Subject: Re: s/pdif signal quality
From: jsrtheta@colorado.net
Date: Mon, 08 Sep 1997 12:34:45 -0600
Message-id: <873739410.10600@dejanews.com>

In article <873531609.7015@dejanews.com>,
  cassidy@math.niu.edu wrote:
>
>   I've read that spdif digital format is "consumer-oriented" and that
> AES/EBU digital is "pro-oriented".  Recently I heard that s/pdif is of
> "significantly lower quality" (than AES/EBU) and that, in particular, the
> lower frequencies are not transmitted as well as the higher ones.  Is
> this True?
>   If it IS true, is there is difference between spdif that is transmitted
> via optical/Toslink vs. coaxial?

I run both AES/EBU and coaxial.  Setup:  Theta Data Basic to Theta Pro
Basic IIIa, one connection through Audio Magic Excalibur silver AES/EBU,
the other through XLO Type 4 digital coaxial.  (There's also a loop
running from the Pro Basic through a Theta TLC and back to the Pro Basic,
but let's forget about that for now.)
AES/EBU is a pro connection, for sure, as it's balanced, but I don't know
that it's necessary for home audio involving short runs.  It does seem to
sound mellower.  I don't think the bass is improved, though.
I have read numerous articles claiming AES/EBU is superior.  Genesis
recommends AES/EBU for their digital lens, and one dealer told me that
AES/EBU is going to become the standard.  I don't know about that, but I
find that I prefer the coaxial XLO.  Could I distinguish in DBT?  I doubt
it, but there is certainly no huge improvement in sound from the Audio
Magic AES/EBU.
Toslink has always been frowned on.  I have AB'd between a toslink and a
coaxial - couldn't hear a difference, but it was a quick test under less
than ideal conditions, so who knows?

From: shlampen@aol.com (ShLampen)
Date: 8 Sep 1997 17:27:30 GMT
Message-id: <19970908172700.NAA12844@ladder01.news.aol.com>

In article <3411E57D.1BBD@nccn.net>, "D.R \"Chris\" Christensen"
<chrisc@nccn.net> writes:

>  It has been said that the S/PDIF for factor can
>introduce more jitter to the digital signal, than in the AES form
>factor.

I don' t think this is true.  Impedance-specific twisted pairs (i.e.
twinax) and coax each have their good and bad points, and so does fiber,
even plastic fiber..

In crosstalk and noise:
Plastic fiber is in first. Fiber is immune to all noise, crosstalk etc.
becuase it is optical.
Twisted-pairs are next. Noise pickup and crosstalk varies with the quality
of the twisted pair, the number of twists per inch and lot sof other factors.
Coax is last because it offers no inherent immunity other than the shield
and, being an unbalanced line, offers no other immunity.

In structural return loss (i.e. impedance stability) and, therefore, bit
error rates:
I believe coax is first.  It is STILL (after 50 years) the most stable
construction with the most stable impedance, the least reflections (return
loss) and therefore, can go farther with AES/EBU than any of the three.
Impedance tolerance rages for +/- 5 ohms (for poor coax) to +/- 3 ohms for
good (we even have one that is +/- 1 1/2 ohms!
Twisted pairs are next.  Common AES/EBU cable is around +/- 10 ohms for
the good stuff (although we have some bonded twisted pairs which are around
+/- 5 ohms, like a poor quality coax!)
Fiber, impedance doesn't apply but there is optical refraction/reflection
within the fiber which does the same damage.  Plastic is terrible in this
regard.  Glass fiber would win everything but it's rare to see in the home
environment, and it's a good question whether the significant cost increase
of conversion in-and-out of optical is worth the benefit.

If it is just jitter...I would guess this order:

Glass fiber
Coax
Twisted pairs
Plastic fiber
Dixie-cup-and-a-string

Steve Lampen
Technology Development Manager
Belden Wire & Cable Co.
www.belden.com
Our 4000-product download catalog is now on-line.
steve.lampen@belden.com

Thema:	Re: S/PDIF sync
Von:	mihartkopf@aol.com (Mihartkopf)
Datum:	23 Oct 1997 06:18:39 GMT

In <344E458B.2777@loop.com>, Matt Smith <thuja@loop.com> writes:

>Can anybody define S/PDIF sync for me. 

S/PDIF is a self-clocing interface which means the sync is included in the
signal itself. The whole datastream is divided into blocks of 192 frames
and each frame contains two subframes. Each subframe holds one sample, so a
frame holds the samples of both channels at a time.
Each subframe consists of 32 bit slots, and the first 4 slots in the
subframe contain the PREAMBLE. The preamble is the sync signal.
The rules say, that at the border of each slot there must be a Hi/Lo or a
Lo/Hi jump. So a digital "0" holds the level for the whole time slot, and a
digital "1" has a jump in between. But always there is a jump at the border
of the time slot.
The preamble is the only signal that may (and must, to be recognized)
violate that rule.
The very first preamble in a block (192 frames) marks a subframe "A" and
is called "Z" and violates the first and the third border jump. The other
two preamble forms ("X" for subframe "A" elsewhere in the block and "Y" for
subframes "B") violate the first and the second border jump. "X" forms a
digital "1" in its last (forth) slot and "Y" forms a digital "0" instead,
but has a jump in the middle of the third slot.
Did you understand? Perhaps you should see a pic. You may download a
better explanation from my website (see signature) at the Electronic Library.

