Mytek Stereo 192

[57gold]

From AudioQuest's website:

 

 

 

I would think that noise could just as easily find its way onto a FW cable and then into an audio system.

 

And like USB, FW sends power down two conductors in a FW cable. Another way for noise to enter an audio system.

 

Just saying...

 

Read the Regen stuff...it's the "packet" noise that is messing with the sound.

 

AQ, a company w/good products, has to pitch real and imagined and possible "noises" and "pollutions" to justify the sale of its many hundreds of $s cable products.

 

Quoted Ayre on the limitations of USB and a studio product company on FW versus USB...neither sell cables.

 

Real answer is for DAC producers to: a) incorporate appropriate technologies into DACs to deal with the many weaknesses of USB (had to chuckle when I read a review that was posted/linked here on CA related to a new model $15K DAC from Lampi that needed a Regen to sound optimal...not sure why $15K can't buy "packet noise" attenuation inside the freakin' box); or b) develop an alternative means of transmission of files to DAC that has integrity.

[kennyb123]

AQ, a company w/good products, has to pitch real and imagined and possible "noises" and "pollutions" to justify the sale of its many hundreds of $s cable products.

 

I think you might want to take those words back if you got to hear the bang for the buck delivered by a single or pair of Jitterbugs in a good system.

 

I don't think their cables are anywhere close to being as good of a value, but they've done a good thing by making the Jitterbug so affordable.

 

 

 

Real answer is for DAC producers to: a) incorporate appropriate technologies into DACs to deal with the many weaknesses of USB

 

They'll get there. The industry is still learning here, which is a good thing in my view. That some creative individuals are coming up with innovative ways to affordably improve audio should be applauded.

 

If FW wasn't already dead for home audio some smart folks would probably be looking into ways to improve it too.

 

I would say that a pair of Jitterbugs plus a REGEN delivered as significant of an upgrade as my last couple DAC upgrades. For $375!

[STC]

It's an analog signal that passes through a USB cable. Additionally the USB cable provides a path though which noise can travel from the computer to the DAC. So it's more than just a problem of "sending data" as you suggest.

 

Light reading for you:

 

Q&A with John Swenson. Part 1: What is Digital? | AudioStream

Q&A with John Swenson. Part 2: Are Bits Just Bits? | AudioStream

 

 

 

 

This sounds like a logical fallacy (argument from authority) as these say nothing about their skills at assembling a high end audio system. That's why I asked about your system. It helps me to understand YOUR skill at assembling a system.

 

 

 

If you believe that then I know for a fact that I can't change your mind. We can just agree to disagree.

 

I use the same approach to select audio gear that I do when picking my favorite coffee, or beer, or chocolate, or whatever.

 

 

 

I used to have a Theta transport and DAC connected to an ARC preamp and Classe power amp. So maybe we aren't that far apart in our tastes.

 

 

 

Not about approval, but about context and frame of reference. You've made wise choices in gear, so you should be able to hear differences in USB cables as long as your signal cabling isn't getting in the way.

 

Too many folks who claim they don't hear difference in cables don't list their systems. So it's impossible to know if they don't hear the difference simply because their system isn't sufficiently resolving. Methinks that's exactly why they don't list their systems.

 

Thanks for the links. There are many other links which say the opposite too.

 

Anyway, do you think if I should stop downloading from LINN and Chesky servers? You see, I stay over 10000 kilometers from those servers and the data from their server to my server are coming from non audiophile grade cables. Could it be this is the reason I am not hearing any difference?

[Miska]

If FW wasn't already dead for home audio some smart folks would probably be looking into ways to improve it too.

 

With the Mytek Stereo192-DSD DAC I'd recommend Mac users to use Firewire and Windows users to use USB.

 

I use it now with my iMac (OSX) using Apple's Thunderbolt-to-Firewire adapter and things work very well. Before that, for long time I was using it on Windows (Mac Mini with BootCamp) through USB and that was working well too. But the opposite can be very tricky, USB was problematic on OS X and the Firewire can be very tricky on Windows.

 

I run the DAC at DSD128 for all source material.

 

 

Technically best connections for external DAC are Ethernet and Thunderbolt. Which one is better depends on your exact requirements. (both are now replacing older connection methods on pro-audio side)

[kennyb123]

Anyway, do you think if I should stop downloading from LINN and Chesky servers? You see, I stay over 10000 kilometers from those servers and the data from their server to my server are coming from non audiophile grade cables. Could it be this is the reason I am not hearing any difference?

