[audiobomber]

Of course they’re not, talk about ridiculous! How many of them even have a degree in computer network protocols? And if they do, where do you think they got it, Head-Fi, Stereophile Magazine or maybe a university? The actual people with the “interest and capabilities” are the scientists and engineers who invented, test and maintain the protocols, the organisations that publish the technical standards and the universities that teach them. Who, apart from deluded audiophiles, wouldn’t know that?

But wait, you’ve already dismissed all universities, science, encyclopaedias, professionals, etc., in favour of those trying to sell you audiophile products. Now that’s dogma!

No, you're wrong. I dismiss the scientists and engineers who think simple static measurements describe sound quality. I follow those who know this is wrong, and design products for sound, not numbers. Like Ed Meitner for example:

"There is a reason for measurement up to the point that you reach a certain level of what I call 'hygiene,'" Ed Meitner said. "Beyond that, you need to perform a voicing that produces 10 octaves of sound. It will always come down to the sound you hear. The measurement is the 'hygiene,' and the listening is the final performance. Only after that do we put our name to it. We hope that we make as transparent a sound—I would say 'no sound'—as possible. We are in the anticoloration business."

"I think some people think that a microphone hears the same way as our ears. So, we have these room correction devices that fool you into believing that they create the perfect listening environment because the microphone tells you so. The same thing happens when you look at your distortion analyzer or whatever makes you think you're okay. You're far from okay."

"There are sonic differences that cannot be measured. I won't say they will never be able to be measured, but as far as we know, there's no current measurements that will guarantee sound. I can give you an example. When solid state amplifiers came on the scene, people hated the sound because they used maximum power distortion measurements [to gauge success]. But all the problems were at zero-crossing distortion, which was a measurement that wasn't really included. As the level was going down, the distortion was going up relative to it. So, it was only after listening that the measurement was finally included. "There will always be a new measurement that will be discovered. For example, the effect of jitter was initially ignored."

Pawlicki delivered the second part of the EMM/Meitner one-two punch. "To reinforce Ed's statement, all the typical classical measurements in audio typically involve steady-state sinewave tests. Music, however, has ... transients and continuous changes. Part of EMM's philosophy is to take care of the constant transients with the MDAT circuit. Of course, we perform the steady-state hygiene measurements to ensure that it always works. But the constantly changing transient part, which is music, needs to be treated a bit differently. That has been our philosophy from the very beginning. To this day, people ignore how continual changes to the signal—transients—are handled in real circuits."

Footnote 1: EMM Labs manufactured more than 100 mobile professional recording systems, including a "Sonoma" DAW (digital audio workstation) with multichannel EMM ADC-8 and DAC-8 converters, which have been used worldwide to record, master, and publish the majority of SACDs. It also manufactured the "Switchman" preamplifier and various D/A converters. Having developed the first complete multichannel DSD playback system from disc to preamplifier, EMM claims that almost every new SACD is made by using its DSD converters somewhere in the production chain.

Footnote 2: Founder/President Ed Meitner received a patent in 1993 for describing the phenomenon that produces program-related jitter and explaining how to remove accumulated music-correlated jitter by reclocking an audio datastream. EMM Labs Director of Sales Shahin Al Rashid told me in an email, "Although re-clocking is now widely used, Ed is chagrined that it is often implemented ineffectively by not following the patent completely. The central point in Ed's patent that is often missed is that because jitter is propagated and amplified through power-supply interactions, each clock-generator and clock-distribution circuit, as well as every re-clocker chip, must have its own stable power-supply, individually regulated and separate from the system power supply." In 2012, Ed Meitner developed the first DACs with sub-picosecond (or femtosecond) clock jitter in their reference clock oscillators.
https://www.stereophile.com/content/meitner-ma3-integrated-da-processor

When Sony needed to transform SACD from an intriguing concept into a sonic reality, they turned to one of the world's leading experts in digital converter design: Ed Meitner. In those early days, Ed's EMM Labs worked hand-in-glove with Sony and Phillips to refine what was possible from the format and to establish the unparalleled sonic purity for which SACD is now known.

The company soon developed the first complete multichannel DSD playback system from disc to preamplifier. And quickly became one of the industry's most respected makers of high-resolution audio recording and playback products. These distinctions were achieved by the unique amalgam of talent here—physicists, engineers and music lovers alike—working as a team to create the most faithful music production and reproduction equipment available. And, in fact, possible.

