Review and Measurements of Benchmark AHB2 Amp

[Wino]

Good choice. I would have gone with the LA4, too, but for my need for 3-4 of them.

I'm happy with the choice. I probably would have gone with the Holo Audio Serene if it wasn't a 6 week wait. But, happy to save the $$$ and wait time.

 

[yanm]

I have compared the AHB2 to my Sunfire Cinema and also to the ATI AT525NC using and A/B speaker switch for instant switching.
I used a MiniDSP SHD USB that with the main channels duplicated and level matched to .1 dB using DSPs. The levels were matched using a Fluke 87V voltmeter playing 0 DBFS 1kHz sine waves.
In both cases, the AHB2 had a cleaner upper end, symbols and sibilants were better resolved.

Initially, I did not have the Sunfire balanced, perhaps it was cold when matched, but it was running a .2 dB louder. The result was the Sunfire was perceived to have better bass. Once properly matched that disappeared.

For personal listening tests, proper level matching is a must. It is easy for users to compare using known preamp levels. The AHB2 has lower gain than most amps, even in high gain mode. This will likely be perceived as less dynamic since the loud sounds are simply not as loud until matched.

- Rich

Interesting that a 0.2dB mismatch (only 2.3% level difference) results in an audible difference. What’s the consensus for proper matching, within 0.1dB or better?

Anyway, it seems that without voltmeter proper level matching is not achievable… I would not trust my sound meter to give better than 0.5dB-accuracy given variation in background noise and all that. I guess it is why that so many people hear differences between audibly transparent gears. I also guess that some DAC manufacturers may use that to their advantage: “let’s make that 4Vrms output a little bit hotter than needed and our DAC will sound better than the competition.”

 

[DonH56]

Level matching to <0.1 dB is the number I have seen accepted (alternatively, 0.1 dB is about the threshold where a difference in level can be detected in a blind test). That is about 1% and virtually impossible to determine without a voltmeter, oscilloscope, spectrum analyzer, or other measuring device. And I agree a most likely source of differences in listening tests among otherwise identical units. It has been a while, but years ago I found unit-to-unit variance greater than 0.1 dB for things like preamps and DACs.

 

[RichB]

Interesting that a 0.2dB mismatch (only 2.3% level difference) results in an audible difference. What’s the consensus for proper matching, within 0.1dB or better?

Anyway, it seems that without voltmeter proper level matching is not achievable… I would not trust my sound meter to give better than 0.5dB-accuracy given variation in background noise and all that. I guess it is why that so many people hear differences between audibly transparent gears. I also guess that some DAC manufacturers may use that to their advantage: “let’s make that 4Vrms output a little bit hotter than needed and our DAC will sound better than the competition.”

Since the preamp is the MiniDSP, it was not over driving the preamp out into either amp since matching was achieved through attenuation.

As @Don56 stated, <0.1 dB seems to be the recognized standard.
In digital audio, preamplification headroom is not a bad thing so long you don't overdrive the amp.

- Rich

 

[Rottmannash]

I put my order in for two AHB2's today. I am still in need of a preamp. Has anyone compared to Topping Pre90 to Holo Serene with the AHB2? While the Serene would look nice with the Holo May, not seeing a reason for the extra dollars over the Pre90.

@Kal Rubinson reviewed the Pre90 in Stereophile.

 

[MaxBuck]

I’m using 4 AHB2s, all bridged in an active biamp. No preamp, just a Meitner MA3 DAC and a Pass XVR crossover. It sounds fabulous, detailed, huge dynamic range, no distortion what so ever. This replaced much more expensive Constellation Centaur monos. For me these were an improvement and the best of a long series of much more expensive amps, including Spectral, Pass, Cello, Levinson and MSB.

Well, yeah, I'd think.

Good grief, what a phenomenal set of components. I am jealous. Now, how about sending the Meitner in to @amirm for measurements?

 

[PGAMiami]

Well, yeah, I'd think.

Good grief, what a phenomenal set of components. I am jealous. Now, how about sending the Meitner in to @amirm for measurements?

Perhaps the most important “component“ in my system is an Audiolense convolution filter designed by Mitch Barnett that runs on a Roon NUC, and also the TADs and RPG room treatments. The difference in electronics are subtle by comparison. That said, the more I listen to the AHB2s, the more impressed I am. They are truly amazing amps. Totally grain free If the source is grain free.

 

[PGAMiami]

Well, yeah, I'd think.

