Software help archive

A read-only archive of old serato.com help threads.

Serato destroys WAV files!

Product
Scratch Live
Version
2.0
Hardware
Rane SL1
Computer
PC
OS
Platform
-
pueblofunky 9:27 PM - 17 December, 2010
Product: scratchlive
Version: 2.0
Hardware: scratchlive-sl1
Computer: pc
OS Version: win764
---

Be careful!

Serato 2.0 + 2.2B4 (and I'm sure all other versions between) are destroying WAV files.

The number of bytes of a WAV file must be EVEN - but after analyzing WAV files the most have an odd number of bytes which is NOT allowed.

Read - taken from www.sonicspot.com

One tricky thing about RIFF file chunks is that they must be word aligned. This means that their total size must be a multiple of 2 bytes (ie. 2, 4, 6, 8, and so on). If a chunk contains an odd number of data bytes, causing it not to be word aligned, an extra padding byte with a value of zero must follow the last data byte. This extra padding byte is not counted in the chunk size, therefor a program must always word align a chunk headers size value in order to calculate the offset of the following chunk.

or netghost.narod.ru

ChunkSize is the length of the data stored in the ChunkData field, not including any padding added to the data.

ChunkData contains data that is WORD-aligned within the RIFF file. If the data is an odd length in size, an extra byte of NULL padding is added to the end of the data. The ChunkSize value does not include the length of the padding.

Subchunks also have the same structure as chunks.

=>

A WAV file MUST HAVE an even number of bytes!

The SMF 10:35 PM - 17 December, 2010
where the hell did that come from??

never witnessed my wav's bein destroyed
pueblofunky 10:47 PM - 17 December, 2010
Yes - they are! As long as you use SSL only - it will work.

But if you use other tagging software like Tag & Rename ...

E.g. try to use Wavelab 5 - open multiple WAV files (with ID3 tags) and store them with another file name (use a test directory !!!).

Wavelab 5 does not know the "id3 " sub-chunk and due a bug it does not store them. But after storing the WAV file - you will only see an even number of bytes as file size. Never and odd number.

That is because all odd chunk + sub-chunks have an extra "00" NULL byte.
pueblofunky 10:49 PM - 17 December, 2010
PS: I've never noticed that because I've never used SSL for tagging - but did it now because SSL 2.0 does not forget unknown ID3 tags anymore ...

But I didn't test it so much and didn't check the file size - and now I've analyzed 20.000 WAV files and the most have an odd number of bytes now.

And I've problems to tag them now with Tag & Rename (which supports more ID3 tags as SSL).

:-(
The SMF 10:55 PM - 17 December, 2010
only got a few wav's, so im not gunna worry really
pueblofunky 7:56 AM - 18 December, 2010
I've just opened the 3 test files (empty, SSL20-analyzed, SSL22B4-analyzed) with Wavelab 7.01 and saved them under another file name.

Wavelab 7 stores the sub-chunk "id3 " at the beginning of the file (as first sub-chunk) - AND:

Original file by SSL 2.0/2.2B: Number stored as length of "id3 ": 1383 (4995 Bytes) - but without an "00" NULL byte to get a word-aligned sub-chunk (= even number of bytes for the file size).

File stored by Wavelab 7: Number stored as length of "id3 ": 1384 (4996 Bytes) - added an "00" NULL byte to be word-aligned (Wavelab 5+7 are able to read the destroyed file (W5 not always as I can remember!))

However - as I can read the specification - "00" should be added but not included in the stored length of the sub-chunk => both are wrong,

but Wavelab 7 has an error detection and is able to read it and stores it correctly (so far - the additional "00" stored in the length does not effect others to read due the internal representation of the ID3 format - which has its own length stored in each frame).
pueblofunky 8:02 AM - 18 December, 2010
From now please read the full story in the beta forum:

www.serato.com

Thanks
pueblofunky 8:43 AM - 5 August, 2011
Hi,

where is the Serato Scratch Live Public Beta 2.3 forum group?

When I click on serato.com I can not find the group.

I've googled around and found this link:

serato.com

It's a public beta (and I can join now) but how can I find a public beta forum group in the future. Is there a link to see when a new public beta is released?

After I've joined the forum group and clicked on serato.com I can see the forum group of course.

Otherwise I can not help you - and this is a "special into deep" action which can not be done by regular software testers.

Thanks
Rane, Support
Zach S 5:13 PM - 5 August, 2011
Thanks for the info:)
I'm going to take this out of the Help needed section as its for people who need help but if this issue still exists I'm going to ask that Serato add something to the FAQ and post a sticky at the top of the help section to let our users know.

It looks like the dev team has been working on this for some time now so hopefully they have fixed it in 2.3.
Serato, Forum Moderator
Samuel S 2:31 AM - 8 August, 2011
Hey pueblofunky,

We usually announce our Public Beta releases through our newsletter, Facebook and Twitter. These updates will usually have a link for the area. You can subscribe to the newsletter here:

serato.com

In terms of your .wav files being marked as corrupt, can you let me know if analyzing these in Scratch Live 2.3 fixes them?

Cheers,

Sam.
pueblofunky 5:21 AM - 8 August, 2011
Hi Sam,

Quote:
You can subscribe to the newsletter here


I get the newsletter and always read them but didn't find anyone about the beta. I don't use Facebook or Twitter. No time. ;-)

Quote:
In terms of your .wav files being marked as corrupt, can you let me know if analyzing these in Scratch Live 2.3 fixes them?


I've replied here: serato.com

However - all works and all seems to be correct so far - except if you re-analyze a corrupt track previously analyzed with 22033 (and before) now with 23058, 23058 shows the indicator that the track is corrupt.

The track is only corrupt during reading (of course) but after re-writing with the new logic the WAV file is CORRECT and the indicator shouldn't be shown of course.

The solution is NOT to re-analyze the track in 22033 - a 2nd re-analyze (or delete from library/import into library) with 23058 is the correct way!

I only have to test Tag & Rename 3.5.7 with all different possibilities because TR includes the padding x00 in the id3 chunk size (to get an even size) so I don't know what will happen if there is an odd size and a new padding x00. I've to report a bug to TR of course. I'd an issue that TR didn't detect the ID3 anymore but I don't know what I've done anymore. ;-)
pueblofunky 10:48 AM - 8 August, 2011
Also not working with 23065.

After analyzing a corrupt WAV file (created by SSL <= 22033) the error indicator will be shown. The new file incl. analyzing data is correct, so the indicator doesn't fit to the original WAV file! If the indicator appears, the original WAV file shouldn't be changed!

A new re-analyzing with 23065 works as workaround, indicator will be removed.

Should be fixed for RC2!
Serato, Forum Moderator
Michael R 10:21 PM - 8 August, 2011
Hey pueblofunky,

Thanks for the feedback, when the developers are free they'll respond to your thread in the public beta area. Please continue this discussion in that thread, as all public beta issues need to be contained to the public beta area.

Cheers :)