What happened to all those A/V receivers with digital amp sections? - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 30 Old 05-21-2012, 03:45 PM - Thread Starter
Senior Member
 
donutfan's Avatar
 
Join Date: Jul 2011
Posts: 390
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 19
About 6-8 years ago, surround receivers with digital amplifers were all the rage (btw I know they aren't truly digital, but am using that term because most people are familiar with it).

But now except for some upscale Pioneers for example, most receivers are still using conventional class A/B analog amps.

What happened? They didn't sound acceptable? Too expensive? Couldn't handle lower impedance speaker loads (I saw this complaint a lot)?
donutfan is offline  
Sponsored Links
Advertisement
 
post #2 of 30 Old 05-21-2012, 03:58 PM
AVS Special Member
 
DonoMan's Avatar
 
Join Date: Feb 2005
Location: Massachusetts
Posts: 3,670
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Too expensive and not enough consumer interest (though I was and continue to be interested in Class D), I'd say. There are FRCD techs available that work better with (say) 4 ohm speakers than others. My personal choice is Hypex UcD, though I have not listened to every single option out there.

"Vintage" is good for wine, not for A/V equipment.

-Dan D.
DonoMan is offline  
post #3 of 30 Old 05-21-2012, 04:23 PM
Senior Member
 
Leo King's Avatar
 
Join Date: Jun 2008
Location: US
Posts: 398
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
About 7 years ago I bought a Panasonic 7.1 AVR with D amp section (or so they called). The sound was very "bright" and it lacked the oomph that conventional A & A/B amps had even when used with a good powered subwoofer. IMO, it's good for a small bedroom or a small den and that's all. I ran it for a few months and put it back in the box. It's still in the mint condition because I didn't use it much.
Leo King is offline  
post #4 of 30 Old 05-21-2012, 04:29 PM
AVS Special Member
 
Venomous's Avatar
 
Join Date: Dec 2004
Location: Huntington Beach/San Diego CA
Posts: 1,198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Pioneer is the only one out there doing it right now.
Venomous is offline  
post #5 of 30 Old 05-21-2012, 06:05 PM - Thread Starter
Senior Member
 
donutfan's Avatar
 
Join Date: Jul 2011
Posts: 390
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 19
Quote:


IMO, it's good for a small bedroom or a small den and that's all.

Yep I notice most HTiBs still use them (and their typically very small speakers don't need much power).
donutfan is offline  
post #6 of 30 Old 05-21-2012, 06:44 PM
Senior Member
 
Leo King's Avatar
 
Join Date: Jun 2008
Location: US
Posts: 398
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Right, I bought the Panasonic SA-XR55 (100W/channel in stereo mode, 20-20,000Hz, 0.09% THD - 100Wx7 @ 1KHz, 0.09% THD) initially for a medium sized living room but despite the 100W rated power, it didn't seem to be adequate for the room so I moved it to a small office where it sounded OK. But I didn't like it mainly because of the bright and dry sound from it. I didn't find the dynamics in the sound like I could hear from the class A, A/B or even B amps.
Leo King is offline  
post #7 of 30 Old 05-21-2012, 07:34 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by Leo King View Post

Right, I bought the Panasonic SA-XR55 (100W/channel in stereo mode, 20-20,000Hz, 0.09% THD - 100Wx7 @ 1KHz, 0.09% THD) initially for a medium sized living room but despite the 100W rated power, it didn't seem to be adequate for the room so I moved it to a small office where it sounded OK. But I didn't like it mainly because of the bright and dry sound from it. I didn't find the dynamics in the sound like I could hear from the class A, A/B or even B amps.

I'd like to correct this, since it's simply misinformation.

This "bright and dry sound" makes no sense from a technology standpoint, nor is it supported by any blind test I've seen. I replaced a relatively expensive Class A amp with the SA-XR57 and it sounded exactly the same....

The SA-XR57 was a great receiver and if wasn't for the fact that I need room correction, I'd still use it.

And its real power rating is more than comparable to current (and much bigger/heavier) A/B receivers:

Five channels driven continuously into 8-ohm loads:
0.1% distortion at 72.6 watts
1% distortion at 104.6 watts

For what it's worth, the SA-Xr57 felt a tad more robust than a recent Marantz SR5006 (Class A/B) I had.

The reality is that Class D hasn't taken off for a variety of reasons:

The manufacturers are happy recycling existing technology designs, which keeps their costs down.

AVRs are already a commodity and Class D can make it very hard to stem that tide, plus it's much harder to romance it (unless it has the B&O pedigree, as the ICE units).

"Real man" buy their amps and receivers by the pound, so selling a light, small unit may work for the pro market, but not for the Best Buy crowd.

