Benchmark AHB2 Review (Updated Measurements)

[fpitas]

If you adopt a basic investment perspective and assuming you bought the unit for $3k in 2014, you could easily sell it today nearly 10 years later for $2k, so the real cost is only $1k or $100 a year (plus inflation), less than a media/ streaming subscription.

That's the upside of buying quality equipment and not constantly swapping things out..

 

[JustJones]

I think they were $2500 in 2014 when they first came out.

Edit. No they were $2995

Benchmark Media AHB2 Stereo Power Amplifier This little unit will change the way you look and listen to amplifiers. Review And Images By Greg Weaver

Benchmark Media AHB2 Stereo Power Amplifier This little unit will change the way you look and listen to amplifiers. Review And Images By Greg Weaver of Enjoy the Music.com

www.enjoythemusic.com

 

[AdamG]

Possibly if it had been written more wryly I would have understood it as humour. But my bad I guess if it was meant to be funny.

I was right there with you and I was not getting the joke. But I am glad it ended well. Happy New Years Team ASR. To another year worshiping SINAD in the Church of ASR under the guidance of our Dear Leader Amir the Magnificent!

Yeah just Happy New Year and enjoy your music your way and we can all learn a little by hanging out with all our incredible intelligent and skilled members. That’s the value. Some of it does rub off on us mere mortals. Everyone brings something to the table and we are all better for it. Cheers Shipmates.

 

[Axo1989]

Possibly if it had been written more wryly I would have understood it as humour. But my bad I guess if it was meant to be funny.

When I first heard that joke I (I think I may have read it here) I thought it somewhat hilarious that NASA would have embarked on a race to the moon and spent squillions at the whim of a film director insisting on authentic mise-en-scène. Not to mention that particular director. Along with the irony of having to go to the moon to fake going to the moon. But having worked in bureaucracy I know the scenario is absurd enough to be realistic, which is also what makes it funny.

I was right there with you and I was not getting the joke. But I am glad it ended well. Happy New Years Team ASR. To another year worshiping SINAD in the Church of ASR under the guidance of our Dear Leader Amir the Magnificent!

I know it was skating on the edge of Poe's law. Generally explaining jokes is shite, so I wouldn't but It's 2024 here already, so why not, and HNY to everyone.

 

[pablolie]

Bryston has always been considered a reliable purchase. ...
{good stuff edited out for brevity]

My experience with warranties is that stuff always happens after they run out. Won't mention the brands, but two of the highly regarded ones in this thread fall squarely in there. Repair costs were kinda fair though.

Far more importantly, as I am logging out of online media, I wish you all and your loved ones a great 2024! Thanks for the conversations!

 

[pseudoid]

Thank you for the 2nd confirmation/testament, @amirm,

This AH2 is the pure definition of a "cash cow"... or at least should be.

 

[Kal Rubinson]

This was to be expected, as the AHB2 does not control the woofers as well as the C298. This shows the tendency with lower impedances:

Damping Factor at 4ohm (63Hz/1kHz/14kHz)
AHB2: 48/45/19
C298: 833/769/559
Source: stereo.de

The reactive load tests of the AHB2 and the M23 (with high probability similar for the C298) also indicate this.

Interesting proposition.

For me, the tell-tale recording was of the Rachmaninov All-Night Vigil. There is a quiet portion at the end of the Part 5. Nunc dimittis where the only the voices are those of the octavists at the bottom of their range. Played over the Blades, the full progression down to the lowest notes is equally clear with either the NAD or the Benchmark. It is the same with the Benchmark and the Estelon but, if I insert the NAD, the very last tones are simply missing. This makes it hard for me to think that there is an increase in bass due to a loss of control by the Benchmark.

Of course, without real-time measurement, I will not argue with you on this.

 

[SorenTyson]

Possibly if it had been written more wryly I would have understood it as humour. But my bad I guess if it was meant to be funny.

I think this part made it clear to be a joke:

…that he insisted on filming on location.

 

[pogo]

Interesting proposition...Of course, without real-time measurement, I will not argue with you on this.

