Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

ZAX files and the new Deva recorder.

134 views
Skip to first unread message

Glenn Sanders

unread,
Jan 25, 2003, 10:56:24 AM1/25/03
to
I would like to clarify exactly what a .ZAX file is and why it is a
benefit to the production sound community.


The .ZAX file format is a data compression format developed for the
same purpose as the .ZIP file format for computer files. It is not a
new audio file format.

It is optimized specifically for audio file data to reduce the space
that the files take up by an average of 3 to 1 depending on audio
content. When the files are "Unzaxed" (hey, I made a new word) they
are convertible into standard file formats at any sample rate.

What this does is save you a lot of time and money. If you have to
create multiple copies of your audio on location, The .ZAX file format
will do it in 1/3 the time of any other recording system. At the end
of the day the last thing you want to do is sit on the set for an
extra hour while the DVD-RAM copies are made. In the case of the Deva,
3 disks can be recorded at the same time in any file format. When they
yell cut at the end of the day you are done. Pop out the disks and go
home.

The .ZAX file format dramatically cuts down the cost of media on a
daily basis. 1 $10 single sided 4.7 gig disk with .Zax files is
equivalent to 5 $15 8cm 1.4 gig per side DVD-RAM disks. This is a
saving of about $65 per day if the maximum of 4.7gigs is utilized!
(All prices are estimated)

It also allows for 10 tracks of audio to be recorded to DVD-RAM in
real time. This is not possible without this solution. It also allows
us to pack 3 times the amount of audio onto the DVD-RAM drive. A 4.7
gig DVD-RAM disk can now hold a full day of high bit rate multitrack
audio. This would take several disk changes using the small profile
8cm DVD-RAM disks.

For postproduction the savings are many. The time required to read the
.ZAX files off of a DVD-RAM can be up to 15 times faster than with
uncompressed files. This is due to the data interleaving required in
location disk recording confusing the operating systems of PCs and
Macs into thinking the files are very fragmented requiring extra time
to read the files from the DVD-RAM and the 3-1 compression advantage.
The files also will take up to 1/3 the space on the server at the post
house.

The New Deva recorder has 1.8 gigaflops of processing power. This is
27 times the power of the current Deva II. This is necessary to
compress the files and record them to 3 different disks at the same
time.

The main points I would like to leave you with are these.

1 You do not have to use the .ZAX format if you do not want to. BWF
and SD2 are still included in the next generation of Deva.

2 The data compression is loss-less and does not effect the audio
content in any way.

3 With a .ZAX file the need to coordinate file format, sample rate and
16/24 bit file type with post is unnecessary.

4 You will save lots of time and money.

5 It is not a new audio format. It is a new audio data compression
format designed specifically for Location recording.

6 The Unzax program to expand the files is included on every Deva
generated disk for free.

7 .ZAX files are internet friendly. Emailing audio to post over high
speed Internet connection is now practical.


I predict that in the future very few location mixers will deliver
uncompressed files to post, as there are only benefits to this
process. The time saved in post as well as on Location will require
the use of a compressed file format. Howy, myself and the staff at
Zaxcom are proud to be the first ones to make it a reality with the
new generation of Deva recorder.


Best Wishes

Glenn

pkur...@earthlink.net

unread,
Jan 25, 2003, 1:28:12 PM1/25/03
to
In article <27b5dc17.03012...@posting.google.com>,
g...@zaxcom.com (Glenn Sanders) wrote:

> 3 With a .ZAX file the need to coordinate file format, sample rate and
> 16/24 bit file type with post is unnecessary.

How does this format help with varying sample rates? Is this not done
with hardware or software sample rate conversion?


> 5 It is not a new audio format. It is a new audio data compression
> format designed specifically for Location recording.

Is the original material still recorded in MARF? If .ZAX is the
origination format, are the numer of tracks limited if choosing to
record in .bwf? Is this format as fault-tolerant as MARF (if you lose
power during a take) or is this why Howy is working on the Flash backup?

All-in-all, this sounds VERY promising.
Peter

Jeff Wexler

unread,
Jan 25, 2003, 3:09:20 PM1/25/03
to
On 1/26/03 4:56 AM, in article
27b5dc17.03012...@posting.google.com, "Glenn Sanders"
<g...@zaxcom.com> wrote:

> I would like to clarify exactly what a .ZAX file is and why it is a
> benefit to the production sound community.
>
> The .ZAX file format is a data compression format developed for the
> same purpose as the .ZIP file format for computer files. It is not a
> new audio file format.