 

You may be right. Your attempt at humor falls flat. Maybe all those kilometers are to blame? :-)

[STC]

You may be right. Your attempt at humor falls flat. Maybe all those kilometers are to blame? :-)

 

I was hoping you and the link you posted would shed some light to it. I guess, now is the time to stop because no answers will be forthcoming.

[kennyb123]

I was hoping you and the link you posted would shed some light to it. I guess, now is the time to stop because no answers will be forthcoming.

 

I would gladly engage in a discussion. Your reply suggested you weren't serious.

[STC]

I would gladly engage in a discussion. Your reply suggested you weren't serious.

 

That was a valid question. Going by your logic, anyone could tell the obvious sound difference between the original files located in LINN server and one downloaded in Sydney from the LINN server.

 

According to your reference, the logic would be , the file reaching Sydney after traveling thousands of kilometers through non audiophile grade cable should be different or at least sufficiently degraded that you would able to hear the cymbal, or sax or drums to sound different. If that wasn't the case, please explain.

 

I am talking about the digital data.

[kennyb123]

That was a valid question. Going by your logic, anyone could tell the obvious sound difference between the original files located in LINN server and one downloaded in Sydney from the LINN server.

 

Would you please provide a quote from me where I suggested such a thing?

 

What you've started is ludicrous and not my position whatsoever.

 

It was another member who suggested that you needed to use a shorter length. I didn't comment on this.

 

 

According to your reference, the logic would be , the file reaching Sydney after traveling thousands of kilometers through non audiophile grade cable should be different or at least sufficiently degraded that you would able to hear the cymbal, or sax or drums to sound different. If that wasn't the case, please explain.

 

Would you please provide quotes from the references I provided that suggested this?

[STC]

Would you please provide a quote from me where I suggested such a thing?

 

What you've started is ludicrous and not my position whatsoever.

 

It was another member who suggested that you needed to use a shorter length. I didn't comment on this.

 

 

 

 

Would you please provide quotes from the references I provided that suggested this?

 

Well.... I said, the logic of your argument...

 

All I said was that I use a $4 cable but was ridiculed by you that I was using a donut tires to drive an expensive car ( or something to the effect).

 

You also mentioned, that the USB cable was sending more than data and referred to some engineering paper. Oops.... Sorry. I meant Audiostream interview.

 

So if transmission wasn't just digital data then by your logic, you should not have any trouble distinguishing the two files. Obviously, the files downloaded in Sydney would be easily distinguished.

 

The answer is ether a Yes or No. If the answer is YES than you are right? We all should stop downloading files as the sound quality would be compromised. If No, then I would appreciate if could explain why it isn't the case.

[Miska]

All I said was that I use a $4 cable but was ridiculed by you that I was using a donut tires to drive an expensive car ( or something to the effect).

 

I recommend trying suitable ferrite beads on your cable. Or getting a HiSpeed certified cable with ferrites, such things are available for around $10.

 

You know there's a reason why ferrites are used on cables...

 

So if transmission wasn't just digital data then by your logic, you should not have any trouble distinguishing the two files. Obviously, the files downloaded in Sydney would be easily distinguished.

 

The answer is ether a Yes or No. If the answer is YES than you are right? We all should stop downloading files as the sound quality would be compromised. If No, then I would appreciate if could explain why it isn't the case.

 

That is getting besides the point because the download doesn't involve D/A conversion. Or even in streaming case it most likely doesn't have notable impact on player computer's EMI/RFI pattern. But may be the case for small "streamer" devices as has been seen from measurements (local vs streamed playback).

 

These things only begin to matter when you have A) computer and B) D/A converter connected to each other. You may want to make sure none of EMI/RFI from (A) gets into mixed signal circuitry of (B) withing at least 120 dB dynamic range. So beware of for example ground currents carrying even one microvolt of noise. As long as you don't have (B) involved you don't need to worry as long as you don't have too much noise for bits to flip in wrong position (for that you need just a bit over 6 dB dynamic range). This has nothing to do with the data transferred between (A) and (B) per se. This is only a 501190 times difference in SNR requirement.

