$500 amplifiers to test and review?

[sonci]

https://www.stereophile.com/content/sonic-impact-model-ta2024-super-t-power-amplifier-measurements
J.A measured the Sonic super T, of course didn't do that good.

 

[andreasmaaan]

https://www.stereophile.com/content/sonic-impact-model-ta2024-super-t-power-amplifier-measurements
J.A measured the Sonic super T, of course didn't do that good.

LOL, I remember the hype and actually had the Dayton knock-off of one of these at one point.

 

[restorer-john]

Many years ago, magazine reviews did make an attempt at testing amps into a simulated complex load, albeit a 2uF in parallel with 8 ohms to simulate an electrostatic loudspeaker, but that doesn't seem to happen these days.

Correct. Most competent reviewers presented oscillograms of square waves at a nominal level into fully resistive and capacitive loads at low, mid and high frequencies. Full tone response and filter plots were the norm, as were power bandwidth tests etc.

My favourite test (one that wasn't done in the past) is a full rated power frequency response sweep. It differs greatly from the usual 1w@8ohms FR and is not as coarse as rated -3dB points for power bandwidth.

 

[restorer-john]

Personally, I can't see the point testing any amplifiers of the switching variety that exhibit high levels of spurious out of band signals. They do absolutely nothing to advance the pursuit of high fidelity and are a challenge to measure in any case. Special 'consideration' must be made for them in terms of filters for test gear. Hardly a level playing field.

An amplifier should be silent and not be an RF transmitter.

 

[andreasmaaan]

Personally, I can't see the point testing any amplifiers of the switching variety that exhibit high levels of spurious out of band signals. They do absolutely nothing to advance the pursuit of high fidelity and are a challenge to measure in any case. Special 'consideration' must be made for them in terms of filters for test gear. Hardly a level playing field.

An amplifier should be silent and not be an RF transmitter.

What impact on fidelity do you believe out of band signals have?

 

[restorer-john]

What impact on fidelity do you believe out of band signals have?

It's really simple. Waveform fidelity is compromised. Show me any PWM/PDM amplifier that can manage a decent 10KHz square wave, at a reasonable voltage swing. Hint: they can't.

A correctly engineered amplifier simply increases the amplitude of the waveform and adds nothing else. Residual noise should be as low as possible. Excess, non-harmonically related spuriae are also unacceptable.

If you are going wideband, there should be no spurious signals, unrelated to the input in that band. IMD products will appear in the audible bandwidth folded down from out of band HF tones (switching frequencies etc). If you are bandwidth limiting (as in the case of PWM/PDM amps), out of band spuriae are simply a byproduct of poor filter design, laziness and general cheapness.

So-called 'digital' amplifiers have infested every part of consumerdom because they are cheap and cheerful. They have improved in the last 20 years, but they are not remotely close in absolute performance, to carefully engineered analogue amplification. If you want an instrumentation-grade amplifier with a ruler flat response, vanishing levels of distortion and noise, and freedom from frequency response anomalies into varying loads, that amplifier will not be a PWM/PDM unit.

 

[andreasmaaan]

It's really simple. Waveform fidelity is compromised. Show me any PWM/PDM amplifier that can manage a decent 10KHz square wave, at a reasonable voltage swing. Hint: they can't.

Ok, but by this standard, we can rule out literally any DS DAC or other DAC that uses an imaging filter from being good enough to be worth measuring. That would seem absurd to me.

A correctly engineered amplifier simply increases the amplitude of the waveform and adds nothing else. Residual noise should be as low as possible. Excess, non-harmonically related spuriae are also unacceptable.

Well no amplifier does this. The point of measurement is to determine how far an amplifier departs from this ideal.

If you are going wideband, there should be no spurious signals, unrelated to the input in that band. IMD products will appear in the audible bandwidth folded down from out of band HF tones (switching frequencies etc). If you are bandwidth limiting (as in the case of PWM/PDM amps), out of band spuriae are simply a byproduct of poor filter design, laziness and general cheapness.

And the measurements will show these and readers can interpret their audibility or offensiveness or otherwise, as is the case with measurements of any other device.

So-called 'digital' amplifiers have infested every part of consumerdom because they are cheap and cheerful. They have improved in the last 20 years, but they are not remotely close in absolute performance, to carefully engineered analogue amplification. If you want an instrumentation-grade amplifier with a ruler flat response, vanishing levels of distortion and noise, and freedom from frequency response anomalies into varying loads, that amplifier will not be a PWM/PDM unit.

IMHO unrealistically lofty standards for a thread entitled "$500 amplifiers to test and review". In this price range, I'd hazard to guess that many of the best-measuring devices would in fact be class D.

 

[Xulonn]

Perhaps I need two amplifiers - one to please my ears and another one to please my oscilloscope?