Best regards
Michael Hartkopf
++++++++++++++++++++++++++++++
data of many many many microphones at
http://members.aol.com/mihartkopf/
++++++++++++++++++++++++++++++

Thema:	Re: AES/EBU == S/PDIF????
Von:	Tomi Holger Engdahl <then@tinasolttu.cs.hut.fi>
Datum:	16 Oct 1997 10:09:08 +0300

"Luis Gustavo Martins" <lgustavomartins@mail.telepac.pt> writes:
 
> Can anyone tell me which are the diferences between the AES/EBU and the
> S/PDIF Digital audio formats, or point me to an apropriate site on the Web
> were I can find the answer?

S/PDIF is a consumer version of professional AES/EBU interface.
Here is comparision table:

                          AES/EBU              S/PDIF
Cabling                   110 ohm shileded TP  75 ohm coaxial or fiber
Connector                 3-pin XLR            RCA (or BNC)
Signal level              3..10V               0.5..1V
Modulation                biphase-mark-code    biphase-mark-code
Subcode information       ASCII ID text        Copy protection info

Some more S/PDIF information is available at
http://www.hut.fi/Misc/Electronics/docs/audio/spdif.html

> Which of these formats is the most used on the consumer type DAT and CD
> players digital audio I/O?

S/PDIF

-- 
Tomi Engdahl (tomi.engdahl@iki.fi, http://www.iki.fi/then/)
I maintain electronics and PC hardware web pages at
http://www.hut.fi/Misc/Electronics/


Thema:	Re: AES/EBU == S/PDIF????
Von:	jfarrington@wesleyan.edu (Jim Farrington)
Datum:	Thu, 23 Oct 1997 19:02:01 GMT

>> S/PDIF is a consumer version of professional AES/EBU interface.
>> Here is comparision table:
>> (snip)

According to the manual for our Marantz 620 CDR, AES/EBU deletes
certain subcodes when recording from DAT, and so they recommend the
SPDIF interface for DAT-CDR recording.  Seems strange to me, but so it
is claimed.

Jim Farrington
Music Librarian
Olin Memorial Library
Wesleyan University
jfarrington@wesleyan.edu

Thema:	Re: AES/EBU == S/PDIF????
Von:	kludge@netcom.com (Scott Dorsey)
Datum:	Thu, 23 Oct 1997 20:31:55 GMT

In article <344f9eb2.280357293@news.wesleyan.edu> jfarrington@wesleyan.edu writes:
>>> S/PDIF is a consumer version of professional AES/EBU interface.
>>> Here is comparision table:
>>> (snip)
>
>According to the manual for our Marantz 620 CDR, AES/EBU deletes
>certain subcodes when recording from DAT, and so they recommend the
>SPDIF interface for DAT-CDR recording.  Seems strange to me, but so it
>is claimed.

The S-PDIF code carries the START-ID information in one of the fields.
The AES/EBU format uses that field for something else.  As a consequence,
if you dub from DAT-> DAT with S-PDIF, the START-IDs are preserved, but
if you dub with AES/EBU, they aren't.  You can guess what happens with
DAT->CDR dubs.
--scott
-- 
"C'est un Nagra.  C'est suisse, et tres, tres precis."


Thema:	Re: AES/EBU == S/PDIF????
Von:	shlampen@aol.com (ShLampen)
Datum:	24 Oct 1997 14:43:14 GMT


I believe I have learned something today!  If this is correct above, then
(aside from signal voltage, which I knew about) it is the subcode
information which makes S/PDIF different from AES3???  Now, many installs
use a balun and convert from AES3 to a coax format (recognized in AES3),
but this is NOT S/PDIF (correct?) because it still includes the ASCII text.

Some of these baluns also drop the level from (I thought it was up to 5
volts in the original AES/EBU) 3.10 volts, down to 1 volt so you can use
analog video routers, DA's etc. Voltage is really only a factor in cable
distance and attenuation, so that's not such a big thing. In fact, AES3 on
coax can go considerably farther than on twisted pairs.

Does this mean, if I made an RCA to BNC adaptor cable, I could not feed an
AES3 input with S/PDIF data (at least not successfully)???? Or vice versa??

Just curious.  Comments please!



Steve Lampen
Technology Development Manager
Belden Wire & Cable Co.
www.belden.com
My new book, "Wire, Cable, and Fiber Optics for Video and Audio Engineers"
has just been published by McGraw-Hill.


Subject:	Re: S/PDIF
From:	kludge@netcom.com (Scott Dorsey)
Date:	Wed, 29 Oct 1997 17:13:09 GMT

In article <34576A4E.5BA23296@ping.be> loic.collignon@ping.be writes:
>Hi!
>I am realising a SPDIF patchbay for my computer (the cardsound is the
>Audiomedia III), but a patchbay with relays, switchable by two rotators
>(IN/OUT) and my questions are: could I do that with relays ? Did I must
>use a special processor to switch the SPDIF instead of relays?
>Thanks a lot.

No, you could use a relay matrix to do it.  You could even buy a standard
relay switching device for video, and use that. 

Pretend it's video and treat it that way.
--scott
-- 
"C'est un Nagra.  C'est suisse, et tres, tres precis."


Subject:	Re: S/PDIF
From:	munster@cww.de (Philipp Pertermann)
Date:	Fri, 31 Oct 1997 15:51:11 GMT

On Wed, 29 Oct 1997 17:54:38 +0100, "Loc COLLIGNON" <angel@ping.be> wrote:

>Hi!(snip)

I use TTL ICs to switch / dispatch S/PDIF signals. That is cheaper than relays
and consumes less power. The only problem that occurs is that the S/PDIF lines
have 75 ohm termination. So you will have to take that in account for the input
/ output stages. Although I don't do the conversion for the output since I
realized that the normal output of a TTL IC seems to do the job as well.
If you're interested I can send you the schematics / layout of my little digital
<-> optical converter with selectable input. It contains a simple and quite
reliable input stage.

But leaving that alone I assume the job could be done with relays as well.

Phil