Impressions: SMSL PA200 GAN FET Class-D Power Amp

[Julf]

I didn't get that just a passionate detailed subjective impression. Regardless the post was over the top attacking choice of speakers etc like they were offended or something.

Yes, and passionate subjective impressions, while welcome on "audiophile" sites, are not really appropriate here.

 

[Scrivs]

I didn't get that just a passionate detailed subjective impression. Regardless the post was over the top attacking choice of speakers etc like they were offended or something.

"passionate detailed subjective impressions" are exactly the way those without the requisite technical knowledge end up making uninformed purchase decisions over and over, ad infinitum.

Isn't that how the OP arrived at his ultimate 1mHz wonder?

 

[Evo42]

Yes, and passionate subjective impressions, while welcome on "audiophile" sites, are not really appropriate here.

Isn't that the point of the impressions thread, if it was just talking about the build quality and inputs that would be pretty boring. There are places here such as the review/measurements threads to stick to the fully objective analysis.

 

[Mort]

just talking about the build quality and inputs that would be pretty boring.

To many of us, personal impressions are the part that's boring. Why would I care about one person's subjective audio stories? I can get that all over the Internet.

 

[Evo42]

To many of us, personal impressions are the part that's boring. Why would I care about one person's subjective audio stories? I can get that all over the Internet.

It was a small part of the overall impressions and they were aware of the audience. It's interesting as no one else got their hands on one yet.

 

[Roland68]

Why such an argumentative, condescending and aggressive rant? Some interesting points but your tone and attitude to someone's initial impression post is poor. What did they do to you to get you so riled up?

EddNog seems to have really hit a sore spot with @Artcore87 .

Why get the NC252MP with 150wpc into 8 ohms when for 695 from Buckeye you can get the ncore amp that has 350w into 8 ohms (and both obviously do more into 4)?

All your statements about switching frequency and analog sound are nonsense.

Also if the B100 wasn't enough, you could just get the B200, tons of power, I guarantee more than these can put out continuously for any length of time. Multiple hypex options with way more power too. nc500x for 995 with 380wpc/700wpc.

If the 90db sinad someone mentioned is accurate (and I don't know if it is I haven't seen it myself) then this smsl amp is FAR inferior to hypex, topping, or purifi. like not even close.

You say you didn't want to try the purifi amp one user suggested to you because "if it sounds the same as the hypex you're not interested". The reason he's saying that is because the hypex and purifi amps, at least in the scientific terms valued by the mindset of this forum, are both so incredibly good that they are essentially technically perfect amplifiers, beyond the threshold of human audibility. The new eigentacht 2 purifi also has more power.

The smsl amp looks cool but I need to see detailed testing of it.

You also could have gotten a Tripath TA3020 based amp that can do ~200w into 8 ohms and 400 into 4. If you value subjective impressions the Tripath amps were often described as sounding almost tube like (very analog)... but they don't exhibit elevated even order harmonics and have extremely low distortion so who knows what that meant lol. I can say my own TA3020 sounds incredible. Transparent, low noise, very "fast" with all the micro-details and micro-dynamics you could want =P The Tripath solution is DIY (buy completed amp board, completed PSU board, but no case and wires that's on you) but you can get them for say 230-300 bucks a pop depending on configuration (that's for a board AND smps combined, or the option of the board+smps in one). You can also run them in a bridged configuration and get way more power if you wanted. they make great killowatt+ level subwoofer amps for this reason, except they're only rated to 4 ohms in that configuration, and I'm guessing THD is worse in this configuration as is usually the case, probably also the case with your SMSL amps.

I was hopeful about these SMSL amps, but if the 90db sinad measurement is indeed accurate, they're out. Absolutely NOTHING beats hypex, purifi and topping b/la series amps for performance, value, and power. Nothing. Period. Well if you're ok with semi-DIY Tripath is still great. We're only talking 2 zeroes in THD instead of 3 zeroes, but I don't think you're gonna hear that... but they're not necessarily load independent. Nor is the noise floor as vanishingly low, although it's still sufficiently low. But IMO, AFAIK, you can't beat ~$230 dollars for 200w 8 ohms 400w 4 ohms at .1% thd from the ta3020 v4d from connexelectronic where value is concerned.