Good grief, what a phenomenal set of components. I am jealous. Now, how about sending the Meitner in to @amirm for measurements?

Meitner MA3 Integrated D/A processor

www.stereophile.com

 

[HarmonicTHD]

Meitner MA3 Integrated D/A processor

www.stereophile.com

Oh brother. After reading this
„There are sonic differences that cannon be measured. I won't say they will never be able to be measured, but as far as we know, there's no current measurements that will guarantee sound.“
I stopped.

 

[PGAMiami]

Oh brother. After reading this
„There are sonic differences that cannon be measured. I won't say they will never be able to be measured, but as far as we know, there's no current measurements that will guarantee sound.“
I stopped.

Ed is a great engineer that truly understands digital and has done great things innovating in DSD. I would expect he knows how to design and measure what’s important to him. But why should he reveal all his insights? So someone can knock him off? Also he doesn’t just tow the audiophile party line. Example His products are one of the few high end products that use switching power supplies even though most audiophiles hate these. And the measurements were not bad, btw.

 

[PGAMiami]

Ed is a great engineer that truly understands digital and has done great things innovating in DSD. I would expect he knows how to design and measure what’s important to him. But why should he reveal all his insights? So someone can knock him off? Also he doesn’t just tow the audiophile party line. Example His products are one of the few high end products that use switching power supplies even though most audiophiles hate these. And the measurements were not bad, btw.

One rule that’s helped me is too look for companies that make gear that’s used by professionals. Benchmark, Emm Meitner, TAD.

 

[HarmonicTHD]

Ed is a great engineer that truly understands digital and has done great things innovating in DSD. I would expect he knows how to design and measure what’s important to him. But why should he reveal all his insights? So someone can knock him off? Also he doesn’t just tow the audiophile party line. Example His products are one of the few high end products that use switching power supplies even though most audiophiles hate these. And the measurements were not bad, btw.

Sure. The MA3 looks like a SOTA and transparent DAC not more nor less than the ones reviewed here.

Reading then what never has been proven (and there is not a shred of evidence in the review) that it „sounds different“ and the rest of the word salad all based on a sighted uncontrolled test, just disqualifies any company and designer or reviewer for me.

The facts provided by JA very much suggest that it sounds like any other SOTA DAC and that no one would be able to tell any difference in an ABX test. If that is worth 10k because one likes the design or for the bragging rights - please by all means go for it.

 

[PGAMiami]

Sure. The MA3 looks like a SOTA and transparent DAC not more nor less than the ones reviewed here.

Reading then what never has been proven (and there is not a shred of evidence in the review) that it „sounds different“ and the rest of the word salad all based on a sighted uncontrolled test, just disqualifies any company and designer or reviewer for me.

The facts provided by JA very much suggest that it sounds like any other SOTA DAC and that no one would be able to tell any difference in an ABX test. If that is worth 10k because one likes the design or for the bragging rights - please by all means go for it.

FIR and IIR filters measure different and sound different. The MA3 uses both and switches from one to another in real time as the music signal changes. This is explained in the review and I believe is unique to Ed’s DACs. The DAC also handles the DSD conversion and jitter impeccably. The net result is that it sounds very, very smooth and yet detailed, AND it measures well. Bad recordings still sound bad as it’s not just artificially smoothing everything.

As a former EE, I’m a believer that we can measure things that affect how products sound. The problem is in knowing what to measure. When new technologies are introduced and are judged using the measurements that were useful for the prior technology, we often end up measuring the wrong things. This happened with the tube to transistor transition when everyone was looking at THD and noise at full power, and ignoring IMD and switching distortion. It also happened with LP to CD when we focused on flat frequency response, wow and flutter, and noise and had not yet understood digital artifacts, poor filter designs and jitter, all things that can be measured but were not an issue with LP. Coincidentally Ed played an important role in the early days of digital understanding and developing solutions to the issues that were unique to digital.

Amir does a fantastic job and I’m sure he‘s forgotten more about engineering than I know. That said, all DACs that measure well don’t sound the same. I’m sure there are alternatives to the MA3 that would also measure and sound great, and even at a cheaper price. But the overall package with the MA3 is very compelling because of how well it handles multiple inputs, Roon, variable volume, remote with discrete input selection and exceptional customer service from Ed’s son, Amadeu.

