Why does SPDIF basically suck?

[Audiophile Neuroscience]

1 hour ago, Superdad said:

 

You mean one that turns itself off at sunset and Friday and won’t turn on until Saturday night? 

 

...or supports the new graphical user interface with laser technology, GUILT

[Marcin_gps]

16 hours ago, Summit said:

 

I see that you totally dismiss my experience and my theory regarding conversions between different digital interfaces. But okay let’s say that you are correct and I like some added coloration. If that’s true it must also be the reason I have a JCAT LAN, don’t you think? I mean my preference for some extra coloration is probably constant. 

 

Don’t worry you are not alone in your thought about the degradation caused by conversions between different digital interfaces or the deprivation made by adding more component in the digital transmission chain. I can only report what I hear and it’s not in line with your or most others in this thread. But it’s not all. Not only do I use a JCAT LAN and a DDC. Those are connected to my ultraRendu, which besides being a RoonReady output converts Ethernet to USB which is a sort of Digital to Digital Converter (DDC) .

 

It's not the same. Replacing a component with sth better always leads to better sound. Adding a component that is not required in the chain - not necessairly. 

 

And saying that ultraRendu converts Ethernet to USB is a huge oversimplification... ultraRendu is a micro PC. 

 

What I was trying to say is that in computer audio there is no need to engage S/PDIF at all unless we have some really old school DAC like Audio Note which lacks USB connectivity.

 

 

[mansr]

18 minutes ago, Marcin_gps said:

Replacing a component with sth better always leads to better sound.

Overgeneralisation of the day.

[Summit]

9 hours ago, Audiophile Neuroscience said:

 

Some years back I started to wonder about changing to computer usb over transport/spdif when I was getting sound quality from a laptop into an RME Babyface DAC that rivaled my Reference Mark Levinson 31.5 > aes/ebu (or coax spdif or AT&T ST fibre) >30.6. I listened extensively to many combos & dacs at various dealers, too many to list, some of which i also auditioned at home including MSB Platinum stack and a UMT transport, tried and purchased Bricasti Mi Dac and then Gryphon Kalliope.

 

Is it really too much to ask for that the ones that claim SQ superiority (or call things crap) of one design over another also state which audio gear the result was made. General statements that DDC or S/PDIF etc are inferior is very hard to take serious if no reference to which gear that been used are made.

[Audiophile Neuroscience]

1 hour ago, mansr said:

Overgeneralisation of the day.

 

I gather you don't like JPlay Mans? If so, we share something in common. I wondered whether you have some association with JRiver (I do not apart from using their software). In fact, who are you? care to share?

 

[Audiophile Neuroscience]

8 minutes ago, Summit said:

 

Is it really too much to ask for that the ones that claim SQ superiority (or call things crap) of one design over another also state which audio gear the result was made. General statements that DDC or S/PDIF etc are inferior is very hard to take serious if no reference to which gear that been used are made.

 

I do not claim anything apart from telling you what I heard on what gear.As I said YMMV.

[Summit]

8 hours ago, miguelito said:

That’s a balanced connection, a different beast. We are talking clocking here. BNC is perfect for that.

 

No no you are mistaken we are debating S/PDIF in this thread which is a digital interface for carrying digital audio signals between devices over either optical or electrical cable. S/PDIF is not made or used to send a pure clock signals between external reference clock and DACs. The clock aspects discussed here is very different from the one that is sent from a reference clocks, the frequency is also different. The only thing they have in common is the cable and connectors. It’s just like with LVDS I2S, which is a totally different digital interface than the HDMI that is used to stream movies. The only thing in common is that they often can use the same HDMI cable for carrying digital audio signals between two devices.

Okay back to basic. S/PDIF is a digital interface developed by Sony and Philips. S/PDIF is a protocol as well as a set of physical layer specifications for carrying digital audio signals between devices over either optical or electrical cable. S/PDIF was developed at the same time as the main AES3 standard, that are used to interconnect professional audio equipment in the professional audio field. S/PDIF is almost identical to AES3 (AES/EBU) at the protocol level and differ in that it has different connectors. With one exception the S/PDIF protocol is identical to AES3. The difference is that channel status bit differs in S/PDIF.

