >>> On Wednesday 16 April 2014 08:44:15 david did opine:
> >>>> On 04/15/2014 03:17 AM, Len Ovens wrote:
> >>>>> On Tue, 15 Apr 2014, James Mckernon wrote:
> >>>>>> On Mon, Apr 14, 2014 at 2:06 AM, Len Ovens <len@email-addr-hidden>
> >
> > wrote:
> >>>>>> I would realy like to stay away from having to use a USB
> >>>>>> or FW audio IF. In fact I would like to be able to
> >>>>>> continue to use my delta 66 for as long as I can before
> >>>>>> I spend more money :) The
> >>>>>>
> >>>>>> Thanks for the useful info in your post. Just to be clear on this
> >>>>>> part: are
> >>>>>> you saying you don't want to switch to USB/FW solely because you
> >>>>>> want to keep using your delta 66, or because you have some
> >>>>>> definite preference for
> >>>>>> PCIe over USB/FW devices? If the latter, I wonder why?
> >>>>>
> >>>>> USB in audio is limited. Getting clear USB ports interrupt wise is
> >>>>> not easy. Audio can not be on a hub or share it's usb with
> >>>>> anything else, but many new MB have no mouse or kb port so the
> >>>>> USB is already being used for that much. The real reason though,
> >>>>> is latency. With the pci the latency can be 1/4 what it can be in
> >>>>> USB or FW. That is the lowest seeting jack for USB or FW is 64/2,
> >>>>> but I can run the d66 at 16/2 with no problem on a well tuned
> >>>>> system. This does make a difference for live work. I know that
> >>>>> 64/2 seems like very good latency (it is) but remember that the
> >>>>> card then adds another ms in each direction as well as the stage
> >>>>> distances on top of that. That is the time it takes the sound to
> >>>>> reach my ear after going through the computer as a processor and
> >>>>> then through the air to my ear. Maybe that is still not worth
> >>>>> worrying about... but even with 30 feet of cord and no digital
> >>>>> delay, I can hear the delay from my playing to the sound reaching
> >>>>> my ear.
> >>>>
> >>>> Interesting. What is the difference between speed of sound in air
> >>>> and the speed of electricity through a cable?
> >>>
> >>> Sound is nominally 720 miles per hour. Rather leisurely IOW.
> >>> A perfect cable is C speed, 258 times faster. But cable (coaxial)
> >>> actually range in speeds between 66% of C for home usable cables, to
> >>> around 98% of C for 9" diameter high power broadcast stuff, C being
> >>> 186,272 miles per second in a vacuum. Thats 298,035.2 kilometers
> >>> per second for the metric folks here.
> >>
> >> Then running your sound from stage to backhouse sound board back to
> >> stage and hearing it through headphones would give no latency at all.
> >
> > For an analogue board, small fraction of a millisecond, for a digital
> > board, anybodies guess. A/D and D/A's are essentially pretty quick,
> > but I'd still put most digital boards above a millisecond.
>
> That's interesting. We're looking into replacing the analog board with a
> digital one.

We replaced an old 12 stereo channel 48 input logitek analog board,
couldn't keep ahead of replacing the CDS channel switches in it because it
was full of 5532's and even with added fans, their life in that heat was
about a year, with a 24 input Macky(SP?). Digital, programmable motorized
gain pots, very low noise. But we had to shut the studio monitors off, the
delay had our talent stumbling over their own shoelaces trying to read the
teleprompter.

I didn't have a way to measure it, other than a dual trace scope and hand
claps which said 16 to 17 ms, but I'd have guessed around 20. Knowing what
to expect, I still couldn't read the opening paragraph of a tv technology
article without stumbling lots of times if the monitor was loud enough to
sound like an echo. The TD can still talk in their ear, but no backfeed
goes out that circuit, its killed when the mics go live.

We have an even newer board now, 48 stereo wide SMPTE out, and I believe
its considerable faster than the Macky was, but I don't think the studio
monitors are used for anything but talkback during commercials yet. I've
been retired since mid 2002, and haven't really tracked that stuff 100%
since.

Cheers, Gene

-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user