Review and Measurements of Benchmark AHB2 Amp

[Soniclife]

Just bumping this with another question:

I was asking ChatGPT about this and I wanted some other input. Apparently the smaller cheaper class D amps like the topping mini 300 can in some case is very frequency response with speaker load. Whereas the benchmark amplifier would be load invariant.

ChatGPT mentions that if there was any effect on the frequency response, it would be most likely a possible rise or boost in some high frequencies in my speakers.

Does that make sense?

I’ve also started to look at some comparison amplifiers like the Fossi z A3 (which seems a bit more powerful)

Do you not read the amplifier reviews here?

 

[MattHooper]

Do you not read the amplifier reviews here?

Thanks for your input, but I have some trouble interpreting some of the graphs and comments. I’ve always been sort of dyslexic when it comes to electronic equipment, related graphs and measurements, etc.
So I could use some help.

For instance on the topping review I see one graph suggesting load invariance and another graph suggesting increasing distortion in high frequencies. I’m not sure how to interpret those together. I only brought the question here because of the conversation that brought up the Topping as a substitute for my receiver currently driving my speakers.

 

[PGAMiami]

Just bumping this with another question:

I was asking ChatGPT about this and I wanted some other input. Apparently the smaller cheaper class D amps like the topping mini 300 can in some case is very frequency response with speaker load. Whereas the benchmark amplifier would be load invariant.

ChatGPT mentions that if there was any effect on the frequency response, it would be most likely a possible rise or boost in some high frequencies in my speakers.

Does that make sense?

I’ve also started to look at some comparison amplifiers like the Fossi z A3 (which seems a bit more powerful)

Yes this is correct. The better, more sophisticated digital amplifiers are designed to avoid this problem where the output is load dependent. AHB2 is not a digital amplifier.

 

[Soniclife]

Thanks for your input, but I have some trouble interpreting some of the graphs and comments. I’ve always been sort of dyslexic when it comes to electronic equipment, related graphs and measurements, etc.
So I could use some help.

You can usually always just skip to Amir's summary where any issue that affects users will be called out, that often not much is said about audible issues is because the good stuff doesn't have them. After this just check the amp has enough power for you.

Frequency response and distortion are separate things, the high frequency distortion looks bad, but will almost never be audible, frequency response is simple, it should be flat up to 20khz, no need to buy an amp that isn't the days.

 

[MattHooper]

You can usually always just skip to Amir's summary where any issue that affects users will be called out, that often not much is said about audible issues is because the good stuff doesn't have them. After this just check the amp has enough power for you.

Frequency response and distortion are separate things, the high frequency distortion looks bad, but will almost never be audible, frequency response is simple, it should be flat up to 20khz, no need to buy an amp that isn't the days.

OK, thanks I will take further questions about the topping to the topping thread.

 

[Roland68]

Just bumping this with another question:

I was asking ChatGPT about this and I wanted some other input. Apparently the smaller cheaper class D amps like the topping mini 300 can in some case is very frequency response with speaker load. Whereas the benchmark amplifier would be load invariant.

ChatGPT mentions that if there was any effect on the frequency response, it would be most likely a possible rise or boost in some high frequencies in my speakers.

Does that make sense?

I’ve also started to look at some comparison amplifiers like the Fossi z A3 (which seems a bit more powerful)

ChatGPT is once again delivering rubbish.

By "cheap Class D amplifiers," I mean the TPA325x without power factor correction (PFFB) or with poorly integrated PFFB.
Infineon MERUS-based amplifiers also exhibit load dependency.

Amplifiers such as the 3E Audio A5/SE, A7/Mono/SE, Topping PA5, PA5 II, Mini 300, PA7, and Sylph Audio TPA3251/55, all with PFFB, do not have load dependency.

A Mini 300 can easily compete sonically and in terms of measurements with some amplifiers costing several thousand euros, and even surpass them. It's sometimes hard to grasp, but highly integrated circuits also have advantages.

