• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Amplifier Bakeoff: Purifi Eval1, McIntosh MA252 & Benchmark AHB2

OP
P

PJ2000

Member
Joined
Aug 19, 2021
Messages
25
Likes
55
Did you use the Low Gain mode on AHB2? If so, I don't think D90SE's output has enough range to drive that (you'll need close to 10V). You will probably need a Pre90 or some other preamp in between them.
No we had to use it in the highest gain mode for as close to parity as possible across the amplifers.
 
OP
P

PJ2000

Member
Joined
Aug 19, 2021
Messages
25
Likes
55
Agreed. This info makes the review a bit untrustworthy. Also imagine the speaker distortion at 110dB SPL sine ..... And it has survived ....
Fair point. Next time we do this we will try factor in all the feedback on potentially uncontrolled variables.

That being said, if there was an error in calibration of levels, that would need to be a systematic error to skew the results in a specific direction vs. adding random 'noise.' Why do you think this invalidates that results and if so how much does it increase the error bars in the results? 5%? 10% 200%?
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,600
Likes
4,905
Location
Prague
Fair point. Next time we do this we will try factor in all the feedback on potentially uncontrolled variables.

That being said, if there was an error in calibration of levels, that would need to be a systematic error to skew the results in a specific direction vs. adding random 'noise.' Why do you think this invalidates that results and if so how much does it increase the error bars in the results? 5%? 10% 200%?

I am only complaining on technical correctness of use of 110dB SPL 1kHz acoustic sine as a method of calibration. Even if you move the microphone like 1cm when using a pure tone in a room, you may get change in measured SPL. That's all. And you take a risk of damage the speaker driver with a continuous sine wave of such level.
I did not participate in your test so I can say nothing more. I can say nothing about consistency of level calibration regarding guesses how many percent of error bars were affected. It was up to you to prepare the best conditions. I wonder if you took your time and efforts with getting several top amplifiers, top speaker, why you did not pay more attention to technical correctness of your test trial.
 
OP
P

PJ2000

Member
Joined
Aug 19, 2021
Messages
25
Likes
55
I am only complaining on technical correctness of use of 110dB SPL 1kHz acoustic sine as a method of calibration. Even if you move the microphone like 1cm when using a pure tone in a room, you may get change in measured SPL. That's all. And you take a risk of damage the speaker driver with a continuous sine wave of such level.
I did not participate in your test so I can say nothing more. I can say nothing about consistency of level calibration regarding guesses how many percent of error bars were affected. It was up to you to prepare the best conditions. I wonder if you took your time and efforts with getting several top amplifiers, top speaker, why you did not pay more attention to technical correctness of your test trial.
Well we did to the best of our knowledge. Clearly where there is an opportunity to improve our testing methodology in any future testing, we will for sure. In this case the Mic was on a fixed stand on a heavy table and it was not moved in any way shape or form. The only motion of the Mic in testing would be air currents, which for a mic with an aperture that size and the wavelength of a 1 khz sine wave shouldn't be material.
 

GD Fan

Senior Member
Forum Donor
Joined
Jan 7, 2020
Messages
492
Likes
681
Well we did to the best of our knowledge. Clearly where there is an opportunity to improve our testing methodology in any future testing, we will for sure. In this case the Mic was on a fixed stand on a heavy table and it was not moved in any way shape or form. The only motion of the Mic in testing would be air currents, which for a mic with an aperture that size and the wavelength of a 1 khz sine wave shouldn't be material.
Sadly these home test threads always end up the same way. Nothing like ASR to make you regret even trying.
 

