Class D: Turns Out it Does Suck!

[sandyk]

8 minutes ago, diecaster said:

 

What do you have against DSD? Most of my music is Redbook but the DSD I do have sounds great on my 2 channel setup.  

 

 Perhaps he is suggesting that DSD/SACD  HF residuals may upset our tweeters just like his Class D modules' 460kHZ residuals  do ?

[diecaster]

2 minutes ago, sandyk said:

 

 Perhaps he is suggesting that DSD/SACD  HF residuals may upset our tweeters just like his Class D modules' 460kHZ residuals  do ?

 

That would not make any sense. There is a big difference between moving noise way up in the frequency range and having an amp that pumps out high frequency noise. 

[wgscott]

27 minutes ago, diecaster said:

 

What do you have against DSD? Most of my music is Redbook but the DSD I do have sounds great on my 2 channel setup.  

 

Nothing, really.  But has a big noise hump that starts around 50kHz or so.

 

Presumably your system, like mine, is unaffected by that frequency.

 

Neither my ears, my speakers or my dogs are sensitive to this.

 

[wgscott]

11 minutes ago, diecaster said:

 

That would not make any sense. There is a big difference between moving noise way up in the frequency range and having an amp that pumps out high frequency noise. 

 

What difference does it make, from the point of view of the speakers (and listener)?

[diecaster]

30 minutes ago, wgscott said:

 

What difference does it make, from the point of view of the speakers (and listener)?

 

It’s a huge difference. That Class D amp is making its own noise at that high frequency and putting it out at about -20dB.  My Class A/B tube amp has rolled off by at least 3dB by 50kHz. Even if the noise was at -60dB, that puts it at -63dB and it is more likely at lower than -80dB. 

[Audiophile Neuroscience]

1 hour ago, diecaster said:

 

IThat Class D amp is making its own noise at that high frequency and putting it out at about -20dB.  My Class A/B tube amp has rolled off by at least 3dB by 50kHz. Even if the noise was at -60dB, that puts it at -63dB and it is more likely at lower than -80dB. 

 

can you characterize the sound of your A/B tube amp compared to your Pioneer solid state Class A amp compared to say any Class D amp or is it not that simple. Might a well implemented class D sound better than a class A or do you feel there are inherent and audible limitations?

I use a pure class A amp which also has variable bias settings for A/B operation. I've never heard a class D amp that I like but not saying they cannot sound fantastic.

[firedog]

Sorry, these last few pages are mostly nonsense.

Techno gobbledy-goop for people trying to turn their personal taste in amps into some kind of objective truth. No decent Class D amp is harmful to anything and doesn't distort or mess with tweeters more than any other amp.

 

Here's the bottom line:

Lots of people - including pro reviewers, experienced audiophiles, musicians, etc., think the better ClassD amps sound fine and are a bargain - high level sound at more reasonable prices than many more expensive conventional amps that sound no better. They call the sound accurate and clean.

Other people think the same amps sound "clinical".

 

It's just taste. Audiophiles will never get consensus on any component. One man's "warm and natural" is another guy's "euphonic distortion"; one guy's "clinical" is another one's "accurate and realistic".

[esldude]

14 minutes ago, Audiophile Neuroscience said:

 

can you characterize the sound of your A/B tube amp compared to your Pioneer solid state Class A amp compared to say any Class D amp or is it not that simple. Might a well implemented class D sound better than a class A or do you feel there are inherent and audible limitations?

I use a pure class A amp which also has variable bias settings for A/B operation. I've never heard a class D amp that I like but not saying they cannot sound fantastic.

I'll give my subjective impressions though you weren't asking me. 

 

Class D when good sounds fast and powerful.  And it doesn't seem to care if you use very nearly all of that power as the sound quality stays consistently clean at all levels. It is capable of depth and space when its on the recording.  It doesn't create it however leading some to think it clinical. 

 

Spectral amps have the speed of class D and maybe more.  They sound powerful and relatively unaffected by how much you use, but not quite to the extent Class D does. 

