We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident. Audio is captured from the tape
machines, run through a lucid 96/24 ADC, and into the S/PDIF ins of an
M-audio delta dio 96/24 sound card. This setup worked fine before the
new OS; now we're also running SF 8. We've made sure to get updated
drivers for the sound card.

The problem is that the audio sounds like it's been captured at 44.1kHz
and then changed to 96kHz. There are crackles that are not present in
original material. I have checked and rechecked the audio card
settings in SF and have tried different settings.

If you have any ideas, I'd greatly appreciate them. Thanks.

You might want to make sure you have the latest SF update. I think its
8.0A or something (on Sony's site). I have SF8 but have not done any
recording with it yet.

--Peter

Kayte wrote:
We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident. Audio is captured from the tape
machines, run through a lucid 96/24 ADC, and into the S/PDIF ins of an
M-audio delta dio 96/24 sound card. This setup worked fine before the
new OS; now we're also running SF 8. We've made sure to get updated
drivers for the sound card.

The problem is that the audio sounds like it's been captured at 44.1kHz
and then changed to 96kHz. There are crackles that are not present in
original material. I have checked and rechecked the audio card
settings in SF and have tried different settings.

If you have any ideas, I'd greatly appreciate them. Thanks.

Kayte wrote:
We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident. Audio is captured from the tape
machines, run through a lucid 96/24 ADC, and into the S/PDIF ins of an
M-audio delta dio 96/24 sound card. This setup worked fine before the
new OS; now we're also running SF 8. We've made sure to get updated
drivers for the sound card.

The problem is that the audio sounds like it's been captured at 44.1kHz
and then changed to 96kHz. There are crackles that are not present in
original material. I have checked and rechecked the audio card
settings in SF and have tried different settings.

If you have any ideas, I'd greatly appreciate them. Thanks.


Do you have the audio properties set to "ASIO" or "Windows Clasic Wave
Driver" as opposed to "Microsoft Sound Mapper"?

Kayte,

We recently put XP on our audio pc

Windows XP needs to be optimized for audio performance. Most important is
disabling all unnecessary background services, but there are many other
tweaks you can apply to improve audio performance. This is what you should
investigate next. One resource is my three-part series from Keyboard
magazine, 5th in the list on my Articles page:

www.ethanwiner.com/articles.html

These articles also list many other resources that I'm sure you'll find
useful.

--Ethan

Chip Borton wrote:

Do you have the audio properties set to "ASIO" or "Windows Clasic Wave
Driver" as opposed to "Microsoft Sound Mapper"?

I tried this and nothing changed.

Ethan Winer wrote:
Kayte,

We recently put XP on our audio pc

Windows XP needs to be optimized for audio performance. Most important is
disabling all unnecessary background services, but there are many other
tweaks you can apply to improve audio performance. This is what you should
investigate next. One resource is my three-part series from Keyboard
magazine, 5th in the list on my Articles page:

www.ethanwiner.com/articles.html

These articles also list many other resources that I'm sure you'll find
useful.

--Ethan


Thanks

"Kayte" wrote in message

Thanks


http://mediasoftware.sonypictures.co....asp?ForumID=3

geoff

"Kayte" wrote in message...

We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident.


Why?!? What made you think you needed XP on a workstation that *worked*?

My suggestion would be to build another box for SF-8 and reconfigure your
old 98 system with SF-7 properly and keep it as a *reliable* back-up box.
We're talking small dollars these days to build a good PC.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

On Mon, 01 Aug 2005 22:22:42 GMT, "David Morgan \(MAMS\)"
wrote:


"Kayte" wrote in message...

We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident.


Why?!? What made you think you needed XP on a workstation that *worked*?

Good point. Maybe the computer can't handle XP.

Julian

On 1 Aug 2005 07:41:50 -0700, "Kayte" wrote:

We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident. Audio is captured from the tape
machines, run through a lucid 96/24 ADC, and into the S/PDIF ins of an
M-audio delta dio 96/24 sound card. This setup worked fine before the
new OS; now we're also running SF 8. We've made sure to get updated
drivers for the sound card.