Unfortunately, the relevant measurements are missing in this case. A good explanation of what is happening can be found here (sadly in German): Link

 

[HarmonicTHD]

This was to be expected, as the AHB2 does not control the woofers as well as the C298. This shows the tendency with lower impedances:

Damping Factor at 4ohm (63Hz/1kHz/14kHz)
AHB2: 48/45/19
C298: 833/769/559
Source: stereo.de

The reactive load tests of the AHB2 and the M23 (with high probability similar for the C298) also indicate this.

You can have a look here, the Excel at the end helps to eyeball the audiability. I am on an iPhone ATM so I didn’t run the numbers.

The stereo de figures seem a bit low compared to Benchmarks own specs at 8 ohm though. But I didn’t look into it further either.

Audio Myth - "Damping Factor Isn't Much of a Factor"

Myth - "Damping Factor Isn't Much of a Factor" Myth - "A Damping Factor of 10 is High Enough" Myth - "All Amplifiers Have a High-Enough Damping Factor" Where did these Myths Originate? These myths seem to trace back to a well-known paper (PDF) written by Dick Pierce. His analysis shows that a...

benchmarkmedia.com

 

[pogo]

he stereo de figures seem a bit low compared to Benchmarks own specs at 8 ohm though.

The DF is determined at STEREO under approximately real operating conditions at 4ohms.
Such measurements are performed with a 1m long cable with a gauge of 1.5 square millimeters, i.e. close to reality, which should also take into account longer cables with higher gauges.
Unfortunately, we also lack the values for 2ohm in reality.

 

[Gringoaudio1]

I think this part made it clear to be a joke:

…that he insisted on filming on location.

Doh! I didn’t read very carefully!!

 

[yanm]

The DF is determined at STEREO under approximately real operating conditions at 4ohms.
Such measurements are performed with a 1m long cable with a gauge of 1.5 square millimeters, i.e. close to reality, which should also take into account longer cables with higher gauges.
Unfortunately, we also lack the values for 2ohm in reality.

Assuming a purely resistive load, and that the amplifier can drive the 2 ohm load, the DF at 2 ohm is simply half of the one at 4 ohm. This is is because DF = LOAD IMPEDANCE / AMPLIFIER OUTPUT IMPEDANCE. Changing the load impedance should not change the output impedance.

Damping Factor at 2ohm (63Hz/1kHz/14kHz)
AHB2: 24/23/9
C298: 416/384/279

By the way, any idea about the discrepancies between the DF measured by STEREO and the ones from the AHB2 datasheet?

• Damping factor at 8ohm (20Hz/1kHz/20kHz): 350 / 254 / 34 (datasheet values)
• Damping factor at 4ohm (20Hz/1kHz/20kHz): 175 / 127 / 17 (linear scaling from 8ohm)
• Damping factor at 2ohm (20Hz/1kHz/20kHz): 87 / 63 / 8 (linear scaling from 8 ohm)

The 20kHz one mostly matches the one at 16kHz from STEREO but the lower frequency ones are quite off.

Edit: Thinking a bit more about it ... If the cable is included in the DF calculation: DF = LOAD IMPEDANCE / (CABLE IMPEDANCE + AMPLIFIER IMPEDANCE). Now, a 1m/1.5mm2 wire is about 10mOhm per km so 10mOhm for 1m. As there are two wires in a speaker cable, that makes 20mOhm in total. The lower the load impedance and the output impedance of the amplifier, the more importance of the cable in the DF including amplifier and cable. That's why the 20kHz data sheet value mostly matches the 16kHz STEREO measurement (in this case, the AHB2 output impedance is 'large': 8Ohm/34 = 235mOhm => the cable has little influence). On the other hand, taking the 20mOhm for the cable and the 254 DF at 8ohm/1kHz from the AHB2 datasheet, I get a DF of 77 at 4ohm/1kHz, which is in the same ballpark than than the value of 45 measured by STEREO. The difference is likely due to the non purely resistive nature of the amplifier-cable-load circuit and the different resistance of the cable used by STEREO with respect to my calculation.

Now though, assuming a purely resistive circuit again, the maximum DF with such a cable of 20mOhm is 200 for a 4ohm load, so how can the C298 achieve a much larger values? Or am I totally misunderstanding something?

