Topic: toslink vs. spdif

[stvgray]

Do most people feel you get better sound when you transfer a digital signal via SPDIF than you do through a Toslink connection?

Steve

[Tommy E.]

I have never noticed a difference, however my playback system is not as nice as some around here.

Tommy

[admkrk]

i think you just have less chance of errors and maybe less noise.     it's still 1s and 0s         

[dklein]

You're probably asking about optical vs. coax - both are s/pdif signals.

Both are transmitting pulses - one electrical, the other optical.  Everything always starts as electrical but it can be converted to optical for transmission, and then turned back into electrical.  Why?  Optical signals are not subject to electrical interference.  Both can experience signal losses over distance (electrical gets weaker, optical gets dimmer) but nothing you'll experience with 10 foot cables.  The quality of cable becomes important in long distance hauls.  Which format is better?  In most cases, no difference. 

AES/EBU, the professional s/pdif is a higher voltage electrical signal on a balanced connection which eliminates any interference and can go even longer distances.  But for us, it's pretty much 6 of one, half dozen of the other.

[Brian]

to me...optical cables feel cheap and unrealiable

edit: i also don't trust an optical signal bouncing all over the cable's "walls" on it's way to the recorder.

[jpschust]

you have significantly less jitter in a good coax cable than in a good optical cable.  not to mention that coax cables are a hell of a lot more durable than optical cables.

[MattD]

you have significantly less jitter in a good coax cable than in a good optical cable.

True, but since the original poster was asking about a transfer, the cable doesn't matter. So long as the data gets from point A to point B, the transfer will be fine.

Now, if you're talking about using the cable to feed a locked D/A converter, then coax is the better format. AES/EBU is even better still, or using any of the above with worclock locked. Jitter does not matter when doing a digital to digital transfer.

[jpschust]

you have significantly less jitter in a good coax cable than in a good optical cable.

True, but since the original poster was asking about a transfer, the cable doesn't matter. So long as the data gets from point A to point B, the transfer will be fine.

Now, if you're talking about using the cable to feed a locked D/A converter, then coax is the better format. AES/EBU is even better still, or using any of the above with worclock locked. Jitter does not matter when doing a digital to digital transfer.

wouldnt jitter potentially lead to digital noise if there was a lot of it? 

[MattD]

Yes, but that would mean that the jitter would be so bad that the signal was interrupted and data stopped making it from point A to point B. 

[jpschust]

Yes, but that would mean that the jitter would be so bad that the signal was interrupted and data stopped making it from point A to point B. 

like, for example, a slight kink in an optical cable?

[dnsacks]

yep, the optical cable equivalent of a bad soldier connection/bad termination on a coax cable.

If kinked optical cables causing diginoise was a statisticly significant problem, you'd think that we'd hear about this a lot more often with all of the folks recording optical in to their jb3s.

[jpschust]

yep, the optical cable equivalent of a bad soldier connection/bad termination on a coax cable.

If kinked optical cables causing diginoise was a statisticly significant problem, you'd think that we'd hear about this a lot more often with all of the folks recording optical in to their jb3s.

don't we hear a ton of problems with the optical recording on the jb3's as it is?  i'm just saying with a more fragile connector the potential for problems is arguably higher.  with a new cable on a digi to digi transfer i wouldn't forsee any problem, but you get my point.

[SparkE!]

you have significantly less jitter in a good coax cable than in a good optical cable.

How so?  The transmission channel (coax or optical) should have almost no effect on jitter.  Jitter is introduced at the source of the signal, not in the medium through which it travels.

  not to mention that coax cables are a hell of a lot more durable than optical cables.

True.  I think you could easily break the end off of an optical cable, but it would take a lot more force to accomplish the same thing on a coax cable.

One thing that's nice about optical cable, though, is that you don't end up sharing your mic's signal ground with the noisy power ground of your recording devices.  With coaxial cable, you run the risk that noise on your ground connection that is generated by the hard drives and tape drive motors in all of the rigs in the recording chain will end up on your mic signal ground.  Any signal on mic ground might as well have been put on the mic signal itself if you are running unbalanced mic cables.  As long as you're running balanced mic cables, then you should not have to worry so much.  But get this: If the main digital output from the source rig is optical, you don't have to worry at all about hard drive or tape motor noise ending up in your recording, no matter whether you are using balanced or unbalanced mic cables.

Ground noise has no way to couple back up the chain if the entrance to the chain is optical.  Think about it.  Optical cable has no ground connection, but coax does and in most rigs, it's part of the same ground structuire that goes in the shield connection to the mics.

So, if durability is your prime concern, go with coaxial.  If noise immunity is your main concern, go with optical.  Neither cable type will hurt the signal jitter.

[jpschust]

i would think that the internal reflections that one faces with light passing through an optical cable would have a significant impact on jitter.

btw nice avatar   gotta love redmeat.

[SparkE!]

i would think that the internal reflections that one faces with light passing through an optical cable would have a significant impact on jitter.

btw nice avatar   gotta love redmeat.

Jitter doesn't become a problem until it is on the order of 100s of nS.  The delay in optical fiber is about 1 nS per foot, so you'd get a 10 nS round trip delay in a single bounce off of each end of the cable in a 5 foot cable.  You can count on the cable to have better than 90% transmissivity at each end of the cable.  That means that the round trip reflected signal will be less than 10% of 10% of the original. (It takes 2 bounces to make it back to the far end of the cable.)  That means that the reflected pulse is less than 1% of its original size.  So, you've got a signal that's delayed 10 nS and only 1 % the size of the original.  That will not make a significant impact on jitter.  Yes, there's an effect, but no, it's not significant.

Oh, and thanks for noticing the avatar.  Redmeat has always been one of my favorites.

[Tim]

don't we hear a ton of problems with the optical recording on the jb3's as it is? 

a year to a year and half ago yes, not anymore.

[Tony B]

Around 20 shows on mine, all through optical, and not a problem with a single one.

I basically treat my toslink cable like I would any...uh..sensitive part of my body...and it's treated me well.

Besides, the fact that it's light carrying the sound trips me out   

[Massive Dynamic]

Here's a link to a related thread from the Oade board.

[dklein]

The jb3 problems were:
1) issues with the 'grey' boxes that were converting coax > optical - they had trouble accurately reading the Sony s/pdif stream
2) optical cables not fully inserting into the JB3 because of that stupid ridge above the input jack

[hexyjones]

No problems with the CO2 in my configuration...nothing obvious anyway...

the only time I got any static - I was using the JB3 analog line in...

[dklein]

The other nice thing about optical is you eliminate the issue of different voltage levels from different manufacturers.