The problem is that the audio sounds like it's been captured at 44.1kHz
and then changed to 96kHz. There are crackles that are not present in
original material. I have checked and rechecked the audio card
settings in SF and have tried different settings.

If you have any ideas, I'd greatly appreciate them. Thanks.

Windows XP is loaden with unnecessary effects, they can all be
switched off. Then there are many "tweak" programs which do additional
optimisations. Win XP Manager is an example. There are also lists of
Windows XP services which could be disabled; this is machine-dependent
and one has to be cautious when disabling some. But nothing helps if
the processor and the low amount of RAM can't handle Windows XP and
Sound Forge 7 and 8.

I am still using SF 6 with XP because I've found that SF 7 and 8 are
as many others applications of today, suffering of elephantism ie.
they are bloated. For example, SF8 took much more to load compared to
SF6. And the few new features, like various metering options, aren't
justifying a change to a higher version.

A good, optimized software, is hard to make but it is still a
blessing. These optimizations, however, are not "Visual This-And-That"
made but sometimes need a good mastering of the machine language and
processor architecture.

Hats off to those who still care and do.

Edi Zubovic, Crikvenica, Croatia

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

David Morgan (MAMS) wrote:
"Kayte" wrote in message...

We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident.


Why?!? What made you think you needed XP on a workstation that *worked*?

My suggestion would be to build another box for SF-8 and reconfigure your
old 98 system with SF-7 properly and keep it as a *reliable* back-up box.
We're talking small dollars these days to build a good PC.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

On Tue, 02 Aug 2005 07:58:59 -0700, Kayte wrote:

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

In what way was win98 unable to connect to the network. As far as my
experience goes I've found win98 supports most standard network protocols.
What are the hardware specs of the pc you upgraded to XP? Maybe it's just
not up to the task cpu and particularly memory wise?

--
Jafar Calley
Producer - http://moonlife-records.com
--------------------------------------
See the latest Mars and Saturn images
http://fatcat.homelinux.org

"Kayte" wrote in message oups.com...

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.


XP is on the office machine, it's bloated and cumbersome. The rest of
my PC boxes (some of them very new and blazingly fast) are all on 98SE.
None of them are networked except my internet box which is on DSL, so
it reacts (slow boot) like it's on a network. Sorry, I can't help much. Perhaps
you could set things up to dual boot and use a 98 partition. I'm also sorry
the higher-ups weren't more concerned about reliability... after all, there's
not much one can't simply burn to CD or DVD to move from PC to PC.

IMHO, networking is for a bunch of office puppies; audio is for dedicated
workstations. Of course, I'm told here pretty regularly that you can do
all sorts of things and run all sorts of software with XP and still get away
with it. But on the other hand, I meet people almost weekly that are
having great difficulty keeping up with their files, folders and even disc
space (especially multiple users) on XP. Hang on the Microsoft XP
groups for a while and you'll see a plethora of non-booting or randomly
shutting-down PCs, and BSOD complaints every day... and XP has
been around for quite a while now - no excuse for that. Wish I could
help more, but I'm keeping my SoundForge and CDArchitect boxes on
Windows 98 for a while longer.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

jafar wrote:
On Tue, 02 Aug 2005 07:58:59 -0700, Kayte wrote:

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

In what way was win98 unable to connect to the network. As far as my
experience goes I've found win98 supports most standard network protocols.
What are the hardware specs of the pc you upgraded to XP? Maybe it's just
not up to the task cpu and particularly memory wise?

--
Jafar Calley
Producer - http://moonlife-records.com
--------------------------------------
See the latest Mars and Saturn images
http://fatcat.homelinux.org

I don't know why they couldn't get it on the network. I didn't see a
problem with keeping it running 98, but my opinion doesn't count
because i'm just the audio person.

But that doesn't really matter because the administrators have declared
that this pc will run XP and that's how it is.
Now it is my job to figure out how we can get any work done.

This PC is pentium 4 2GHz, 512 MB ram. That should be plenty.

