Review and Measurements of Schiit Lyr Tube Headphone Amplifier

[Palladium]

If that poor kid went on to become an engineer, founded, built, and ran a successful multi-million dollar company based on their engineering which grew in sales every year, but was being judged by that mediocre kid who is neither an engineer nor founded or runs any company but for some reason doesn't feel like an arse judging those who have. Although I suppose we all know that reason ...

Oh, that "movie reviewers are not qualified to review them because they didn't make movies" argument.

Yeah, I also wonder whats up with the the clearly misguided hostility against a company who makes overpriced crappy audio products. I dunno, maybe they can stop it by not making overpriced crappy audio products?

 

[leonroy]

This thread is degenerating. One can respect quality measurements which show gear performing above or below par but the random jibes at Schiit audio undermine the high bar this forum is supposed to set.

 

[mindbomb]

@amirm

Is the bandwidth for the thd+n measurements 20khz?

 

[amirm]

@amirm

Is the bandwidth for the thd+n measurements 20khz?

Almost. It is 22 hz to 22 kHz. It is the default by Audio Precision and I left it as such.

 

[FrivolsListener]

This thread is degenerating. One can respect quality measurements which show gear performing above or below par but the random jibes at Schiit audio undermine the high bar this forum is supposed to set.

Yeah, there's a bit of this going on.

@amirm, I have a hunch: What does your line voltage where you test equipment measure at? Do you have a true RMS meter with reasonable accuracy to measure your mains or filtered voltage?

 

[amirm]

@amirm, I have a hunch: What does your line voltage where you test equipment measure at? Do you have a true RMS meter with reasonable accuracy to measure your mains or filtered voltage?

Did you just ask a Chinese chef if he has a wok??? Of course I have an RMS meter. Last I checked, I have a dozen of them if not more! I also have a differential probe so can use my scope safely to capture the waveform. What good is making any such measurement?

 

[FrivolsListener]

Did you just ask a Chinese chef if he has a wok??? Of course I have an RMS meter. Last I checked, I have a dozen of them if not more! I also have a differential probe so can use my scope safely to capture the waveform. What good is making any such measurement?

I judge a Chinese chef by his results, not by his tools. I've had good stir fry prepared in a very western skillet.

I want to know an accurate measurement of your mains voltage for your DUT. I assume you are near your gear now or will be in the next 24 hours. Or, the next time you test something plugged into mains, please record the mains voltage.

Again, I'm playing a hunch, based on something written on another website when seeing . . . similar behavior with a project in the works.

 

[Dimitri]

based on something written on another website when seeing . . . similar behavior with a project in the works

Sharing a link might be nice.

 

[FrivolsListener]

Sharing a link might be nice.

Being a newcomer, I'm not sure how sharing links to other audio forums is taken here. That said, here you go:

https://www.head-fi.org/threads/sch...bable-start-up.701900/page-2009#post-14104452

https://www.head-fi.org/threads/sch...bable-start-up.701900/page-2072#post-14148020

Now, some context. Jason Stoddard, on his ongoing blog, started a project designed to show how tradeoffs in engineering a "simple project" would look. In that first post, he posts a spectrum plot of a 1 kHz tone, with a noise level that looks like Amir's plots here of Schiit gear, along with a whole bunch of (what did I do wrong?). (For comparison, he also plots that same tone through a Jotunheim. His measurements show noise at levels 120 dB down from the signal.)

The second post, he starts with, "I am an idiot," and says that his problem was that the LM317 regulators were running below their minimum regulatable voltage (oversimplified). He suggests several changes he can make and then the tradeoffs.

Now, I'm a fan of Schiit. Why? Because I've had their stuff and they haven't sounded awful. I have two Magni 3's neither of which hum or emit smoke with any phone. I had a Magni before that (still have it; it was reduced to third-tier service now). I have a Schiit Gungnir Multibit and hear amazing sound from it. I've had some cheaper Schiit products that didn't sound as good (but weren't bad).

In short, what Amir measures, I can't hear. I'm operating on the assumption that neither Amir nor Schiit are stupid nor evil, and so my OCD makes me want to dig into what could be wrong.

In pursuit of this, I picked up a Topping D-30 and had it in my heavy-use critical system to hear how it sounded compared to the Gungnir.

So, after I saw the post about the undervoltage causing the LM317s to drop out of regulation, I wondered what Amir is doing.

 

[junki]

In short, what Amir measures, I can't hear. I'm operating on the assumption that neither Amir nor Schiit are stupid nor evil, and so my OCD makes me want to dig into what could be wrong.

Measurements are mostly diagnostic. 99% of DACs are such that 99% of people can't tell the difference between them, regardless of how they measure. Measuring equipment are orders of magnitude more sensitive than human ears. Headphones and speakers themselves have orders of magnitude more distortion than your DAC or AMP. Old-school R-2R DACs (like your Gumby) has not and still do not measure nearly as cleanly as Delta-sigma DACs. However, many careful listeners who care about both sound and measurement, such as Stereophile, attest they sound more "organic" or "natural".

