large files in sound forge - quantizing

[Kayte]

I'm working in sound forge with a wav file that is 3.5 hours long and
cannot be split up (it's eventually going to be encoded in real and
quicktime for streaming on the internet). It was captured at 96/24 and
it is too big to be saved right now (I believe 2 gigs is the max for
wav files). I'm going to be resampling to 44.1kHz or possibly 22.5. I
believe it will still be too big if I take it down to 16. Does anybody
know of another solution besides going to 8-bit or splitting it up?

Thanks,

Kayte

[Arny Krueger]

"Kayte" > wrote in message
ups.com

> I'm working in sound forge with a wav file that is 3.5
hours
> long and cannot be split up (it's eventually going to be
> encoded in real and quicktime for streaming on the
internet).
> It was captured at 96/24 and it is too big to be saved
right
> now (I believe 2 gigs is the max for wav files). I'm
going to
> be resampling to 44.1kHz or possibly 22.5. I believe it
will
> still be too big if I take it down to 16. Does anybody
know
> of another solution besides going to 8-bit or splitting it
up?

Stereo 44/16 files run about 10.8 mB per minute. That's
about 3 hours and 9 minutes in a 2,048 megabyte (2 GB) file.
So, no you're not going to hit the 3:30 mark in 2GB at
44/16.

8 bits is pretty egregious-sounding. ;-(

If you can do it, the best thing to do at this point is to
sacrifice sample rate. If 32k samples/sec is and available
option, it cabn sound remarkably good. 22.5 can work if the
track doesn't need a lot of crispness to work artistically.
After all, 22.5 is not that much worse than typical 16 mm
optical sound.

Other not-all-that not bad options that may or may not be
available is 12 bits per sample, at least available on some
Macs.

[EricK]

Kayte wrote:
> I'm working in sound forge with a wav file that is 3.5 hours long and
> cannot be split up (it's eventually going to be encoded in real and
> quicktime for streaming on the internet). It was captured at 96/24 and
> it is too big to be saved right now (I believe 2 gigs is the max for
> wav files). I'm going to be resampling to 44.1kHz or possibly 22.5. I
> believe it will still be too big if I take it down to 16. Does anybody
> know of another solution besides going to 8-bit or splitting it up?

Sample Rate conversion to 48kHz will cut the file size in half. That
should get you below 2 gigs. I personally would also reduce the bit-rate
to 16bit. Your final destination is a streaming format. In My opinion,
you don't need such a high resolution file at this point in the game.
The smaller you can get the file, the quicker you can render any
processing that you plan on applying.

--
Eric

Practice Your Mixing Skills
www.Raw-Tracks.com
www.Mad-Host.com

[daz[at]roughdiamondmarketing[dot]com]

"Kayte" > wrote in message
ups.com...
> I'm working in sound forge with a wav file that is 3.5 hours long and
> cannot be split up (it's eventually going to be encoded in real and
> quicktime for streaming on the internet). It was captured at 96/24 and
> it is too big to be saved right now (I believe 2 gigs is the max for
> wav files). I'm going to be resampling to 44.1kHz or possibly 22.5. I
> believe it will still be too big if I take it down to 16. Does anybody
> know of another solution besides going to 8-bit or splitting it up?

what about saving as wav64 - didn't sonic foundry invent this storage
'format' specifically to address this problem ?

I don't know all the techy stuff, but as far as i'm aware, its a regular
wave file with 64 bit headers enabling files over 2gb .... (those in the
know can elaborate here ....)

daz

[Mike Rivers]

In article om> writes:

> I'm working in sound forge with a wav file that is 3.5 hours long and
> cannot be split up (it's eventually going to be encoded in real and
> quicktime for streaming on the internet). It was captured at 96/24 and
> it is too big to be saved right now

So if it's too big to be saved, what's its present status? Floating
around in the air waiting for a box that it will fit in?

Honestly, I don't know why people do this to themselves. Hopefully
next time you'll work out a way to break up your recording in
manageable chunks. If you really want someone to sit at his computer
and listen to your magnum opus for three and a half hours, you can
always stitch the pieces together for the final production.

> (I believe 2 gigs is the max for
> wav files). I'm going to be resampling to 44.1kHz or possibly 22.5. I
> believe it will still be too big if I take it down to 16.

Well, you can't do any of that until you save the file. If you really
can't save it, you're sunk.

For next time, I'd suggest that you use a multi-track program for
recording. Assign your sound card to two tracks. Start recording in
stereo (I guess you're talking stereo here) on Track 1. Then after an
hour or so, start recording on Track 2 while Track 1 is still
recording. After a minute or so (which should give you a good place to
splice), stop recording on Track 1 and save that file. An hour of
stereo at 24/96 is just under 2 GB so it should be capable of being
saved on just about any system. Then an hour later, start recording on
Track 1 again.



--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Laurence Payne]

On 8 Jul 2005 06:41:05 -0700, "Kayte" > wrote:

>I'm working in sound forge with a wav file that is 3.5 hours long and
>cannot be split up (it's eventually going to be encoded in real and
>quicktime for streaming on the internet). It was captured at 96/24 and
>it is too big to be saved right now (I believe 2 gigs is the max for
>wav files). I'm going to be resampling to 44.1kHz or possibly 22.5. I
>believe it will still be too big if I take it down to 16. Does anybody
>know of another solution besides going to 8-bit or splitting it up?

If it's too big to be saved, where is it now?

Split it. Why not? It can be seamlessly stitched together again
later, wen it's in a more manageable format.