We are ordering 512MB more ram.

I've been using XP for a couple of years now and am much happier with it
than W2K or W98. Its much more stable, and most newer software will
likely require it (including SF8, looking at Sonys specified minimum
requirements).

It may not run so good on older systems, as it is somewhat more resource
intensive. You can disable unneeded things as others have suggested and
that will help some.

You should make sure that they did a clean install (and not an upgrade),
and only install the things that you really need.

Take a look in the registry under the key:
HKLM\Software\Microsoft\Windows\CurrentVersion\Run

And make sure there isn't anything in there that is not needed. These
entries cause additional processes to start up at boot time. Its also
the typical place that viruses and other parasites put their stuff in.
Just be careful with edits, as there is no undo. In unsure, export that
area of the registry to a file so you can restore.

Use the explorer and look under:
C:\Documents and Settings\All Users\Start Menu\Programs\Startup

And make sure there is nothing there that is not needed; this is another
place where things get started from at boot time.

Also look at services that can be turned off, like Messaging, Indexing,
etc. Make sure no antivirus stuff is running.

And there are a number of GUI "effects" things that can be turned off.

My main PC actually has (2) boot partitions, one with a minimal install
used for audio work, and business/play stuff on the other one. You might
want to explore that option too.

--Peter

Kayte wrote:
The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

David Morgan (MAMS) wrote:

"Kayte" wrote in message...


We recently put XP on our audio pc which had been running sound forge 7
on windows 98 without incident.


Why?!? What made you think you needed XP on a workstation that *worked*?

My suggestion would be to build another box for SF-8 and reconfigure your
old 98 system with SF-7 properly and keep it as a *reliable* back-up box.
We're talking small dollars these days to build a good PC.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

"Kayte" wrote in message oups.com...


jafar wrote:
On Tue, 02 Aug 2005 07:58:59 -0700, Kayte wrote:

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

In what way was win98 unable to connect to the network. As far as my
experience goes I've found win98 supports most standard network protocols.
What are the hardware specs of the pc you upgraded to XP? Maybe it's just
not up to the task cpu and particularly memory wise?

--
Jafar Calley
Producer - http://moonlife-records.com
--------------------------------------
See the latest Mars and Saturn images
http://fatcat.homelinux.org

I don't know why they couldn't get it on the network.

They didn't try hard enough... they wanted to be 'trendy', up-to-date,
and able to look into your computer. They disregarded all concern
for reliable audio in this pursuit - but that's just my humble opinion.

I didn't see a
problem with keeping it running 98, but my opinion doesn't count
because i'm just the audio person.

There wasn't a problem. Networking was the problem, and they
apparently didn't try very hard to link up with 98.

But that doesn't really matter because the administrators have declared
that this pc will run XP and that's how it is.

I'm sorry you're having to go through this. I have a client that just went
to XP and I'm having to learn more of it than I care to.

Now it is my job to figure out how we can get any work done.

Optimize....

http://www.jakeludington.com/ask_jak...and_video.html

Optimize. http://www.tascamgiga.com/pdf/optimizing-xp-and-2k.pdf

Optimize. http://musicxp.net/ (Do NOT click on any of the links in
the article... just use the menu at the left).

http://www.tunexp.com/tips/work_with...audio_quality/

http://www.soniccontrol.com/tech/mid...tweaking.shtml

Shut down services you don't need.... (optimize)
http://www.beemerworld.com/tips/servicesxp.htm



They probably made you use an 'upgrade' disc... so now it's virtually
impossible to remove it. Plus, now that it's installed, you've missed
a lot of opportunities for set-up preferences, optimizing, dual booting,
etc.. Best of luck, I'm sure you will make it work.

This PC is pentium 4 2GHz, 512 MB ram. That should be plenty.

I'm posting now on a box that's exactly this spec. It's basically a surf
toy, but there are a few audio programs on there and software for
supporting web sites, pretty much nothing more... just the 98SE OS.
If you whittle things down to just what you need, it should be fine...
although I don't think there's any reason it won't take a gig of RAM.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

thanks for the links-- I think this computer is now properly optimized.