 

Good class A amps shouldn't care about power used, but they do seem to fog up just a bit when much of it is used. Sometimes that fog is heard as additional space and dimensionality.  They don't fall apart the way some class A/B amps do when pressed into using lots of the available power.  Some of the finest A/B amps don't leave much to choose vs class A. 

 

Push-pull tube amps are a different thing altogether much more effected by what they are connected to.  You'll just have to listen and hear it for yourself. 

[sandyk]

6 minutes ago, esldude said:

but they do seem to fog up just a bit when much of it is used. Sometimes that fog is heard as additional space and dimensionality.

 The only way that is likely to happen is when they are almost at their designed power specifications and there is a large increase in distortion products as they enter clipping. This is similar to what you may hear with some digital gear where distortion or low level wideband noise appears to result in false treble.

[esldude]

1 minute ago, sandyk said:

 The only way that is likely to happen is when they are almost at their designed power specifications and there is a large increase in distortion products as they enter clipping. This is similar to what you may hear with some digital gear where distortion or low level wideband noise appears to result in false treble.

As someone with ESL's, I can probably manage that with most amps.  I'm rather sensitive to even a little bit of actual clipping however.  I don't drive my amps to clipping.  I hear others do it all the time without noticing and it drives me up the wall. 

 

I once had an SS McIntosh amp with the Power Guard circuitry.  A 752 which is 75 wpc.  It was marginal for my Acoustat speakers.  Power Guard was a comparator which would activate if distortion reached .5% and was optically coupled into the amp circuitry to reduce peaks and prevent actual clipping.  Driven ridiculously hard it would act as a compressor. 

 

If you pushed it some getting the indicators to illuminate it would fog up some, and you could manipulate volume to get a handle on that sound.  This minor fogging if you barely activated the circuit is rather similar to what I hear with class A amps under heavy load.  And not dissimilar to powerful push pull tube amps under heavy load.  

 

So the fog I am referring to is not like the wideband brightening of poor digital gear.  Nor an amp driven to actual clipping. 

[Audiophile Neuroscience]

4 minutes ago, esldude said:

I'll give my subjective impressions though you weren't asking me. 

 

Class D when good sounds fast and powerful.  And it doesn't seem to care if you use very nearly all of that power as the sound quality stays consistently clean at all levels. It is capable of depth and space when its on the recording.  It doesn't create it however leading some to think it clinical. 

 

Spectral amps have the speed of class D and maybe more.  They sound powerful and relatively unaffected by how much you use, but not quite to the extent Class D does. 

 

Good class A amps shouldn't care about power used, but they do seem to fog up just a bit when much of it is used. Sometimes that fog is heard as additional space and dimensionality.  They don't fall apart the way some class A/B amps do when pressed into using lots of the available power.  Some of the finest A/B amps don't leave much to choose vs class A. 

 

Push-pull tube amps are a different thing altogether much more effected by what they are connected to.  You'll just have to listen and hear it for yourself. 

 

I'm not sure what you mean by fog up Dennis. It certainly doesn't seem to suggest a positive trait despite maybe resulting in extra space and dimensionality.

 

My experience of many A/B amps is that they sound a little leaner and can impart an impression of more etch/outline and detail leading to descriptions of clinical if you don't like it and spacious ,detailed an open if you do like it. It can also give the impression of more rhythmic drive I think if transients are crisply reproduced

 

Many Class A otoh sounds often warmer and fuller relatively speaking sometimes described as more musical if you like it and less detailed if you dont.Sometimes being described as more "tube-like"

 

So its all subjective obviously. You may value different qualities or 'colors' over others

 

These are also definitely just generalizations or overall character impressions and so I have no fixed belief one amp will be better than another (except arguably with expectation bias).I suspect it all comes down to implementation and quality of components,  etc

 

My current amp has massive power reserves (500,000uf toroidal transformers) and for my taste is the best I have heard to date at just sounding neutral, getting out of the way of the music, imparting no signature.I have no idea if that has anything to do per se with how it 'uses power'.