 

[Sokel]

Assuming a purely resistive load, and that the amplifier can drive the 2 ohm load, the DF at 2 ohm is simply half of the one at 4 ohm. This is is because DF = LOAD IMPEDANCE / AMPLIFIER OUTPUT IMPEDANCE. Changing the load impedance should not change the output impedance.

Damping Factor at 2ohm (63Hz/1kHz/14kHz)
AHB2: 24/23/9
C298: 416/384/279

By the way, any idea about the discrepancies between the DF measured by STEREO and the ones from the AHB2 datasheet?

• Damping factor at 8ohm (20Hz/1kHz/20kHz): 350 / 254 / 34 (datasheet values)
• Damping factor at 4ohm (20Hz/1kHz/20kHz): 175 / 127 / 17 (linear scaling from 8ohm)
• Damping factor at 2ohm (20Hz/1kHz/20kHz): 87 / 63 / 8 (linear scaling from 8 ohm)

The 20kHz one mostly matches the one at 16kHz from STEREO but the lower frequency ones are quite off.

Edit: Thinking a bit more about it ... If the cable is included in the DF calculation: DF = LOAD IMPEDANCE / (CABLE IMPEDANCE + AMPLIFIER IMPEDANCE). Now, a 1m/1.5mm2 wire is about 10mOhm per km so 10mOhm for 1m. As there are two wires in a speaker cable, that makes 20mOhm in total. The lower the load impedance and the output impedance of the amplifier, the more importance of the cable in the DF including amplifier and cable. That's why the 20kHz data sheet value mostly matches the 16kHz STEREO measurement (in this case, the AHB2 output impedance is 'large': 8Ohm/34 = 235mOhm => the cable has little influence). On the other hand, taking the 20mOhm for the cable and the 254 DF at 8ohm/1kHz from the AHB2 datasheet, I get a DF of 77 at 4ohm/1kHz, which is in the same ballpark than than the value of 45 measured by STEREO. The difference is likely due to the non purely resistive nature of the amplifier-cable-load circuit and the different resistance of the cable used by STEREO with respect to my calculation.

Now though, assuming a purely resistive circuit again, the maximum DF with such a cable of 20mOhm is 200 for a 4ohm load, so how can the C298 achieve a much larger values? Or am I totally misunderstanding something?

You have to include connectors and the rest in the calculation which is not insignificant relative to the cable itself.
Not that will make any big difference but it's nice to include all parameters.

 

[beagleman]

Yes and no. We really have no idea what sort of levels @Kal Rubinson was pushing through the AHB-2/Estolon Aura combination and consequently what current demands were being made (or not).

I mean, you can run just about any amplifier into a <1R loudspeaker at a low(ish) level and not trip a protector/blow a fuse or bake a set of OPTs. With the Benchmark, it's a case of how high the demand is, for how long- transients are not an issue- continuous demands in BTL will shut it down fast.

But I do agree with all your points Don. The Estolon seems to have a rather extreme narrow range where the load is a bit difficult, but there's been plenty of loudspeakers worse in that regard and plenty of amplifiers that would take such a load also in their stride.

I am interested in exactly what was happening with the NAD vs the Benchmark as both are capable amplifiers. It would be very interesting to investigate wouldn't it?

The other somewhat polar opposite issue, any time an amp is designed to run really low impedances reliably, it has to by that very design, be limited in power when powering speakers of more normal impedances.

Something to do with lots of current will work well in low impedances but voltage will be lower in that type of design, Unless a variable voltage rail is employed for higher impedance loads.

 

[Endibol]

This was to be expected, as the AHB2 does not control the woofers as well as the C298. This shows the tendency with lower impedances:

Damping Factor at 4ohm (63Hz/1kHz/14kHz)
AHB2: 48/45/19
C298: 833/769/559
Source: stereo.de

The reactive load tests of the AHB2 and the M23 (with high probability similar for the C298) also indicate this.

First of all, I wish all ASR-members a happy and healthy 2024!

The "Damping factor" difference between the NAD298 and the AHB2 is not the reason for the difference in bass production Kal heard. In the following I try to explain this in a (simplified) way.