On 2 Aug 2005 07:58:59 -0700, "Kayte" wrote:

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

Maybe its time to tell the higher ups you need a new computer with
lots of RAM and processor speed. I would at least explain to them the
system that used to work just fine will never work the same even if
you do solve your initial problems, the extra overhead of XP will
speciously slow down your workstation.

Julian

"Kayte" wrote in
oups.com:



I don't know why they couldn't get it on the network. I didn't see a
problem with keeping it running 98, but my opinion doesn't count
because i'm just the audio person.


Most corporate IT support staff set standards for machines they'll _permit_
to attach to their network. Certainly where I worked, Windows 98 SE
machines were banned from network connection several years ago (some still
found in labs, and probably embedded in other machines). Much as I've raged
at IT folks, there are actually pretty good network stability and security
reasons for some restrictions, including this one.

This PC is pentium 4 2GHz, 512 MB ram. That should be plenty.


I run SF6 on a 3.2 GHz, hyperthreaded Pentium 4 with 2M cache and 1Gbyte
RAM under Windows XP Pro. Unless Sonic Foundry or Sony pigged it badly
going from version 6 to 8, I think you will find the main benefit to more
RAM will be the larger disk cache it enables. The practical benefit is
that you can do more edits, on larger audio files, before the editing
requires reference to an actual disk copy rather than coming back out of
the cache. That should _not_, however relate to functional problems. If
the files you edit are short, and your number of edits limited, you'd not
even notice the difference (nothing helps like more RAM when you need it,
and nothing is more useless than more RAM when you don't).

I've seen not issues, however, I never run audio directly into the machine.
I used to run S/PDIF in (to an Audiophile 2496), which is a real-time
transfer. I always took the machine off the network, rebooted, and turned
off _everything_ before doing that import. Now I run Firewire in, which
is just another asynchronous thing, and don't worry about it.

Good luck,
Peter A. Stoll

(used to run Sound Forge 4ish on a Windows 98SE machine, running SF6 on
this XP machine is way, way better).

If
the files you edit are short, and your number of edits limited, you'd not
even notice the difference (nothing helps like more RAM when you need it,
and nothing is more useless than more RAM when you don't).

The files I edit are long (15 minutes up to 3 hours), and I do a small
to moderate amount of editing. I'll bet I could use more RAM...


I've seen not issues, however, I never run audio directly into the machine.
I used to run S/PDIF in (to an Audiophile 2496), which is a real-time
transfer. I always took the machine off the network, rebooted, and turned
off _everything_ before doing that import. Now I run Firewire in, which
is just another asynchronous thing, and don't worry about it.


Could you explain this to me? What are the issues that running S/PDIF
into the sound card causes?

Thanks

"Kayte" wrote in
oups.com:


The files I edit are long (15 minutes up to 3 hours), and I do a small
to moderate amount of editing. I'll bet I could use more RAM...


Yes, I bet it will help for you, though on the longer files full-file
edits (such as a full-file level change) you are going to go to disk
anyway unless your motherboard will take more RAM than I guess is likely.
With my Gigabyte of RAM, 10-minute 24/44.1 files go like lightning, but
my usual concert-length files (60 minutes) us the disk anyway.

If you go from 512 to 1G, you'll be just about tripling the working RAM
available for useful work (XP chews about a 1/4 Gb as the price of
admission, if you or your IT folks have you running extra stuff it could
be worse).



Could you explain this to me? What are the issues that running S/PDIF
into the sound card causes?


Running S/PDIF in is just like running audio in for the important aspect
that it is not a proper communications protocol at all. The data just
come along at their own clip ("real time") and if the receiver is busy
for a moment, there is no way to say "retry" or "I did not get that" or
even "error, don't trust this file". This means that if the processes
running on the CPU don't visit the buffer(s) soon enough, you are
guaranteed error (though some applications seem to keep track and attempt
to show you such problems--I recall SF showed me import glitches once
when I was not disciplined about cleaning the system of distractions
first.

