>> Initially I will start out with regular ones since this only gets used
ones or twice a week for a couple of hours, but as it gets more and more
use and they depend on it more and more I will then move them towards the AV
grade disk.
Well, if t-calc is a problem, you'll know it pretty quick because you'll
have drop-outs in the recording. But it the drives you use work, then there
is no need to go to AV grade. It's not a matter of wearing out, it's that
you can't record without dropping bits. So if you get cheap drives with big
buffers and/or put them in a stripe array (RAID 0), you should be fine. You
might consider putting them in a RAID 0+1 array, so if you lose a drive, you
don't mess up the church service. Lots of small, cheap drives should work
better than one big cheap drive.
Ken Elliott
=====================
-----Original Message-----
From: slug@nks.net [mailto:slug@nks.net] On Behalf Of Chuck Hast
Sent: Tuesday, January 09, 2007 9:08 PM
To: slug@nks.net
Subject: Re: [SLUG] Audio storage and manipulation
On 1/9/07, Ken Elliott <kelliott11@cfl.rr.com> wrote:
> >> OK... What is thermal calc, I assume that it has to do with
temperature.
>
> Yes. Ever so often, a drive will stop accepting I/O commands in order
> to recalculate the position of the head, relative to the platter. As
> the platter and actuator arms warm up, they expand and that throws off
> the position. In most applications, this is unnoticed. But in a
> real-time application, you lose the data while this is going on. The
> general method was to put a larger buffer on the drive and continue to
> accept I/O commands while T-Calc was being performed. There are other
> methods, but you get the idea.
>
> >> Also what is a AV rated drive, is that AV as in Audio Visual?
>
> Yes. All it means is that it will meet it's rated write speed
> continuously, and that the write speed is fast enough for an
> audio/visual recording application. One trick is to get a large
> drive, and partition is into several small drives (or perhaps just
> one). Since the average access time is calculated at the time it
> takes to move the heads from the centre of partition to the edge of
> the partition, a smaller partition will speed up seek time. On
> 'famous' drive is actually a large drive with the firmware set to restrict
the movement of the head to a small area.
>
> One thing to watch for. Really large drives (consumer-grade) are now
> so dense that they _always_ using error correction. If you want a
> robust drive, get the smallest capacity drive you can get. The
> exception _might_ be high-end commercial SCSI drives, that are 3x as
> expensive and are far more robust. My workstation has three 70GB
> 15,000 rpm drives hooked to an Ultra 320 SCSI interface and they scoot!
>
OK, that gives me a much better idea along with the data that Steve sent.
I can see where the recalc could be a problem, and of course proper
buffering would deal with it except in the cases where the data moving
in/out of the disk was such a high volumn that the system would not know
where to put it or the delay for the recalc would case a drop in the
outgoing data stream.
I will certainly take the HD issues into account. Initially I will start out
with regular ones since this only gets used ones or twice a week for a
couple of hours, but as it gets more and more use and they depend on it
more and more I will then move them towards the AV grade disk.
Now to just figure out which software will meet my needs.
-- Chuck Hast -- KP4DJT -- To paraphrase my flight instructor; "the only dumb question is the one you DID NOT ask resulting in my going out and having to identify your bits and pieces in the midst of torn and twisted metal." ----------------------------------------------------------------------- This list is provided as an unmoderated internet service by Networked Knowledge Systems (NKS). Views and opinions expressed in messages posted are those of the author and do not necessarily reflect the official policy or position of NKS or any of its employees.----------------------------------------------------------------------- This list is provided as an unmoderated internet service by Networked Knowledge Systems (NKS). Views and opinions expressed in messages posted are those of the author and do not necessarily reflect the official policy or position of NKS or any of its employees.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 20:30:35 EDT