 

 

[esldude]

19 minutes ago, Audiophile Neuroscience said:

 

I'm not sure what you mean by fog up Dennis. It certainly doesn't seem to suggest a positive trait despite maybe resulting in extra space and dimensionality.

 

My experience of many A/B amps is that they sound a little leaner and can impart an impression of more etch/outline and detail leading to descriptions of clinical if you don't like it and spacious ,detailed an open if you do like it. It can also give the impression of more rhythmic drive I think if transients are crisply reproduced

 

Many Class A otoh sounds often warmer and fuller relatively speaking sometimes described as more musical if you like it and less detailed if you dont.Sometimes being described as more "tube-like"

 

So its all subjective obviously. You may value different qualities or 'colors' over others

 

These are also definitely just generalizations or overall character impressions and so I have no fixed belief one amp will be better than another (except arguably with expectation bias).I suspect it all comes down to implementation and quality of components,  etc

 

My current amp has massive power reserves (500,000uf toroidal transformers) and for my taste is the best I have heard to date at just sounding neutral, getting out of the way of the music, imparting no signature.I have no idea if that has anything to do per se with how it 'uses power'.

 

 

I've told before of doing serial amp testing.  One amp is loaded with power resistors and feeds the amp on the speakers.  Few amps are without character.  I've actually not done that test with any class D.  Spectrals can be inserted or removed and you will not know it.  According to the Swedish AES some Brystons can manage the trick.  Usually you won't need blind testing the differences are clear sighted as long as levels are matched. 

 

If some audio club wanted a great meeting, doing some serial testing would be a very neat and educational thing to do.  Get as clean and transparent an amp on the speakers they can manage and feed it with several different amps set to unity gain by the output loading.  Switch them in and out.  I like to know the results of doing that with some of the Hypex based amps.  Could be they are undetectable. 

[sandyk]

35 minutes ago, esldude said:

So the fog I am referring to is not like the wideband brightening of poor digital gear.  Nor an amp driven to actual clipping. 

 Dennis

 The attached is what I am referring to with the large increase in distortion just BEFORE it actually reaches clipping.

I have never experienced what you are referring to, but then I never use any amplifier close to maximum power into a low impedance or difficult load as some Electrostatics can be, also with quite a high capacitance IIRC.

Alex

[esldude]

1 minute ago, sandyk said:

 Dennis

 The attached is what I am referring to with the large increase in distortion just BEFORE it actually reaches clipping.

 

Alex

I wouldn't call 1% or less a large increase.  But I think we are thinking of the same thing having seen your latest post.  Yes, that is about where fogginess comes in.  Class A if much good does this smoothly enough on peaks it isn't going to stick out like a sore thumb and can enhance some aspects for some people's taste. 

[sandyk]

6 minutes ago, esldude said:

I wouldn't call 1% or less a large increase.  But I think we are thinking of the same thing having seen your latest post.  Yes, that is about where fogginess comes in.  Class A if much good does this smoothly enough on peaks it isn't going to stick out like a sore thumb and can enhance some aspects for some people's taste. 

 Dennis

 With the amplifier that I showed, it is also possible by increasing the supply voltage rails and markedly improve the PSU current capabilities to do up to 30W in Class AB without the huge increase in distortion shown.

[esldude]

2 minutes ago, sandyk said:

 Dennis

 With the amplifier that I showed, it is also possible by increasing the supply voltage rails and markedly improve the PSU current capabilities to do up to 30W in Class AB without the huge increase in distortion shown.

I owned a Pass Aleph 3 at one time.  30 watts class A, not capable of anything else.  Single ended FET. 

 

It could sound very, very good with a speaker it could handle.  It sounded like class B with most speakers most people might own.  Forget anything with ports including a little Radio Shack Linaeum LX5 speaker.  With Quads it was pretty good if you could contain yourself on the volume.  A good match with Quad 57s too.  Fantastic on K-horns or LaScalas.  Maggies were a bit too much drain on the limited current capability.  This same idea might be the best headphone amp one could imagine repackaged for that use.  Of course I've used a couple SET's on phones and find them good for that purpose.  But I don't like listening over phones. 