This is oversimplifying things only a little bit.
Ryan1 is offline  
post #8 of 30 Old 05-21-2012, 07:44 PM
Senior Member
 
Leo King's Avatar
 
Join Date: Jun 2008
Location: US
Posts: 398
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Yep, it's only my opinion and you're entitled to your own. That's why I said the above was IMO, I didn't say that's what everyone else would hear because definitely, different people don't hear/feel the same.
Leo King is offline  
post #9 of 30 Old 05-21-2012, 08:12 PM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 247
Quote:
Originally Posted by Ryan1 View Post

I'd like to correct this, since it's simply misinformation.

This "bright and dry sound" makes no sense from a technology standpoint, nor is it supported by any blind test I've seen. I replaced a relatively expensive Class A amp with the SA-XR57 and it sounded exactly the same....

The SA-XR57 was a great receiver and if wasn't for the fact that I need room correction, I'd still use it.

Actually, it could sound different than other amps, because it has a fairly high output impedance. So speakers with a very non-linear impedance curve will interact with the amp. That's, I think, why a lot of the tube guys went gaga over them. They genuinely did sound like triodes, because they had similar output impedance.

That said, when I had my XR55 I did a level-matched single-blind test against my previous stack of kit (Marantz AV600 + Adcom GFA-5800 + Adcom GFA-2535) on a set of speakers with fairly linear impedance curves (KEF Q15/Q95C up front, KEF Q-Compact surrounds) and could not tell a difference.

Also, it's worth noting that the better Pioneers still have Class D amps. They don't have the size/mass benefits of the old Pannies, though.

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
post #10 of 30 Old 05-21-2012, 08:36 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by Leo King View Post

Yep, it's only my opinion and you entitle to your own. That's why I said the above was IMO, I didn't say that's what everyone else would hear because definitely, different people don't hear/feel the same.

With all due respect, it's not a matter of "opinion."

Amp technology is well understood and as log as an amp measures flat within the audible band and does not distort audibly, then it will sound the same as another amp which does the same.

There is no magic about it, despite what you may read in some reviews. And the ear is easily fooled by preconceived ideas, by a price tag, or by a difference in volume.

As to "power," here are the test numbers for the most recent 100 WPC Class A/B AVR reviewed by Home Theater (Yamaha Aventage RX-A1010):

Five channels driven continuously into 8-ohm loads:
0.1% distortion at 53.5 watts
1% distortion at 65.3 watts

And again, here are the numbers for the Class D SA-XR57:
0.1% distortion at 72.6 watts
1% distortion at 104.6 watts

The Yamaha actually claims 110WPC and does considerably better in two channel mode, but in the real world both are going to perform just fine in a normal home environment, with normally efficient speakers. (BTW, my SA-XR57 fed two Gallo Solo speakers and a powered sub in a roughly 60'+ by 20"+ room with a rather high ceiling, in addition to serving 7.1 duty in my TV room (it had a built-in A-B speaker selector), and I never felt it was short of power.)

And again, these are facts and numbers, not opinions.
Ryan1 is offline  
post #11 of 30 Old 05-21-2012, 08:52 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by Venomous View Post

Pioneer is the only one out there doing it right now.

Rotel also does it, search for the RSX-1562.

But like the Pioneer, it's overly big, heavy and (IMO) overpriced.

Quote:
Originally Posted by DS-21 View Post

Actually, it could sound different than other amps, because it has a fairly high output impedance. So speakers with a very non-linear impedance curve will interact with the amp. That's, I think, why a lot of the tube guys went gaga over them. They genuinely did sound like triodes, because they had similar output impedance.

...

Sure, but a speaker with a very non-linear impedance will interact with any amp that can't handle its dips.

So if a speaker dips to, say 2 ohms at the low end and the amp can't deliver the constant voltage to it, then we may end up with some very hot interaction and one can definitely hear it (or end up hearing nothing).

But that's very different than hearing, under normal conditions, stuff like "bright and dry" or "smooth and airy...."
Ryan1 is offline  
post #12 of 30 Old 05-21-2012, 09:22 PM
Senior Member
 
Leo King's Avatar
 
Join Date: Jun 2008
Location: US
Posts: 398
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Ryan1 View Post

With all due respect, it's not a matter of "opinion."

Amp technology is well understood and as log as an amp measures flat within the audible band and does not distort audibly, then it will sound the same as another amp which does the same.

There is no magic about it, despite what you may read in some reviews. And the ear is easily fooled by preconceived ideas, by a price tag, or by a difference in volume.

As to "power," here are the test numbers for the most recent 100 WPC Class A/B AVR reviewed by Home Theater (Yamaha Aventage RX-A1010):

Five channels driven continuously into 8-ohm loads:
0.1% distortion at 53.5 watts
1% distortion at 65.3 watts

And again, here are the numbers for the Class D SA-XR57:
0.1% distortion at 72.6 watts
1% distortion at 104.6 watts

The Yamaha actually claims 110WPC and does considerably better in two channel mode, but in the real world both are going to perform just fine in a normal home environment, with normally efficient speakers. (BTW, my SA-XR57 fed two Gallo Solo speakers and a powered sub in a roughly 60'+ by 20"+ room with a rather high ceiling, in addition to serving 7.1 duty in my TV room (it had a built-in A-B speaker selector), and I never felt it was short of power.)

