Benchmark AHB2 design

[mdunjic]

Please link me. Before you do, carefully read what I stated: "significant level of switching noise at higher frequencies"

"higher frequencies" = ? ... maybe somewhat vague imo

One example of excellent review => https://www.audiosciencereview.com/...-3-channel-purifi-amplifier-review-2nd.43834/

Search results for query: class D

www.audiosciencereview.com

 

[pablolie]

This is also incorrect. Class D amps suffer significant level of switching noise at higher frequencies, where as the AHB2 does not. That is the reason why John Siau has his disdain for class D.

Ncores and similar "suffer" from nothing. They measure just as well. See this (and it wouldn't look like this if Class D had such a noise issue, would it?)

Did you ever consider that someone affiliated with a company might be a tad biased towards their own design, both for engineering pride and commercial differentiation reasons?

 

[mdunjic]

The THX AAA patent was filed back in 2017. Which amp uses this technology? You brought up tube amps and Hegel? They are some of the worst performing amps in terms of THD+N/SINAD.

Quad 405 / 405-2 / 306 / 606 / 707 / 909 and modern Quads using their "current dumping" feed-forward technology implementation ARE NOT TUBE AMPs lol ... please take a look

 

[PristineSound]

"higher frequencies" = ? ... maybe somewhat vague imo

One example of excellent review => https://www.audiosciencereview.com/...-3-channel-purifi-amplifier-review-2nd.43834/

Search results for query: class D

www.audiosciencereview.com

Ncores and similar suffer from nothing. They measure just as well. See this (and it wouldn't look like this if Class D had such a noise issue, would it?)
View attachment 479567

This is why when Amir stack ranks based solely on 1kHz is misleading. This is only for a 1kHz test tone!

Take a look at the 15kHz test tone.

And this is precisely why everyone now thinks these Hypex and Purifi are the second coming of Jesus.

 

[pablolie]

This is why when Amir stack ranks based solely on 1kHz is misleading. This is only for a 1kHz test tone!

Take a look at the 15kHz test tone.

And this is precisely why everyone now thinks these Hypex and Purifi are the second coming of Jesus.

They are perfectly linear up to 20kHz, so no, Amir's tests are not "misleading".

 

[PristineSound]

They are perfectly linear up to 20kHz, so no, Amir's tests are not "misleading".

What do you mean by "linear" the frequency response or the distortion profile is consistent?

 

[Matias]

This is why when Amir stack ranks based solely on 1kHz is misleading. This is only for a 1kHz test tone!

Take a look at the 15kHz test tone.

And this is precisely why everyone now thinks these Hypex and Purifi are the second coming of Jesus.

Good old FAQ #7.

'Class D FAQ'

Jan 20, 2021

This topic is dedicated to Frequently Asked Questions about class D amplifiers. If you have been directed here the reason is that it is easier to concentrate the frequent answers here instead of repeating them in several topics.


Q1: Class D power ratings cannot be true. How can these small and light "toy amplifiers" generate all these watts??

A1: The key word is efficiency. That means less power is wasted as heat, and as a consequence, the manufacturers don't have to attach heavy heatsinks to the amplifier and build a large chassis. But these watts and amperes are all true...

• Matias
• class d faq
• Replies: 163
• Forum: Stereo and Multichannel Amplifier Reviews

• 

 

[mdunjic]

This is why when Amir stack ranks based solely on 1kHz is misleading. This is only for a 1kHz test tone!

Take a look at the 15kHz test tone.

And this is precisely why everyone now thinks these Hypex and Purifi are the second coming of Jesus

Pretty good thread it seems

'Class D amplifier's THD+N above 1KHz test result? | Simplify'

Jan 3, 2025

Seeing multitone test output on reviews, specially for class D amps (like TPA3255, etc.), THD+N % below 1KHz isn't as good as 1KHz or below.
Would you see it -55/60/70/80 DB for higher freq. vs -100/102 DB for mid/bass/lower freq.? How and What would you interpret it into your conclusion?

.......with the fact that most of us probably own many of such amps with this performance and probably are daily drivers for most of us with acceptable audio output.

How would the SINAD chat look if the test was taken at 15 KHz rather than 1 KHz?

• Guddu
• Replies: 16
• Forum: General Audio Discussions

• 

 

[mdunjic]

This is why when Amir stack ranks based solely on 1kHz is misleading. This is only for a 1kHz test tone!

Take a look at the 15kHz test tone.

And this is precisely why everyone now thinks these Hypex and Purifi are the second coming of Jesus

Good old FAQ #7.

'Class D FAQ'

Jan 20, 2021

This topic is dedicated to Frequently Asked Questions about class D amplifiers. If you have been directed here the reason is that it is easier to concentrate the frequent answers here instead of repeating them in several topics.


Q1: Class D power ratings cannot be true. How can these small and light "toy amplifiers" generate all these watts??

A1: The key word is efficiency. That means less power is wasted as heat, and as a consequence, the manufacturers don't have to attach heavy heatsinks to the amplifier and build a large chassis. But these watts and amperes are all true...

