is a S/DIF a S/DIF a S/DIF?

  • Thread starter Thread starter adam
  • Start date Start date
A

adam

New member
I just got a Dell computer w/ a Turtle Beach Sound Card. I can buy a S/DIF card for it for <$99 - optical/RCA. I will be using it in conjunction with my fostex FD8 that has an optical s/pdif out. Will the quality be the same on this inexpensive soundcard's s/pdif connection as with a more expensive card. I won't be doing any analog recording w/ with computer (for now) - i record everything to the fd8 and then transfer to computer to mix and fool around with. Thanks for your help.
 
If the optical ports work on either side like they should
(and why shouldn't they?)
transfers by this method should be 100% verbatim front to back. The only difference I've heard anyone stand their ground on about S/PDIF vs. AES/EBU vs. ADAT are reliability of AES/EBU or the extra channels in the ADAT format. You only notice reliability problems when the system fails. And I've yet to see a bogus S/PDIF or AES/EBU transfer.
What peripheral equipment (if any) do you use with the FD8 to consider the package ready to capture a live gig?
I'm interested in mics/pres/effects/etc.
That ain't bad for a daughtercard with both optical and coaxial dogital I/O. I just yanked the Montego that came with my DELL.
It's in a box somewhere....
 
Actually doc, I have read some stuff about corrupt digital transfers using S/PDIF and AES/EBU. The things I read suggest that ANY data transfer will produce some kind of error in the copied data.

Good cable helps keep the errors down to mostly unnoticable levels.

Also, soundcards have a way of introducing noise into any wire connection. So, it stands to figure that this noise can corrupt digital data too.

I gotta save these damn links on this kind of stuff to share. I just read them, then move on.... :(

Ed
 
Drstawl - I've been fooling around w/ the FD8 for a year now, I was new to home recording before I got it. I've gotten comfortable with using it to record various guitars, vocals and other odd instruments (e.g. ancient Casio SK-1 w/ weird effects). As for effects/tools, I have a Quadraverb and nanocompressor, guitar stuff usually goes through a old rocktron progap (i love this thing - even though i shouldn't). I have an art tube preamp for additional warmth on vocals, mic'ed acoustic guitar, couple of differnet pedals like a rocktron Tsunami. For mics I shure 58 and a cheap audi technica that sounds good, even though it shouldn't. I've liked the results and look forward to using my new computer to arrange and mix. That's 1 thing the fd8 isn't great for. Thanks for th reply.
 
Well Ed- You know I'd love to see those articles. I've never heard a bogus digital transfer.
Digital audio transfer works or it doesn't.
Although almost every analog transfer has its own characteristic degradation.
 
okay doc, I had to do some searching. But first, I suppose that I should have said something like "digital copies MIGHT be corrupt" etc....

AES/EBU and S/PDIF formats transmit digital data up around like 6MHZ. This is in the realm of most video data. So, where the problem can start is when the cable that is used to transmit digital data using one of those two formats is not of the highest quality, or designed to allow an "even" transfer across the whole length of the cable. The result is can be one of two things; jitter and errors. Jitter is taken care of when the recording device plays back the jitterfull recording and reclocks it, about the only time this may make any difference would be if you where using say one of wanna be mastering boxes like the TC Finalizer Express in between two DAT players using the digital I/O's. If there is enough jitter, it would tend to make you adjust the processing is ways you may not need to because the jitter was causing you to hear the sound differently then it would be played back on the recieving DAT. So, one should always playback the recording to assure that it sounds the way they want it to. You cannot in a lot of cases trust what you are hearing while you are actually recording.

Errors can be a few things. Certain bits can be lost during the transfer. You may or may not notice this without ultra high quality converters for playback. Most digital gear now has "error correction" circuitry that will "fill in the gaps" to a certain degree. Now, this should not be confused with the error corrected recording being "an exact digital copy" of the original because obviously it is not. The error correction circuit basically took the sample before the error and the one after the error and "filled in the gap" with an average between the two. Other errors can be waveforms that are different than the original. If it is bad enough, it will just go onto tape as nothing the D/A converter can detect and would create a dropout in the recording. If this dropout is long enough, it will be audible. DROP OUTS ARE A WAY OF LIFE IN DIGITAL, THEY HAPPEN ALL THE TIME.