Why on earth did you record at 24/96 if it was going to end up being
streamed on the 'net?

[Arny Krueger]

"Mike Rivers" > wrote in message
news:znr1120832308k@trad
> In article
> om>
> writes:
>
>> I'm working in sound forge with a wav file that is 3.5
hours
>> long and cannot be split up (it's eventually going to be
>> encoded in real and quicktime for streaming on the
internet).
>> It was captured at 96/24 and it is too big to be saved
right
>> now
>
> So if it's too big to be saved, what's its present status?

Possibly it was saved in wav64 format. I've been
interpreting the question as relating to how the file would
be exported for inclusion in a video file.

[Peter Larsen]

Kayte wrote:

> I'm working in sound forge with a wav file that is 3.5 hours
> long and cannot be split up

Sure it can, you can always recombine. The simplest split is lengthwise,
ie. save left and right tracks independently.

> (it's eventually going to be encoded in real and quicktime
> for streaming on the internet). It was captured at 96/24 and
> it is too big to be saved right now (I believe 2 gigs
> is the max for wav files).

No, the max size for the .wav formar is 4 GB. Some faulty software
reportedly thinks it is 2 GB. Do not use faulty software.

> I'm going to be resampling to 44.1kHz

You're at 24 bit 96 kHz now, on the computer it is probably 32 bit 96
anyway, go to 32-48 and stay there until the file is processed, then
dither down to 16 bits. For streaming you might as well downsample to 32
kHz, leaves less work for the bit-reduction algorithm. Otoh many DA
converters sound better at 48 kHz, so that is in favour of staying at 48
kHz. The choice is likely to be "political" rather than technical.

> or possibly 22.5.

Sounds like AM. Don't.

> I believe it will still be too big if I take it down to 16.
> Does anybody know of another solution besides going to 8-bit
> or splitting it up?

Split and re-combine, or split and stream in chapters. The latter could
well be the most user friendly.

> Kaytea


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

[Mike Rivers]

In article > writes:

> No, the max size for the .wav formar is 4 GB. Some faulty software
> reportedly thinks it is 2 GB. Do not use faulty software.

How does one know what software is faulty in this respect? Wait until
it screws you, or screws someone who reports it someplace where you'll
see the warning?

All software is faulty, and should be avoided at all costs. It's just
too risky to not be able to see what's going on at all times.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Scott Dorsey]

In article <znr1121003143k@trad>, Mike Rivers > wrote:
>
>All software is faulty, and should be avoided at all costs. It's just
>too risky to not be able to see what's going on at all times.

That, in short, is the whole point of having the source. It can be open
source, it can be proprietary licensed source, whatever. But you need
to be able to look inside.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Kayte]

The file is saved as 15 separate wav files; obviously I have to turn
the computer off sometimes. What I want to do is be able to save it so
that I can shift it to the encoding computer (via a network drive) and
turn it into streaming files.

You know what, I don't know why people would want to listen to this
especially. I certainly don't want to hear it, but it's what my
organization got a grant to do and what they're paying me to make sure
happens. It's a recording of speakers at a conference and for some
reason people want access to it.

[Kayte]

It's saved as 15 separate files. It can't go on the web this way so
eventually it will need to be saved and converted.

[Kayte]

was not my choice. it was recorded on a little tape deck and somebody
else captured it and put it on a bunch of cds and gave it to my
organization to deal with.

[Kayte]

Thanks. My orders were to keep it as a 3.5 hour single file, as this
is the most convenient for the people who are going to be using it. It
does not have to be crystal clear, it just has to be good enough to
understand the speech. I will try it at 32/16 and see if that will do
it.

[Mike Rivers]

In article om> writes:

> The file is saved as 15 separate wav files; obviously I have to turn
> the computer off sometimes. What I want to do is be able to save it so
> that I can shift it to the encoding computer (via a network drive) and
> turn it into streaming files.
>
> It's a recording of speakers at a conference and for some
> reason people want access to it.

I'd think that those listening to the streaming audio would want to be
able to select individual speakers rather than listen to one
continuous six hour long file (or whatever it is). Why not just encode
each speaker's talk individually and post them as separate files?
Somebody will thank you for it. And if you post only one really long
stream, someone will ask if you can break it up.




--
I'm really Mike Rivers - )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Kayte Revitte]

Mike Rivers wrote:
> In article om> writes:
>
> > The file is saved as 15 separate wav files; obviously I have to turn
> > the computer off sometimes. What I want to do is be able to save it so
> > that I can shift it to the encoding computer (via a network drive) and
> > turn it into streaming files.
> >
> > It's a recording of speakers at a conference and for some
> > reason people want access to it.
>
> I'd think that those listening to the streaming audio would want to be
> able to select individual speakers rather than listen to one
> continuous six hour long file (or whatever it is). Why not just encode
> each speaker's talk individually and post them as separate files?
> Somebody will thank you for it. And if you post only one really long
> stream, someone will ask if you can break it up.
>
>
>
>
> --
> I'm really Mike Rivers - )
> However, until the spam goes away or Hell freezes over,
> lots of IP addresses are blocked from this system. If
> you e-mail me and it bounces, use your secret decoder ring
> and reach me here: double-m-eleven-double-zero at yahoo

You know what? It's really not my decision. And nobody would thank me
they would probably tell me to do it over again, this time how they
wanted it. Like I said, this isn't my personal project, this is
something I'm doing for an organization. I'm "just" the audio person
here, I make no creative decisions. I actually have gotten the file
saved in whole as a 16 bit 32khz wav. And it's 3.5 hours not 6.