Jeff Wexler asks:
I understood from the beginning that the .zax file format was a data file
loss-less compression format, but I am still confused about certain things.
If the .zax format is used, does it just apply to the data put down on the
main internal hard drive? Data that goes out to the other 2 possible media,
the DVD-RAM drive and a possible drive on the Firewire, what data/audio
format are they in?

When you say the de-compression software will be on "every Deva disk" does
this mean that we would deliver, lets say, a DVD-RAM disk with .zaxed bwf
files and in post they "un-zax" these files using the software resident on
the DVD-RAM disk and they are left with fully compatible .bwf files as
always?

I think the confusion just comes in trying to figure out what the workflow
will be. It is quite an achievement in my opinion to allow with the new
DEVAS this degree of flexibility and compatibility.
>
Regards, Jeff Wexler

coff...@webtv.net

unread,
Jan 25, 2003, 2:43:57 PM1/25/03
to
Commercial/Clarification:

I have been getting questions about .Zax.

Clearly understand that DEVA III, IV & V include it automatically for
free, along with all of today's normal formats. You may never use that
setting.....it just adds a possible future post fix as DEVA looks down
the road ahead.

John Coffey C.A.S.
http://www.coffeysound.com

Glenn Sanders

unread,
Jan 25, 2003, 6:30:12 PM1/25/03
to
> How does this format help with varying sample rates? Is this not done
> with hardware or software sample rate conversion?

The .ZAX file expander has a sample rate conversion algorithum built
in. If you record at 48.0 Khz and output a .ZAX file to the DVD. The
file expander can for example convert the file to a .BWF file sampled
at 48.048 Khz.

> Is the original material still recorded in MARF?

Yes it is.


> If .ZAX is the
> origination format, are the numer of tracks limited if choosing to
> record in .bwf? Is this format as fault-tolerant as MARF (if you lose
> power during a take) or is this why Howy is working on the Flash backup?

The Deva internal hard disk will have the same fault tolerance as the
DevaII. You will be able to remove power during the recording process
and not lose any audio.

>
> All-in-all, this sounds VERY promising.
> Peter

Thanks Peter.

Jeff Wexler asks:
>I understood from the beginning that the .zax file format was a data
file
>loss-less compression format, but I am still confused about certain
things.
>If the .zax format is used, does it just apply to the data put down
on the
>main internal hard drive? Data that goes out to the other 2 possible
media,
>the DVD-RAM drive and a possible drive on the Firewire, what
data/audio
>format are they in?

The .ZAX is an output function for now. We may use it on the internal
disk to triple the internal storage of the Deva.

The 3 file choices available in the Deva 3 for output will be .SD2
.BWF and .ZAX. The internal 5.25 inch DVD ram and the Fire wire disk
can be different file types recorded at the same time.


>When you say the de-compression software will be on "every Deva disk"
does
>this mean that we would deliver, lets say, a DVD-RAM disk with .zaxed
bwf
>files and in post they "un-zax" these files using the software
resident on
>the DVD-RAM disk and they are left with fully compatible .bwf files
as
>always?

You would deliver a .ZAX file to post. Post would then decide what
file format they needed to use. The unzax software lets post make
anything they need from your .ZAX files.


>I think the confusion just comes in trying to figure out what the
workflow
>will be. It is quite an achievement in my opinion to allow with the
new
>DEVAS this degree of flexibility and compatibility.

The workflow will be no different than with current file outputs. Just
faster and more efficent.


Regards, Jeff Wexler

Best Wishes

Glenn

Jeff Wexler

unread,
Jan 25, 2003, 6:42:25 PM1/25/03
to
On 1/26/03 12:30 PM, in article

> You would deliver a .ZAX file to post. Post would then decide what
> file format they needed to use. The unzax software lets post make
> anything they need from your .ZAX files.
>

Last question on this issue as I do not wish to take up any more of Glenn's
valuable time discussing all the details that I am sure will be handled by
users in the future, not just interested parties on this newsgroup; the
question is, it appears that if a disk is delivered to post with both .zax
files and the necessary conversion and un-zaxing utility software, will this
software be useable, compatible, with both Mac and PC systems running
current OS?

JW

Kurt Albershardt

unread,
Jan 25, 2003, 8:18:14 PM1/25/03
to
Glenn Sanders wrote:
>
> The .ZAX file expander has a sample rate conversion algorithum built
> in. If you record at 48.0 Khz and output a .ZAX file to the DVD. The
> file expander can for example convert the file to a .BWF file sampled
> at 48.048 Khz.

