Digital transfer of DAT tapes to computer

[Hugh Robjohns]

That won't work, the sample rate via the S/PDIF will keep changing nad the computer will either keep stopping or flag files with the wrong sample rate.

If the tapes really are of mixed format, then going analogue is the sensible approach -- the universal sample rate converter -- as everything will end up with the same sample rate in the one project. Much easier to deal with, and the quality loss will be negligible via any decent modern A-D converter which will be an order of magnitude better than those of the original DAT, or the DAT's D-A.

hugh

[chris...]

That won't work, the sample rate via the S/PDIF will keep changing nad the computer will either keep stopping or flag files with the wrong sample rate.

Right. Makes sense.

I guess I'm keen on digital transfer to avoid the hassle of setting analog input levels. Could record at 24bit, but that seems wasteful of space, given the task in hand. So more likely I'll record at 16bit, and set the level carefully. (Tho' I guess I could record at 24bit, normalise, then dither down to 16bit, without noticeably affecting things).

Try to not laugh much, but the only A/D converter I currently have to hand is Yamaha 01V mixer from 1998 (going ADAT to computer). Not the latest+greatest converters, but will probably do.

Edit - also have a little Sony PCM-M10 flash recorder with line-in, which is 10 years newer than the DAT, so might just have better converters.

[ef37a]

Surely this is no different from importing a mixture of 44.1kH and 48kHz files?

I am fairly sure I could do this in Samplitude then SAVE everything at 44.1kHz.

My son regulary records things from BBC R3 on the web in realtime and saves it as a 44.1kHz 16bit.wav which it assuredly is not originally! Same goes for Youtube sound.

But then I have probably got it all wrong!

Dave.

[Hugh Robjohns]

I guess I'm keen on digital transfer to avoid the hassle of setting analog input levels.

But it couldn't be easier! The absolute peak level is a known quantity, and any modern A-D will have more headroom capability than you could ever want while still having a lower noise floor than the DAT machine's D-A. So all you have to do is find the loudest bit on tape, set the record level of your interface to make that peak to a comfortable level -- say peaks to -4dBFS or so, and let it get on with it!

Could record at 24bit, but that seems wasteful of space

Totally trivial and irrelevant concern with typical hdd sizes today. I'd record 24 bit with headroom. Tweak the material as you want, and then bounce to 16 bit if you feel the need, having removed the surplus headroom.

Edit - also have a little Sony PCM-M10 flash recorder with line-in, which is 10 years newer than the DAT, so might just have better converters.

Probably so, yes. The O1V's converters weren't its greatest strength.... but even so, I reckon you'd be hard pressed to hear any significant deficiencies compared to the 12-bit non-linear long play DAT mode.

Hugh

[ken long]

Not when the DAT device is acting as the master clock, Dave - which is the only way to digitally migrate the data.

[Hugh Robjohns]

My son regulary records things from BBC R3 on the web in realtime and saves it as a 44.1kHz 16bit.wav which it assuredly is not originally!

Most DAWs will automatically sample rate convert files of one sample rate when dropped into a project of a different sample rate. And files can be saved to a different sample rate by passing through an SRC, which is an automatic bounce process in many DAWs.

Usually, though, the interface is told what sample rate to run at by the DAW project... But the interface also has to synchronise itself to the incoming clock rate if recording a digital source.

Some interfaces have SRCs on the digital inputs to deal with such incompatibilities, but most don't. So it comes down to a specific setup and whether the project rate is always the master, in which case who knows what will happen, or whether the project sample rate is forced to change on the fly to follow the input rate (in which case which bit of the project is at the right rate?)!

hugh

[ef37a]

My son regulary records things from BBC R3 on the web in realtime and saves it as a 44.1kHz 16bit.wav which it assuredly is not originally!

Most DAWs will automatically sample rate convert files of one sample rate when dropped into a project of a different sample rate. And files can be saved to a different sample rate by passing through an SRC, which is an automatic bounce process in many DAWs.

Usually, though, the interface is told what sample rate to run at by the DAW project... But the interface also has to synchronise itself to the incoming clock rate if recording a digital source.

Some interfaces have SRCs on the digital inputs to deal with such incompatibilities, but most don't. So it comes down to a specific setup and whether the project rate is always the master, in which case who knows what will happen, or whether the project sample rate is forced to change on the fly to follow the input rate (in which case which bit of the project is at the right rate?)!

hugh

Err? Right! Well it sounds ok. When he is done "analysing"it all gets squashed down to MP3 for archiving.

Dave.

[Tomás Mulcahy]

I recently had a couple of tapes with 44.1kHz and 48kHz on the same tape. I was unable to seek out the recordist and smack him for such a mortal sin.

I used a borrowed Tascam DAP-2 (?) sending S/Pdif to M Audio FW1814 into Pro Tools. The M Audio would make a horrible buzzing noise when the SR changed. So I would simply stop recording, and load up a new Pro Tools session at the rate displayed on the Tascam. I edited out the nasty buzz afterwards.

Doing it this way bypasses all converters which as Hugh points out doesn't make much difference. There is a slight difference though! In this case it was worth it. Or at least, it kept me happy! But if I had a lot of long tapes at mixed SR, I would use an audio interface with built in SRC, such as the Pro Tools HD192. Life's too short to monitor such long transfers!

[James Perrett]

I recently had a couple of tapes with 44.1kHz and 48kHz on the same tape. I was unable to seek out the recordist and smack him for such a mortal sin.

I had something similar this week - probably recorded in different studios. I simply put the different sample rates on different tracks in Reaper. Reaper allows the soundcard sample rate to override the session sample rate and it automatically converts the sample rate on the fly if the file sample rate is wrong.

I'll finally render everything together and make sure that I use the highest quality sample rate conversion setting (which Reaper says in non real time but it goes at 8X on my machine).

James.

[Tomás Mulcahy]

Another example of why Reaper is far ahead of Pro Tools!