To take things a step further, you need the better Purifi/Hypex amplifiers, a Benchmark AHB2, or a Topping B100/B200/LA90 (D).

Whether that will actually be audible is questionable and will likely depend heavily on the rest of the audio chain.

 

[kyuu]

Yes this is correct. The better, more sophisticated digital amplifiers are designed to avoid this problem where the output is load dependent. AHB2 is not a digital amplifier.

Class D designs are not digital amplifiers either.

 

[antcollinet]

Apparently the smaller cheaper class D amps like the topping mini 300 can in some case is very frequency response with speaker load. Whereas the benchmark amplifier would be load invariant.

That only applies to amps without post filter feedback. Does not apply to the topping amps (including the mini 300)

take care when asking for facts from AI. AI hallucinates up to 100% of 'facts'.

 

[Sal1950]

Whether that will actually be audible is questionable

And then in the big picture, isn't that what's really important here?
Ultimately much of the numbers and graphs, and the comparison of them is way outside my, and a large percentage of our members, technical wheelhouse.
We as a objective community need to put together some solid, reasonably supportable numbers that the layman can understand without going back to college for a degree. It's nothing that can be addressed in this dedicated review thread but something we should definitely start to address in the near future. JMHO

and will likely depend heavily on the rest of the audio chain

Another bugaboo that will also need clarification! In my mind I can think of a dozen if, ands, or buts, on the interrelation of components. But I do believe that once a "transparent wheelhouse" is determined for each individual component the only relationships left to worry over are input/output impedance's, and drive levels.

 

[etc6849]

A great tear-down video featuring John Saiu popped up on my YouTube feed:

 

[Matias]

Interesting how they use mu-metal shielding between the power supply and the amp. Also the transfer of heat through those bars into the heatsinks.

 

[etc6849]

Definitely some solid engineering choices by some of the top designers in the world. I have forgotten how many years they spent developing it with THX, but it was something like a decade.

Interesting how they use mu-metal shielding between the power supply and the amp. Also the transfer of heat through those bars into the heatsinks.

 

[Martini]

Why do I wish their modules were inside my Genelec 8351s

 

[etc6849]

Those are nice speakers! I'd never go back to consumer level gear... pro-audio is the only way to go. I actively tri-amped my bed speakers, plus cross to 8 subs. A bit overkill, but more Benchmark AHB2's = more better.

Running 16 AHB2's in there now, with 15 channels for bed + 6 for height speakers... 29 unique DSP/DAC channels including the 8 subs, with my RME sound card in my PC acting as the volume control. I'm still waiting on Trinnov to ship me the Altitude CI I paid in full for 11 weeks ago though, but at least I found VST plug-ins to upconvert to play with my height speakers.

Why do I wish their modules were inside my Genelec 8351s

 

[TheWalkman]

Great video tear down. Clearly the amp is very well thought out.

That said, for several years, I lusted over owning a Benchmark amp.

Three years ago, I pulled the trigger on a Purifi based Audiophonics amp instead and haven’t looked back.

Though the Audiophonics is Class D, it offers the same clean performance as the Benchmark, but it has far fewer components than the Benchmark, meaning, hopefully, greater longterm reliability, and if cost less than half the price of the Benchmark. My intentions are that the Audiophonics is likely my last amp, so the peace of mind of fewer components and reliability is important to me.

I’m sure Benchmark owners are very satisfied with their amps- as well they should - but my days of lusting over the AHBs are long gone.

 

[etc6849]

If it's this Purifi module, the tests look outstanding. However, having an extra 4-5dB of dynamic range for 30W and below makes the Benchmark worth it for my tri-amp configuration as I'm driving very efficient midrange horns fed directly from the amp.

I'd hear a hiss at my MLP with the Purifi modules. Is this worth an extra ~$1k/channel, it could be depending on your setup and number of speakers.