The most important point though, and your most egregious mistake, is using the LS50 Meta speakers for your use case, i.e. cranking them, blasting them with the doors open, thinking you need multiple hundreds of watts for them. Your amp isn't the problem, your speakers are. Those are TINY PATHETIC ANEMIC speakers and very inefficient, and you're trying to make them do something they were not made to do. As soon as you start pushing the excursion of those TINY drivers their distortion shoots way up, because the woofer cone IS the waveguide for the tweeter... which is why KEF themselves are smart enough to cross them over to dedicated woofers in their real/serious speakers, you know... the larger ones. Have you not seen erins audio corner's klippel testing at various SPL levels, the distortion, multi-tone distortion, and compression graphs? WHY?! You need bigger speakers, more sensitive speakers. Those things aren't staying clean much higher than 86db, they're breaking 3% at what, the next test is 91 I think? Dude come on. Size matters. Those speakers are so overrated it's not even funny. The KEF coaxes have their strengths, I won't deny that at all, great for nearfield, low level listening, off-axis listening where you don't want a narrow sweet spot, fine. But you need to high pass those and cross them at like 100hz which is garbage, you don't want a sub playing up that high there's so many downsides to that setup. Get some real speakers that you can cross a sub to lower, and that are easily another 3db or 6db more sensitive and require half or a quarter of the power for the same volume, and have half the distortion at high levels. Derrrr I need 300 watts because my inefficient 5.25" speakers don't get loud enough... seriously what?! KEF makes great speakers that are a bad value if you spend enough money (blade 2 metas anybody?), but their itty bitty entry level stuff is just a bad value and there's so many better options unless you absolutely MUST have a coax for some reason... even then I think there are better options. At least you didn't get the q150, one of the lamest speakers to ever exist only suitable for surround channels. The LS50 is a nice speaker it just isn't made for what you want, it has EXTREME limitations, VERY early rolloff, no bass and no real output capability, not if you really want to crank it like I do, and like it sounds like you do... it's the wrong pick. It's about 4-16 times too small, in terms of driver surface area or cabinet volume, take your pick.

I absolutely agree with you about the LS50 (Meta), but that applies to many compact speakers, especially when operated without a subwoofer and/or high-pass filter.
These speakers are very often played too loudly and overloaded by their owners. This is also evident in the pictures of damaged and destroyed LS 50 (Meta) speakers, where you can clearly see how the cones have been bent and then torn apart due to constant overloading. And these aren't just 10-20% overloads, but massive overloads without any sense or reason.

But that fundamentally has nothing to do with these amplifiers.
I'm also one of those people who can't do anything with Hypex/Purifi amplifiers. I've heard more than 10 of the high-quality models, but I don't get the "I want one" feeling.
Last year, two friends came to see me and once again wanted to convince me of their (expensive) Hypex/Purifi amplifiers. What was the outcome? Both of them left with their heads hanging low and ordered the Sabaj A30a. Both then spent several weeks comparing and ultimately kept the A30a. I have to say that neither of them has a problem buying an amplifier for €50,000 if they think it would be better.

But that's exactly the point. Of course, I pay attention to measurements first, and they should be good. But the final decision is always made after I listen to a device.
And you can't discuss that.

 

[goryu]

EddNog seems to have really hit a sore spot with @Artcore87 .




But that's exactly the point. Of course, I pay attention to measurements first, and they should be good. But the final decision is always made after I listen to a device.

Which just proves that some people prefer distortion to accuracy, place more import on factors other than objective performance, and/or are letting their biases make their decisions.

 

[Deleted member 72219]

EddNog seems to have really hit a sore spot with @Artcore87 .



I absolutely agree with you about the LS50 (Meta), but that applies to many compact speakers, especially when operated without a subwoofer and/or high-pass filter.
These speakers are very often played too loudly and overloaded by their owners. This is also evident in the pictures of damaged and destroyed LS 50 (Meta) speakers, where you can clearly see how the cones have been bent and then torn apart due to constant overloading. And these aren't just 10-20% overloads, but massive overloads without any sense or reason.

