Digital transport brawler

[Andrewjvt]

I've never noticed any difference in sound quality regardless of source Blu-ray digital out TV digital out Laptop usb

Only the level in the TV is much lower for some reason.

I'm going to do some real time demos now taking everything on board that's been said.

You are welcome to come and try some alternatives here.

Thanks for the offer.
I will wait on David's reply as he is close and I'm already a customer of his but if he does not have time I'll contact you for a trip.

 

[davidf]

For a demo of laptop v cheap transport v fancy transport?

For one minute I'll agree that my asynchronous dac is dependent on a quality stream of data to perform well. I'm up for a test of various products. Best guy I can think of is David as he's the closest and has convert/innous and bluesound Oh and also electro stuff.

Perhaps we can arrange this some time I'd like to come over and check your place out anyway. Are you up for it?

I presume we’re talking a Saturday? Unavailable for the next two Saturdays, and also working at the North West Audio Show next month. The systems available would be a more ‘down to earth’ (Moon Neo Ace/Hegel Rost etc) rather than high end unless I can borrow a nice DAC laden amplifier from someone...

But yes, I don’t mind hosting something. A bit of preparation time would be ideal though, as I wouldn’t want to just ‘do it on the day’ so to speak - I’d want some sort of structure to the event so we all know exactly what we’re doing and how we’re doing it.

 

[Andrewjvt]

For a demo of laptop v cheap transport v fancy transport?

For one minute I'll agree that my asynchronous dac is dependent on a quality stream of data to perform well. I'm up for a test of various products. Best guy I can think of is David as he's the closest and has convert/innous and bluesound Oh and also electro stuff.

Perhaps we can arrange this some time I'd like to come over and check your place out anyway. Are you up for it?

I presume we’re talking a Saturday? Unavailable for the next two Saturdays, and also working at the North West Audio Show next month. The systems available would be a more ‘down to earth’ (Moon Neo Ace/Hegel Rost etc) rather than high end unless I can borrow a nice DAC laden amplifier from someone...

But yes, I don’t mind hosting something. A bit of preparation time would be ideal though, as I wouldn’t want to just ‘do it on the day’ so to speak - I’d want some sort of structure to the event so we all know exactly what we’re doing and how we’re doing it.

Don't stress as this is not urgent and I can do week days
But I'm also swamped at work for many weeks.
I've got the **** quality laptop if you have the quality zennith to compare.

 

[newlash09]

Thanks So presuming we all have the same dac that is very very good

In your opinion does the digital transport (laptop, auralic, innous etc) Effect the sound quality seems it just sends the info to the dac?

Like all things, it depends.  It would be a safe assumption that the data received by the DAC will be identical no matter the transport - unless the software on the transport is messing with it (I believe both WIndows and Linux varients will do this unless properly configured).

If your DAC is extracting its clock from the input stream, then it could be subject to the quality of the clock in the transport, but frankly if this is the case you should buy another DAC. There is no reason these days to slave a DAC to the clock on the input stream, and if yours does, I don't think it belongs in the 'Audiophile' category.

It is possible that you are getting more than you bargained for from the transport.  Noisy supply rails in the transport may export themselves along the common ground connection to the DAC, although any decent DAC should be able to filter this out.  Toslink gets around this problem, but comes with its own catalogue of issues.

Were I in the DAC business, I would use async USB, with galvanic isolation at the DAC.  In this case, I would be 100% sure that all transports (from a RasPi to a £10,000 audiophool system) would sound the same (subject to my comments about Windows /Linux).

That was one informative post

 

[MajorFubar]

For years I read WHF like a disciple, then this one time I they had a review of a really expensive transport, mid 1990s it was, can't remember the exact price but it was heading towards five figures, which was extortiante then. I knew the review was utter ******** when they talked about better crisper bass, clearer treble, wider deeper soundstage, and all the other legacy redundant superlatives from their experience of reviewing analogue front-ends for decades, none of which could possibly exist as differences between one transport and another unless one of them was passing the signal through a digital signal processor of some kind. Left me a bit deflated actually, like being let down by a reliable friend you trusted, and was a minor contributing factor in my dropping away from buying the mag regularly in 1995ish. Though my change of life and their shift of focus towards AV were by far the bigger factors.