It definitely is to me. I say this after noting that the manufacturer state the midrange is claimed to have a raw sensitivity of 110dB. https://www.stereophile.com/content/klipsch-palladium-p-39f-loudspeaker

Further, none of these options existed when the AHB2 came out. It was revolutionary in 2014, and still is to this day.

Great video tear down. Clearly the amp is very well thought out.

That said, for several years, I lusted over owning a Benchmark amp.

Three years ago, I pulled the trigger on a Purifi based Audiophonics amp instead and haven’t looked back.

Though the Audiophonics is Class D, it offers the same clean performance as the Benchmark, but it has far fewer components than the Benchmark, meaning, hopefully, greater longterm reliability, and if cost less than half the price of the Benchmark. My intentions are that the Audiophonics is likely my last amp, so the peace of mind of fewer components and reliability is important to me.

I’m sure Benchmark owners are very satisfied with their amps- as well they should - but my days of lusting over the AHBs are long gone.

 

[Roland68]

If it's this Purifi module, the tests look outstanding. However, having an extra 4-5dB of dynamic range for 30W and below makes the Benchmark worth it for my tri-amp configuration as I'm driving very efficient midrange horns fed directly from the amp.

I'd hear a hiss at my MLP with the Purifi modules. Is this worth an extra ~$1k/channel, it could be depending on your setup and number of speakers.

It definitely is to me. I say this after noting that the manufacturer state the midrange is claimed to have a raw sensitivity of 110dB. https://www.stereophile.com/content/klipsch-palladium-p-39f-loudspeaker

Further, none of these options existed when the AHB2 came out. It was revolutionary in 2014, and still is to this day.


View attachment 535266

Given the abundance of active systems and highly efficient loudspeakers—such as those from Avantgarde Acoustic—it surprises me that there are no power amplifiers specifically designed to be ultra-low-noise within the 0.5 to 10-watt range (or up to 20 or 30 watts), and further optimized for impedances between 2 and 8 ohms.
This would be particularly beneficial in the high-frequency and midrange regions.

 

[Timmeon]

A great tear-down video featuring John Saiu popped up on my YouTube feed:

This is the kind of content that makes me want to support a company. Tear apart your own product and explain your engineering decisions. Love this.

Thanks for sharing.

 

[Hayabusa]

Given the abundance of active systems and highly efficient loudspeakers—such as those from Avantgarde Acoustic—it surprises me that there are no power amplifiers specifically designed to be ultra-low-noise within the 0.5 to 10-watt range (or up to 20 or 30 watts), and further optimized for impedances between 2 and 8 ohms.
This would be particularly beneficial in the high-frequency and midrange regions.

Noise level is normally a noise floor ( look at these familiar distortion slopes always matching the absolute level), so you just need that low noise floor.
The AHB-2 has -106 at 1 watt, so a speaker with 106dB/1m sensitivity still is ok with 0dB spl noise level at 1 m listening distance, let alone normal listening distances.

 

[Frank Dernie]

Given the abundance of active systems and highly efficient loudspeakers—such as those from Avantgarde Acoustic—it surprises me that there are no power amplifiers specifically designed to be ultra-low-noise within the 0.5 to 10-watt range (or up to 20 or 30 watts), and further optimized for impedances between 2 and 8 ohms.
This would be particularly beneficial in the high-frequency and midrange regions.

I have a pair of Tune Audio Anima horns which have a 109dB/watt spec, confirmed by measurement by a French magazine FWIW.
I was most concerned about noise but actually have driven them with Devialet D-Premier, Classic 16.0 40 wpc push-pull valve amplifier and a Job INTegrated and none gave audible noise at my listening distance of 4.5 m, and the Job, which is now its normal amp, has inaudible noise even with my ear against the mouth of either mid or treble horns.
I haven't tried a class-D amp but have been tempted to connect the headphone output of a portable player just to hear, but haven't bothered to make up a cable.