But that fundamentally has nothing to do with these amplifiers.
I'm also one of those people who can't do anything with Hypex/Purifi amplifiers. I've heard more than 10 of the high-quality models, but I don't get the "I want one" feeling.
Last year, two friends came to see me and once again wanted to convince me of their (expensive) Hypex/Purifi amplifiers. What was the outcome? Both of them left with their heads hanging low and ordered the Sabaj A30a. Both then spent several weeks comparing and ultimately kept the A30a. I have to say that neither of them has a problem buying an amplifier for €50,000 if they think it would be better.

But that's exactly the point. Of course, I pay attention to measurements first, and they should be good. But the final decision is always made after I listen to a device.
And you can't discuss that.

Well yeah, I mean...if I had the budget, I'd just get the Benchmark and call it a day!

-Ed

 

[Sokel]

If the 90db sinad someone mentioned is accurate (and I don't know if it is I haven't seen it myself) then this smsl amp is FAR inferior to hypex, topping, or purifi. like not even close.

The "someone" me only quote SMSL's own specs, so...



...and that's almost the sweet spot it's much worst going up or down:



No clean SINAD chart though at 4 Ohm which usually can get a penalty.

(I wouldn't worry about Edd, next month will have other amps )

 

[Roland68]

Which just proves that some people prefer distortion to accuracy, place more import on factors other than objective performance, and/or are letting their biases make their decisions.

That's your opinion, I have a different one. No problem for me.

Just a question: if device 1 has better measurements, would you prefer it to device 2, even if you like device 2 better in your setup?

Prejudices are fundamentally excluded in our comparisons. We started with blind tests 25 years ago as part of a project and have perfected them to such an extent that we have been advising manufacturers in this field for some time now. All devices have the same requirements; any physical influences are reduced to a minimum (but the same for all devices), the volume is meticulously measured and controlled both by voltage and sound measurements at four frequencies.

Everyone can see whether device 1, device 2, or device X is active, but there's no way to determine which device is currently running; it's only displayed at the end of the run. With each new run, the entire system is randomly reassigned.
This means that any prejudices, preferences and other preferences for a device are irrelevant.

 

[Mort]

I would love to hear more about the benefits of GAN FET. Is that benefit measurable?
Are we potentially in opamp territory?

I ask that earnestly. I'm not real strong on audio engineering.

 

[Sokel]

I would love to hear more about the benefits of GAN FET. Is that benefit measurable?
Are we potentially in opamp territory?

I ask that earnestly. I'm not real strong on audio engineering.

Here you go:

Is GaN worth it? - ICEpower

icepoweraudio.com

I believe they sum things nicely.

 

[Roland68]

I would love to hear more about the benefits of GAN FET. Is that benefit measurable?
Are we potentially in opamp territory?

I ask that earnestly. I'm not real strong on audio engineering.

The biggest advantage is active components, such as transistors and other semiconductors, with lower heat generation, smaller space requirements, higher switching power, faster switching times, and greater energy efficiency.

Whether this has an advantage in typical measured values is questionable; so far, I haven't seen any major differences that could be attributed to it.
But in terms of heat generation and reduced size, it is an advantage.

Otherwise, just read up on it; it's not such a new topic: gallium nitride (GaN).

 

[goryu]

That's your opinion, I have a different one. No problem for me.

Just a question: if device 1 has better measurements, would you prefer it to device 2, even if you like device 2 better in your setup?

Prejudices are fundamentally excluded in our comparisons. We started with blind tests 25 years ago as part of a project and have perfected them to such an extent that we have been advising manufacturers in this field for some time now. All devices have the same requirements; any physical influences are reduced to a minimum (but the same for all devices), the volume is meticulously measured and controlled both by voltage and sound measurements at four frequencies.

Everyone can see whether device 1, device 2, or device X is active, but there's no way to determine which device is currently running; it's only displayed at the end of the run. With each new run, the entire system is randomly reassigned.
This means that any prejudices, preferences and other preferences for a device are irrelevant.

If two amps are listened to in a properly designed, set up, and executed experiment, if they both amplify the input signal without audible levels of distortion throughout the audible frequency spectrum, have a flat frequency response through the audible spectrum and have a response independent of load, they will not have any audible differences. Any preference for one over the other would be subjective and based on non audible factors. If you disagree, please share the possible audible factors (audible differences) that would contribute in such a scenario to a subjective preference.