Today, EMM Labs converter systems are the de facto DSD reference of the recording industry. Currently, almost every new SACD in production is being made with our DSD converters. The analogue mixing consoles Ed designed back in the early 1970's for Olive Electrodynamics Company, for example, are still sought after today. So too are his Museatex and Melior audiophile products. His original thinking in the field of audio design led to several patents including:

• Reproduction equipment for digital audio
• Very low jitter clock recovery from serial audio data
• Adaptive digital audio interpolation system
• Magnetic pickup preamplifier
• Retrofitable CD player system

Other notable accomplishments include creating the world's first VCA-controlled preamplifier. The first commercially available cryogenically treated cables. The legendary BiDat converter. The coveted MTR-101 amplifier.
https://www.emmlabs.com/about.html


PS I own the coveted Meitner MTR-101 monoblocks, as well as a pair of STR-50 stereo amps.

 

[audiobomber]

Impression of: iFi iPower Elite + iPurifier3 + Ground cable

I prefere Stock PSU and usb cable + ground cable.

Maybe its because the RME ADi-2 DAC FS is crap anyway?

Im not impressed with this DAC in general. So im maybe biased

Optical or USB its the same.

Are you using the RME headphone amp? According to this guy, the headphone amp is weak:
The ADI-2 DAC’s headamp has a decidedly low-fi sound.
https://audiophilestyle.com/ca/reviews/rme-adi-2-dac-fs-review-r852/

 

[bfreedma]

No, you're wrong. I dismiss the scientists and engineers who think simple static measurements describe sound quality. I follow those who know this is wrong, and design products for sound, not numbers. Like Ed Meitner for example:

"There is a reason for measurement up to the point that you reach a certain level of what I call 'hygiene,'" Ed Meitner said. "Beyond that, you need to perform a voicing that produces 10 octaves of sound. It will always come down to the sound you hear. The measurement is the 'hygiene,' and the listening is the final performance. Only after that do we put our name to it. We hope that we make as transparent a sound—I would say 'no sound'—as possible. We are in the anticoloration business."

"I think some people think that a microphone hears the same way as our ears. So, we have these room correction devices that fool you into believing that they create the perfect listening environment because the microphone tells you so. The same thing happens when you look at your distortion analyzer or whatever makes you think you're okay. You're far from okay."

"There are sonic differences that cannot be measured. I won't say they will never be able to be measured, but as far as we know, there's no current measurements that will guarantee sound. I can give you an example. When solid state amplifiers came on the scene, people hated the sound because they used maximum power distortion measurements [to gauge success]. But all the problems were at zero-crossing distortion, which was a measurement that wasn't really included. As the level was going down, the distortion was going up relative to it. So, it was only after listening that the measurement was finally included. "There will always be a new measurement that will be discovered. For example, the effect of jitter was initially ignored."

Pawlicki delivered the second part of the EMM/Meitner one-two punch. "To reinforce Ed's statement, all the typical classical measurements in audio typically involve steady-state sinewave tests. Music, however, has ... transients and continuous changes. Part of EMM's philosophy is to take care of the constant transients with the MDAT circuit. Of course, we perform the steady-state hygiene measurements to ensure that it always works. But the constantly changing transient part, which is music, needs to be treated a bit differently. That has been our philosophy from the very beginning. To this day, people ignore how continual changes to the signal—transients—are handled in real circuits."

Footnote 1: EMM Labs manufactured more than 100 mobile professional recording systems, including a "Sonoma" DAW (digital audio workstation) with multichannel EMM ADC-8 and DAC-8 converters, which have been used worldwide to record, master, and publish the majority of SACDs. It also manufactured the "Switchman" preamplifier and various D/A converters. Having developed the first complete multichannel DSD playback system from disc to preamplifier, EMM claims that almost every new SACD is made by using its DSD converters somewhere in the production chain.

Footnote 2: Founder/President Ed Meitner received a patent in 1993 for describing the phenomenon that produces program-related jitter and explaining how to remove accumulated music-correlated jitter by reclocking an audio datastream. EMM Labs Director of Sales Shahin Al Rashid told me in an email, "Although re-clocking is now widely used, Ed is chagrined that it is often implemented ineffectively by not following the patent completely. The central point in Ed's patent that is often missed is that because jitter is propagated and amplified through power-supply interactions, each clock-generator and clock-distribution circuit, as well as every re-clocker chip, must have its own stable power-supply, individually regulated and separate from the system power supply." In 2012, Ed Meitner developed the first DACs with sub-picosecond (or femtosecond) clock jitter in their reference clock oscillators.
https://www.stereophile.com/content/meitner-ma3-integrated-da-processor

When Sony needed to transform SACD from an intriguing concept into a sonic reality, they turned to one of the world's leading experts in digital converter design: Ed Meitner. In those early days, Ed's EMM Labs worked hand-in-glove with Sony and Phillips to refine what was possible from the format and to establish the unparalleled sonic purity for which SACD is now known.

