[LAU] digital voodoo: master fader should be set at 0db

[Arnold Krille]

On Wednesday 14 April 2010 14:56:53 Paul Davis wrote:
> On Wed, Apr 14, 2010 at 12:23 AM, Rick Green <rtg at aapsc.com> wrote:
> >   Last weekend, I was running FOH for a couple of concerts, and the
> > artist had hired a professional recording engineer to lay down a
> > multitrack of the show for possible commercial release.  THey had a few
> > 'special surprise guests' showing up, so I kept bringging out 'just one
> > more' microphone until the channel count was up to 22, I think.  After
> > sound check, the recording engineer expressed some trepidation that his
> > external hard drive could handle all that bandwidth.  I asked him if he
> > had pushed record and tried it during soundcheck, and he said "Of course,
> > but you know when they get excited and start playing loud during the
> > show, it fills up those bits pretty quick, and maybe then the firewire800
> > won't be fast enough."
> 
> good story.
> 
> what's ironic though is that its now reasonably well documented that
> if the disk drive is in the line of fire when they start to play loud,
> it really will be unable to keep up. this has nothing to do with bit
> rates, but is (probably) caused by the the vibrations causing read
> failures which necessitate a lot of retrys, thus slowing down the
> effective streaming bandwidth of the disk. if the disk is kept out the
> way of direct incoming sound, the issue goes away.
> 
> yes, really!

For all who won't believe Paul: That is a fact kind of proven by (I think) Sun 
engineers who wondered about disk performance in their testing environment 
being better than in the real server room. Then they shouted at the disks and 
it got worse even more. They did a video on that where the show the error-
rates going up just when they shout at the disk.

Vibrations in the audible frequency make your disk loose track while 
reading/writing.

So you don't really want to get a mouse-pad to protect against vibrations but 
want to enclose the whole disk-array in a sound-prove case. Of course it 
doesn't hurt to decouple that mechanically against deep-frequency vibrations.

And the next time you shout at your computer because of disk-errors: That is 
what makes it worse:-)

Arnold
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
Url : http://lists.linuxaudio.org/pipermail/linux-audio-user/attachments/20100414/76a09f89/attachment.pgp