 

[Andrewjvt]

For years I read WHF like a disciple, then this one time I they had a review of a really expensive transport, mid 1990s it was, can't remember the exact price but it was heading towards five figures, which was extortiante then. I knew the review was utter ******** when they talked about better crisper bass, clearer treble, wider deeper soundstage, and all the other legacy redundant superlatives from their experience of reviewing analogue front-ends for decades, none of which could possibly exist as differences between one transport and another unless one of them was passing the signal through a digital signal processor of some kind. Left me a bit deflated actually, like being let down by a reliable friend you trusted, and was a minor contributing factor in my dropping away from buying the mag regularly in 1995ish. Though my change of life and their shift of focus towards AV were by far the bigger factors.

I feel you totally on this point.

 

[lpv]

http://archimago.blogspot.co.uk/2016/08/musings-do-audiophile-computer-based.html?m=1

 

[lpv]

Have a look at this and let us know what you think

internet, I see this HW 2.2" on the board), or he used some remarkably poor switching power supply, I see no evidence of a problem personally. Maybe he should take a moment and use the testing equipment in the background of the video and see what he finds... IMO, there's no reason to bring up questionable pet theories like the "water organ" analogy and claims of "voltage of the power supply not constant" as the rationale for vague subjective impressions. What evidence is there to support the claim of "stunning" improvement such that he's suddenly "enthusiastic" to compose a review!?"

innous put 5 linear power supplies in their statement 11000 euro box ( they could at least make the box looks nice for that money)

https://benchmarkmedia.com/blogs/application_notes/152143111-audio-myth-switching-power-supplies-are-noisy

we are not new in this hobby and by now we should all have some links to post here with the relevant measurements Andrew was asking for?

 

[davidf]

The Statement eight linear supplies.

 

[Andrewjvt]

The Statement eight linear supplies.

£1350 for each

 

[insider9]

It's great to quote specs and measurements but could you tell me how does jitter of 50 picoseconds sound? How much of an impact does it have on sound quality? What if it was only 5 pico seconds? What differences would you hear? Secondly how would it show up on acoustic measurements?

Until we understand how specific measurements translate to what we hear it's probably safe to say they're as informative as values of capacitors in power supply.

 

[lpv]

it would be great if you can post some relevant measurements too..

 

[Andrewjvt]

Of jitter should be handled by the dac not the transport.

 

[insider9]

What measurements are you talking about? If you go through my post I post measurements. These are the ones I do on my own and understand what they mean.

 

[insider9]

Of jitter should be handled by the dac not the transport.

But often isn't. Assuming best case scenario isn't what's happening in real life. That's why you're going round in circles. Andyjm mentioned even software could have impact on jitter yet you seem intent on going with only the facts that are convenient. And that reinforce your view point. Not everyone is using async USB. Without asynchronous transmission cables could also make a difference.

You know I'm as much as you against the foo but if you want to know why people use transports like Innuos don't assume best case scenario. Most don't have DACs like you do and we don't live in an ideal world.

 

[Andrewjvt]

Of jitter should be handled by the dac not the transport.

But often isn't. Assuming best case scenario isn't what's happening in real life. That's why you're going round in circles. Andyjm mentioned even software could have impact on jitter yet you seem intent on going with only the facts that are convenient. And that reinforce your view point. Not everyone is using async USB. Without asynchronous transmission cables could also make a difference.

You know I'm as much as you against the foo but if you want to know why people use transports like Innuos don't assume best case scenario. Most don't have DACs like you do and we don't live in an ideal world.

Totally of track now regards the purpose of this thread.
It was supposed to be about different models of digital transports and if they are worth it if your dac is doing a good job already or the dac is designed well.
This thread has nothing to do with different models of dac design and definitely is not asynchronous v synchronous. This thread was not about my dac either.
It was about the rip of costs of digital transport and all the different models that exist.
Also seems people can't differentiate between the 2 products either.
I feel like I've asked a question and recieved loads of answers for different questions and hardly anyone has understood the question(Apart from major and Andyjm that did understand)
Answers that deal with file format, streaming services and even Netflix and now dac operations.
I don't do smiley icons but if I did it would be the one smashing his head through a window.

Darius I need to do that thing with the password protection and lose the password.