The company soon developed the first complete multichannel DSD playback system from disc to preamplifier. And quickly became one of the industry's most respected makers of high-resolution audio recording and playback products. These distinctions were achieved by the unique amalgam of talent here—physicists, engineers and music lovers alike—working as a team to create the most faithful music production and reproduction equipment available. And, in fact, possible.

Today, EMM Labs converter systems are the de facto DSD reference of the recording industry. Currently, almost every new SACD in production is being made with our DSD converters. The analogue mixing consoles Ed designed back in the early 1970's for Olive Electrodynamics Company, for example, are still sought after today. So too are his Museatex and Melior audiophile products. His original thinking in the field of audio design led to several patents including:

• Reproduction equipment for digital audio
• Very low jitter clock recovery from serial audio data
• Adaptive digital audio interpolation system
• Magnetic pickup preamplifier
• Retrofitable CD player system

Other notable accomplishments include creating the world's first VCA-controlled preamplifier. The first commercially available cryogenically treated cables. The legendary BiDat converter. The coveted MTR-101 amplifier.
https://www.emmlabs.com/about.html


PS I own the coveted Meitner MTR-101 monoblocks, as well as a pair of STR-50 stereo amps.

Marketing material from a vendor that has absolutely nothing to do with the discussion of USB and/or Ethernet. Well done...

 

[gregorio]

I follow those who know this is wrong, and design products for sound, not numbers.

Thanks for all that marketing, it certainly puts Wikipedia, all the world’s universities, scientists and engineers in their place.

BTW, you do know that DACs don’t output any sound and that if you don’t give them any numbers they don’t output anything all? What do you think digital audio is, if it’s not a measurement? The answer to that should be interesting! Jeez

G

 

[hakunamakaka]

Of course they’re not, talk about ridiculous! How many of them even have a degree in computer network protocols? And if they do, where do you think they got it, Head-Fi, Stereophile Magazine or maybe a university? The actual people with the “interest and capabilities” are the scientists and engineers who invented, test and maintain the protocols, the organisations that publish the technical standards and the universities that teach them. Who, apart from deluded audiophiles, wouldn’t know that?

Why do you need to understand low level algorithms and code of such protocols ? Actually nearly no one does apart engineers developing it. All you have to know is how certain protocol operates and decide if it fits your needs. There is no need for a diploma to get a good understanding, especially in the days of internet. Bold and incorrect statement

 

[gregorio]

Why do you need to understand low level algorithms and code of such protocols ? Actually nearly no one does apart engineers developing it. All you have to know is how certain protocol operates and decide if it fits your needs. There is no need for a diploma to get a good understanding, especially in the days of internet. Bold and incorrect statement

Apart from the last sentence, you seem to have entirely agreed with me!

So who are the “very best people with the interest and capabilities”? Audiophile designers who don’t even “need a diploma” in network protocols or those most highly qualified scientists and engineers who invented and maintain those protocols and the universities which award the diplomas (and degrees)? As it’s obviously the latter, my statement was neither bold nor incorrect, it’s an obvious simple fact.

G

 

[bfreedma]

Why do you need to understand low level algorithms and code of such protocols ? Actually nearly no one does apart engineers developing it. All you have to know is how certain protocol operates and decide if it fits your needs. There is no need for a diploma to get a good understanding, especially in the days of internet. Bold and incorrect statement

If a simple internet search was all that was needed to understand the 802 protocols, then the entire market for "Ethernet enhancement products" would instantly vanish as everyone would realize they simply can't perform as claimed.

There's a reason that those who understand Ethernet don't purchase these products or support the vendors that produce them. The 802 standards ensure that any "needs" required to ensure data integrity are already accounted for. There is no reason for an end user to go off standard or even be concerned about protocols, regardless of how many vendors with products to sell state otherwise.

Anyone who argues against that doesn't understand the technology they are arguing about.