And again, these are facts and numbers, not opinions.

With all due respect, it's not all about the numbers on papers, it's also how they sound in real life to the individuals . That's the reason some people love Yamaha receivers, some like Pioneer, NAD or Denon.

30+ years ago, I loved to talk about numbers like you do now.
Leo King is offline  
post #13 of 30 Old 05-21-2012, 09:30 PM
AVS Special Member
 
JHAz's Avatar
 
Join Date: Mar 2009
Posts: 4,042
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 119 Post(s)
Liked: 166
Oh boy. And me without any popcorn.

FWIW the effects of a higher output impedance are visible with low output power, which is to say they have zero to do with running out of power at lower impedances. Stereophile's testing is at 2.83 volts (a watt at 8 ohms) and their simulated speaker load shows small but real FR inacuracies into that load with almost every amp.
JHAz is offline  
post #14 of 30 Old 05-21-2012, 09:39 PM
Senior Member
 
Leo King's Avatar
 
Join Date: Jun 2008
Location: US
Posts: 398
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by JHAz View Post

Oh boy. And me without any popcorn.

Want some popcorn from me? No free movie, though.
Leo King is offline  
post #15 of 30 Old 05-21-2012, 09:56 PM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,963
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 150 Post(s)
Liked: 133
The Panasonic XR series (designed by TI) was an interesting breakthrough when released..
However after about 1-2 years they developed a high frequency oscillation that was known for destroying tweeters..
Seemed to be linked when used with low sensitvity 4 Ohm loudspeakers.

I still use 1 in the garage, works good for background music when working on the Mopars..

Just my $0.02....
M Code is offline  
post #16 of 30 Old 05-21-2012, 10:46 PM
AVS Special Member
 
Venomous's Avatar
 
Join Date: Dec 2004
Location: Huntington Beach/San Diego CA
Posts: 1,198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Quote:
Originally Posted by Ryan1 View Post

Rotel also does it, search for the RSX-1562.

But like the Pioneer, it's overly big, heavy and (IMO) overpriced.

I didn't know Rotel made a class d integrated, pretty cool. The new pioneer elites coming out are going to be overpriced. Not sure if their size remain the same as last generation. Audibly, I heard no diff between the class d and the class a/b AVRs. It's justl odd to see class d's more expensive in home audio compared to how class d dominates car audio with sub amps and are relatively inexpensive.
Venomous is offline  
post #17 of 30 Old 05-21-2012, 11:17 PM
Senior Member
 
OldSchoolMETAL's Avatar
 
Join Date: Nov 2011
Location: Chicago area/Indiana
Posts: 404
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I wish Stereo Review was still around, they actually did tests that actually impacted sound. They tested a receiver/amps impedance out and used reactive loads to test for frequency response since no speaker is a giant resistor that won't interact with the amps/receivers output.

They also test damping factor to see how well the receiver/amp controlled the back motion of the motor (speaker ), and it's ability to keep ringing from turning the sound into mud.

They also tested dynamic output along with RMS output. Being dynamic output in a music signal can generally be 2.5X it's steady state signal and much above with classical or movie sound tracks, dynamic output actually matters much more.

They also tested low impedance receiver/amp output. They test the normal 4 ohm output that many do not, and even tested 3 ohm and 2 ohm output since with a "REAL" reactive sound speaker, it's impedance can drop by 1/2 with certain frequencies, from it's nominal impedance. So an 8 ohm speaker could drop to 4 ohms and a 4 ohm speaker could drop to 2 ohms. Not every loud speaker will drop by 1/2 but very many do.

Sound and vision's and Home Theater mags test results mean nothing in how a receiver will sound. They test frequency response with a resistive, not reactive, load. They do not test dynamic output at all. They do not test 4 ohm output with all channels driven.

Following Ohms Law, if an Amp clips into 4 ohms at 100 watts, to keep it from clipping with 8 ohm speakers that dip into 4 ohm loads with certain frequencies, the amp will only be able to put out 50 watts into 8 ohms.

Being that most of these receivers are already current tapped out with an all channels driven configuration, I would not be surprised if they couldn't even match their 8 ohm power output with a 4 ohm load. Maybe 80% at best.

So if you have an amp that's clipping at 80 watts with 7 channels driven (this would be a mid end to a high end receiver), at 80% of that 80 watts, would give you 64 watts into 4 ohm loads. This would equate to an 8 ohm output of 32 watts/ channel so that the receiver doesn't clip into the loud speaker's 4 ohm dip.