• Matias
• class d faq
• Replies: 163
• Forum: Stereo and Multichannel Amplifier Reviews

• 

Right here

 

[jmdesignz2]

This is also incorrect. Class D amps suffer significant level of switching noise at higher frequencies, where as the AHB2 does not. That is the reason why John Siau has his disdain for class D.

“Suffer” ? “Significant levels of switching noise ”?

Show me where the Hypex ncx500 amp has audibly worse performance due to these

 

[PristineSound]

Here is the audibly challenge once again, where did I say it was audible? I challenge anyone here to take a screenshot of my saying such switching noise is audible and show it to me.

"Then why the hell are you bringing it up?" My answer: Why the hell not? It's a matter of engineering discussion, if it's not audible and therefore we don't bring it up, then why the hell do we even have ASR? Why are we even measuring after 100dB SINAD when common cited level of distortion audibility in music is .5%?

 

[Matias]

Switching noise covered on FAQ #4. Please just read the entire post and its links and save us the trouble.

'Class D FAQ'

Jan 20, 2021

This topic is dedicated to Frequently Asked Questions about class D amplifiers. If you have been directed here the reason is that it is easier to concentrate the frequent answers here instead of repeating them in several topics.


Q1: Class D power ratings cannot be true. How can these small and light "toy amplifiers" generate all these watts??

A1: The key word is efficiency. That means less power is wasted as heat, and as a consequence, the manufacturers don't have to attach heavy heatsinks to the amplifier and build a large chassis. But these watts and amperes are all true...

• Matias
• class d faq
• Replies: 163
• Forum: Stereo and Multichannel Amplifier Reviews

• 

 

[pablolie]

Here is the audibly challenge once again, where did I say it was audible? I challenge anyone here to take a screenshot of my saying such switching noise is audible and show it to me.

When you say "suffer significant level of switching noise at higher frequencies", it "seems" to imply it must be audible, otherwise why call it "significant" rather than "insignificant"?

In any case, it seems we have gone full circle here and we agree that the difference is inaudible... which incidentally was my sole claim from the start.

 

[jmdesignz2]

When you say "suffer significant level of switching noise at higher frequencies", it "seems" to imply it must be audible, otherwise why call it "significant" rather than "insignificant"?

In any case, it seems we have gone full circle here and we agree that the difference is inaudible... which incidentally was my sole claim from the start.

The difference in amps is audible when used outside their linear range, even if it is only on transient/momentary dynamic peaks of 106dB+

You don’t lose the ability to hear differences just because playback is loud. What changes at 106 dB is safety, not basic discrimination. People can still pick out ~1 dB shifts at high levels, and it’s even easier to hear when an amp clips, compresses, or shifts the spectrum—those come through as changes in timbre and impact, not just loudness.

Now look at the numbers for this setup: 87 dB (1 W/1 m) speakers around 6 Ω that dip toward 4 Ω, a 10-foot seat costs ~9–10 dB, and two speakers give ~3 dB back. Hitting 106 dB peaks at the seat needs roughly 370 W per channel. Into 4 Ω that’s about 38.4 Vrms (~54 Vpeak) and 9.6 Arms (~13.6 Apeak). A Benchmark AHB2 is ~190 W/4 Ω, which equates to ~27.6 Vrms (~39 Vpeak) and ~6.9 Arms (~9.8 Apeak). Those peaks simply ask for more voltage and current than the AHB2 can provide, so it clips or limits, which you hear immediately as flattened transients, edgier highs, and reduced slam. Real speakers make it tougher: an impedance trough near 3–4 Ω spikes current, and 2–3 dB of power compression from hot voice coils hikes the electrical demand another ~1.6–2×. An amp with more headroom—e.g., a Buckeye NCx500 around 600 W/4 Ω (~49 Vrms, ~12.3 Arms)—keeps those same peaks inside its linear envelope, so it stays clean and uncompressed.

Yes, 106 dB isn’t safe for long, but the AUDIBLE differences between an amp that clips/approaches clipping and one that doesn’t are very obvious at that level.

 

[mdunjic]

Here is the audibly challenge once again, where did I say it was audible? I challenge anyone here to take a screenshot of my saying such switching noise is audible and show it to me.

"Then why the hell are you bringing it up?" My answer: Why the hell not? It's a matter of engineering discussion, if it's not audible and therefore we don't bring it up, then why the hell do we even have ASR? Why are we even measuring after 100dB SINAD when common cited level of distortion audibility in music is .5%?

Sure ... AHB2 is undeniably an excellent amplifier ... could even be considered the best today ... in league of its own as you said ... I do agree. As I said I have been admirer of feed-forward concept for almost 20 years, owning most of Quad's 'current dumping' power amps, which are using feed-forward, over that whole time

However, AHB2 is costing princely 3500 USD a piece, delivering 100W/ch into 8 Ohms / 190W into 4 Ohms and max 29A of current per ch ... if I want more power and headroom, I have to bridge them (getting limited to mainly using 8 Ohm nominal impedance speakers to be safe) and I need to have 2 of them ... for princely cost of 7000 USD

OR, for 'only' 2600 USD, I can have pair of Purifi 9040 monoblocks, delivering min 3x that power per ch, 40A of current each, and with puristically worse distortion at 15KHz than AHB2, but which is completely inaudible, eh?