Galliardist

Active Member
Joined
Jun 26, 2021
Messages
292
Likes
301
Location
Sydney. NSW, Australia
Makes sense. I've heard in another board that it is always recommended to have an analog preamp in between a given source and a power amplifier. This applies whether the DAC source has a preamp mode or not.
This is audiophile nonsense. Really, the feed from a good DAC like the RME ADI-2 is just perfect and does not need anything. A modern DAC with volume control is quite simply nothing other than a traditional preamplifier, but now with digital inputs. In the case of the ADI-2 it even has traditional goodies like tone and balance controls, filters, dynamic loudness and a lot more.
There's no golden rule here. Sometimes a separate preamp will make sense, sometimes not. If your system is single source and provides a good volume control, though, it makes sense to choose a power amp with sufficient gain that the preamp isn't needed.

I've always used line level sources and integrated amps, which makes me a dinosaur - I've seen the alternatives go wrong so often though, so I guess I'm just playing safe.
 

acetogen

Member
Joined
May 30, 2021
Messages
94
Likes
30
There's no golden rule here. Sometimes a separate preamp will make sense, sometimes not. If your system is single source and provides a good volume control, though, it makes sense to choose a power amp with sufficient gain that the preamp isn't needed.

I've always used line level sources and integrated amps, which makes me a dinosaur - I've seen the alternatives go wrong so often though, so I guess I'm just playing safe.
If you don't mind me asking, what is the minimum impedance that a preamp output should have over that of the input for the power amplifier, assuming one is using separate components? Thanks
 

Galliardist

Active Member
Joined
Jun 26, 2021
Messages
292
Likes
301
Location
Sydney. NSW, Australia
If you don't mind me asking, what is the minimum impedance that a preamp output should have over that of the input for the power amplifier, assuming one is using separate components? Thanks
See my second sentence - for a real world practical answer I'll pass and let someone with more experience answer. If nobody comes I'll come back later with something
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
13,420
Likes
20,900
It's the other way around. Usually the power amp input should be 10x the pre amp output impedance. You'd also like to see the output impedance not be over 300 ohms usually.
 

Galliardist

Active Member
Joined
Jun 26, 2021
Messages
292
Likes
301
Location
Sydney. NSW, Australia
It's the other way around. Usually the power amp input should be 10x the pre amp output impedance. You'd also like to see the output impedance not be over 300 ohms usually.
Thanks. I've just found the Bookmark feature, so I'll just quote this next time it comes up and not look so dumb :D
 
Last edited:

Galliardist

Active Member
Joined
Jun 26, 2021
Messages
292
Likes
301
Location
Sydney. NSW, Australia
All the actual recordings we listened to and scored were generally no higher than 90-95 dB. We did try maxing out all the amps to see where they would clip and even the McIntosh could drive the Magicos to decent levels. Frankly none of them were loud enough that any of the amplifiers were straining in any obvious way. The fun piece we used to intentionally tax the amplifiers was 'Poem of Chinese Drum, Hok-man Yim.'
I'm assuming you listened for clipping subjectively. If you did, you may miss the issue where the amplifier doesn't audibly clip or harden, but the bass is not being properly driven (where things will first show up with the Magico).
You'd miss that with your test track, where you'd feel the drum beats from the mid-bass, and with the calibration at 1kHz
 

MediumRare

Major Contributor
Forum Donor
Joined
Sep 17, 2019
Messages
1,537
Likes
1,799
Location
Chicago
Makes sense. I've heard in another board that it is always recommended to have an analog preamp in between a given source and a power amplifier. This applies whether the DAC source has a preamp mode or not.
LOL, you are really funny.
 

acetogen

Member
Joined
May 30, 2021
Messages
94
Likes
30
LOL, you are really funny.
Thank you.
I saw your picture of the wind in Iowa City. First, it has no reference and no explanations, second, it is probably for surface wind tracking, not what is needed to decide whether wind turbines (80 m ~ 260 feet) are useable over there; -third, it does prove my point, sort of, because a maximum of 16 mph (the grey area) is not enough to power most wind turbines, fourth, SE Iowa has much less wind currents than the NW and SW parts of the state where the turbines are. As a comparison today's weather in Lincoln NE had an average of 15 mph, there are no turbines in SE Nebraska as far as I know. I will post the maps later in the discussion. However you are correct that Iowa is at 55-60% wind utilization for electricity, so EVs have a real net effect on emissions. That is not the case here in NE, where coal+natural gas is still 71%, so hybrids are still a better option.
 