By contrast, just about any computer communications protocol includes
provisions for noticing whether the data came across OK, and for
requesting and getting retries from the source if they did not.
Sometimes a provision for error correction allows small errors to be
perfectly corrected without a resend.

Think about it--if Operating System installs were subject to transmission
error at the rate folks see on audio input to some systems, the PC would
never even boot! Unreliable as PC's are as devices, most data
transmission they are involved in is flawless or declares itself invalid.

The protocols come in layers, so just because the data move across USB or
Firewire or ... is not proof it is a robust transmission, but at least it
is possible. With pure audio in or S/PDIF, about all you can hope is
that an application on the receiving end will notice it waited too long
and warn you not to trust the result. But against noise error on the
physical transmission link, you are defenseless.

So I'll "capture" and audio source on LP or cassette or even DAT by
playing it into my SD722 offline, then transfer the 722 file either by
Firewire (IEEE 1394) or by pulling out the CF card and putting it in a
reader on the PC. Avoiding electrical noise is one reason, but avoiding
corrupt transfers (and not needing to "rig for silent running" the PC by
turning everything off) is a stronger reason for me.

Peter A. Stoll
retired computer guy
singer
amateur sound recording guy

On 2 Aug 2005 07:58:59 -0700, "Kayte" wrote:

The higher ups ("they") decided they needed to be able to connect the
audio pc to the network, which required it to have XP.

My network includes XP and W98 machines. Why couldn't theirs?

Have you yet shared the specs of this machine? IS it underpowered for
XP?

On Tue, 02 Aug 2005 16:50:47 GMT, "David Morgan \(MAMS\)"
wrote:

Hang on the Microsoft XP
groups for a while and you'll see a plethora of non-booting or randomly
shutting-down PCs, and BSOD complaints every day... and XP has
been around for quite a while now - no excuse for that.

To be fair, you'd have seen just as many (or very likely more) on W98
groups when that was the mainstream os. And, in both cases, you're
only seeing the complaints, not the satisfied customers.

The bloat is one issue. But it's configurable. People who attempted
an in-place upgrade to XP from 98, maybe on an imperfect installation
or inadequate hardware are another. Imperfect XP drivers,
particularly from niche-market suppliers, are yet another (though now
mostly resolved). But it's silly to denounce XP as an unstable os.

On 2 Aug 2005 10:13:42 -0700, "Kayte" wrote:

This PC is pentium 4 2GHz, 512 MB ram. That should be plenty.

That seems plenty for XP and SoundForge. Which is only a two-track
recorder, after all.

On Tue, 02 Aug 2005 15:19:45 -0500, "Peter A. Stoll"
wrote:

I run SF6 on a 3.2 GHz, hyperthreaded Pentium 4 with 2M cache and 1Gbyte
RAM under Windows XP Pro. Unless Sonic Foundry or Sony pigged it badly
going from version 6 to 8, I think you will find the main benefit to more
RAM will be the larger disk cache it enables. The practical benefit is
that you can do more edits, on larger audio files, before the editing
requires reference to an actual disk copy rather than coming back out of
the cache. That should _not_, however relate to functional problems. If
the files you edit are short, and your number of edits limited, you'd not
even notice the difference (nothing helps like more RAM when you need it,
and nothing is more useless than more RAM when you don't).


Indeed. If you stuff your machine full of RAM, offline processing of
large files will speed up noticeably. But recording and playback
won't be any different. Audio buffers are measured in KB not MB.

On 2 Aug 2005 13:37:48 -0700, "Kayte" wrote:

I've seen not issues, however, I never run audio directly into the machine.
I used to run S/PDIF in (to an Audiophile 2496), which is a real-time
transfer. I always took the machine off the network, rebooted, and turned
off _everything_ before doing that import. Now I run Firewire in, which
is just another asynchronous thing, and don't worry about it.


Could you explain this to me? What are the issues that running S/PDIF
into the sound card causes?

And why would it be different to running analogue into the sound card?
One way the card's ad converters have a job to do, one way they
don't. Why would the computer care?