Review Tests are such a disappointment. They offer no real world results but just enough filler for those without the knowledge, to be enticed by them.

While they do offer some insight in how an amp does react, it doesn't tell the whole story.

An amp with a high output impedance, low damping factor, smooth 'resistive' frequency output, and high 8 ohm rms output power, could be the biggest under performing piece of crap receiver out there in terms of sound quality. BUT it will measure extremely good under Sound and Vision's and Home Theater Mags current tests.
OldSchoolMETAL is offline  
post #18 of 30 Old 05-22-2012, 01:09 AM
Senior Member
 
OldSchoolMETAL's Avatar
 
Join Date: Nov 2011
Location: Chicago area/Indiana
Posts: 404
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by Ryan1 View Post

I'd like to correct this, since it's simply misinformation.

This "bright and dry sound" makes no sense from a technology standpoint, nor is it supported by any blind test I've seen. I replaced a relatively expensive Class A amp with the SA-XR57 and it sounded exactly the same....

The SA-XR57 was a great receiver and if wasn't for the fact that I need room correction, I'd still use it.

And its real power rating is more than comparable to current (and much bigger/heavier) A/B receivers:

Five channels driven continuously into 8-ohm loads:
0.1% distortion at 72.6 watts
1% distortion at 104.6 watts

For what it's worth, the SA-Xr57 felt a tad more robust than a recent Marantz SR5006 (Class A/B) I had.

The reality is that Class D hasn't taken off for a variety of reasons:

The manufacturers are happy recycling existing technology designs, which keeps their costs down.

AVRs are already a commodity and Class D can make it very hard to stem that tide, plus it's much harder to romance it (unless it has the B&O pedigree, as the ICE units).

"Real man" buy their amps and receivers by the pound, so selling a light, small unit may work for the pro market, but not for the Best Buy crowd.

This is oversimplifying things only a little bit.

Not impressed with Technic's/Panasonics Digital amp but really liked their H+ class amps. I used to own the Technics SA-GX790. Definitely wasn't a high end amp by today's standards at the time for surround sound, but it did pretty good for Stereo. It's down falls were that it's rear speakers wired internally in series, not parrallel as well as when selecting both A & B main speakers at the same time. It also was a lower current amp as it's output diminished greatly is surround sound. It only had 4 channel amps as the rear speakers were wired in series internally to "1" amp. This did cause frequency response issues. It was a 4.0 Surround Receiver with only Dolby Prologic.
OldSchoolMETAL is offline  
post #19 of 30 Old 05-22-2012, 02:21 AM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by OldSchoolMETAL View Post

...So if you have an amp that's clipping at 80 watts with 7 channels driven (this would be a mid end to a high end receiver), at 80% of that 80 watts, would give you 64 watts into 4 ohm loads. This would equate to an 8 ohm output of 32 watts/ channel so that the receiver doesn't clip into the loud speaker's 4 ohm dip....

And in the real world, none of this would likely matter.

First, the vast majority of people do not use 4 ohm speakers other than in their cars.

Second, that 32 WPC number is rather meaningless (although it can still be reasonably loud), since I can't think of many scenarios where you'd be driving all 7 channels at full blast.

Third, Class D amps can generally handle difficult loads (4 ohm and less) considerably better than most A/B amps, so I am not even sure why we are discussing it as if it's a negative for Class D.

Fourth, the subject of this thread is why there aren't more Class D amps and I've stated my theory, although I'd be curious as to what M Code has to say.

Fourth, I am curious why you are not impressed with the Panasonics (the amps are actually TI, as M Code mentions). Please do not tell me that you find them bright and dry.

Quote:
Originally Posted by M Code View Post

The Panasonic XR series (designed by TI) was an interesting breakthrough when released..
However after about 1-2 years they developed a high frequency oscillation that was known for destroying tweeters..
Seemed to be linked when used with low sensitvity 4 Ohm loudspeakers.
...

Interesting, I didn't know that. I also don't know why people still buy 4 ohm speakers, unless they have some peculiar voltage restrictions....

My SA-XR-57 was rock solid, although the Gallos are rather efficient (and they have CDT tweeters). I also still have a stereo Tripath amp, the 100 WPC AudioDigit AD-CT100-HCLI, which uses the TAA4100 chip and still works great.
Ryan1 is offline  
post #20 of 30 Old 05-22-2012, 04:45 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 1178
Quote:
Originally Posted by donutfan View Post

About 6-8 years ago, surround receivers with digital amplifers were all the rage (btw I know they aren't truly digital, but am using that term because most people are familiar with it).

But now except for some upscale Pioneers for example, most receivers are still using conventional class A/B analog amps.

What happened? They didn't sound acceptable? Too expensive? Couldn't handle lower impedance speaker loads (I saw this complaint a lot)?

