ATI Battle Class AB vs Class D Amplifier Shootout

Irvrobinson

Irvrobinson

Audioholic Spartan
And how many high quality amplifiers that use negative feedback are so poorly designed that they become unstable and oscillate? I've heard of none.

Based on the measurements, I have no doubt that the AHB2 is a very fine amplifier. And that's not really the point of this thread, is it? The real question is, why did you hear such an easily detectable difference between the two ATI amplifiers?
 
P

PENG

Audioholic Slumlord
May be a few of us can get together one day to do a DBT ourselves!! Who volunteer to organize one? Do it in Toronto and I'll bring at least 2 of my preamp, amps, and AVR.
 
Irvrobinson

Irvrobinson

Audioholic Spartan
May be a few of us can get together one day to do a DBT ourselves!! Who volunteer to organize one? Do it in Toronto and I'll bring at least 2 of my preamp, amps, and AVR.
Toronto just might be my favorite city to visit, so on that promise alone I could be talked into it. But I am not fond of participating in audio DBTs.
 
Paul DS

Paul DS

Full Audioholic
So far I have had 3 class D receivers (Panasonic) and I have been very pleased with every one of them. I saw no difference whatsoever in audio quality between the class D and A/B receivers. If given half a chance, I prefer a class D amp. They are a lot lighter and use far less electricity.
 
mtrycrafts

mtrycrafts

Seriously, I have no life.
Yes, why an audible difference indeed. I only read that it was a single blind comparison, no other protocol or steps taken mentioned: how level matched, how switched, how the signal was fed and a few more.
 
RichB

RichB

Audioholic Field Marshall
From the article:
FLAC files were played via Roon playing to the Oppo UDP-205 connected via balanced inputs to the Oppo HA-1. The HA-1 provides fully balanced preamplification.
Level matching was done using voltage measurements while connected to the speakers playing 250Hz and 1kz tones. It was then verified using pink noise, REW and the UMIK-1 microphone.

- Rich
 
mtrycrafts

mtrycrafts

Seriously, I have no life.
From the article:


Level matching was done using voltage measurements while connected to the speakers playing 250Hz and 1kz tones. It was then verified using pink noise, REW and the UMIK-1 microphone.

- Rich
Thanks.How did your switch the 205 between the two amps?
How close was that voltage for level matching? How was it dialed in to such close tolerance?
 
RichB

RichB

Audioholic Field Marshall
Thanks.How did your switch the 205 between the two amps?
How close was that voltage for level matching? How was it dialed in to such close tolerance?
I used the HA-1 for volume control which has .5 dB steps.
They were within the 1 digit of precision (2.8x on both). Basically, I got lucky and they matched well within the .5 dB adjustment range.
This can be seen in the near field measurements.

Originally, I split the XLR outs of the HA-1 to both amps and switched on speaker leads and adjusted the volume levels.
However, the AT4002 showed some roll-off after 10kHz. @PENG suggested this could be a difference in input impedance and changed to also move the XLR's which seemed to fix that problem. All measurements used are taken durin the same session to assure that there were no changes in the mic position.

Both amps were placed on an table to make switching easy. A towel was placed over the HA-1 to hide the volume levels and the amp connections could not be seen from the listening position.
StackRear.jpg

- Rich
 
Last edited:
P

PENG

Audioholic Slumlord
I used the HA-1 for volume control which has .5 dB steps.
They were within the 1 digit of precision (2.8x on both). Basically, I got lucky and they matched well within the .5 dB adjustment range.
This can be seen in the near field measurements.

Originally, I split the XLR outs of the HA-1 to both amps and switched on speaker leads and adjusted the volume levels.
However, the AT4002 showed some roll-off after 10kHz. @PENG suggested this could be a difference in input impedance and changed to also move the XLR's which seemed to fix that problem. All measurements used are taken durin the same session to assure that there were no changes in the mic position.

Both amps were placed on an table to make switching easy. A towel was placed over the HA-1 to hide the volume levels and the amp connections could not be seen from the listening position.
View attachment 29635
- Rich
Did you score 10/10?:D
 
RichB

RichB

Audioholic Field Marshall
Did you score 10/10?:D
I think I missed one :) It's actually not a pleasant procedure.
My friend got 5 out of 5 right, after asking to switch more than once per amp.
I did not impose any more on him. He is not an audiophile but has a good set of ears.
He was also very good at picking bi-amped and single amped on the Salon2's (driving by the AT6000's).

I have had my preferences after long-term listening sessions between amps of linear powered class A/B amps, but was never confident that I could do well in an SBT/DBT test. This is not the case with the AT522NC, I have no problems, they have a different signature than the AT4000 amps.
My expectation is that a discerning listener to could identify the difference.

In the video world, displays of two different technologies can be calibrated to very exacting standards but no one would say that you cannot tell the difference. Each technology has its strengths and weaknesses. I believe in measurements, I just don't believe that everything that should be measured to make them indistinguishable has been measured. I also believe that many amps are similarly good,

I will take Bruno's Putzey's advice, report the audible difference but not proffer a hypothesis as to why.