"Damping factor" is merely a (confusing) definition of the output impedance of the amplifier (nominal speaker impedance/output impedance) and its name hints to a phenomenon where it has (some) influence: the damping of the loudspeaker cone.

Let's have a look at this cone damping:
Suppose the amplifier sends a pure sign wave to the loudspeaker causing the cone to move with a certain frequency. Suppose now the amp suddenly stops feeding the loudspeaker with this signal. Ideally the cone would need to stop moving at once as well. It doesn't do this however due to the inertia of the cone, voice coil and other mechanical components attached to it. Fortunately there always is some purely mechanical damping due to mechanical resistance of the spider etc. which slows down the movement.

There is an additional source of damping: The moving voice coil in the permanent magnetic field produces a voltage that tends to create a current and accompanying magnetic force in the opposite direction of its movement. This is Lenz's Law acting and helps to dampen the movement. The magnitude of this damping is determined by the current that's able to flow; the larger, the faster the damping.
This means that the total of the impedances of the electrical circuit formed by the amplifier output, speaker cable, passive cross over and voice coil needs to be as low as possible.

To see which of these elements are important we need to look at the relative magnitudes of the impedances. For simplicity I assume that the impedance of the loudspeaker itself is in the order of a few Ohms. The speaker cable adds a few hundredths of Ohms max. In order to substantially influence the total impedance of the circuit the output impedance should be in the same order of magnitude as the loudspeaker impedance or larger. This probably used to be the case in the era of tube amps, but with modern transistor amps the output impedance is a few tenths of an Ohm max in most situations.
The conclusion is that with modern transistor amplifiers the main factor determining the amount of electromagnetic cone damping is the impedance of the voice coil/cross over circuit in the speaker itself: The influence of the output impedance is negligible.
This is an important reason to abandon the term "Damping Factor" and shows that whatever difference Kal heard between the NAD and the Benchmark was not caused by the difference in output impedance (damping factor).

The second and most important aspect where the output impedance plays a role lies in the fact that it's part of a voltage divider circuit between the amp output and the loudspeaker. Because of the fact that the loudspeaker impedance is complex and therefore varying with frequency it is important to keep the combination of speaker cable and output impedance sufficiently low. As mentioned in this thread above this is described in an excellent article by John Siau of Benchmark Audio

Audio Myth - "Damping Factor Isn't Much of a Factor"

Myth - "Damping Factor Isn't Much of a Factor" Myth - "A Damping Factor of 10 is High Enough" Myth - "All Amplifiers Have a High-Enough Damping Factor" Where did these Myths Originate? These myths seem to trace back to a well-known paper (PDF) written by Dick Pierce. His analysis shows that a...

benchmarkmedia.com

In view of the above it is my strong wish that the industry stops specifying the output impedance as "Damping Factor". Perhaps once a brilliant marketing concept to distinguish transistor from tube amps, but a source of confusion and misunderstanding today.

 

[peng]

First of all, I wish all ASR-members a happy and healthy 2024!

The "Damping factor" difference between the NAD298 and the AHB2 is not the reason for the difference in bass production Kai heard. In the following I try to explain this in a (simplified) way.

"Damping factor" is merely a (confusing) definition of the output impedance of the amplifier (nominal speaker impedance/output impedance) and its name hints to a phenomenon where it has (some) influence: the damping of the loudspeaker cone.

Let's have a look at this cone damping:
Suppose the amplifier sends a pure sign wave to the loudspeaker causing the cone to move with a certain frequency. Suppose now the amp suddenly stops feeding the loudspeaker with this signal. Ideally the cone would need to stop moving at once as well. It doesn't do this however due to the inertia of the cone, voice coil and other mechanical components attached to it. There always will be some purely mechanical damping due to mechanical resistance of the spider etc. which slows down the movement.

There is an additional source of damping: The moving voice coil in the permanent magnetic field produces a voltage that tends to create a current and accompanying magnetic force in the opposite direction of its movement. This is Lenz's Law acting and helps to dampen the movement. The magnitude of this damping is determined by the current that's able to flow; the larger, the faster the damping.
This means that the total of the impedances of the electrical circuit formed by the amplifier output, speaker cable, passive cross over and voice coil needs to be as low as possible.