Switchmode power amplifiers have a built-in technical issue - the high source impedance that their output filter networks create at high audio frequencies. The long term solution to this problem includes higher switching frequencies, but that requires solid state devices that are as yet too expensive or haven't even been devised. Costs are coming down, but AFAIK a pure analog output stage is still less expensive.

Automated system integration tools such as Audessy would seem to provide an effective solution for the HF source impedance problem by equalizing actual room response. But licensing and providing the DSP power, and the microphone with known response to implement it cost money. There is also a consumer education and acceptance issue.

The first stage of implementation of switchmode technology to audio power amps is in the power supply where there are few if any unsolved problems. However, costs are only just now approaching the break even point.
arnyk is offline  
post #21 of 30 Old 05-22-2012, 05:46 AM
AVS Special Member
 
Zen Traveler's Avatar
 
Join Date: Feb 2005
Location: Texas
Posts: 4,371
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 269 Post(s)
Liked: 371
Quote:
Originally Posted by Ryan1 View Post


...Sure, but a speaker with a very non-linear impedance will interact with any amp that can't handle its dips.

So if a speaker dips to, say 2 ohms at the low end and the amp can't deliver the constant voltage to it, then we may end up with some very hot interaction and one can definitely hear it (or end up hearing nothing).

+1 My speakers are "8 Ohm Compatible" but drop down to 2.8 Ohms in certain frequencies.
Zen Traveler is offline  
post #22 of 30 Old 05-22-2012, 08:28 AM
Senior Member
 
OldSchoolMETAL's Avatar
 
Join Date: Nov 2011
Location: Chicago area/Indiana
Posts: 404
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by Zen Traveler View Post

+1 My speakers are "8 Ohm Compatible" but drop down to 2.8 Ohms in certain frequencies.

Yes, but he either failed to read that in my response or doesn't understand the science because loud speakers are not resistors of a constant
Ohm measurement.
OldSchoolMETAL is offline  
post #23 of 30 Old 05-22-2012, 04:59 PM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 247
Quote:
Originally Posted by Ryan1 View Post

Rotel also does it, search for the RSX-1562.

But like the Pioneer, it's overly big, heavy and (IMO) overpriced.

Agreed, though if Pioneer had modern room correction (say, ARC or Trinnov) I would likely have bought one.

Quote:
Originally Posted by Ryan1 View Post

Sure, but a speaker with a very non-linear impedance will interact with any amp that can't handle its dips.

While others have hopefully clarified your understanding of the relationship between amplifier output (aka source) impedance and frequency response, I think some graphs of different amps into Stereophile's "simulated speaker load" (NHT?) may nonetheless be instructive.

The dashed line is the FR of a really crappy amp, a "high end" solid state model, with high output (aka source) impedance, the Theta Digital Intrepid.



And here's the same graph for a competently-designed amp from Bryston (black line):



On a lot of speakers, those amps will sound different because they will have different frequency responses in the midrange.

For comparison, here's an ICEpower amp from Bel Canto:



And a conrad johnson tube amp:



Quote:
Originally Posted by Ryan1 View Post

But that's very different than hearing, under normal conditions, stuff like "bright and dry" or "smooth and airy...."

Actually, those kind of descriptions generally correlate to small differences in level or small FR changes.

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
post #24 of 30 Old 05-23-2012, 05:58 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by DS-21 View Post

...

Actually, those kind of descriptions generally correlate to small differences in level or small FR changes.

Hm, not so sure about this. I do understand the relationship between impedance and frequency response, but I also understand that it rarely makes a practical difference, at least not with well designed speakers and amps.

These are the same kind of graphs that are used to sell four figures speaker cables to the more gullible among us.

Sure there are frequency response changes in the graphs, but whether they are audible is entirely another matter. The ability of the human ear to hear FR deviations varies depending on the particular frequency, but my understanding is that generally in speaker design +/- 3dB FR deviations are considered to be at the threshold of perception.

The graphs from Stereophile (which is, BTW, an all too frequent purveyor of all sorts of audiophile mythology used to sell the product appearing on its pages) appear to be within such tolerances and should be inaudible, with the likely exception of the Conrad Johnson.

This jives well the inability to tell apart different (but designed within spec) amps in double blind tests.

BTW, The Audio Critic (which is a much, much more trustworthy publication than Stereophile, IMO) uses the PowerCube test, which does something similar to what you see in the Stereophile graphs.
Ryan1 is offline  
post #25 of 30 Old 05-23-2012, 06:52 PM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 247
Quote:
Originally Posted by Ryan1 View Post

Hm, not so sure about this. I do understand the relationship between impedance and frequency response, but I also understand that it rarely makes a practical difference, at least not with well designed speakers and amps.

Well, that's a bit of a tautology, because a "well designed amp" will have very low output impedance. In the real world, many amps (including the Panny XR55) do not.