 

[gimmeheadroom]

Are you using the RME headphone amp? According to this guy, the headphone amp is weak:
The ADI-2 DAC’s headamp has a decidedly low-fi sound.
https://audiophilestyle.com/ca/reviews/rme-adi-2-dac-fs-review-r852/

I have gotten a lot of heat for saying the amp is not suitable for high impedance dynamics, but that's my experience. I think the amp quality is quite good, it drives LCD-2C beautifully and LCD-3 very well. It just doesn't have enough voltage for 300 ohm Sennheisers to bring out what they're capable of. Anyway, I don't use the headphone output on mine. I did at first but it was too annoying having the DAC slide all over the table.

 

[hakunamakaka]

Apart from the last sentence, you seem to have entirely agreed with me!

So who are the “very best people with the interest and capabilities”? Audiophile designers who don’t even “need a diploma” in network protocols or those most highly qualified scientists and engineers who invented and maintain those protocols and the universities which award the diplomas (and degrees)? As it’s obviously the latter, my statement was neither bold nor incorrect, it’s an obvious simple fact.

G

You misunderstand the concept of the engineers who are maintaining these protocols and other side who uses them to transmit data. The ones developing it do not care for all audio related use case scenarios.

If a simple internet search was all that was needed to understand the 802 protocols, then the entire market for "Ethernet enhancement products" would instantly vanish as everyone would realize they simply can't perform as claimed.

There's a reason that those who understand Ethernet don't purchase these products or support the vendors that produce them. The 802 standards ensure that any "needs" required to ensure data integrity are already accounted for. There is no reason for an end user to go off standard or even be concerned about protocols, regardless of how many vendors with products to sell state otherwise.

Anyone who argues against that doesn't understand the technology they are arguing about.

Exactly you do not need such a deep level of understanding in underlying behavior. Neither a degree to get things working. User can go off the standard and it's up to the vendor to decide if it's worth the time and resources to adjust.

I build and maintain storage infra for large enterprises and when big money is involved customer can dictate rules to the vendor even if the build goes off the standard. I know many use cases where vendor fw upgrades were released just to fit large customers non-standard builds.

None of the audiophile companies could push development on the other end to such extent

 

[bfreedma]

You misunderstand the concept of the engineers who are maintaining these protocols and other side who uses them to transmit data. The ones developing it do not care for all audio related use case scenarios.



Exactly you do not need such a deep level of understanding in underlying behavior. Neither a degree to get things working. User can go off the standard and it's up to the vendor to decide if it's worth the time and resources to adjust.

I build and maintain storage infra for large enterprises and when big money is involved customer can dictate rules to the vendor even if the build goes off the standard. I know many use cases where vendor fw upgrades were released just to fit large customers non-standard builds.

None of the audiophile companies could push development on the other end to such extent

I suspect that the people who wrote the custom firmware were actual experts and not people who learned how to do it via 5 minutes of Googling.

The people selling solutions to problems that don't actually exist certainly aren't experts - that's one thing we do agree on.

 

[gregorio]

You misunderstand the concept of the engineers who are maintaining these protocols and other side who uses them to transmit data. The ones developing it do not care for all audio related use case scenarios.

That depends on which protocols. Obviously the developers of say the AES3 protocol do care for audio related use because it’s specifically an audio protocol maintained by the Audio Engineering Society. In the case of say Ethernet then they wouldn’t care about audio related use because there aren’t any audio related issues. There are issues, that Ethernet already deals with, but nothing specific to audio data.

G

 

[hakunamakaka]

I suspect that the people who wrote the custom firmware were actual experts and not people who learned how to do it via 5 minutes of Googling.

The people selling solutions to problems that don't actually exist certainly aren't experts - that's one thing we do agree on.

People writing firmware are good at coding, but they are not experts on how it behaves. An output of someones piece of code feeds to another persons developed functions and is pushed further in large development teams. Software architects are the ones that plans and sets rules on how everything connects and works, but they don't care for the stuff that's not within their goal.

Uber smart ones that can handle everything while developing complex stuff are rarity and exception to the rule.

That depends on which protocols. Obviously the developers of say the AES3 protocol do care for audio related use because it’s specifically an audio protocol maintained by the Audio Engineering Society. In the case of say Ethernet then they wouldn’t care about audio related use because there aren’t any audio related issues. There are issues, that Ethernet already deals with, but nothing specific to audio data.

G

Yes, but how many consumers uses AES3 or I2S ? Majority sits with PC's and for that purpose uses USB which was not created with the thought for audio use thus bringing challenges to the chain. Will it be audible depends on DAC and source, to my surprise it was on my setup and this may relate to poor dac design/noisy usb. I don't have a patience and time to look for root cause, the DDC simply fixed an issue to the degree that I find sound enjoyable. Anyone stacking DDC between clean source and DAC is most likely seeing imaginary improvements, but DDC use case is for troublesome setups.