[STC]

I recommend trying suitable ferrite beads on your cable. Or getting a HiSpeed certified cable with ferrites, such things are available for around $10.

 

You know there's a reason why ferrites are used on cables...

 

 

 

That is getting besides the point because the download doesn't involve D/A conversion. Or even in streaming case it most likely doesn't have notable impact on player computer's EMI/RFI pattern. But may be the case for small "streamer" devices as has been seen from measurements (local vs streamed playback).

 

These things only begin to matter when you have A) computer and B) D/A converter connected to each other. You may want to make sure none of EMI/RFI from (A) gets into mixed signal circuitry of (B) withing at least 120 dB dynamic range. So beware of for example ground currents carrying even one microvolt of noise. As long as you don't have (B) involved you don't need to worry as long as you don't have too much noise for bits to flip in wrong position (for that you need just a bit over 6 dB dynamic range). This has nothing to do with the data transferred between (A) and (B) per se. This is only a 501190 times difference in SNR requirement.

 

Thank you for giving a very detailed explanation.

 

I did use the ferrite beads when I was controlling the volume through my laptop. I think it made a difference but that was with 1.5 meter USB cable. However, when I bypassed the internal laptop soundcard and disabled the volume control in JRiver I noticed slight improvement in the highs without the beads. I didn't bother to find out why except reading in some engineering forum that my observation was probably correct.

 

Now about the second part of your post which is too complicated for me to understand.

 

I still cannot get it why it is possible to transfer the the audio file without any loss in the audio quality thousands over miles but unable to transfer a few more meters to the DAC chip without any degradation.

 

Ahh... If only someone could invent a small device to reconstruct the data packet reaching the DAC, just like how it gets reconstructed at the laptop...or that already existed?

[wwaldmanfan]

I have a Kimber Kable B-Bus Cu USB cable. This was one of the first "audiophile" USB cables on the market. It is also one of the few I've seen with ferrite beads. Is it possible that, besides absorbing noise, they might affect the signal? On my system, when I was using it, I think that there was a slight improvement in SQ after removing the beads.

 

[Miska]

I still cannot get it why it is possible to transfer the the audio file without any loss in the audio quality thousands over miles but unable to transfer a few more meters to the DAC chip without any degradation.

 

Because things get sensitive only at edges between digital and analog domain, thus at A/D and D/A conversion points. Plus of course any plain analog path is sensitive. As long as you stay completely within digital domain things are simpler.

 

If you want to achieve best possible results, the conversion point is very sensitive to any disturbances. A/D and D/A converters operate at this sensitive boundary. Digital circuits generate high frequency switching noise you absolutely want to keep out of the analog signal. However, your DAC needs to deal with both digital and analog signal at the same time, within the same box in a very close proximity.

 

Ahh... If only someone could invent a small device to reconstruct the data packet reaching the DAC, just like how it gets reconstructed at the laptop...or that already existed?

 

That's what the DACs do, but the problem is not the data packet. Problem is that the data packet's analog representation is unwanted noise from the analog domain's point of view... (any real world digital data transmission has it's analog representation too)

 

To reach 120 dB SNR the DAC needs to isolate (noise of) the digital side from the analog side with 1:1000000 (one-to-million) ratio while perfectly converting the information delivered as digital data to analog domain.

[Miska]

I have a Kimber Kable B-Bus Cu USB cable. This was one of the first "audiophile" USB cables on the market. It is also one of the few I've seen with ferrite beads. Is it possible that, besides absorbing noise, they might affect the signal? On my system, when I was using it, I think that there was a slight improvement in SQ after removing the beads.

 

As long as it doesn't cause transmission errors, it doesn't matter if it affects the digital signal. I have only one "audiophile" USB cable (Supra USB cable). Everything else is USB 2 HiSpeed certified ordinary stuff, with or without ferrite beads. There are probably many audiophile USB cables that wouldn't pass HiSpeed certification.

 

Based on my measurements, ferrite beads have varying amount of positive effect, but I have not measured any negative effects.

 

My Firewire cable from Apple TB-FW adapter to Mytek DAC is Lacie's flat 9-6 cable. Has been working perfectly.

[STC]

Because things get sensitive only at edges between digital and analog domain, thus at A/D and D/A conversion points. Plus of course any plain analog path is sensitive. As long as you stay completely within digital domain things are simpler.