Furthermore, pause to consider for a moment how few "well designed speakers" exist in consumer audio. If it has a tweeter with a 4" faceplate mounted in a 180deg waveguide (i.e. flushed into a baffle) it ain't well designed.

Quote:
Originally Posted by Ryan1 View Post

These are the same kind of graphs that are used to sell four figures speaker cables to the more gullible among us.

That's simply not rational.

Differences in FR are audible. That's why controlled listening tests demand a 0.1dB level match. If you have one amp that has an FR variance into a given pair of loudspeakers of less than 0.2dB over the midrange (and less elsewhere) and another amp that has an FR swing into those same loudspeakers of over 0.5dB, a good listener may well be able to distinguish them.

Low output impedance, furthermore, is not something only found on the Brystons and McIntoshes of the world. Any well-designed Class AB chipamp or discrete amp amplifier will have very low output impedance. Many tube amps will not. Many Class D amps will not.

Do you remember "Sideshow Bob" Carver's original Sunfire amps?

His tagline for them was something like "a 300-Watt amp with the soul of a 9-Watt triode." They had two taps. They sounded different on many speakers. Connect to one of them, and it behaved a lot like a badly-designed tube amp. (There are also good tube amps, remember!) The difference was just that the crappy-sounding tap had a resistor that raised the output impedance.

Quote:
Originally Posted by Ryan1 View Post

The ability of the human ear to hear FR deviations varies depending on the particular frequency, but my understanding is that generally in speaker design +/- 3dB FR deviations are considered to be at the threshold of perception.

The first part is true, the second part is widely off the mark.

Quote:
Originally Posted by Ryan1 View Post

The graphs from Stereophile (which is, BTW, an all too frequent purveyor of all sorts of audiophile mythology used to sell the product appearing on its pages) appear to be within such tolerances and should be inaudible, with the likely exception of the Conrad Johnson.

Note that I never claimed the amps for which I posted graphs would always be found to be audibly different.

It depends on the load presented by the speaker.

But also, don't throw the baby out with the bathwater. Stereophile contains a lot of voodoo idiocy, yes. It also contains a great deal of useful data on how audio gear behaves. More than any other publication of which I know. I recommend you check out my article, How to Read Stereophile Without Wanting to Throw it Across the Room.

Quote:
Originally Posted by Ryan1 View Post

This jives well the inability to tell apart different (but designed within spec) amps in double blind tests.

Note that there have been plenty of valid amp tests (controlled, blind) showing differences. See, for instance, here.

Quote:
Originally Posted by Ryan1 View Post

BTW, The Audio Critic (which is a much, much more trustworthy publication than Stereophile, IMO) uses the PowerCube test, which does something similar to what you see in the Stereophile graphs.

The PowerCube is an interesting measurement. I wish all reviews of amps had a PowerCube measurement. But it measures peak voltage under various loads and phase angles. That is fundamentally different from measuring FR into a simulated loudspeaker load.

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
post #26 of 30 Old 05-23-2012, 07:11 PM
AVS Special Member
 
YeuEmMaiMai's Avatar
 
Join Date: Apr 2002
Posts: 1,076
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Ryan1 View Post

With all due respect, it's not a matter of "opinion."

Amp technology is well understood and as log as an amp measures flat within the audible band and does not distort audibly, then it will sound the same as another amp which does the same.

There is no magic about it, despite what you may read in some reviews. And the ear is easily fooled by preconceived ideas, by a price tag, or by a difference in volume.

As to "power," here are the test numbers for the most recent 100 WPC Class A/B AVR reviewed by Home Theater (Yamaha Aventage RX-A1010):

Five channels driven continuously into 8-ohm loads:
0.1% distortion at 53.5 watts
1% distortion at 65.3 watts

And again, here are the numbers for the Class D SA-XR57:
0.1% distortion at 72.6 watts
1% distortion at 104.6 watts

The Yamaha actually claims 110WPC and does considerably better in two channel mode, but in the real world both are going to perform just fine in a normal home environment, with normally efficient speakers. (BTW, my SA-XR57 fed two Gallo Solo speakers and a powered sub in a roughly 60'+ by 20"+ room with a rather high ceiling, in addition to serving 7.1 duty in my TV room (it had a built-in A-B speaker selector), and I never felt it was short of power.)

And again, these are facts and numbers, not opinions.

Just to add to the above
Yamaha specs are given in multiple formats: each channels rated capacity + 2 channel actual capacity.

You can find Yamaha's a1010 specs below. The bolded one is the important one as it tells you total power available to the 7 amps....

Rated Output Power (1kHz, 2ch driven) 120W (8 ohms, 0.9% THD)
Rated Output Power (20Hz-20kHz, 2ch driven) 110W (8 ohms, 0.06% THD)


Dynamic Power per Channel (8/6/4/2 ohms) 135/165/210/280 W


34W per channel (7 channels driven)

48W per chanel (5 channels driven)

Obviously Yamaha under rates their stuff slightly as the really have a total of 270W of amp power available based upon what that unit was measured at.

http://usa.yamaha.com/products/audio..._u/?mode=model

Strong or weak in the end we are all dead
YeuEmMaiMai is offline  
post #27 of 30 Old 05-23-2012, 09:57 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by DS-21 View Post

...
That's simply not rational.