[Summit]

1 hour ago, Marcin_gps said:

 

It's not the same. Replacing a component with sth better always leads to better sound. Adding a component that is not required in the chain - not necessairly. 

 

And saying that ultraRendu converts Ethernet to USB is a huge oversimplification... ultraRendu is a micro PC. 

 

What I was trying to say is that in computer audio there is no need to engage S/PDIF at all unless we have some really old school DAC like Audio Note which lacks USB connectivity.

 

 

 

Charming, and if I or someone else get better SQ result by adding an Offramp 5, Singxer SU-1, Berkeley Alpha USB, or are using the coax from something like an Aurender, Auralic Aries, are we only fooling ourselves or have tin ears?

The crapping on S/PDIF and DDCs here is IMO so low, the tech explanations given for the added coloration so vague and real apples to apples comparisons with named gear almost none existing to motivate me to continue this discussing.

BTW, manufactures that says that all but their design is crap and are calling their costumers fools for using what they think sound best will not be supported by me.

[Marcin_gps]

1 minute ago, Summit said:

 

Charming, and if I or someone else get better SQ result by adding an Offramp 5, Singxer SU-1, Berkeley Alpha USB, or are using the coax from something like an Aurender, Auralic Aries, are we only fooling ourselves or have tin ears?

The crapping on S/PDIF and DDCs here is IMO so low, the tech explanations given for the added coloration so vague and real apples to apples comparisons with named gear almost none existing to motivate me to continue this discussing.

BTW, manufactures that says that all but their design is crap and are calling their costumers fools for using what they think sound best will not be supported by me.

 

That's fine. I just shared my opinion, no need to get defensive. 

 

[mansr]

27 minutes ago, Audiophile Neuroscience said:

I gather you don't like JPlay Mans?

I dislike them no more than anyone else selling overpriced miracle components with zero evidence that they make a difference.

 

27 minutes ago, Audiophile Neuroscience said:

If so, we share something in common. I wondered whether you have some association with JRiver (I do not apart from using their software). In fact, who are you? care to share?

I have no association with JRiver or any other company in the audio sector. Recently, I have officially taken on the role as maintainer of Sox. This is mentioned on my profile here. If you want to know more, Google me. I'm not hard to find.

[adamdea]

13 hours ago, sandyk said:

 So did another couple of friends of mine.

However, their performance was WAY below that of their claims and those of most reviewers. However, they improved markedly with a good low noise dual rail external power supply instead of the A.C. wallwart, the addition of voltage regulators for the analogue section, and as you can see in the photo, a .3PPM 24.576MHZ TCXO powered by a variation of the PFM " Flea"  low noise supply to it.

However , they did a pretty decent job in the Digital area.

I had the full 3 pack with the PSU (trasnformer) and valve thing. I think the dac marked the point (perhaps at the latest, we might have got there before) where you could get pretty much state of the art measured performance (certainly at least pretty much as far as 16/44 will go) for peanuts. It also marked the beginning of the dac come back (they had fallen largely out of favour as there isn;t much point using a dac instead of a cdp unless you have more than one digital source).

I gave up on the valve thing pretty soon as I concluded it just added 1dB of gain and some crackling.

[adamdea]

2 hours ago, mansr said:

Overgeneralisation of the day.

I read that statement as "when I think I am getting something better than I can be predicted to enjoy the thing I think is better in line with my expectations" or "the effect of expectation bias on my enjoyment is reassuringly linear" .

I wish I could say the same as then it would be fairly easy to predict whether it was worth spending money on things I quite fancy buying.

[miguelito]

51 minutes ago, Summit said:

No no you are mistaken we are debating S/PDIF in this thread which is a digital interface for carrying digital audio signals between devices over either optical or electrical cable.