- Rich
 
lovinthehd

lovinthehd

Audioholic Jedi
I think I missed one :) It's actually not a pleasant procedure.
My friend got 5 out of 5 right, after asking to switch more than once per amp.
I did not impose any more on him. He is not an audiophile but has a good set of ears.
He was also very good at picking bi-amped and single amped on the Salon2's (driving by the AT6000's).

I have had my preferences after long-term listening sessions between amps of linear powered class A/B amps, but was never confident that I could do well in an SBT/DBT test. This is not the case with the AT522NC, I have no problems, they have a different signature than the AT4000 amps.
My expectation is that a discerning listener to could identify the difference.

In the video world, displays of two different technologies can be calibrated to very exacting standards but no one would say that you cannot tell the difference. Each technology has its strengths and weaknesses. I believe in measurements, I just don't believe that everything that should be measured to make them indistinguishable has been measured. I also believe that many amps are similarly good,

I will take Bruno's Putzey's advice, report the audible difference but not proffer a hypothesis as to why.

- Rich
What is the difference, though, i.e. how would you describe it? Is it significant at all?
 
Irvrobinson

Irvrobinson

Audioholic Spartan
I think I missed one :) It's actually not a pleasant procedure.
My friend got 5 out of 5 right, after asking to switch more than once per amp.
I did not impose any more on him. He is not an audiophile but has a good set of ears.
He was also very good at picking bi-amped and single amped on the Salon2's (driving by the AT6000's).

I have had my preferences after long-term listening sessions between amps of linear powered class A/B amps, but was never confident that I could do well in an SBT/DBT test. This is not the case with the AT522NC, I have no problems, they have a different signature than the AT4000 amps.
My expectation is that a discerning listener to could identify the difference.

In the video world, displays of two different technologies can be calibrated to very exacting standards but no one would say that you cannot tell the difference. Each technology has its strengths and weaknesses. I believe in measurements, I just don't believe that everything that should be measured to make them indistinguishable has been measured. I also believe that many amps are similarly good,

I will take Bruno's Putzey's advice, report the audible difference but not proffer a hypothesis as to why.

- Rich
Just expressing a personal view, Rich, these results are not believable. I have driven Salon2s single-amped and passively bi-amped with an AT3005, and it is inconceivable that the difference could be detected most of the time, unless there was a level difference, or the AT6000s were somehow defective, which I doubt. Bi-amping involves splitters from the pre-amp to the amp channels, and it seems to me that the load differences caused by the splitters electrical effects (lower impedance, requiring more current to achieve the same level) affecting the pre-amp's frequency response is more plausible than an AT6000 channel not being able to effectively drive a relatively easy load like a Salon2, even to very loud levels. And I know from personal experience that detecting differences is much more difficult at high levels than at reasonable levels.

As for the AT522NC having a discernible signature, either it has a significant distortion, or it has a different frequency response. Selectivity of 5/5, while not statistically significant, still strikes me as implausible with only distortion signature differences between the amps. And if the differences aren't just distortion it means one or both of the amplifiers are defective.

One thing we are agreed on, audio difference testing is not pleasant. ;-)
 
Last edited:
RichB

RichB

Audioholic Field Marshall
Just expressing a personal view, Rich, these results are not believable. I have driven Salon2s single-amped and passively bi-amped with an AT3005, and it is inconceivable that the difference could be detected most of the time, unless there was a level difference, or the AT6000s were somehow defective, which I doubt. Bi-amping involves splitters from the pre-amp to the amp channels, and it seems to me that the load differences caused by the splitters electrical effects (lower impedance, requiring more current to achieve the same level) affecting the pre-amp's frequency response is more plausible than an AT6000 channel not being able to effectively drive a relatively easy load like a Salon2, even to very loud levels. And I know from personal experience that detecting differences is much more difficult at high levels than at reasonable levels.

As for the AT522NC having a discernible signature, either it has a significant distortion, or it has a different frequency response. Selectivity of 5/5, while not statistically significant, still strikes me as implausible with only distortion signature differences between the amps. And if the differences aren't just distortion it means one or both of the amplifiers are defective.

One thing we are agreed on, audio difference testing is not pleasant. ;-)
The other change when bi-amping is the separation of the crossovers.
Kevin Voecks has stated that there can be increased clarity in bi-amping. Anyway OT to the this thread.