Translating what is measured to what is audible is something quite difficult. Some things, such as distortion, noise levels, channel imbalance, etc. are audible if they're severe enough. This level of audible severity are orders of magnitude higher than what would look tolerable on measurement devices.

Ultimately the question of what sounds good isn't very much related to measurement. There are subtle differences in the way difference source equipment sounds, and holistic distortion as a sum of measured and unmeasured parts come together to form a unit's sound. Sound distorted in certain ways can very well sound better, wider, deeper, better imaged, whatever-word-you-want-to-use than completely undistorted sound. This is why the entire industry of DSPs exist, as an extreme example.

So who knows. Maybe some equipment combinations add something subtle back into the sound of certain tracks that was lost during recording & mastering, resulting in a holistically more organic/natural sound. Maybe some tracks sound worse and others sound better. It's all just dancing at that point.

But we do know less jitter is better than more jitter. We do know less THD is better than higher THD. We can say "this unit measures better." That means it reproduces the recorded audio format more faithfully. That doesn't mean your brain will interpret the waves coming out of your headphones to be more pleasing.

 

[Wombat]

S

Measurements are mostly diagnostic. 99% of DACs are such that 99% of people can't tell the difference between them, regardless of how they measure. Measuring equipment are orders of magnitude more sensitive than human ears. Headphones and speakers themselves have orders of magnitude more distortion than your DAC or AMP. R-2R DACs (like your Gumby) has and still do not measure nearly as cleanly as Delta-sigma DACs. However, many careful listeners who care about both sound and measurement, such as Stereophile, attest they sound more "organic" or "natural".

Translating what is measured to what is audible is something quite difficult. Some things, such as distortion, noise levels, channel imbalance, etc. are audible if they're severe enough. This level of audible severity are orderds of magnitude higher than what would look tolerable on measurement devices.

Ultimately the question of what sounds good isn't very much related to measurement. There are subtle differences in the way difference source equipment sounds, and holistic distortion as a sum of measured and unmeasured parts come together to form a unit's sound. Sound distorted in certain ways can very well sound better than completely undistorted sound. This is why the entire industry of DSPs exist, as an extreme example.

'Sound distorted in certain ways can very well sound better than completely undistorted sound'. To some individuals. To others it is just different or seen for what it is, distorted.

 

[FrivolsListener]

Measurements are mostly diagnostic. 99% of DACs are such that 99% of people can't tell the difference between them, regardless of how they measure. Measuring equipment are orders of magnitude more sensitive than human ears. Headphones and speakers themselves have orders of magnitude more distortion than your DAC or AMP. Old-school R-2R DACs (like your Gumby) has not and still do not measure nearly as cleanly as Delta-sigma DACs. However, many careful listeners who care about both sound and measurement, such as Stereophile, attest they sound more "organic" or "natural".

Translating what is measured to what is audible is something quite difficult. Some things, such as distortion, noise levels, channel imbalance, etc. are audible if they're severe enough. This level of audible severity are orderds of magnitude higher than what would look tolerable on measurement devices.

Ultimately the question of what sounds good isn't very much related to measurement. There are subtle differences in the way difference source equipment sounds, and holistic distortion as a sum of measured and unmeasured parts come together to form a unit's sound. Sound distorted in certain ways can very well sound better, wider, deeper, better imaged, whatever-word-you-want-to-use than completely undistorted sound. This is why the entire industry of DSPs exist, as an extreme example.

So who knows. Maybe some equipment combinations add something subtle back into the sound of certain tracks that was lost during recording & mastering, resulting in a holistically more organic/natural sound. Maybe some tracks sound worse and others sound better. It's all just dancing at that point.

But we do know less jitter is better than more jitter. We do know less THD is better than higher THD. We can say "this unit measures better." That doesn't mean it sounds better.

Yes. I understand that. But my experiences with Gungnir were revealing: What I thought was a good recording, showed no better. My initial reaction was, "Well, I can't hear it." It showed itself in a couple unexpected recordings, and I had a, "Where did THAT come from?" moment. Occasionally, I still do.

And, yes, I know many people can't tell a cell-phone DAC from an expensive piece of equipment. My login name was chosen with the acknowledgement that my ears aren't what they used to be, and so I couldn't call myself a serious listener anymore.

 

[amirm]

Being a newcomer, I'm not sure how sharing links to other audio forums is taken here. That said, here you go:

https://www.head-fi.org/threads/sch...bable-start-up.701900/page-2009#post-14104452

https://www.head-fi.org/threads/sch...bable-start-up.701900/page-2072#post-14148020

Hmmm. I still don't know what you are trying to troubleshoot. Measuring wall voltage is not going to inform you about any issues without instrumentation.