Seeing that most of those stellar AHB2 measurements (on the link I provided) are being done with lowest 9db gain setup (professional studio setup mainly), while for Purifi with 20db gain, my simple and rational engineering mind made a very easy and simple choice ... it's going to be Purifi and I can tell you ... it's an excellent high end sound I am getting out of those Buckeye puppies ... and I am not against Benchmark ... I admire them and in fact happily use their DAC3 B DAC and LA4 pre-amp as a front end to my Buckeyes (I now bypass all Buckeye front-end buffer boards and drive Purifi modules directly with LA4 professional signal levels) ... but I have to admit ... having 2 AHB2s was nonsensical expense for home usage ... unless you are professional studio which doesn't mind paying extra premium

It's a tough competition in power amp arena, primarily for domestic use, that Benchmark is facing these days I must say

 

[pablolie]

....

Yes, 106 dB isn’t safe for long, but the AUDIBLE differences between an amp that clips/approaches clipping and one that doesn’t are very obvious at that level.

With the Watts on offer that the amps under discussion here support, driving an amp into clipping may well indicate very poor system planning/matching, imo. I compared them playing at a spirited 90-something dB or so. Nowhere near clipping range, so it is utterly immaterial -and yes, inaudible- in my particular environment. And I'd like to hear from anyone that has ever driven an AHB2 or M22 (etc) into clipping....

 

[RichB]

FIFY.

If it doesn't cause any audible problem, it is by definition insignificant.

Perhaps, not but it could be system dependent where some systems modulate into the audible range.

Any class-D without ultra-sonic noise will be touted by the manufacturer when it is achieved.

What matters to me is the first watt, where I listen 99% of the time, accurate clip meters, and bullet-proof protection.

- Rich

 

[pablolie]

...
What matters to me is the first watt, where I listen 99% of the time, accurate clip meters, and bullet-proof protection.
...

<- This! I can't recall *ever* driving any of the amps I have ever owned into clipping, but yes, solid protection is a safety feature (although the amps we discuss in here will probably blow up many speakers before they get to clipping, probably...).

 

[PristineSound]

FIFY.

If it doesn't cause any audible problem, it is by definition insignificant.

When you say "suffer significant level of switching noise at higher frequencies", it "seems" to imply it must be audible, otherwise why call it "significant" rather than "insignificant"?

In any case, it seems we have gone full circle here and we agree that the difference is inaudible... which incidentally was my sole claim from the start.

I don't know why you would automatically tied the word "significant" to audibility. You are assuming, I won't repeat the saying: "Do you know what happens when you. . .blah blah blah" And yes, swinging from 105dB SINAD to 55dB SINAD is not only significant, I should of say "very significant."

 

[PristineSound]

Sure ... AHB2 is undeniably an excellent amplifier ... could even be considered the best today ... in league of its own as you said ... I do agree. As I said I have been admirer of feed-forward concept for almost 20 years, owning most of Quad's 'current dumping' power amps over that time

However, AHB2 is costing princely 3500 USD a piece, delivering 100W/ch into 8 Ohms / 190W into 4 Ohms and max 29A of current per ch ... if I want more power and headroom, I have to bridge them (getting limited to mainly using 8 Ohm nominal impedance speakers to be safe) and I need to have 2 of them ... for princely cost of 7000 USD

OR, for 'only' 2600 USD, I can have pair of Purifi 9040 monoblocks, delivering 3x that power per ch, 40A of current each, and with puristically worse distortion at 15KHz than AHB2, but which is completely inaudible, eh?

Seeing that most of those stellar AHB2 measurements (on the link I provided) are being done with lowest 9db gain setup (professional studio setup mainly), while for Purifi with 20db gain, my simple and rational engineering mind made a very easy and simple choice ... it's going to be Purifi and I can tell you ... it's an excellent high end sound I am getting out of those Buckeye puppies ... and I am not against Benchmark ... I admire them and in fact happily use their DAC3 B DAC and LA4 pre-amp as a front end to my Buckeyes (I now bypass all Buckeye front-end buffer boards and drive Purifi modules directly with LA4 professional signal levels) ... but I have to admit ... having 2 AHB2s was nonsensical expense for home usage ... unless you are professional studio which doesn't mind paying extra premium

It's a tough competition in power amp arena, primarily for domestic use, that Benchmark is facing these days I must say

Nothing you said is disagreeable.

I have both.

I am not here to talk about which one is better by taking all into consideration, I don't favor one over the other, otherwise, I wouldn't have both.

But only from an engineer feat perspective, I doubt there will be another amp that can top the AHB2. I am hopeful that I will be proven wrong for the sake of pushing the envelope of engineering.