Skip to main content

Okay, so I was recording a soprano soloist on a recital yesterday with the Aurora 8 piped to both my computer (via AES) and the HD24 (non-XR) via lightpipe. The computer crashed in the recital, so I was stuck using the backup.

The sample rate was 96kHz.

Total track count was 6.

When I tried to copy these individual files from the Fireport onto the computer, it copied all 6 files that showed up, but the resulting files were as such:

Channel 1 (left piano) occupied both files 1 and 2 (identical files, both identified as 24/96 by the computer.)

Channel 2 (right piano) occupied files 3 and 4 (again, identical files, both identified by the system as 24/96).

Channel 3 (Flank left) occupied files 5 and 6 (again, identical files. .)

Channels 4-6 did not copy to the computer.

I spoke with Alesis tech support for about an hour today and they were baffled (all the way up to the point where the phone *magically* disconnected).

Strangely enough, I can digitally copy all of the files back into the computer by piping them BACK through the Aurora 8 and into the computer, but obviously, this is not exactly time conscious.

Does any one know what the h*ll happened and what I need to do?

My fireport has the latest firmware as does my HD24.

Cheers!

J.

Comments

anonymous Sat, 03/24/2007 - 12:15

Wow, nope, never heard of that one before. At least you can keep them in the digital realm when tracking them into your computer. It may take some time, but all is not lost!

I'd take that drive and reformat it to get rid of any errors, then do a hard reset on your HD-24, if that's possible. I've never tried it on mine!

Boswell Mon, 03/26/2007 - 05:13

Hi Jeremy,

Yeah, the 96KHz can be a pain if the two manufacturer's interpretations of the U2 bit (double data rate) are slightly different.

However, I doubt whether your two files for each channel are bit-identical. I would guess that one contains the even-numbered samples and the other has the odd-numbered. They would sound identical as 48KHz files, because the time-shift of 1 sample is inaudible.

I have had 96KHz recordings arrive as two files per channel when copied via the FireWire adapter, but always as 2x the number of files, that is, all the original channels have been represented. Equipment that correctly deals with the U2 bit and the double data rate, e.g. a Yamaha 01V96 and the HD24, usually give retrievable files. The problem seems to arise when you set the HD24 into 96KHz mode, but it doesn't correctly see the U2 bit in the incoming data stream, so during file transfer it gets confused as to whether the data is N 96KHz files or 2N 48KHz files.

It might be worth trying the FTP file transfers via the ethernet link to see if that correctly gives you double data rate files, but you may spend several days waiting for the transfer to complete, and it would be quicker to replay the files in real time via ADAT through the Aurora as you mention.

Have you tried posting your problem on the Yahoo HD24 tech group? I've had rather variable responses from that board, but it's worth a try.

Cucco Mon, 03/26/2007 - 12:54

That was my thought too about the interleaved bits, but each file shows as 96kHz in both the computer and the HD24. (Not to say that it's not possible to show this, but it's not, as Spock would say, logical).

I could try the FTP, but gosh that would take a LONG time and if futile would just be a monumental waste of time.

I'm gonna keep trying things and see what works...
I'll keep ya posted.

Boswell Mon, 03/26/2007 - 15:05

Cucco wrote: That was my thought too about the interleaved bits, but each file shows as 96kHz in both the computer and the HD24. (Not to say that it's not possible to show this, but it's not, as Spock would say, logical).

Yes, that's what appears to happen when it gets confused - 96KHz in the headers but files constructed of 48KHz samples.

My theory goes something like creating a 96KHz song in the HD24 writes a header on the song with 96KHz set in it, but if the U2 bit is not seen in each of the ADAT streams that go to make up the song at the time of creating the song, the actual data is not flagged as double data rate, and the FireWire transfer firmware gets confused. Since being sensitized to this possibility, I have always carefully set 96KHz on the ADCs and started up my lightpipes before creating a new song, and haven't had the problem since. However, it may be just coincidence, as I didn't keep a record of exactly what I had done on the previous occasions that the recorded data showed strange 96KHz-related file effects. I could always replay the song on the HD24 correctly, so it was an inconvenience at most and not a disaster.

x

User login