 

[FrivolsListener]

Hmmm. I still don't know what you are trying to troubleshoot. Measuring wall voltage is not going to inform you about any issues without instrumentation.

Jason found his problem with a Variac. I wonder if your readings -- which are far worse than what others find -- could be due to undervoltage. Measuring the mains voltage might show a low voltage, something which linear power supplies might have trouble with, but a switching supply would not. Likewise, a Variac might suddenly cause your hum, noise and other issues with Schiit gear disappear.

 

[amirm]

Jason found his problem with a Variac. I wonder if your readings -- which are far worse than what others find -- could be due to undervoltage. Measuring the mains voltage might show a low voltage, something which linear power supplies might have trouble with, but a switching supply would not. Likewise, a Variac might suddenly cause your hum, noise and other issues with Schiit gear disappear.

I am pretty sure the "buzz" he is talking about is mechanical noise from the wall-wart transformer in his situation, not audio noise. Also, the Lyr in this thread has built-in power supply, not external unit so this would not apply to it at all.

 

[FrivolsListener]

I am pretty sure the "buzz" he is talking about is mechanical noise from the wall-wart transformer in his situation, not audio noise. Also, the Lyr in this thread has built-in power supply, not external unit so this would not apply to it at all.

No, I was talking about the noise you measure in Schiit gear. Which looked suspiciously like what Jason measured in the misbehaving, unregulated circuit.

Doesn't that look familiar?

Compare with (in the same post), his measurement of a 1 kHz tone in Jotunheim:

...which reminds me less of your measurements and more of Atomicbob's ones.

So...is your voltage too low, or at least out of spec compared to what Schiit specifies?

This is not a question in a vacuum: Japanese mains is 100 V, not 120 (+/- 10%). Many modern electronics wouldn't care. Stuff with linear power supplies -- like all Schiit gear -- would.

 

[amirm]

So...is your voltage too low, or at least out of spec compared to what Schiit specifies?

I am not home so can't measure until late tomorrow. And what does Schiit specify?

 

[March Audio]

No, I was talking about the noise you measure in Schiit gear. Which looked suspiciously like what Jason measured in the misbehaving, unregulated circuit.

Doesn't that look familiar?

Compare with (in the same post), his measurement of a 1 kHz tone in Jotunheim:

...which reminds me less of your measurements and more of Atomicbob's ones.

So...is your voltage too low, or at least out of spec compared to what Schiit specifies?

This is not a question in a vacuum: Japanese mains is 100 V, not 120 (+/- 10%). Many modern electronics wouldn't care. Stuff with linear power supplies -- like all Schiit gear -- would.

Considering the amount of equipment Amir has measured that do not show any issues I think it would be unlikely that low mains voltage is the issue. Also you would really need to measure the output of the regulator to ascertain at what point it drops out.

A design should really have plenty of latitude for supply variation. Yes that may increase dissapation and heat in a linear regulator but it's all really very, very basic part of psu design. So if it is the case, then it's just another example of Schitt design

 

[Blumlein 88]

No, I was talking about the noise you measure in Schiit gear. Which looked suspiciously like what Jason measured in the misbehaving, unregulated circuit.

Doesn't that look familiar?

Compare with (in the same post), his measurement of a 1 kHz tone in Jotunheim:

...which reminds me less of your measurements and more of Atomicbob's ones.

So...is your voltage too low, or at least out of spec compared to what Schiit specifies?

This is not a question in a vacuum: Japanese mains is 100 V, not 120 (+/- 10%). Many modern electronics wouldn't care. Stuff with linear power supplies -- like all Schiit gear -- would.

Look, Schiit has been given plenty of chances and plenty of rope on issues with their gear. Quite simply if it measures like Schiit, then it is Schiit. We don't need to make special accommodations for Schiit to end up giving even then rather middling performance results. Which in some cases they still can't.

What should be noticeable is:

Even poor gear may not be obviously bad upon listening.

Carefully building a rep for quality and building expectations for something special can overcome some very poor performing gear. People still swear by it, jump in to immediately recommend it to other people and are oblivious to the fact even at lower prices the gear they are getting isn't worth it.

And this after about 3 public sources have gotten similar very poor measured results, and still some wish to go the extra mile and give some extra benefit of doubt. Just drop that Schiit and get with the program. Schiit is as Schiit does. By now they have earned careful audiophiles looking at them with a jaundiced eye. It is entirely appropriate when considering any Schiit purchase to assume the stuff is substandard until very well proven otherwise. And this must be by third parties. You can't just let them make this Schiit up.

 

[FrivolsListener]

I am not home so can't measure until late tomorrow. And what does Schiit specify?

According to the web site, 115 or 230 vac.