I agree that, if a computer has been doing other work, a reboot is
good practice before audio recording. As is disabling the network.
But often I forget, and it seems to make no difference :-)

On Tue, 02 Aug 2005 16:24:52 -0500, "Peter A. Stoll"
wrote:

10-minute 24/44.1 files go like lightning, but
my usual concert-length files (60 minutes) us the disk anyway.

Even with longer files, you find more RAM lets off-line processing be
done in bigger chunks. It's quicker.

Laurence Payne wrote in
:

On 2 Aug 2005 13:37:48 -0700, "Kayte" wrote:


And why would it be different to running analogue into the sound card?
One way the card's ad converters have a job to do, one way they
don't. Why would the computer care?


I agree-the same problems are presented by analog audio-in and S/PDIF in.
That was my point in distinguishing these from IEEE 1394 (Firewire) or USB
transfers. Real-time transfers with no proper protocol, and no mechanism
for the receiving side to tell the sending side "wait--I'm not ready for
more" or "send that last bit again, I did not get it" pose a different
problem than proper transfers supported by a proper protocol.

Peter A. Stoll

Laurence Payne wrote in
:

On Tue, 02 Aug 2005 15:19:45 -0500, "Peter A. Stoll"
wrote:

I run SF6 on a 3.2 GHz, hyperthreaded Pentium 4 with 2M cache and
1Gbyte RAM under Windows XP Pro. Unless Sonic Foundry or Sony pigged
it badly going from version 6 to 8, I think you will find the main
benefit to more RAM will be the larger disk cache it enables. The
practical benefit is that you can do more edits, on larger audio
files, before the editing requires reference to an actual disk copy
rather than coming back out of the cache. That should _not_, however
relate to functional problems. If the files you edit are short, and
your number of edits limited, you'd not even notice the difference
(nothing helps like more RAM when you need it, and nothing is more
useless than more RAM when you don't).


Indeed. If you stuff your machine full of RAM, offline processing of
large files will speed up noticeably. But recording and playback
won't be any different. Audio buffers are measured in KB not MB.

But if your "recording" is done on another machine, and gets to the PC by
Firewire or USB transfer, and if audio only leaves the machine by file
transfer or CD/DVD burning, then there is _no_ recording exposed, and the
playback is only monitoring, and not in the actual final product sound path
at all. I do hear occasional glitches in my playback, as I don't turn
everything off for processing work, and don't defrag. Most likely the PC
failed to find the next sector on the disk fast enough. If I'm worried I'm
hearing something in the actual material, I just click back on the waveform
and listen again.

Peter Stoll

Peter A. Stoll wrote:
"Kayte" wrote in
oups.com:


The files I edit are long (15 minutes up to 3 hours), and I do a small
to moderate amount of editing. I'll bet I could use more RAM...


Yes, I bet it will help for you, though on the longer files full-file
edits (such as a full-file level change) you are going to go to disk
anyway unless your motherboard will take more RAM than I guess is likely.
With my Gigabyte of RAM, 10-minute 24/44.1 files go like lightning, but
my usual concert-length files (60 minutes) us the disk anyway.

If you go from 512 to 1G, you'll be just about tripling the working RAM
available for useful work (XP chews about a 1/4 Gb as the price of
admission, if you or your IT folks have you running extra stuff it could
be worse).



Could you explain this to me? What are the issues that running S/PDIF
into the sound card causes?


Running S/PDIF in is just like running audio in for the important aspect
that it is not a proper communications protocol at all. The data just
come along at their own clip ("real time") and if the receiver is busy
for a moment, there is no way to say "retry" or "I did not get that" or
even "error, don't trust this file". This means that if the processes
running on the CPU don't visit the buffer(s) soon enough, you are
guaranteed error (though some applications seem to keep track and attempt
to show you such problems--I recall SF showed me import glitches once
when I was not disciplined about cleaning the system of distractions
first.

By contrast, just about any computer communications protocol includes
provisions for noticing whether the data came across OK, and for
requesting and getting retries from the source if they did not.
Sometimes a provision for error correction allows small errors to be
perfectly corrected without a resend.