To see which of these elements are important we need to look at the relative magnitudes of the impedances. For simplicity I assume that the impedance of the loudspeaker itself is in the order of a few Ohms. The speaker cable adds a few hundredths of Ohms max. In order to substantially influence the total impedance of the circuit the output impedance should be in the same order of magnitude as the loudspeaker impedance or larger. This probably used to be the case in the era of tube amps, but with modern transistor amps the output impedance is a few tenths of an Ohm max in most situations.
The conclusion is that with modern transistor amplifiers the main factor determining the amount of electromagnetic cone damping is the impedance of the voice coil/cross over circuit in the speaker itself: The influence of the output impedance is negligible.
This is an important reason to abandon the term "Damping Factor" and shows that whatever difference Kai heard between the NAD and the Benchmark was not caused by the difference in output impedance (damping factor).

The second and most important aspect where the output impedance plays a role lies in the fact that it's part of a voltage divider circuit between the amp output and the loudspeaker. Because of the fact that the loudspeaker impedance is complex and therefore varying with frequency it is important to keep the combination of speaker cable and output impedance sufficiently low. As mentioned in a thread before this is described in an excellent article by John Siau of Benchmark Audio (

Audio Myth - "Damping Factor Isn't Much of a Factor"

Myth - "Damping Factor Isn't Much of a Factor" Myth - "A Damping Factor of 10 is High Enough" Myth - "All Amplifiers Have a High-Enough Damping Factor" Where did these Myths Originate? These myths seem to trace back to a well-known paper (PDF) written by Dick Pierce. His analysis shows that a...

benchmarkmedia.com

).

In view of the above it is my strong wish that the industry stops specifying the output impedance as "Damping Factor". Perhaps once a brilliant marketing concept to distinguish transistor from tube amps, but today a source of confusion and misunderstanding.

You may be a little late to the party lol.. Others have tried to explain such concepts to him, iirc at least one even showed him the math, but there's no help to him, apparently but I still hope to be wrong, may be some day he would be convinced. It seems that as soon as he grab ahold of some buzz words (not really just buzz obviously but to him it really has become just buzz words) such as DF, and recently, "EPDR", he would unlikely change his mind or let go regardless what explanation/addition info were offered to help him understand more on the topics. Since he got that EPDR thing figured out (I doubt he really figured that out), he's pushing or at least suggesting Amir to include those power cube tests for every power amp, even the sad light weight class AB AVRs review/measurements lol... As such, why he didn't bother suggesting Amir to include measuring output impedance on every power amp test is puzzling, may be he really still don't understand his favorite DF thing is really, or virtually an output impedance thing??

 

[snapsc]

This thread started out as an update to the AHB2 review. It then moved in the direction of alternatives to the Benchmark amp that also measure very well, may or may not sound different from the others and may or may not be a better value. The alternative amps mentioned include both Hypex and Purifi versions.

Since ASR values measurements and manufacturers who provide measurements and since there are no Gan amps in the discussion, I'd like to ask/propose that the Orchard Audio Stereo Ultra be considered for discussion/comparison as well. Yes, it costs more than some of the other amps..but less than the Benchmark...and by all accounts, the measurements are equally excellent...(and maybe slightly better when it comes to THD+D vs frequency). And, maybe less important to some, but like the other amps, the anecdotal descriptions of the sound are also quite good.

Starkrimson® Ultra Amplifier Modules -- 250/500W into 8/4ohm

Here is what a customer wrote on another forum. "To compare with another well-regarded Class D amp, I did some track by track comparisons with the Purifi-powered NAD c298. The NAD brings a warmer signature than the Ultra, with more prominent bass output. Like the Ultra, the NAD provides great...

www.audiosciencereview.com

 

[Mtbf]

An independent review is always a good method to verify claims being made. The scientific community relies on this process seems to work pretty well.

Your “Valhalla” has already been reached years ago. The rest is GAS.

 

[pogo]

Or am I totally misunderstanding something?

This topic is not trivial and good information can be found here, but it certainly does not cover everything:

'High Performance Audio Power Amplifiers' by Ben Duncan
7.4.3 Damping factor & 2.3.2 What speakers are looking for