Are there choices in SRC quality available, given the time-vs-quality
tradeoffs in SRC algorithms?


Martin Harrington

unread,
Jan 25, 2003, 9:20:22 PM1/25/03
to
Why did they "reinvent the wheel"?
WinZip works perfectly with audio files.
I've proved it many times.
No loss whatsoever.
--
Martin Harrington
www.lendanear-sound.com
"Glenn Sanders" <g...@zaxcom.com> wrote in message
news:27b5dc17.03012...@posting.google.com...

William Sarokin

unread,
Jan 25, 2003, 9:24:47 PM1/25/03
to
I think it was so that the 'zaxxed'; files can be created as multiple
formats. The utility gives you the choice of opening the audio file as a
BWF mono or poly or as an SD2 file, it also allows you to change the
sampling rate.
Billy Sarokin
"Martin Harrington" <lend...@ozemail.com.au> wrote in message
news:GBHY9.130$0K6....@nnrp1.ozemail.com.au...

Kurt Albershardt

unread,
Jan 25, 2003, 9:35:37 PM1/25/03
to
Martin Harrington wrote:
> Why did they "reinvent the wheel"?
> WinZip works perfectly with audio files.
> I've proved it many times.
> No loss whatsoever.

Did they add an audio file setting to WinZip? Used to be that only
WinRAR (amongst the general-purpose file compression utilities) had this
feature.

So now we have (at least) the following lossless compression formats:

SHN
APE
RAR
OSQ (WaveLab)
FLAC
OFR
ZAX


Hrmmpf....

Glenn Sanders

unread,
Jan 25, 2003, 11:10:19 PM1/25/03
to
Jeff Wexler <jws...@earthlink.net> wrote in message news:<BA598831.888E%jws...@earthlink.net>...

100% compatable. No problem.

Best Wishes

Glenn


>
> JW

Martin Harrington

unread,
Jan 26, 2003, 12:08:43 AM1/26/03
to
I don't think there is any need to add an audio file setting to WinZip.
As far as it is concerned, it's a file....
It works...I often do it...always uncompresses to original.
--
Martin Harrington
www.lendanear-sound.com
"Kurt Albershardt" <ku...@nv.net> wrote in message
news:10435485...@nnrp1.phx1.gblx.net...

Jeff Wexler

unread,
Jan 26, 2003, 12:32:15 AM1/26/03
to
On 1/26/03 6:08 PM, in article u3KY9.171$0K6....@nnrp1.ozemail.com.au,
"Martin Harrington" <lend...@ozemail.com.au> wrote:

> I don't think there is any need to add an audio file setting to WinZip.
> As far as it is concerned, it's a file....
> It works...I often do it...always uncompresses to original.
> --

Jeff Wexler comments:
I think Martin may have missed something here, or I don't really understand
this so well, that the implementation of the .zax format (which is totally
an option and is provided "for free") provides other utility conversion
functions beyond just loss-less data compression, and these conversion
routines also relate to audio and audio file formats, sample rates and so
on. I believe that the people at Zaxcom are quite aware that there are other
"off the shelf" compression schemes but that these may not provide all the
facility they are providing us with.

Also, I might add, even with something as ubiquitous as WinZip or even
Stuffit, there have been a fair number of compatibility issues amongst the
various dominate computer platforms, Mac OS and Windows.

JW

Martin Harrington

unread,
Jan 26, 2003, 1:43:24 AM1/26/03
to
>I think Martin may have missed something here,

Quite possible...WinZip doesn't do anything other than straight compression
so I guess that Zax is more complex and dedicated.
--
Martin Harrington
www.lendanear-sound.com


"Jeff Wexler" <jws...@earthlink.net> wrote in message

news:BA59DA2F.88B4%jws...@earthlink.net...


> On 1/26/03 6:08 PM, in article u3KY9.171$0K6....@nnrp1.ozemail.com.au,
> "Martin Harrington" <lend...@ozemail.com.au> wrote:
>
> > I don't think there is any need to add an audio file setting to WinZip.
> > As far as it is concerned, it's a file....
> > It works...I often do it...always uncompresses to original.
> > --
>
> Jeff Wexler comments:

Kurt Albershardt

unread,
Jan 26, 2003, 3:44:39 AM1/26/03
to
Jeff Wexler wrote:
>
> Also, I might add, even with something as ubiquitous as WinZip or even
> Stuffit, there have been a fair number of compatibility issues amongst the
> various dominate computer platforms, Mac OS and Windows.

Stuffit is a Mac-centric format for sure.
.ZIP is at least supported on several OS's by a multitude of vendors.