Think about it--if Operating System installs were subject to transmission
error at the rate folks see on audio input to some systems, the PC would
never even boot! Unreliable as PC's are as devices, most data
transmission they are involved in is flawless or declares itself invalid.

The protocols come in layers, so just because the data move across USB or
Firewire or ... is not proof it is a robust transmission, but at least it
is possible. With pure audio in or S/PDIF, about all you can hope is
that an application on the receiving end will notice it waited too long
and warn you not to trust the result. But against noise error on the
physical transmission link, you are defenseless.

So I'll "capture" and audio source on LP or cassette or even DAT by
playing it into my SD722 offline, then transfer the 722 file either by
Firewire (IEEE 1394) or by pulling out the CF card and putting it in a
reader on the PC. Avoiding electrical noise is one reason, but avoiding
corrupt transfers (and not needing to "rig for silent running" the PC by
turning everything off) is a stronger reason for me.

Peter A. Stoll
retired computer guy
singer
amateur sound recording guy

Thank you. I am making a list of issues to bring up with my
supervisors about the setup, and this will go on it.

Laurence Payne wrote:

...snip...

To be fair, you'd have seen just as many (or very likely more) on W98
groups when that was the mainstream os. And, in both cases, you're
only seeing the complaints, not the satisfied customers.

The bloat is one issue. But it's configurable. People who attempted
an in-place upgrade to XP from 98, maybe on an imperfect installation
or inadequate hardware are another. Imperfect XP drivers,
particularly from niche-market suppliers, are yet another (though now
mostly resolved). But it's silly to denounce XP as an unstable os.

Bloat issues are only re-configurable if allowed by the
corporate network/IP folks. In my experience with corporate
networks the support folks tend to want all computers on the
network to look and act the same to aid in remote administration
and such.

I had to build and maintain my own lab networks to meet my
experimental needs. The corporate folks wouldn't let me on
their net with a non standard computer. Most data had to
be moved via the sneaker net.

The OP indicated similar corporate sys-admin problems.
[YMMV]

Later...

Ron Capik
--

On Wed, 03 Aug 2005 08:07:31 -0500, "Peter A. Stoll"
wrote:

I agree-the same problems are presented by analog audio-in and S/PDIF in.
That was my point in distinguishing these from IEEE 1394 (Firewire) or USB
transfers. Real-time transfers with no proper protocol, and no mechanism
for the receiving side to tell the sending side "wait--I'm not ready for
more" or "send that last bit again, I did not get it" pose a different
problem than proper transfers supported by a proper protocol.

You'd be making a good case for using a dedicated device for audio
capture except that we all DO routinely achieve glitchless audio input
using quite un-esoteric techniques on multi-purpose computers.

On Wed, 03 Aug 2005 08:12:20 -0500, "Peter A. Stoll"
wrote:

I do hear occasional glitches in my playback, as I don't turn
everything off for processing work, and don't defrag. Most likely the PC
failed to find the next sector on the disk fast enough. If I'm worried I'm
hearing something in the actual material, I just click back on the waveform
and listen again.

If you're just playing back a stereo track in SoundForge, I'm very
surprised you're getting any glitches at all. SF has a very adequate
audio buffering system, and two audio tracks are a very undemanding
job for a modern hard drive.

On Wed, 03 Aug 2005 13:34:14 GMT, Ron Capik
wrote:

Bloat issues are only re-configurable if allowed by the
corporate network/IP folks. In my experience with corporate
networks the support folks tend to want all computers on the
network to look and act the same to aid in remote administration
and such.

I had to build and maintain my own lab networks to meet my
experimental needs. The corporate folks wouldn't let me on
their net with a non standard computer. Most data had to
be moved via the sneaker net.

