Digital out

[tyflos]

My error. Its teflos. I must be blind.

Actually, the greek word for 'blind' is tyflos. Or to give it properly, τυφλός.

[tyflos]

My error. Its teflos. I must be blind.

Actually, the greek word for 'blind' is tyflos. Or to give it properly, τυφλός.

Absolutely!
Nice to see you have an interest in the Greek language.
You can now use it to study the New Testament. God Bless!

Another thing I've discovered is audible difference using different coax cables.

The first cable I used is the yellow composite RCA cable that come with standard budget dvd players, and the second is a high quality hi-bandwidth RCA audio cables.

I've read that there is an impedance difference between video and audio RCA cables. 75ohms and 50 ohms.

The yellow video cable has a higher noise floor.
The audio cable had a lower noise floor and sweeter high frequency sound. Maybe I'm not reading between the lines of what you guys say properly, but I think interconnects be it optical or electrical make a good difference. I'm happy now.

Great discussion guys!  Let's remember to stay on topic, though.

I've read that there is an impedance difference between video and audio RCA cables. 75ohms and 50 ohms.

Yes, there is a characteristic impedance difference - and in theory you could use either type, as long as they are correctly terminated.

The yellow video cable has a higher noise floor.

This is quite irrelevant - we are talking about a binary digital signal that crosses a threshold way above the noise level...

The audio cable had a lower noise floor and sweeter high frequency sound. Maybe I'm not reading between the lines of what you guys say properly, but I think interconnects be it optical or electrical make a good difference. I'm happy now.

The noise floor would be identical - it doesn't have a choice. But if you have a mismatched coax termination, then you will get reflections back down the coax, which could have all sorts of strange effects on the signal timing - if you recall, I said earlier that

I suppose that if you wanted to be really picky about it, you'd have to say that the coax was potentially worse - but only because there are far more things to go wrong.[/list:u]and this is precisely what I was alluding to. It is possible to have a similar situation with the lightpipe, but in practice this is far less likely to happen, because of the way toslink connectors are made. What would happen in this case would be that a small misalignment, or piece of muck at the optical transition would cause internal reflections of a similar nature to those in the coax. And I will also agree that some toslink lightpipes are 'better' than others - especially in internal reflection characteristics. Misalignment is basically not going to happen, by virtue of the design - but the muck thing certainly could. You should keep the ends of your toslink lightpipes clean!

But you should also try not to confuse what you want to hear with what's actually happening. If there are jitter differences caused effectively by either reflections, or different types of clock recovery (more likely), it is well established that they will manifest themselves as a slight 'veiling' of the sound. And since this is always related to clock recovery, it is always possible to reclock the signal with a more stable clock and recover it completely. It's well researched with DAT machine digital signals (which are S/PDIF) - you can render a digital stream virtually unlistenable to by adding jitter to the clock - and if you reclock the unlistenable stream, you can restore it completely.

There may be an impedance difference, but that has nothing to do with the one being audio, the other being video. Each cable has a characteristic impedance, the use is something that we humans associate with it. As far as technical use is concerned, this is ruled by standardisation. So for video you should use 75Ohm. And the same for spdif! This is what Steve means by "as long as they are correctly terminated". The termination is inside your receiver and was defined by the spdif standard, you have no choice about it. It determines the cable to use.

The characteristic impedance for analog audio is undefined as you are far below the wavelenght of your signals, so you end up with a lumped impedance. Unless your cd player is a few km away from your amplifier. Another reason for not using analog rca leads for digital audio.

It is not because you read that your rca audio cable has a higher bandwidth that it influences the high frequency of the analog result. That is determined by the sample rate of the spdif (most probably 44.1kHz). And you don't need silly bandwidth to pass spdif: a couple of MHz is enough. A good video cable will pass several GHz for the lenghts used in home setups.

There may be an impedance difference, but that has nothing to do with the one being audio, the other being video. Each cable has a characteristic impedance, the use is something that we humans associate with it. As far as technical use is concerned, this is ruled by standardisation. So for video you should use 75Ohm. And the same for spdif! This is what Steve means by "as long as they are correctly terminated". The termination is inside your receiver and was defined by the spdif standard, you have no choice about it. It determines the cable to use.

That's why I said 'in theory'... but the effects of reflections, even down short bits of cable, are very real. I have a KVM switcher here that I had to completely rewire to keep the RGB video transmission paths intact and correctly terminated, and even now, if you use it with different types of cable either side of the switch, it can ring so badly that you can't read text on the screen. And that's on cables that are supposed to have the same characteristic impedance! Okay, the frequencies are somewhat higher, but nevertheless, it is still an issue with short cables - even at a couple of megs. But as you say, quite irrelevant at audio frequencies.

Another problem with destructive interference in S/PDIF cables is that also in theory, some lengths of incorrectly terminated cable are going to be more susceptible than others. But there's not a lot you can do about this, because S/PDIF can run at several different clock speeds - which is what makes all the difference. Okay, it's not a perfect world - but I really don't think that it's quite the problem that teflon thinks it is...