 

If you want to achieve best possible results, the conversion point is very sensitive to any disturbances. A/D and D/A converters operate at this sensitive boundary. Digital circuits generate high frequency switching noise you absolutely want to keep out of the analog signal. However, your DAC needs to deal with both digital and analog signal at the same time, within the same box in a very close proximity.

 

 

 

That's what the DACs do, but the problem is not the data packet. Problem is that the data packet's analog representation is unwanted noise from the analog domain's point of view... (any real world digital data transmission has it's analog representation too)

 

To reach 120 dB SNR the DAC needs to isolate (noise of) the digital side from the analog side with 1:1000000 (one-to-million) ratio while perfectly converting the information delivered as digital data to analog domain.

 

 

Hi Miska,

 

Thanks again for the reply. I hope you understand that I am coming from a different angle, i.e., after extensive listening tests I am unable to distinguish the 1 meter and 5 meter USB cables. Some links are useful to reassure me such as this Archimago's Musings: MEASUREMENTS: USB Cables for Audio DACs. [2013-06-18 UPDATE] .... but being and audiophile, I always look for convincing evidence to the contrary. $4 cable for $1500 DAC looks ridiculous. I am aware that you too using non audiophile cables and therefore we are on the same page. I now understand that digitally it is possible to preserve/duplicate copy but problems could start at the conversion stage.

 

You said

Because things get sensitive only at edges between digital and analog domain, thus at A/D and D/A conversion points. Plus of course any plain analog path is sensitive. As long as you stay completely within digital domain things are simpler.

 

So as long as the signal stays in digital domain it shouldnt matter? Digital data traveling thousands of miles under EMI/RFI exposure still somehow managed to remain in pristine condition when it reaches my pc. The PC is capable of ignoring the errors attributed to EMI/RFI and reconstruct the file exactly as the original. We all know that and I am just repeating it here. The problem starts when the conversion takes place.

 

Since I do not have any technical knowledge of how a DAC works, my curious mind is asking:-

 

1) The signal is still in the digital form when it arrives at the DAC chip pin. Correct?

 

2) Somehow, at this point the pollution matters as the DAC is incapable of distinguishing noise from data. A PC could do it but not a DAC?

 

3) To prevent (2) happening, we need to ensure the signal to arrive as pristine as possible and therefore expensive or "better" cables are needed (according to some).

 

4) The weakest link is the interface between the PC and the DAC.

 

Going back to Mytek 192 DAC, it was sold with a standard USB cable. The guarantee that came with their product was

 

"-guaranteed bit transparency from driver to DAC chip". Am I missing something here?

[Miska]

1) The signal is still in the digital form when it arrives at the DAC chip pin. Correct?

 

Yes...

 

2) Somehow, at this point the pollution matters as the DAC is incapable of distinguishing noise from data. A PC could do it but not a DAC?

 

No, it is fully capable of correctly determining the data, but extra noise carried by the data and ground traces may sneak over to the analog side since the DAC chip is a mixed signal chip. It has digital input side and analog output side. You don't want anything unrelated to carry over from digital side to the analog side.

 

So this has nothing to do with correctness of the data, but everything to do with non-data related signal carried over the wire.

 

3) To prevent (2) happening, we need to ensure the signal to arrive as pristine as possible and therefore expensive or "better" cables are needed (according to some).

 

I wouldn't say it that way. I would say that we want to minimize unrelated (non-data) components on the digital signal which naturally has also it's analog representation as a square wave.

 

4) The weakest link is the interface between the PC and the DAC.

 

I would say the weakest link is the DAC chip where digital and analog side are less than a millimeter away from each other on the same piece of silicon while requiring at least one-to-million isolation from each other.

 

Going back to Mytek 192 DAC, it was sold with a standard USB cable. The guarantee that came with their product was

 

"-guaranteed bit transparency from driver to DAC chip". Am I missing something here?

 

What you are missing is consideration of the mixed signal nature of DAC.

[STC]

.,,,,,

No, it is fully capable of correctly determining the data, but extra noise carried by the data and ground traces may sneak over to the analog side since the DAC chip is a mixed signal chip. It has digital input side and analog output side. You don't want anything unrelated to carry over from digital side to the analog side.