 

[HarmonicTHD]

FIR and IIR filters measure different and sound different. The MA3 uses both and switches from one to another in real time as the music signal changes. This is explained in the review and I believe is unique to Ed’s DACs. The DAC also handles the DSD conversion and jitter impeccably. The net result is that it sounds very, very smooth and yet detailed, AND it measures well. Bad recordings still sound bad as it’s not just artificially smoothing everything.

As a former EE, I’m a believer that we can measure things that affect how products sound. The problem is in knowing what to measure. When new technologies are introduced and are judged using the measurements that were useful for the prior technology, we often end up measuring the wrong things. This happened with the tube to transistor transition when everyone was looking at THD and noise at full power, and ignoring IMD and switching distortion. It also happened with LP to CD when we focused on flat frequency response, wow and flutter, and noise and had not yet understood digital artifacts, poor filter designs and jitter, all things that can be measured but were not an issue with LP. Coincidentally Ed played an important role in the early days of digital understanding and developing solutions to the issues that were unique to digital.

Amir does a fantastic job and I’m sure he‘s forgotten more about engineering than I know. That said, all DACs that measure well don’t sound the same. I’m sure there are alternatives to the MA3 that would also measure and sound great, and even at a cheaper price. But the overall package with the MA3 is very compelling because of how well it handles multiple inputs, Roon, variable volume, remote with discrete input selection and exceptional customer service from Ed’s son, Amadeu.

Filters, IMD, DSD etc. or whatever engineering solutions a designer applies, in the end what counts is what comes out the back no matter how the result was achieved and if humans can hear it. Beliefs, opinions, credentials, claims etc simply don’t matter, facts do and no one has passed a blind ABX Test of DACs/AMPs in this category. If you or the manufacturer can provide facts to the opposite I will be happy to recant and buy.

Maybe this helps, provided you are even open to consider.

Audibility thresholds of amp and DAC measurements

Introduction A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC...

www.audiosciencereview.com

Audio Blind Testing - You Are Doing It Wrong! (Video)

Just uploaded a video on the basics of performing controlled tests in audio. It was motivated by saving myself text in having to write all of this down when telling someone who to do these tests right. And as a counter to a few online personalities to keep saying they do "blind" tests yet when...

www.audiosciencereview.com

 

[PGAMiami]

Your tone does not merit a reply

 

[Wino]

Received my AHB2 x2 today. LA4 arrives on Wednesday. Excited to try it all out. Any recommended sensitivity for Holo Audio May DAC and Dynaudio speakers? I will be using balanced connection. Figured I would start with 22 and move up to 14.2. In the end, is this subjective or some math I should be doing?

 

[RichB]

Received my AHB2 x2 today. LA4 arrives on Wednesday. Excited to try it all out. Any recommended sensitivity for Holo Audio May DAC and Dynaudio speakers? I will be using balanced connection. Figured I would start with 22 and move up to 14.2. In the end, is this subjective or some math I should be doing?

I found these specs here: https://www.stereophile.com/content/holoaudio-may-level-3-da-processor-measurements

The May's maximum output level at 1kHz in NOS, OS, and OS/PCM modes was 5.8V from the balanced output.

The AHB2 mid sensitivity setting will permit the Holo to drive the amp to maximum output.

- Rich

 

[Wino]

Perfect, thanks!

I've read most of this thread, but may have missed the info I should be looking for. Is there some article or post that would help determine the best setting? Just want to educate myself Regardless, thanks.

 

[PGAMiami]

I found these specs here: https://www.stereophile.com/content/holoaudio-may-level-3-da-processor-measurements



The AHB2 mid sensitivity setting will permit the Holo to drive the amp to maximum output.

- Rich

I am using mine at the lowest sensitivity setting, and at this setting the maximum volume on even the quietest recordings is louder than I ever listen to. You most probably don’t need to drive the amps to full power, especially with the bridged amps. A lower sensitivity setting will results in a lower signal to noise, assuming the source is not clipping. With the LA4, I believe the manual recommends using the lowest setting.

 

[RichB]

I am using mine at the lowest sensitivity setting, and at this setting the maximum volume on even the quietest recordings is louder than I ever listen to. You most probably don’t need to drive the amps to full power, especially with the bridged amps. A lower sensitivity setting will results in a lower signal to noise, assuming the source is not clipping. With the LA4, I believe the manual recommends using the lowest setting.

Correct and also use the lowest AHB2 sensitivity setting with the LA4. My comment was for the Holo while waiting for the LA4.

- Rich