Most likely different frequency response into load, less or more linear distortion. The oft measured steady-state non-linear distortion number favor the AT522NC.
Skepticism is good, of course, if your are curious listen to one yourself, at home :p

- Rich
 
RichB

RichB

Audioholic Field Marshall
What is the difference, though, i.e. how would you describe it? Is it significant at all?
it was described in the article. More emphasis on cymbals and sibilants from the AT522NC driving the Revel M20's and Salon2's.
The AT522NC is a very dynamic amplifier. These amps reacted differently to my speakers. That's all I can say.
Whether that is believable or not is up to you. ;)

In the family vacation home, I have 4 M20's and a C20 driven by a circa 1999 Sunfire Cinema amp (200 WPC).
The home is 5 hours away and we rent. There is no air conditioning. So low heat and reliability are an absolute must.
The AT500NC amps are at the top of the list to replace this amp should it ever die.
At this point, I beginning to think it never will :)

- Rich
 
P

PENG

Audioholic Slumlord
it was described in the article. More emphasis on cymbals and sibilants from the AT522NC driving the Revel M20's and Salon2's.
The AT522NC is a very dynamic amplifier. These amps reacted differently to my speakers. That's all I can say.
Whether that is believable or not is up to you. ;)

In the family vacation home, I have 4 M20's and a C20 driven by a circa 1999 Sunfire Cinema amp (200 WPC).
The home is 5 hours away and we rent. There is no air conditioning. So low heat and reliability are an absolute must.
The AT500NC amps are at the top of the list to replace this amp should it ever die.
At this point, I beginning to think it never will :)

- Rich
I think it could be your set up. Let's just look at this logically for a moment. You said the reason for the >10 kHz dip was found and fixed if I understood right. That means if you were to plot those REW graphs again, the curves for the 522NC and the 6000 should be so close to each other that even with say 1/24 smoothing, they will overlap and appear as one curve.

So if the mic couldn't hear the difference, how could human? I agree the mic does not hear the same way we do, but we are talking about comparison hearing so regardless, it is a fact that the mic could not hear a difference that is of any significance. Moving your head a few inches should result in more difference than the amps could cause, and that could be one of the reasons (now I am half joking:)).

Regarding the "amps reacted differently to your my speakers..", I think we can confidently rule that out as being the cause, because unless you guys listened to deafeningly loud spl and sat far enough from the M20, either amp would likely be cruising at less than 2 watts average, (and I am guessing much less) that's 200 watts during the rare 20 dB peaks that might be possible on cymbal crashes if the recording is really good. The 522 is rated 300 W inito 4 ohms full bandwidth, that's 600 W peak. The M20 only dips to 4.4 ohm minimum, phase angle is reasonably benign, near 0 degree from 5K and up. Something else was causing the easily audible difference, that's all I can say.
 
RichB

RichB

Audioholic Field Marshall
I think it could be your set up. Let's just look at this logically for a moment. You said the reason for the >10 kHz dip was found and fixed if I understood right. That means if you were to plot those REW graphs again, the curves for the 522NC and the 6000 should be so close to each other that even with say 1/24 smoothing, they will overlap and appear as one curve.

So if the mic couldn't hear the difference, how could human? I agree the mic does not hear the same way we do, but we are talking about comparison hearing so regardless, it is a fact that the mic could not hear a difference that is of any significance. Moving your head a few inches should result in more difference than the amps could cause, and that could be one of the reasons (now I am half joking:)).

Regarding the "amps reacted differently to your my speakers..", I think we can confidently rule that out as being the cause, because unless you guys listened to deafeningly loud spl and sat far enough from the M20, either amp would likely be cruising at less than 2 watts average, (and I am guessing much less) that's 200 watts during the rare 20 dB peaks that might be possible on cymbal crashes if the recording is really good. The 522 is rated 300 W inito 4 ohms full bandwidth, that's 600 W peak. The M20 only dips to 4.4 ohm minimum, phase angle is reasonably benign, near 0 degree from 5K and up. Something else was causing the easily audible difference, that's all I can say.
I think I let Bruno weigh in here:

S&V: Conventional wisdom says a great amplifier has no sonic character of its own and, beyond boosting the signal level, is transparent in the audio chain. But does the nature of high-quality Class D amplification—perhaps through the absence of distortions found in other circuit topologies—produce a particular sonic signature or specific attributes you could describe?

BP: Well, if the amplifier is truly great that’s absolutely right. Sonic signatures are what you get when you approach the same ideal from different angles. There are a few distortion mechanisms conspicuously missing in Class D, mostly those related to the input stage of a Class A(B) solid-state amplifier and nonlinear capacitances. Those are also missing in valve [tube] amplifiers so it’s quite common for people to notice that a Class D amplifier is somehow reminiscent of valve amplification in terms of “sweetness” for want of a better word.
I’ve heard several reports of valve aficionados ditching their glassware and switching to Ncore. All I can conclude from that is that those people clearly weren’t actively seeking the distortion of valves as many believe, but instead had a legitimate beef with certain sonic aspects common to most solid-state designs.
That’s one thing I have to explain again and again to my fellow doubters: when audiophiles report a particular listening experience, that experience is real. Trust that. Just don’t trust the explanation they proffer.

Read more at https://www.soundandvision.com/content/bruno-putzeys-head-class-d-page-2#5TDIFka5TBygy69F.99
I don't think is discussing unicorn distortion, but I suppose it is possible.
Want to borrow my AT522NC ? (I am serious).

- Rich
 
RichB

RichB

Audioholic Field Marshall
Something else like 'Expectation Bias' in the brain.

Or somebody cheated.
I absolutely expected these to sound the same.
The reality led to many hours setting up and listening.
It is what it is, unless it isn't :p

- Rich
 
newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top