Losslessly compressing PCM audio data requires something other than
Lempel-Ziv to get any sort of reasonable ratio.


David Spearritt

unread,
Jan 26, 2003, 5:01:02 PM1/26/03
to
In the past I have experienced occasional errors uncompressing zipped
audio files, mind you some of them have been long files, ie > 2hrs, so I
always archive uncompressed original files. I have had the new OSQ in
Wavelab but have been reluctant to use it, also have had one failed restore
from a Wavelab archive from spanned 10 CDR's. So its uncompressed archives
for me at present.

David

"Kurt Albershardt" <ku...@nv.net> wrote in message

news:10435706...@nnrp1.phx1.gblx.net...

Kurt Albershardt

unread,
Jan 26, 2003, 5:18:41 PM1/26/03
to
David Spearritt wrote:
> In the past I have experienced occasional errors uncompressing zipped
> audio files, mind you some of them have been long files, ie > 2hrs, so I
> always archive uncompressed original files.

My experience with RAR and SHN is that they will extract up to and even
around a bad block, making them much more robust than ZIP. I have a
couple of SHNs which never downloaded completely and they will open and
extract the available files up to and including the available portion of
the last .WAV up to the stop point. Pretty cool, and IMO a minimum
requirement for any archiving format. They also add the assurance of
checksums to the process--WAV/AIFF files can have corruption during
copies that never gets spotted until someone listens to the file, often
too late.

Martin Atias

unread,
Jan 29, 2003, 11:46:41 PM1/29/03
to
Correct me if I'm wrong, and forgive me if I state the obvious, but in order
to compress a file with .Zip or >rar, don't you need to get the files into a
computer first? That would require an uncompressed files transfer (not to
mention the computer).

I think the beauty here is that the compression application is resident in
the DEVA machine, and can apply the compression to the files laid down on
the main and secondary drives instantly. This in addition to the other
functional benefits of sample rate conversion during the decompress, etc.

Marty Atias

Kurt Albershardt

unread,
Jan 30, 2003, 12:20:16 AM1/30/03
to
Martin Atias wrote:
> Correct me if I'm wrong, and forgive me if I state the obvious, but in order
> to compress a file with .Zip or >rar, don't you need to get the files into a
> computer first? That would require an uncompressed files transfer (not to
> mention the computer).

That's just an artifact of the systems we use. Most (maybe all) of the
lossless formats could be implemented in recorder firmware--it's just math.

Richard Crowley

unread,
Jan 30, 2003, 12:29:40 AM1/30/03
to

"Kurt Albershardt" <ku...@nv.net> wrote in message
news:10439039...@nnrp2.phx1.gblx.net...

But L-Z (used in *zip, dunno about stuffit?) is not particularly efficient
for the kind of data found in PCM audio files, is it?


Kurt Albershardt

unread,
Jan 30, 2003, 12:35:46 AM1/30/03
to
Richard Crowley wrote:
>> That's just an artifact of the systems we use. Most (maybe all) of the
>> lossless formats could be implemented in recorder firmware--it's just
>> math.
>
> But L-Z (used in *zip, dunno about stuffit?) is not particularly efficient
> for the kind of data found in PCM audio files, is it?

Which is why .ZIP achieves very little compression on PCM audio files.
It's also poor at recovering damaged archives.

RAR uses something else when you check the 'use multimedia compression'
option.


howy

unread,
Jan 30, 2003, 11:56:17 AM1/30/03
to
>> But L-Z (used in *zip, dunno about stuffit?) is not particularly
efficient
>> for the kind of data found in PCM audio files, is it?
>
> Which is why .ZIP achieves very little compression on PCM audio files.
> It's also poor at recovering damaged archives.
>
> RAR uses something else when you check the 'use multimedia compression'
> option.

The .ZAX compression scheme has a number of features built into it
that are required for professional audio use.

A lossless compression algorithm can perform much better if it
understands the content of the file it is trying to compress. The ZAX
system can perform proper prediction on the audio waveforms instead of
treating the file as semi-random data bytes. This results in much
better compression ratios as well as faster CPU performance.

If you wanted to playback a portion of a zipped file you would have to
decompress the whole file first. The ZAX format would allow someone to
cue to a timecode and start playing audio immediately without
decompressing the file.

If a few bytes of a zipped file were corrupted the result could be the
loss of the entire audio file. The ZAX system isolates corrupted audio
data so any damage that occurred to the media would minimize the
damage to the audio content.

-howy
zaxcom engineer

0 new messages