 

[andyjm]

It's great to quote specs and measurements but could you tell me how does jitter of 50 picoseconds sound? How much of an impact does it have on sound quality? What if it was only 5 pico seconds? What differences would you hear? Secondly how would it show up on acoustic measurements?

Until we understand how specific measurements translate to what we hear it's probably safe to say they're as informative as values of capacitors in power supply.

Going back to my analogy where the waveform was being plotted out on a piece of graphpaper, and the Y values are the sample size and the X values are assumed to be equal increments of X (1,2,3,4,5... and so on) then it is possible to calculate how much the X axis has to be in error to be equivalent to the Y value being in error by 1/(65536*2) - in other words the resolution of the analogue output being reduced from 16 bits to 15 bits. This seems a good place to start for the point that jitter matters.

Unfortunately there is more to it than that. Jitter is a random process, and the distribution of jitter seems to matter. I will try and find the link to the study, but entirely random jitter is far less detectable than jitter that is linked to the signal in some way. S/PDIF in particular can be subject to 'code correlated jitter' where the density of 1s vs 0s in the data bitstream makes the detection of the clock slightly early or late. So the jitter is then linked to the music and apparently much easier for a listener to detect.

As for measurement, jitter is tricky to measure directly without some very fancy gear. Most approaches measure the effects of jitter, not the jitter itself. Jitter is a bit like modulation in that the instantantaneos frequency of the jittered signal is altered by the jitter. So if you play a pure sinewave of a given frequency through a DAC, and look at the output on a spectrum analyser, any frequency that shows up in the output that isn't the original sinewave is the result of jitter (amongst other things).

If you do the maths, then about a 100ps of clock error at 10KHz or 20ps of clock error at 20KHz is equivalent to 1/2 LSB of a 16bit resolution.

I didn't pay the fee, but apparently this paper has the derivations: http://www.aes.org/e-lib/browse.cfm?elib=6111

This is the point jitter begins to effect the overall resolution of the system, that is not the same as the point at which it becomes detectable. Other studies suggest 250nS is the threshold of detection, but as I mentioned above it depends what type of jitter you are talking about. It also depends what jitter numbers you use - its a distribution, so are you quoting peak, RMS, average.....

 

[paulkebab]

this thread with interest but as you say Andrew, tits up.

 

[davidf]

The Statement eight linear supplies.

£1350 for each

Well, at least you get two well made cases, SSD hard drives, CD reader/ripper, USB and Ethernet connections, Innuos’ bespoke software, mains and Ethernet cables, box and packaging, warranty, and great back up service for free. Not bad, you usually have to pay for everything else

 

[nopiano]

Of jitter should be handled by the dac not the transport.

But often isn't. Assuming best case scenario isn't what's happening in real life. That's why you're going round in circles. Andyjm mentioned even software could have impact on jitter yet you seem intent on going with only the facts that are convenient. And that reinforce your view point. Not everyone is using async USB. Without asynchronous transmission cables could also make a difference.

You know I'm as much as you against the foo but if you want to know why people use transports like Innuos don't assume best case scenario. Most don't have DACs like you do and we don't live in an ideal world.

...and in the Aurender review they were using a Chord DAC, which by most reckonings is about as good as DACs get. They reported differences in sound between the CD source and the same file replayed from the server.

But you’re correct, Andrew, they didn’t compare with replaying from, say, a home NAS at a tenth of the price.

I suspect there may be some psychological background, rooted in our analogue experience. That implies if you improve each item, there is a cumulative effect of improvement. It may be that with a very good DAC, those other variables cease to be meaningful. Therefore your belief that Innuos or Aurender bits are unlikely to be any sparklier than regular NAS bits may be true. Or that NAS bits are good enough if the DAC is very good.

 

[insider9]

It's great to quote specs and measurements but could you tell me how does jitter of 50 picoseconds sound? How much of an impact does it have on sound quality? What if it was only 5 pico seconds? What differences would you hear? Secondly how would it show up on acoustic measurements?

Until we understand how specific measurements translate to what we hear it's probably safe to say they're as informative as values of capacitors in power supply.

Going back to my analogy where the waveform was being plotted out on a piece of graphpaper, and the Y values are the sample size and the X values are assumed to be equal increments of X (1,2,3,4,5... and so on) then it is possible to calculate how much the X axis has to be in error to be equivalent to the Y value being in error by 1/(65536*2) - in other words the resolution of the analogue output being reduced from 16 bits to 15 bits.  This seems a good place to start for the point that jitter matters.