Ok read my answers... I said that the issues with clocking would be solved if you also added a word clock connection like dCS has. Those connections work on BNC connectors. 

 

I was not talking about the SPDIF signal itself, which likely would better transmit over a balanced connection than a single-ended one.

[Audiophile Neuroscience]

20 minutes ago, adamdea said:

2 hours ago, mansr said:

Overgeneralisation of the day.

I read that statement as "when I think I am getting something better than I can be predicted to enjoy the thing I think is better in line with my expectations" or "the effect of expectation bias on my enjoyment is reassuringly linear" .

I wish I could say the same as then it would be fairly easy to predict whether it was worth spending money on things I quite fancy buying.

 

As much as I am not a fan of JPlay or the man in question, I think you guys are reading a tad too much into his statement.

[mansr]

1 minute ago, Audiophile Neuroscience said:

As much as I am not a fan of JPlay or the man in question, I think you guys are reading a tad too much into his statement.

There are a few regulars here, our JPlay friend one of them, whose posts almost invariably have an undertone of "you should all buy my stuff." While I understand the desire to promote one's business, it really has no place in a thread such as this one.

[Panelhead]

32 minutes ago, miguelito said:

Ok read my answers... I said that the issues with clocking would be solved if you also added a word clock connection like dCS has. Those connections work on BNC connectors. 

 

I was not talking about the SPDIF signal itself, which likely would better transmit over a balanced connection than a single-ended one.

  That may explain AES delivering better sonics  in my one unit sample. I have only used one dac with coax and AES to allow a comparison.

[Audiophile Neuroscience]

Mans, I looked you up. It's official, I am old

[adamdea]

2 hours ago, Audiophile Neuroscience said:

 

As much as I am not a fan of JPlay or the man in question, I think you guys are reading a tad too much into his statement.

I admit I was looking at this statement in isolation. That just is how I read it (btw nothing to do with Jplay in my case)

5 hours ago, Marcin_gps said:

Replacing a component with sth better always leads to better sound.

[adamdea]

1 hour ago, Audiophile Neuroscience said:

Mans, I looked you up. It's official, I am old

...and I feel like a dumb monoglot

I have to admit to being slightly disappointed (and impressed as hell) that Måns' English does not have the heavy Swedish (?) accent I had imagined in my head when I read his posts.

[miguelito]

2 hours ago, Panelhead said:

That may explain AES delivering better sonics  in my one unit sample. I have only used one dac with coax and AES to allow a comparison.

I would not be surprised if a balanced connection would also improve on the clock sync problem as well as having a cleaner signal (common mode is rejected by construction).

[adamdea]

9 minutes ago, miguelito said:

I would not be surprised if a balanced connection would also improve on the clock sync problem as well as having a cleaner signal (common mode is rejected by construction).

Hmmm I seem to remember from way back someone arguing AES3 was i inferior, but I can't remember why!

[mansr]

22 minutes ago, adamdea said:

...and I feel like a dumb monoglot

I have to admit to being slightly disappointed (and impressed as hell) that Måns' English does not have the heavy Swedish (?) accent I had imagined in my head when I read his posts.

Did you find a video of me or what?

[adamdea]

1 minute ago, mansr said:

Did you find a video of me or what?

Is that you optimising for linux on arm at youtube.?

[Summit]

3 hours ago, miguelito said:

Ok read my answers... I said that the issues with clocking would be solved if you also added a word clock connection like dCS has. Those connections work on BNC connectors. 

 

I was not talking about the SPDIF signal itself, which likely would better transmit over a balanced connection than a single-ended one.

 

What issues with clocking are you talking about and with which specific audio gear have this problem occurred?

 

How does a word clock connection like DCS  change how the S/PDIF clock sync work?

 

If the problem with clocking would be true for S/PDIF. Which it’s not! Why would it be better to transmit over a balanced connection rather than a single-ended one? Please do tell?

[mansr]

34 minutes ago, adamdea said:

Is that you optimising for linux on arm at youtube.?

I've done a couple of talks on the topic.