 

So this has nothing to do with correctness of the data, but everything to do with non-data related signal carried over the wire.

 

 

 

I wouldn't say it that way. I would say that we want to minimize unrelated (non-data) components on the digital signal which naturally has also it's analog representation as a square wave.

 

 

 

I would say the weakest link is the DAC chip where digital and analog side are less than a millimeter away from each other on the same piece of silicon while requiring at least one-to-million isolation from each other.

 

 

 

What you are missing is consideration of the mixed signal nature of DAC.

 

Miska, thank you so much for your patience and for the explanation. Now, I see the logic.

 

I will try a shorter cable with and without the ferrite beads. Of course, I will be doing the listening expecting to hear difference. Hope that would not cloud my judgment.

 

Thanks once again.

[STC]

I tried with the original Mytek USB cable. Can't really tell if there was an improvement. No ferrite clamps used.

 

Left the original Mytek cable on USB 2.0 and connected the $4 cable to USB1.1. All files down or up sampled to 96kHz.

 

Sound driver used was Stereo 192 DSD DAC ( direct sound ). Internal sound card disabled.

 

Pin 1 of both cables taped over.

 

As I was listening to James Taylor's Hourglass - Little more time with you, I could hear the chorus woman and probably another two persons quiet clearly. That's with the $4 five meter cable. Never noticed it before. Since I have heard it now, I am sure I would be hearing it with other cables too. ?

[Miska]

Left the original Mytek cable on USB 2.0 and connected the $4 cable to USB1.1. All files down or up sampled to 96kHz.

 

That USB 1.1 interface will certainly give horrible results due to clocking (it is not asynchronous, so the clock is derived from the data stream timing).

 

Sound driver used was Stereo 192 DSD DAC ( direct sound ). Internal sound card disabled.

 

What do you use as player? DirectSound is the least direct way of playback on Windows since Vista, despite the name which is just because it is related to DirectX gaming framework. With DirectSound you have Windows audio engine mangling with the data.

[STC]

That USB 1.1 interface will certainly give horrible results due to clocking (it is not asynchronous, so the clock is derived from the data stream timing).

 

 

I am using JRiver MC20. The "Stereo 192 DSD DAC ( direct sound )" must be a new feature (the name is exactly how it was displayed in the option). Never seen it before. I am using Windows10. I also noticed that the upsampling automatically engaged and couldnt be disabled with this option.

 

The signal flow chart in page 38 of Mytek 192 manual not very helpful. BTW, I use digital volume control. Not sure if that changes anything.

 

 

 

[kennyb123]

That USB 1.1 interface will certainly give horrible results due to clocking (it is not asynchronous, so the clock is derived from the data stream timing).

 

Just to be clear - are you saying that the Mytek's USB #1 input is not asynchronous? I suggested that was the case here a while ago, and someone told me I was wrong.

[STC]

I am going back to the original setup. I.e, Mytek ASIO driver > USB 2.0. With the direct sound (windows or Mytek's driver), I am hearing loud pops occasionally. To prevent the pops, I need to restart the laptop. With USB2.0 and Mytek's ASIO, I never had this problem.

 

Subjectively, I think Mytek ASIO driver (USB 2.0) is slightly louder than direct sound (windows or Mytek's driver). Can anyone confirm this? It is possible to perceive better sound due to the louder level of USB2.

[Miska]

I am using JRiver MC20. "The Stereo 192 DSD DAC ( direct sound )" must be a new feature (the name is exactly how it was displayed in the option). Never seen it before. I am using Windows10. I also noticed that the upsampling automatically engaged and couldnt be disabled with this option.

 

The signal flow chart in page 38 of Mytek 192 manual not very helpful. BTW, I use digital volume control. Not sure if that changes anything.

 

Yes, that signal flow chart is not useful in this context because what matter is the signal flow inside Windows operating system.

 

You should stick to preferably ASIO...

[Miska]

Just to be clear - are you saying that the Mytek's USB #1 input is not asynchronous? I suggested that was the case here a while ago, and someone told me I was wrong.

 

Yes, the USB 1.1 input is not asynchronous, it is old style interface for backwards compatibility in cases where USB 2.0 cannot be used. The USB 2.0 input is asynchronous.

 

On Windows I'd recommend to stick to USB 2.0 and ASIO driver. On Mac I'd recommend the FW interface instead.