And what the hell is a "cheerful" amplifier?

Anyway, your comments showing your horror at imperfect waveforms prompted me to start a thread in the General Forum so as to not overly pollute this "amplifier suggestions" thread

 

[restorer-john]

Ok, but by this standard, we can rule out literally any DS DAC or other DAC that uses an imaging filter from being good enough to be worth measuring. That would seem absurd to me.

Not true. Most competently designed D/As, whether they be DS or Multibit have little or no out of band spuriae of significance. The key word is significance. Amplifiers with hundreds of millivolts or RF out of band are poorly designed. Full stop.

Well no amplifier does this. The point of measurement is to determine how far an amplifier departs from this ideal.

How many PWM amplifiers have you measured vs conventional analog amplifiers? Consider conventional distortion meters have to have outboard filters as do digital A/D based analysers as there is simply so much out of band crap to deal with. Again, simply faulty design.

And the measurements will show these and readers can interpret their audibility or offensiveness

No, measurements are measurements. No direct interpretation on 'offensiveness' can be drawn from measurements. They serve to analyse the objective performance and the 'offensiveness' is purely subjective and may vary from person to person- hence a completely useless metric.

IMHO unrealistically lofty standards for a thread entitled "$500 amplifiers to test and review". In this price range, I'd hazard to guess that many of the best-measuring devices would in fact be class D.

I don't believe so at all. Having spent a considerable period of my life selling audio equipmnet, prior to the widespread adoption of PWM/PDM amplifiers, and the rest of my life repairing, testing, restoring, and rebuilding gear, I'd hazard a guess completely opposite. I've had many ClassT/PDM/PWM 'digital' amplifiers either across my bench or to play with, and they are all ultimately flawed in one way or another. Yet on the other hand, I have literally hundreds of amplifiers (yes) that are essentially without flaws and they are all solid state Class AB amplifiers.

I don't want an interference producing product in my audio system and there are plenty of conventional amplifiers that can be purchased for under $500 with none of those flaws.

 

[restorer-john]

And what the hell is a "cheerful" amplifier?

Cheap and cheerful. It is an English saying. Basically means inexpensive and fun to play with but shouldn't be taken seriously or considered state of the art in any way.

Music is a collection of waveforms. Nothing more and nothing less.

 

[andreasmaaan]

Not true. Most competently designed D/As, whether they be DS or Multibit have little or no out of band spuriae of significance. The key word is significance. Amplifiers with hundreds of millivolts or RF out of band are poorly designed. Full stop.

These DACs rely on filters to remove these out of band spuriae, and these filters prevent them from accurately reproducing a 10Khz square wave. Your post #146 stated that any device incapable of reproducing a "decent 10KHz square wave" should not be considered for measurement.

Consider conventional distortion meters have to have outboard filters as do digital A/D based analysers as there is simply so much out of band crap to deal with. Again, simply faulty design.

What evidence do you have that this "out of band crap" is audible?

No, measurements are measurements. No direct interpretation on 'offensiveness' can be drawn from measurements. They serve to analyse the objective performance and the 'offensiveness' is purely subjective and may vary from person to person- hence a completely useless metric.

Interpretations (not absolute conclusions) about audibility and offensiveness can certainly be drawn from measurements, and considerable research has gone into determining audibility thresholds, as well as effects on listener preference, of various kinds of distortion and other spuriae.

I don't believe so at all. Having spent a considerable period of my life selling audio equipmnet, prior to the widespread adoption of PWM/PDM amplifiers, and the rest of my life repairing, testing, restoring, and rebuilding gear, I'd hazard a guess completely opposite. I've had many ClassT/PDM/PWM 'digital' amplifiers either across my bench or to play with, and they are all ultimately flawed in one way or another. Yet on the other hand, I have literally hundreds of amplifiers (yes) that are essentially without flaws and they are all solid state Class AB amplifiers.

Which class D and class AB amps in the $500 price range have you measured? Do you have those measurements to post, out of interest?

 

[Xulonn]

Cheap and cheerful. It is an English saying. Basically means inexpensive and fun to play with but shouldn't be taken seriously or considered state of the art in any way.

I like that saying - I have had a fair number of such components over the years. I have also owned a good number of competently engineered mid-level gear. But I've only had a few components which would be considered SOTA - and certainly nothing that would be accepted by hard-core audiophiliacs as high-end state of the art.

Music is a collection of waveforms. Nothing more and nothing less.

Well, I willl certainly have to disagree with you on that - music can be a subjective source of great human pleasure - if you look beyond the physics of it.

 

[andreasmaaan]

Well, I willl certainly have to disagree with you on that - music can be a subjective source of great human pleasure - if you look beyond the physics of it.