So, how does one help to assure that jitter and errors don't mess up the audio?

First, not a lot can be done about a device that has high amounts of jitter on playback. You just need a new unit. Also, some gear has circuitry that corrects jitter problems. These are of course present on higher quality pieces of gear and is one of the reasons for the higher price.

The best way to alleviate jitter and errors during digital transfers is to use quality cable that matches the specifications for the digital transfer standard you are using. This usually means using video coax cable for the transfer. AES/EBU calls for 110ohm cable on high quality XLR-3 connectors. S/PDIF calls for 75ohm cable on either RCA or better yet BNG connectors. By not using cable that meets the ohm load requirements for the used format, the wire will create errors THAT CAN BE AUDIBLE. Better safe then sorry though.

Regular mic and intrument quality cable is not really recommended for digital transfers because it is not generally designed to allow frequencies up to 6MHZ to pass through. So, video coax cable is the desired type of cable. Hell, when I was reading around, the pro's just recommend going to Radio Shack and buying their inexpensive coax cable for use.

The length of the cable obviously has an effect too. It is recommended to keep digital transfer cables as short as possible because longer runs increase the impedence of the load, thus, can create errors. For S/PDIF, up to about 3 meters is supposed to be okay if the cable is good. For AES/EBU, you can extend to maybe 100 meters if the wire is really good. But for runs of this length, it is recommended to use an inline impedence transformer to keep the signal strong.

So, there are many links that state the AES/EBU and S/PDIF standards for digital transfers. Here is one http://www.hut.fi/Misc/Electronics/docs/audio/spdif.html . There are countless other with the same exact information.

Basically, during a digital transfer, you are writing data to the tape. The integrety of the wire plays a roll in the accuracy of the new copy. The newer digital gear allows for a little error, but, you still get better results with better wire as it tends to allow smooth transmission of the data. To say though that a digital copy is "exact" is sort of misleading. It may not sound any worse depending upon how good your playback system is (D/A converters, amp, monitors, etc....) but to say the none of the bits where dropped and error correction was not applied to make up for this is not true.

Anyway, I still haven't found the technical document I read about what happens during digital transfers. It was obscure for sure and just not something that you will find because it dealt mostly with the supposed "exact" copy idea. Even old Bob Katz agrees that a digital copy will "sound" exact to your ears if minimum standards are kept.

Anyway, when I find it I will post it. But for now, keep cable length's short, and use wire that is rated in ohmage for the standard you are using and all should be well and good with your digital copies that sound exact, but are not.... :D

Ed
 
This from the link you gave regarding S/PDIF.....
Just what I found when I replaced $40
S/PDIF cable with $5 video cable.

The output impedance is standard 75 ohm, so ordinary coaxial cable designed for video applications can be used. The minimal input level of S/PDIF interface is 200 mVtt which allows some cable losses. There is no real need for special quality cable as long as the cable is made of 75 ohm coaxial cable (a
good video accessory cable works also as good S/PDIF cable).

[This message has been edited by drstawl (edited 04-16-2000).]
 
is there a difference between a phono plug
and an RCA plug??

the reason i ask is i tried to make my own Coaxial lead & used Phono plugs but the lead does not work at all? (0 sound and i mean o!)

i know phono leads work to a certain degree but i need 75ohm RCA (so ive been told!)

any help would be great !!

otherwise im gonna have to get my wallet out and buy a proper lead, ouch!
 
Yes the RCA is a phono type plug, but the coax cable itself must be for VIDEO to have the proper characteristics for S/PDIF.
The proper cable is ~$5 at Target.
That didn't hurt that bad, now did it?
 
Back
Top