Well I've changed my heart and mind and I'm going back to analog rca after reanalysis. The music sounds more spacious. (Listening cd is Billy Joel Greatest hits)I haven't got a well defined set of knowledge with electronics but I've have heard rumors that some sony dvd players upsample 44.1khz to 48khz. I know I'm getting way off topic but i think upsampling lifts the roll off frequency point a bit higher of the anti-aliasing filter and reduces phase distortion which is out of the audio band that does have an effect on audio quality within the audio band. (that word again timbre).

As to whether coax sounds better than optical on my system - Let it be.
Next CD player will be an SACD/DVD player full stop and that will have to be analog rca cables theres no other choice if I'm going to keep the same receiver.

It seems to me that if you like the Analog RCA better that it is the difference in the quality of the D/A converter being used.  I would suppose if you have a high quality sound card the D/A converter in that card would be much better than the one used in the speakers (especially if they are somewhat budget quality and not studio monitor quality).

On the question of the difference between SPDIF (digital coax) and TosLink (optical digital) the signals are identical.  You could have a ground loop created by the SPDIF cable that is impossible on the TosLink because SPDIF connects the ground of one machine to the other and TosLink does not.  The created signal would be the same (if the same D/A converter is used) but the output might have a hum in it (created in between the output of the D/A converter and the amp, in the low level section) because of the ground loop.

Sorry if this is too technical but it's just the engineer in me talking.  BTW, the TosLink or SPDIF jitter should be identical also (unless the receiving unit wasn't set right for one connection or the other).

It seems to me that if you like the Analog RCA better that it is the difference in the quality of the D/A converter being used.  I would suppose if you have a high quality sound card the D/A converter in that card would be much better than the one used in the speakers (especially if they are somewhat budget quality and not studio monitor quality).

I get the impression that there are two inputs available on the same box - in which case the D-A will be identical.

On the question of the difference between SPDIF (digital coax) and TosLink (optical digital) the signals are identical.  You could have a ground loop created by the SPDIF cable that is impossible on the TosLink because SPDIF connects the ground of one machine to the other and TosLink does not.  The created signal would be the same (if the same D/A converter is used) but the output might have a hum in it (created in between the output of the D/A converter and the amp, in the low level section) because of the ground loop.

It is possible that you could introduce a ground loop with a coax S/PDIF, certainly, but it is not unknown for the coax input to be transformer-isolated - for instance, this is what happens with a lot of stand-alone converters (M-audio, Midiman), and may well happen with some recievers as well - for precisely this reason.

BTW, the TosLink or SPDIF jitter should be identical also (unless the receiving unit wasn't set right for one connection or the other).

In the majority of cases, what actually limits the jitter is the stability of the clock recovery circuit. This is commonly achieved with a PLL of some description, and is entirely dependent on the internal reference clock in the reciever. I've carefully avoided saying this until now - but this is the real reason that there's no difference in perceived quality with different S/PDIF connections as a rule - the only way that the interconnect can affect the sound is by screwing up the clock recovery. Now, I'm not saying that this is impossible - but it really would be a pretty poor design if it did!

 but it is not unknown for the coax input to be transformer-isolated

Same goes for outputs, some use transformers there.

Next CD player will be an SACD/DVD player full stop

May I know what you are considering for that? I'm also looking out that way for the end of the year.

 I have a KVM switcher here that I had to completely rewire to keep the RGB video transmission paths intact and correctly terminated,

Quite a different problem, video at that point is analog. Those kvm switches are a pain, and so are video extension leads.[/quote]

Quite a different problem, video at that point is analog. Those kvm switches are a pain, and so are video extension leads.

No, technically it's an identical problem. The cable neither knows nor cares whether a signal is digital or analog - its termination behaviour in terms of reflections alters not one jot. Back in the late 50's, George Philbrick (one of the originators of the modern op-amp) pointed out to one of his employees that 'digital is just a special case of analog'. It was true then, and just as true now...

From your point of view you are absolutely correct. I was refering to to fact that for reflections to be audible in a digital bitstream, you need quite an influence on the analog waveform. In the video, the analog waveform is the data, and any deviation is visible imediately.

In the digital case, a lot depends on design decisions later in the recovering of clock/data, but in the analog case, you can do whatever you want, it isn't going to make much difference.

Next CD player will be an SACD/DVD player full stop

May I know what you are considering for that? I'm also looking out that way for the end of the year.

I'm considering anything that will become cheaper with similar specifications than the Sony NS999ES of course!
Some people might argue till the cows come home but my opinion is sony( or some high profile Jap companies) have the potential to make the best cd transports in the world. We should have a track to track access time specification competion test. he he  

I was looking at a Philips one. I have had less than stellar experience with the after sales service of Sony. But they are rather hard to find here.