What if it were rephrased to: "Recorded music is a collection of waveforms. Nothing more and nothing less"?

 

[DonH56]

Class D has come a long way since the early days and many modern class D designs sound very good IMO. Higher-frequency switching and advances in circuit design have solved many issues to provide better sound and lower output impedance over the audio band. That said, ultrasonic noise in the MHz range is still a problem, so you are heating up your tweeters and placing a large, high-powered EMI/RFI noise source in your listening room where RF energy can be coupled into the rest of your components. I'd love to try one of the new ATI or other class D amplifiers, but chances are my next amp will be a Benchmark or Bryston more conventional amplifier.

FWIWFM - Don

p.s. I put my technical analysis side to sleep and just enjoy the music/movie in the media room. Similarly I have to focus on the music and not how my lips vibrate when playing a song on my trumpet. We're on a technical forum so my little pea brain automatically goes there when I post here. Does not mean my entire world is waveforms and measurements and I strongly suspect the same is true of John. I think berating someone for participating technically and bringing up technical points in a technical forum is distasteful and misguided. Whatever...

 

[andreasmaaan]

I hope it's clear I wasn't attempting to berate anyone! I'm sorry @restorer-john that it appears to have come across that way. You clearly have greater technical expertise with electronics than I do (which is just enough to use them to design active speakers).

I do share Don's opinion of modern class D amps though, based both on measurements and on listening; my aim was simply to argue that they're a type of amp that's likely to be competitive at the $500 price point, and thus worthy of being considered for testing.

 

[Xulonn]

What if it were rephrased to: "Recorded music is a collection of waveforms. Nothing more and nothing less"?

That works, but even better, music recordings are collections of waveforms.

When I use the word "music" I am not talking simply about waveform. However I'm veering into being pedantic, and will let this go.

 

[Thomas savage]

Can we get back to amp suggestions please , this thread has a very specific purpose.

 

[FrantzM]

Hi

I have had my share of expensive to very expensive audio components. I am at a point where I deeply feel wronged when I pay a lot (subjective) for something that can be had for a lot less (again, subjective)... It is clear to me , now that $10,000 amplifiers are an affectation. OTOH how much should one spend for an amp? What are the threshold of measurements at which amps differences are inaudible? For the record, I don't believe that all amps sound the same ... I need to conduct some blind experiments that would invalidate my beliefs.
I am interested in Class D because of their substantial efficiency (85% is not uncommon). Wasting watt is not an option for me since I am essentially Off grid, Solar with back-up generator. Moreover it seems that building an amp these days can be as simple as a case, a Power supply (SMPS likely) and ICEPower, Hypex or Ncore amplifier modules... even if one doesn't go the DIY route, $500 looks like a good price threshold for a 2-channel amp with at least 2 x 50 watts @ 8 ohms... low THD and IM and >100 dB SNR...
Following this thread with great interest....

 

[tinger]

Just ordered the Arcam rHead from the UK. $230ish delivered.
I'll send it to Amir when I get it.

 

[Xulonn]

I am interested in Class D because of their substantial efficiency (85% is not uncommon). Wasting watt is not an option for me since I am essentially Off grid, Solar with back-up generator. Moreover it seems that building an amp these days can be as simple as a case, a Power supply (SMPS likely) and ICEPower, Hypex or Ncore amplifier modules...

Which is why I suggested the Audiophonics Hypex amp for testing. Sold in France, with shipping, the 75wpc @ 8Ω power amp should cost less than $400 delivered in the U.S. , so it fits Amir's criteria. Audiophonics appears to have an excellent range of components, kits, and finished products, and like Parts Express in the U.S., the bottom of that range is uper el-cheapo class-D T-amps and it goes up from there. Although I have not compared their entire ranges of product offerings, Audiophonics seems to go higher in terms of more expensive and "properly engineered and designed" electronics. The Audiophonics Hypex amps are deep and narrow - just enough space for the modules with a few slots for cooling. Plus, the 5-12v trigger input will appeal to some who like a bit of "automation."

The Hypex modules have a excellent reputation, and are used by a number of mainstream high-fidelity audio amplifier manufacturers. Since these companies used to design their circuits from scratch , I really don't understand what value some of them can add to their amps to justify high prices, since the amplification - and even the power supply on some - are now completely self-contained in the Hypex modules.

I am not only interested in what Amir will come up with for a suite of tests, but also how our resident "hands-on" experts and "technical purity" forum members will respond. Like you, although I am aware that more traditional Class A and AB amps can produce truly excellent sound and specs, I realize that they are energy hogs, especially Class-A. I seems logical that Class-D designers and engineers are continuing to work to improve performance, electrical efficiency, reduce unwanted artifacts, while keeping "fidelity" high.