Last edited:

acetogen

Member
Joined
May 30, 2021
Messages
94
Likes
30
It's the other way around. Usually the power amp input should be 10x the pre amp output impedance. You'd also like to see the output impedance not be over 300 ohms usually.
Indeed. My preamp has a 600 ohms output in balanced mode, my monoblocks operate at 44k ohms. Forgot about that
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
13,420
Likes
20,900
Indeed. My preamp has a 600 ohms output in balanced mode, my monoblocks operate at 44k ohms. Forgot about that
Yep, you have nothing to worry about. The 600 ohms would only matter if you ran high capacitance cabling a long distance. Not likely to ever matter.
 

phoenixdogfan

Major Contributor
Joined
Nov 6, 2018
Messages
1,845
Likes
1,961
Location
Nashville
I own a Purifi Eigentact and love it, but these results seem dubious at best.

Two things immediately jump out:

(1) The testers listened to one song then switched and listened again after switching amps. That's not how you do A-B testing. Humans have a very limited long term memory for sound. Comparing two sounds more than a few seconds apart will be invalid b/c you can't listen to something two minutes ago, and compare it to something else you are currently listening to b/c whatever was listened to two minutes ago is no longer in the memory buffer--only impressions of that earlier listening will remain.

(2) Level matching must be done by electrical output, not acoustic output.

Again, I have an Eigentact. I'm looking at it right now as it's playing "Why So Serious?" on The Dark Knight soundtrack. Sounds awesome, and definitely not the club foot of my system, but I don't feel qualified to say it's either better or worse than the Benchmark and McIntosh. I'm sure those other two amps would be performant as well.
 

Galliardist

Active Member
Joined
Jun 26, 2021
Messages
292
Likes
301
Location
Sydney. NSW, Australia
Thank you.
I saw your picture of the wind in Iowa City. First, it has no reference and no explanations, second, it is probably for surface wind tracking, not what is needed to decide whether wind turbines (80 m ~ 260 feet) are useable over there; -third, it does prove my point, sort of, because a maximum of 16 mph (the grey area) is not enough to power most wind turbines, fourth, SE Iowa has much less wind currents than the NW and SW parts of the state where the turbines are. As a comparison today's weather in Lincoln NE had an average of 15 mph, there are no turbines in SE Nebraska as far as I know. I will post the maps later in the discussion. However you are correct that Iowa is at 55-60% wind utilization for electricity, so EVs have a real net effect on emissions. That is not the case here in NE, where coal+natural gas is still 71%, so hybrids are still a better option.
Interesting... from work done in Australia, EVs still benefit us in all states except Victoria, despite our electricity still being over 70% fossil fuel in NSW. Emissions from coal fired stations vary widely according to the coal burnt (decent in NSW, brown coal in Victoria has higher emissions), as well as the age, design and efficiency of the power stations. So it's quite possible with 71% fossil fuel for an EV to be better than a conventional engine (remember internal combustion engines are pathetically inefficient) - I presume that would also be the case for a hybrid, but have never seen numbers.
 

Gorgonzola

Senior Member
Joined
Jan 27, 2021
Messages
407
Likes
508
Location
Ontario, Canada
@PJ2000, take heart. The main constituency here a ASR will inevitably reject any & all amplifier comparisons that suggest differences are audible -- it's doctrine. Most consider it "settled science" that difference between properly functioning amps operating within specified parameters, are inaudible.

Don't get me wrong: various objections to your approach expressed in this thread are valid, however we should understand that the "confirmation bias" operating around here is that differences are inaudible.

Personally I'm skeptical that I could hear the differences between the three amps in question in rigorous blind test but I'm openminded.
 
Top Bottom