If one or both of two amps are listened to in a properly designed, set up, and executed experiment, if one or both amplify the input signal with audible levels of distortion at some point in the audible spectrum, and/or have a varying frequency response -load dependent response or otherwise, there may indeed be audible differences. Whether or not one or the other is preferred would be a subjective response based on audible factors that may or not be shared by others.

I prefer amps that live up to the definition of an amplifier: they take a small signal and make it larger without adding or subtracting anything. They are a blank canvas to which I can add distortion to, and/or design a frequency response that I prefer. When I order a steak, I prefer the option to add my own salt and pepper to my own taste rather than have someone else season it for me.

 

[Julf]

Just a question: if device 1 has better measurements, would you prefer it to device 2, even if you like device 2 better in your setup?

As you wrote yourself, "This means that any prejudices, preferences and other preferences for a device are irrelevant."

If both devices are acoustically accurate/transparent (as most modern, properly designed amps are), the "preference" is just random and irrelevant.

 

[RoA]

If two amps are listened to in a properly designed, set up, and executed experiment, if they both amplify the input signal without audible levels of distortion throughout the audible frequency spectrum, have a flat frequency response through the audible spectrum and have a response independent of load, they will not have any audible differences. Any preference for one over the other would be subjective and based on non audible factors. If you disagree, please share the possible audible factors (audible differences) that would contribute in such a scenario to a subjective preference.

If one or both of two amps are listened to in a properly designed, set up, and executed experiment, if one or both amplify the input signal with audible levels of distortion at some point in the audible spectrum, and/or have a varying frequency response -load dependent response or otherwise, there may indeed be audible differences. Whether or not one or the other is preferred would be a subjective response based on audible factors that may or not be shared by others.

I prefer amps that live up to the definition of an amplifier: they take a small signal and make it larger without adding or subtracting anything. They are a blank canvas to which I can add distortion to, and/or design a frequency response that I prefer. When I order a steak, I prefer the option to add my own salt and pepper to my own taste rather than have someone else season it for me.

Surely slew rate/rise time can influence how an amplifier is perceived. This is (probably) what to some makes these cheap chinese chip amps sound bad. The money/scope for good, adequate components is simply not there.

 

[Julf]

Surely slew rate can influence how an amplifier is perceived

Only if it is so low that it can't reproduce 20 kHz at full power/amplitude. The amps that can't are very rare these days.

 

[Roland68]

As you wrote yourself, "This means that any prejudices, preferences and other preferences for a device are irrelevant."

You're completely taking my statement out of context. This statement only referred to our blind test scenario, in which these prejudices, predilections, and other preferences for a device are completely ruled out because no one knows which device they're listening to.

If both devices are acoustically accurate/transparent (as most modern, properly designed amps are), the "preference" is just random and irrelevant.

Nevertheless, with different devices that, for example, according to the ASR test, should be very similar (which they actually are), we have very clear results of 8/10 to 10/10 for one of two or three devices. This also applies to repeat tests.
That doesn't have to please anyone, and it doesn't change the result. Even the most extreme skeptic has to notice that in our tests.

But to be clear, I'm talking about small differences, sometimes just nuances, not these phrase-laden statements about differences like worlds apart, completely different, or something like that.
And these differences have become significantly smaller, especially in the last 5-10 years, which is a good thing. We checked this again last year with older devices.

 

[Julf]

You're completely taking my statement out of context. This statement only referred to our blind test scenario, in which these prejudices, predilections, and other preferences for a device are completely ruled out because no one knows which device they're listening to.

But a proper (double) blind test is the only way to find out if you hear a difference or not. Without it, your preference might be (and probably are) due to factors not related to the sound.

Nevertheless, with different devices that, for example, according to the ASR test, should be very similar (which they actually are), we have very clear results of 8/10 to 10/10 for one of two or three devices. This also applies to repeat tests.

I would love to fully review your test setup and protocol.

 

[Roland68]

But a proper (double) blind test is the only way to find out if you hear a difference or not.

Have you read post #130? We've been doing that for a long time, but we've done a lot more.

Without it, your preference might be (and probably are) due to factors not related to the sound.

How would that work if the devices aren't identifiable? What kind of preference for factors other than sound are you talking about?

I would love to fully review your test setup and protocol.

That's no problem at all. A few personal conversations, documentation, notarized contracts including a binding non-disclosure agreement with a security deposit, and a six-figure investment (€ and $ are accepted) in the project.