Differences in FR are audible. That's why controlled listening tests demand a 0.1dB level match. If you have one amp that has an FR variance into a given pair of loudspeakers of less than 0.2dB over the midrange (and less elsewhere) and another amp that has an FR swing into those same loudspeakers of over 0.5dB, a good listener may well be able to distinguish them.

...

You are mixing up audible Volume/Amplitude level deviations (which are indeed considered audible at 0.1dB variance) and Frequency Response deviations.

Frequency Response variance is generally considered fine at +/-3dB (that's why on the back of many speakers you'll find just such qualification to the stated frequency response of the box).

A quick search will find plenty of confirmation for the above statement, such as this excerpt from Polk's Paul DiComo:

"A big improvement would be a frequency response number that also includes the amplitude tolerance, expressed as "XHz-YkHz +/- 3dB." This tells you that the amplitude of the speaker's response relative to frequency does not deviate more than 3 Decibels from the center line. The "plus or minus 3dB" spec is regarded as a standard of sorts. The theory is that 3dB differences are "just perceptible," so a speaker whose response curve lies within that tolerance window is a reasonably accurate speaker."

I have seen claims that Frequency Response deviation of +/-1dB can be audible, but even this, I believe, applies to only a fairly narrow frequency bandwidth, corresponding roughly to human speech (while for instance low Frequency Response deviations may not audible even at much greater deviations).

Quote:
Originally Posted by DS-21 View Post

...
Note that there have been plenty of valid amp tests (controlled, blind) showing differences. See, for instance, here.
...

I beg to disagree.

Your ABX tests link, of which I am well aware, shows that people cannot tell the difference between vastly different solid state amps (even though most are about 3-4 decades old now, the Dynaco Stereo 400 was built in the 1970s), but they can tell the difference between olid state and wacky tube amps (oh, and the 1980 solid state Audio Research D120, which the footnote specifically states "was unstable when clipped, which proved audible.")
Ryan1 is offline  
post #28 of 30 Old 05-23-2012, 11:54 PM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 247
Quote:
Originally Posted by Ryan1 View Post

You are mixing up audible Volume/Amplitude level deviations (which are indeed considered audible at 0.1dB variance) and Frequency Response deviations.

I'm not mixing up anything.

What do you think frequency response is, except for variance of level over frequency?

Quote:
Originally Posted by Ryan1 View Post

Frequency Response variance is generally considered fine at +/-3dB

No.

Furthermore, there's a material difference between ±3dB with a smooth middle decade (100-10k Hz) and rolled off extremes, and ±3dB with a roller-coaster midrange. The human ear is much more sensitive to midrange variance than it is to extreme bass or treble variance.

So a ±3dB tolerance rating doesn't tell us anything useful. And certainly nothing material to amplifiers unless they are extraordinary incompetent in design.

Or, unless they roll off very early. For instance, the tweakoid voodoo website "6moons" got its start singing the praises of this $30 plastic "T-amp." Which, to be fair, measured and sounded a lot like a very expensive but bad SET: high output impedance, extremely early bass rolloff - an F3 of like 70Hz - and so on. That amp certainly sounded different from a "mid-fi" solid state amp, because it was a manifestly lower-fidelity part!

Also, should Polk ever decide to release a competently-designed loudspeaker, I might pay more attention to what their people have to say.

Quote:
Originally Posted by Ryan1 View Post

Your ABX tests link, of which I am well aware, shows that people cannot tell the difference between vastly different solid state amps (even though most are about 3-4 decades old now, the Dynaco Stereo 400 was built in the 1970s), but they can tell the difference between olid state and wacky tube amps (oh, and the 1980 solid state Audio Research D120, which the footnote specifically states "was unstable when clipped, which proved audible.")

Let's be less sloppy with our words, OK?

Nobody has yet demonstrated an ability to distinguish between two nonbroken amps that have low noise, flat frequency response, low output impedance, and low distortion, and are operating within their design parameters in any listening test by their ears alone.

When any one of those conditions is not met, then there may be audible differences between amps. And most Class D amps (I believe the Hypex designs are an exception) have relatively high output impedance.

You should take note of the words of Arny Krueger above. Mr. Krueger was one of the main forces behind ABX audio equipment testing. Here's what he wrote in this thread:

Quote:
Originally Posted by arnyk View Post

Switchmode power amplifiers have a built-in technical issue - the high source impedance that their output filter networks create at high audio frequencies. ***
Automated system integration tools such as Audessy would seem to provide an effective solution for the HF source impedance problem by equalizing actual room response. ***

(And he's entirely right: the FR problems caused by output, aka source, impedance are eminently solvable with EQ. Even automated routines.)