[Shadders]

14 hours ago, wgscott said:

I may have mis-understood too, but even if it is not attenuated at all, no speakers that I have ever seen (let alone mine or anything else commercially available) will see that frequency, let alone respond to it.

Hi,

The crossover of the speaker will not be a bandpass for the tweeter, it will be purely a high pass filter. There will be no inductors in series with the tweeter, so the -15dB (as per the graphs on the website tests) is applied via the cable to the tweeter.

 

Whether the tweeter can convert this into a sound wave is not the issue. The question is, will the tweeter be affected by the significant energy at 460kHz ???

 

Is this what characterises the sound of class D in general - the high frequency carrier applied to the tweeter ???

 

if you use a 22kHz measurement bandwidth filter on a good class A, or class A/B amplifier - it will have the same THD as class D - essentially the noise floor of the amplifier.

 

So, if frequencies above 22kHz are not heard - and the THD for the class A/B and class D are the same, then why are they reported different sounding ???

 

Recall Bob Carver experiment - he emulated a known amplifier with a generic amplifier by setting the specific order harmonics of the generic amplifier to be the same as the proposed amplifier in question. The expert reviewers could not tell the difference between the two. So his experiment indicates that it is the distortion profile that gives an amplifier its specific sound.

 

Again, if both class A/B and class D amplifiers have the same THD profile from 20Hz to 20kHz using the 22kHz measurement bandwidth, then why are they reported to sound different ???, and class D all sound similar ???

 

Regards,

Shadders.

[NOMBEDES]

I had a "higher end" (for whatever that is worth,) German made,  AVM class D amp.  Sounded great to me.  Of course I can not hear anything over 8Khz,   So I guess I am lucky.  In fact, most amps sound the same to me.  But then why listen to an amp when you should be listening to the music.

[Ralf11]

12 hours ago, esldude said:

I'll give my subjective impressions though you weren't asking me. 

 

Class D when good sounds fast and powerful.  And it doesn't seem to care if you use very nearly all of that power as the sound quality stays consistently clean at all levels. It is capable of depth and space when its on the recording.  It doesn't create it however leading some to think it clinical. 

 

Spectral amps have the speed of class D and maybe more.  They sound powerful and relatively unaffected by how much you use, but not quite to the extent Class D does. 

 

Good class A amps shouldn't care about power used, but they do seem to fog up just a bit when much of it is used. Sometimes that fog is heard as additional space and dimensionality.  They don't fall apart the way some class A/B amps do when pressed into using lots of the available power.  Some of the finest A/B amps don't leave much to choose vs class A. 

 

Push-pull tube amps are a different thing altogether much more effected by what they are connected to.  You'll just have to listen and hear it for yourself. 

 

Thanks!

 

Any thoughts on the PS Audio Class D gain cell (non-Class D initial or input stage) sound?

 

esp. if it get fed by a tubed pre-amp (ARC LS25 Mk II)....

[mansr]

2 minutes ago, Ralf11 said:

Any thoughts on the PS Audio Class D gain cell

I am inherently suspicious of anything marketed with made-up terminology.

[esldude]

4 minutes ago, Ralf11 said:

 

Thanks!

 

Any thoughts on the PS Audio Class D gain cell (non-Class D initial or input stage) sound?

 

esp. if it get fed by a tubed pre-amp (ARC LS25 Mk II)....

Haven't heard the gain cell.

As I recall the ARC was pretty clean. With just a touch of lit up from within mid-range.

[Ralf11]

ah yes - that is the warm tubey goodness (tho there is less phaat in the newer ones)

[esldude]

Looking at stereophile measures the Gain cell is the PSAudio input stage feeding an ice power amp. The only obvious differences are more noise than ice power amps usually have.

[esldude]

Was not thinking of warmness on the ARC. More a slight highlighted mid-range with good extended lows and full range treble.

[wgscott]

11 hours ago, Shadders said:

 

Again, if both class A/B and class D amplifiers have the same THD profile from 20Hz to 20kHz using the 22kHz measurement bandwidth, then why are they reported to sound different ???, and class D all sound similar ???

 

Price tag.