Unfortunately there is more to it than that.  Jitter is a random process, and the distribution of jitter seems to matter.  I will try and find the link to the study, but entirely random jitter is far less detectable than jitter that is linked to the signal in some way.  S/PDIF in particular can be subject to 'code correlated jitter'  where the density of 1s vs 0s in the data bitstream makes the detection of the clock slightly early or late. So the jitter is then linked to the music and apparently much easier for a listener to detect.

As for measurement, jitter is tricky to measure directly without some very fancy gear.  Most approaches measure the effects of jitter, not the jitter itself.  Jitter is a bit like modulation in that the instantantaneos frequency of the jittered signal is altered by the jitter.  So if you play a pure sinewave of a given frequency through a DAC, and look at the output on a spectrum analyser, any frequency that shows up in the output that isn't the original sinewave is the result of jitter (amongst other things).   

If you do the maths, then about a 100ps of clock error at 10KHz or 20ps of clock error at 20KHz is equivalent to 1/2 LSB of a 16bit resolution.

I didn't pay the fee, but apparently this paper has the derivations:  http://www.aes.org/e-lib/browse.cfm?elib=6111

This is the point jitter begins to effect the overall resolution of the system, that is not the same as the point at which it becomes detectable.  Other studies suggest 250nS is the threshold of detection, but as I mentioned above it depends what type of jitter you are talking about.  It also depends what jitter numbers you use - its a distribution, so are you quoting peak, RMS, average.....

 

 

Very informative, thanks Andyjm.

 

[Strictly Stereo]

Of jitter should be handled by the dac not the transport.

Why? Some would argue that simpler converters have a more natural sound.

 

[andyjm]

Of jitter should be handled by the dac not the transport.

Why? Some would argue that simpler converters have a more natural sound.

All clock circuits have jitter. Transporting the clock via wires or optical cables introduces further jitter.

The best place to have the clock cicuitry is right next to the D2A converter chip on the circuit board in the DAC. The same clock circuit placed anywhere else will be worse.

 

[Strictly Stereo]

All clock circuits have jitter. Transporting the clock via wires or optical cables introduces further jitter.

The best place to have the clock cicuitry is right next to the D2A converter chip on the circuit board in the DAC. The same clock circuit placed anywhere else will be worse.

The DAC on its own is of little use to anyone. To play music you need a source too. The source also requires a clock. The best option is to have both the source and the DAC share or sync to the same clock. This is pretty much the standard in professional music production. The next best alternatives are to have very precise clocks in both the source and the DAC, or some means of correcting errors in the signal sent from the source.

There are some people who prefer the sound quality of simpler DAC designs. Not all of these offer asynchronous USB, buffering/reclocking of the input signal or other methods for side-stepping or correctly errors. Such DACs perform best when fed with a low jitter input.

My point is simply that there is more than one way to put together a system.

 

[andyjm]

All clock circuits have jitter. Transporting the clock via wires or optical cables introduces further jitter.

The best place to have the clock cicuitry is right next to the D2A converter chip on the circuit board in the DAC. The same clock circuit placed anywhere else will be worse.

The DAC on its own is of little use to anyone. To play music you need a source too. The source also requires a clock. The best option is to have both the source and the DAC share or sync to the same clock. This is pretty much the standard in professional music production. The next best alternatives are to have very precise clocks in both the source and the DAC, or some means of correcting errors in the signal sent from the source.

There are some people who prefer the sound quality of simpler DAC designs. Not all of these offer asynchronous USB, buffering/reclocking of the input signal or other methods for side-stepping or correctly errors. Such DACs perform best when fed with a low jitter input.

My point is simply that there is more than one way to put together a system.

As I mentioned, the best place to have the clock is right next to the D2A chip. Putting the same clock somewhere else will be worse.

Slaving a DAC to a clock generated further upstream is a legacy of a time when flow control and async digital data transfer was expensive. Nowdays chipsets to do this cost pennies. How much did that last USB memory stick cost you bought?

If you choose to buy a DAC that relies upon a synchronous approach, then of course you are dependent on the quality of the input clock, but it will never have the same jitter performance as a DAC wth a local clock.