We use a Silicon Graphics DDS-3 drive and have had great success
migrating DAT tapes with a free, open source osx application called
DatXtract. We've never had an issue with the drive itself, and it's
currently being used with a 733 MHz G4 (10.4) w/ SCSI card and works
just fine as a dedicated DAT transfer workstation. DATXtract
makes .aiff files, retains timecode information, and produces very
helpful error logs that we refer to during our workflow and are also
great to have around for preservation purposes. Files are made in
less than real-time (though not by much) and although monitoring is
not an option, error logs are reliable for our quality control
checks. If the DDS-3 is unable to read a damaged tape section, we
can always play out from our Sony PCM-R500 through an MBox (we
capture with Audacity), which does allow us to convert content that
the DDS-3 can't read. Our rate of tape failure with the DDS-3 is
fairly low, but this, of course, is also dependent on the condition
of your collection.
- Nicole Martin
On Jan 21, 2009, at 5:52 PM, Tom Fine wrote:
> Hi Scott:
> This is all very helpful. Thank you.
> -- Tom Fine
> ----- Original Message ----- From: "Scott D. Smith"
> <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Wednesday, January 21, 2009 1:33 PM
> Subject: Re: [ARSCLIST] Some DAT questions
> A few thoughts and comments on your questions(in no particular order):
> 1. We experimented with a couple of different combinations of
> drives and software a few years ago in an attempt to do direct
> data extracts from DAT tapes. I would have to say that we didn't
> have much luck. I think that there would have to be a lot more
> work done on both the software and the data drives to come up with
> a workable solution. Not saying it can't be done, it's that the
> drives were not really intended to work in the fashion required
> for audio.
> 2. We have generally used professional DAT decks such as the Sony
> 7030 and 7050 machines, since most of the work we do involves time
> code. We have used the AES/EBU digital outputs to feed external D/
> A converters with good success. The only issue to keep in mind
> with these machines is that they aren't very tolerant of tapes
> which are recorded out spec. There is a very tight set of
> tolerances that machine will accept as being valid, beyond which
> it simply will mute. There are a adjustments which can be made to
> some parts of the circuit (especially the servo and tracking),
> which will allow for a certain range of error, but I wouldn't go
> messing with these unless you thoroughly understand how the
> machine works. We routinely keep 4 different decks in operation,
> as I have found that some machines will simply not play certain
> tapes, no matter how far we try to go with tweaking. I have even
> run into incompatibilities between decks from the same manufacturer.
> 3. Because of issues with errors, I would recommend monitoring the
> tapes real-time, and keep the deck error monitoring display
> active. The pro machines also have a setting for which you can
> specify the level of errors which will result in an ERROR flag
> being indicated.
> 4. I don't see any particular reason to make an analog output
> copy, unless you want to archive in analog (some archives we have
> done work with have requested analog versions along with digital,
> but I think this is going by the wayside as the issues with
> digital storage are resolved).
> 5. In general, there is no compelling reason to store DAT (or any
> other digital based tape) tails out, except for the fact that the
> tape should always be exercised prior to playback. If it is tails
> out, it means it will force the user to have to rewind prior to
> playback. Any labels left in the box should be removed, but I
> wouldn't try to deal with any labels on the tape itself, unless
> there are starting to bleed significantly. Solvents such as Goo-
> Gone are extremely difficult to deal with, as they will tend to
> migrate over the entire shell, and make matters worse.
> 6. During the period that DAT was in regular use for production,
> we found the Sony, Maxell and BASF tapes to be generally fairly
> reliable. We did encounter problems with the Ampex tapes, mostly
> related to the shell, and also had major problems with an off-
> brand called "DIC/DAT". We had some occasional oxide shedding
> issues as well, which required cleaning of the tapes. We have also
> encountered tapes which were wound and stored with incorrect tape
> tension, which resulted in problems related to the tape geometry,
> which is a real horror to deal with.
> In general, I have to say I'm glad to see this format go away. Back
> in the 80's, I nearly got booed off the stage at an AES seminar on
> the subject of DAT for pro use, when I pointed out to the audience
> that the format was never intended for professional use. It was
> only because it failed to find a market with the consumer that
> Sony, in their infinite wisdom, decided to try and recoup their
> development costs by trying to adapt the format for pro use.
> We are now living with the consequences.
> Scott D. Smith
> Chicago Audio Works, Inc.
> Quoting Tom Fine <[log in to unmask]>:
>> Hello All:
>> I'd like to tap the collective brain (picture a non-invasive tap,
>> a Vulcan mind-meld) about DATs, with eyes to a medium-sized (several
>> hundred DAT tapes, all dating from the 1990's) transfer project.
>> Answers gleaned from personal experience most appreciated.
>> 1. does anyone on-list have experience with "ripping" audio DATs
>> directly to hard drive via a DAT data-tape drive? If so, what OS,
>> software and results are you getting? Is there a favored source
>> for the
>> appropriate drive?
>> 2. one key reason I was thinking for doing these DATs
>> digital-to-digital would be the real-time monitoring, so I could note
>> the locations of any dropouts or glitches. Assuming I'll find
>> some, are
>> there any hidden tricks or tips to fixing them or is that audio
>> lost on
>> a damaged segment of tape?
>> 3. if done digital-to-computer, I'm assuming SPDIF, but does anyone
>> have personal experience indicating either coax or optical is
>> preferable? I was thinking optical, given the sometimes strange
>> grounding issues of a computer.
>> 4. is there anything to be gained by running a simultaneous
>> analog-to-computer? I'm thinking, no, but I'm also thinking, I sure
>> don't have all the answers so there may be something unknown to me
>> 5. once a DAT has been transfered, what is the proper storage
>> method? I
>> was thinking, don't rewind it and make sure to store it in its
>> protective case. I was also thinking, if there is a label-sticker
>> in the box, take it out since the glue sometimes gets gooey or oily
>> over time.
>> 6. finally, are there any DAT brands/types with known sticky-shed
>> problems? Most of these DATs are BASF, but some are Ampex branded.
>> I am
>> not at all sure that Ampex manufactured its own DATs, they may have
>> resold Japanese tape.
>> Thanks in advance for any tips/advice gleaned from personal
>> -- Tom Fine