The OP indicated similar corporate sys-admin problems.
[YMMV]

It's hard to understand why even the most obdurate corporate sys-admin
couldn't tolerate a stand-alone computer performing a specialist task.
I suspect subtext to this story :-)

Laurence Payne wrote in
:

On Wed, 03 Aug 2005 08:12:20 -0500, "Peter A. Stoll"
wrote:

I do hear occasional glitches in my playback, as I don't turn
everything off for processing work, and don't defrag. Most likely the
PC failed to find the next sector on the disk fast enough. If I'm
worried I'm hearing something in the actual material, I just click
back on the waveform and listen again.

If you're just playing back a stereo track in SoundForge, I'm very
surprised you're getting any glitches at all. SF has a very adequate
audio buffering system, and two audio tracks are a very undemanding
job for a modern hard drive.


Quite right, the drives are far faster than needed for this task (two RAID
pairs of 120 Gbyte Western Digital 8Mbyte buffer drives, I put the CF temp
files on one, and the data on the other, though once I've done a processing
pass or two it is probably going temp to temp). Most likely XP was off
attending to something else during the period when it would have needed to
be chasing the pointer tree to find the next sectors.

I think that sort of thing is precisely the point of this thread. If I
assure the system does not have distractions it has no trouble at all
keeping up. But XP can get distracted for longer than is consistent with
error-free real-time audio in or out quite easily--even on extremely fast
systems.

Peter A. Stoll

Laurence Payne wrote in
:


You'd be making a good case for using a dedicated device for audio
capture except that we all DO routinely achieve glitchless audio input
using quite un-esoteric techniques on multi-purpose computers.


I'll disengage with you on this point here.
Peter

"Peter A. Stoll" wrote in message...

This means that if the processes
running on the CPU don't visit the buffer(s) soon enough, you are
guaranteed error (though some applications seem to keep track
and attempt to show you such problems--I recall SF showed me
import glitches once when I was not disciplined about cleaning the
system of distractions first.

With pure audio in or S/PDIF, about all you can hope is
that an application on the receiving end will notice it waited too long
and warn you not to trust the result.

Avoiding electrical noise is one reason, but avoiding
corrupt transfers (and not needing to "rig for silent running" the PC by
turning everything off) is a stronger reason for me.


Hi Peter,

Sorry to 'snippet' your responses, but I wanted to highlight the very
reasons that I can't trust XP until I learn a whole lot more about it.

I don't follow the corrupt transfer logic, however. I've been transferring
audio into (DEDICATED) workstations on both the Mac and PC varieties
for over 18 years, and there have been less than a handfull of anomalies
combined from both platforms during that time.


--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.artisan-recordingstudio.com

"Laurence Payne" wrote in message ...
On Wed, 03 Aug 2005 08:12:20 -0500, "Peter A. Stoll"
wrote:

I do hear occasional glitches in my playback, as I don't turn
everything off for processing work, and don't defrag. Most likely the PC
failed to find the next sector on the disk fast enough. If I'm worried I'm
hearing something in the actual material, I just click back on the waveform
and listen again.

If you're just playing back a stereo track in SoundForge, I'm very
surprised you're getting any glitches at all. SF has a very adequate
audio buffering system, and two audio tracks are a very undemanding
job for a modern hard drive.

I've been on SF since 4.0 and using Win98SE have never encountered a
glitch on a dedicated workstation. I have SF4.xx, SF6.xx and SF7 running
on a variety of Win98 boxes now... but they are dedicated, streamlined boxes.

DM

"Laurence Payne" wrote in message...

But it's silly to denounce XP as an unstable os.


Didn't do that... but lot's of people *are* crashing and burning. I have
access to two XP boxes now, neither have had problems. But neither
of them are running audio of a professional nature. I'm still afraid of
it as a platform for being able to guarantee my work, as I really don't
understand it's bloat and service processes yet. Couple that with the
fact that I have several 98 workstations on very fast, dedicated PCs,
that are performing superbly, and I have no reason to make any changes.

I never understood all of the complaints in this group about 98, as I was
apparently lucky enough to get first installs (no upgrades or patches are
needed on dedicated audio boxes) that worked immediately and continue
to work.

What's apparently a *big* problem over in the XP groups right now, are
folks who are adding SP-2 and coming back totally dead in the water.

DM