My point is simply that you're a little bit too cavalier in dismissing sonic differences that are plausible, just because so many claims of sonic differences between audio parts are utter nonsense and frankly cannot be true. It's an understandable reaction, to some extent, but it's not consistent with the available data. I can set up a blind test (assuming level matching at 1kHz, rather than broadband EQ) that will show differences between an XR55 and a standard Class AB amp. I can also set up a blind test that does not. Without clipping or current-limiting being the issue. It's trivial: just pick a speaker with a high impedance and a lot of variation in impedance.

You also seem to have missed above that I noted I was personally unable to tell the difference between my Panasonic XR55 and a very tall stack of high-performance for the day separates. The material factor there was that my front three speakers at the time (KEF Q15/Q95C) had fairly benign impedance curves. I used the XR55 for about 3 years. I switched only because of room correction, advanced loudness compensation programs, and the advent of digital SACD/DVD-A multichannel music transmission over HDMI (as opposed to a snake's nest of analog interconnects). My old XR55 currently drives the in-wall speakers in my sister's house. Still going strong, at 7-odd years old!

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
post #29 of 30 Old 05-26-2012, 06:54 PM
AVS Special Member
 
Ryan1's Avatar
 
Join Date: Apr 2001
Posts: 1,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 36
Quote:
Originally Posted by DS-21 View Post

...

What do you think frequency response is, except for variance of level over frequency?
...
The human ear is much more sensitive to midrange variance than it is to extreme bass or treble variance.

So a ±3dB tolerance rating doesn't tell us anything useful. And certainly nothing material to amplifiers unless they are extraordinary incompetent in design.
...
Nobody has yet demonstrated an ability to distinguish between two nonbroken amps that have low noise, flat frequency response, low output impedance, and low distortion, and are operating within their design parameters in any listening test by their ears alone.

When any one of those conditions is not met, then there may be audible differences between amps. And most Class D amps (I believe the Hypex designs are an exception) have relatively high output impedance.
....
You also seem to have missed above that I noted I was personally unable to tell the difference between my Panasonic XR55 and a very tall stack of high-performance for the day separates. The material factor there was that my front three speakers at the time (KEF Q15/Q95C) had fairly benign impedance curves. I used the XR55 for about 3 years. I switched only because of room correction, ....

He, it seems we actually agree on most things (particularly on the part I have bolded in the quote above), just use different words....

Of course I am speaking of non-broken, well designed amps, which are operating within their design parameters, with equally non-broken, well designed speakers. There are always exceptions, some at the very high end "audiophile" end, but then people who like "tube sound" are not looking for amplification neutrality. BTW, ClassD amps seem to have inherently low output impedance (one of the reasons they are found powering subs).

Speaking of things we have in common, I see that apart from being a Panasonic SA-XR5X fan, you are also a Trinnov fan, which I have recently become. I just wish the R-972 was a Class D box, instead of a 50lbs behemoth....

Peace, brother.
Ryan1 is offline  
post #30 of 30 Old 05-27-2012, 10:23 AM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 247
Quote:
Originally Posted by Ryan1 View Post

BTW, ClassD amps seem to have inherently low output impedance (one of the reasons they are found powering subs).

This is your error. Typical Class D amps don't have inherently low output impedance. Rather, due to the necessary low-pass filter (output filter), they have variable output impedance, rising with frequency. (Some of the Hypex designs may be an exception.) So the HF response is going to be variable, based on the impedance in the treble of the speakers used. And that is audible. One can fix that kind of problem with EQ, which is why Arny suggested Class D amps are best for closed systems such as active speakers, or systems that auto calibrate using full-band room correction.

As for subs, Class D amps are often used simply because they're cheaper (output filter not an issue if you're already filtering them at sub 200Hz) and swing a whole lot of voltage and current efficiently. But the designs used for sub amps typically would sound like garbage (because of the corner of the output filter) when used beyond the bass.

And yes, Trinnov is effing awesome. I've used all of the serious RC technologies available in AVR's now (had two boxes with Audyssey MultEQ XT, currently have an Anthem AVR with ARC in the main system, and an R-972 with Trinnov powering both a bedroom 2.1-channel system and a nearfield 5.1-channel system, and providing correction for both systems) Trinnov is the first system that got the mains/sub integration right. With both Audyssey and ARC, I've had to do some post-hoc changes to get it audibly right. A box that combined the best features available modern AVR's: Sherwood's Trinnov room correction, Anthem's QuickMeasure ultility, Class D amps, and the XR55's form factor (combined with lack of bugs and a classy appearance) would be a box I'd pay a lot of money to deploy.

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
Reply Receivers, Amps, and Processors



Forum Jump: 

Posting Rules