 

[bfreedma]

People writing firmware are good at coding, but they are not experts on how it behaves. An output of someones piece of code feeds to another persons developed functions and is pushed further in large development teams. Software architects are the ones that plans and sets rules on how everything connects and works, but they don't care for the stuff that's not within their goal.

Uber smart ones that can handle everything while developing complex stuff are rarity and exception to the rule.



Yes, but how many consumers uses AES3 or I2S ? Majority sits with PC's and for that purpose uses USB which was not created with the thought for audio use thus bringing challenges to the chain. Will it be audible depends on DAC and source, to my surprise it was on my setup and this may relate to poor dac design/noisy usb. I don't have a patience and time to look for root cause, the DDC simply fixed an issue to the degree that I find sound enjoyable. Anyone stacking DDC between clean source and DAC is most likely seeing imaginary improvements, but DDC use case is for troublesome setups.

You seem to be just as familiar with Agile/DevOps as you are with Ethernet/802 and USB.

Without a wireframe and detailed technical objectives of current and required behavior, how does the firmware developer know what to code and what Acceptance Testing success criteria needs to look like? Anyone writing firmware (other than very minor narrow changes) for an Ethernet switch will absolutely know the relevant segments of the 802 standards. In a mid size or larger company, there would need to be Change Documentation written and Change Management Process followed - can't write those if you don't understand the requirements and technology in scope.

Hate to break it to you, but as a software architect, not only do I care for how standards are implemented, I need to know enough to confirm the design and it's execution is aligned to the overall goals of the development initiative while still following standards for protocols, APIs, etc. Otherwise, instead of a heterogeneous product, you end up with a bunch of disconnected functionality that can't properly support the required workflow. Can't wait until the end of a 12 month development cycle to find out some crucial bit of code doesn't work...

Bottom line:
If you have a noisy DAC, its broken - get it fixed
If you have a noisy USB, put a $10 powered USB hub in the chain

No need for a multitude of fixes for problems that either don't exist or should be solved at the source.

 

[Andrew_WOT]

If you have a noisy USB, put a $10 powered USB hub in the chain

Can you please explain how adding powered USB hub helps with electrical noise, is it supposed to break ground loop or provide some sort of galvanic isolation?
Thanks

 

[gregorio]

Yes, but how many consumers uses AES3 or I2S ?

Not many but then S/Pdif is just a derivation of AES3 and is used by a lot of consumers.

Majority sits with PC's and for that purpose uses USB which was not created with the thought for audio use thus bringing challenges to the chain.

Obviously that is incorrect. How could the USB specifications include the UAC (Universal Audio Class) sections if there were no “thought for audio use”? There are of course always “challenges” but if a company is designing an audiophile USB DAC and can’t meet those challenges, they’re incompetent. Especially as many cheaper, non-audiophile DACs do meet those challenges. And if that’s the case, it would be the exact opposite of what @audiobomber claimed, that “Audiophile designers, many of whom are engineers, are the very people with the interest and capabilities.”

Will it be audible depends on DAC and source, to my surprise it was on my setup and this may relate to poor dac design/noisy usb

Exactly! A consumer USB DAC is most likely to be connected to a Personal Computer and virtually all PCs have noisy USB outputs. If an audiophile USB DAC designer doesn’t know of (and address) this obvious “challenge” they’re incompetent, especially as many far cheaper USB DACs do meet this challenge.

So, either you have a seriously incompetently designed/faulty DAC or, you have some serious issue with your PC (a fault or an incorrect setting) which puts it well beyond a typically noisy PC.

Anyone stacking DDC between clean source and DAC is most likely seeing imaginary improvements,

Agreed on the “most likely seeing imaginary improvements”, although that should read: “between a clean or typically noisy source”.

but DDC use case is for troublesome setups.

Surely, DDC use case is not for “troublesome setups” but is defined by it’s actual name; DDC (digital to digital converter), to convert from one digital format/protocol to another? That’s certainly why I bought a DDC about 25 years ago.

I don't have a patience and time to look for root cause, the DDC simply fixed an issue to the degree that I find sound enjoyable.

Yes, that seems to be a peculiarly audiophile approach. If you have a faulty DAC, buy another piece of audiophile equipment that’s more expensive than simply replacing the DAC with a non-faulty/competently designed one. Or just change the PC setting if that’s the cause.

G