Class D: Turns Out it Does Suck!

[Ralf11]

36 minutes ago, GUTB said:

euphonic glow

 

 

...

[cjf]

The sound of many Class D amps need to be looked/listened to from a different angle than what many people are used to. The beauty of a "stale/neutral" sounding amp is that it allows for much greater flexibility in your upstream/downstream component selection that will be paired with the Class D amp.

 

Being able to add flavor before or after the amp knowing that the amp itself is your known good reference in terms of neutrality is very useful when it comes to fine tuning the sound of your system to suit your own tastes

 

The final sound of a system can become very unpredictable when you have flavored components, some very heavily seasoned, at each position before, at and after the amps. What you end up with is a crap shoot and anybodies guess usually resulting to additional component changes and more dollars spent to get what you want.

 

I've owned several Class D amps over the last 5yrs or so. Some of which are considered the best available right now and cost in the 5 figure range. I greatly enjoyed my time with them. IME, The one place the Class D amps that I have used still fall short is their ability to play nice/cope with other components who's load and or behavior is less than nice and or perfect. IMO, It seems that some of these amps are only tested via ideal computer simulations in a lab or with a small handful of friendly partnering components and not so much with real world pairing possibilities that are much less friendly. If your using other equipment that is fairly friendly (probably 90% of what is available falls into this classification) then you will have nothing to worry about.

 

I've written at length about my various experiences and experiments here and elsewhere while trying to find a happy combo using the Class D amps I've owned. I eventually had success but the road was long and expensive. I've since moved on to a seemingly immune to anything bad set of Pass Labs XA Mono's and intend to take them to my grave.

[Ralf11]

The constant switching of Class D made me ‘Mildly Nauseous’ About the Electrons.  But It’s Not Enough to really worry about.

 

 

[STC]

6 hours ago, cjf said:

IME, The one place the Class D amps that I have used still fall short is their ability to play nice/cope with other components who's load and or behavior is less than nice and or perfect. IMO, It seems that some of these amps are only tested via ideal computer simulations in a lab or with a small handful of friendly partnering components and not so much with real world pairing possibilities that are much less friendly.

 

IME, I find Class-D CrownXLS is better at driving the difficult ESL Sound Lab than my other audiophile approved Amp. I am not naming the amp because mine was rather outdated and the newer ones probably are probably better. Having said that, I was told that Sound Lab uses Crown as their in-house Amp, a fact I came to know after I acquired the Crown amp which I bought to check whether a 300 Watter would be a better replacement to my 250W Amp.

[STC]

Little bit of info from Audio Precision about square wave and their importance. Although, I am not sure why they measured with 4 volt input for receiver. Isn't that too high? 

 

Square Wave

[semente]

15 hours ago, gmgraves said:

 

 

Without knowing the conditions, one can only make general comments: The top one shows a sharp high frequency peak in the treble with good and symmetrical slew rate (rise and fall times). The bottom one shows a broader peak with slower but still relatively symmetrical slew. Other than that, we have no idea of the frequency at which the snapshots were made, or what part of which kind of circuit, these square waves represent. If you don't understand what technical measurements mean, your better-off paying no attention to the measurements in a review and just read the subjective evaluations. 

 

If he doesn't understand what technical measurements mean he should learn.

 

Subjective evaluations are...er...subjective(ly meaningless)?

[semente]

12 hours ago, esldude said:

As long as we are posting deceptive squarewave graphs out of context, I offer this small signal 10 khz square wave:

 

 

Thoress 300B monoblocks which according to JA:

 

Overall, the Thöress 300B monoblock measures well for a SET amp.

 

It manages almost (not quite) 10 watts at 3% THD.  So I am not sure it could do anything other than a small signal square wave.  Price is only $12,995 per pair. I guess these don't suck like some stupid class D designs. 

 

They didn't blow much of the budget on beautiful build quality either. Kept it where it counts....producing the kind of special distortion people crave. 

 

 

 

Yet at 3m with @manisandher speakers you'll get 0.2% THD+N at 95.5dB.

Not too bad.

 

The problem is that some people believe it's possible to use flea-power S.E.T.s with speakers spec'ing less than 100dB sensitivity...

 

R

[esldude]

3 hours ago, STC said:

 

IME, I find Class-D CrownXLS is better at driving the difficult ESL Sound Lab than my other audiophile approved Amp. I am not naming the amp because mine was rather outdated and the newer ones probably are probably better. Having said that, I was told that Sound Lab uses Crown as their in-house Amp, a fact I came to know after I acquired the Crown amp which I bought to check whether a 300 Watter would be a better replacement to my 250W Amp.

I have heard quite a variety of amps on my or other people's Soundlabs.  The second best at driving them with aplomb was a bridged pair of Classe 25's.  That is right about 1 kilowatt of honest real power on them.  It gave them some dynamic and other qualities not usually heard from them.  The Wyred4Sound ST500 amp I use is better than that on them.  I don't know if it would play them as plainly loud, but in other ways it is faster, better, tighter and simply cleaner sounding.   Better than other large amps I have heard on them, better than push-pull triodes of large size and better than a horrendously expensive OTL that I heard on them.  At the time I was also replacing a well regarded 250 watt class A/B amp

 

I had no idea Soundlab uses Crown amps.  

 

[manisandher]

25 minutes ago, semente said:

Yet at 3m with @manisandher speakers you'll get 0.2% THD+N at 95.5dB.

Not too bad.

 

What can I say? After decades of owning all sorts of speaker/amp combos, my current setup is easily the most musically satisfying I've ever managed to achieve. It's the end-game for me.

 

Mani.

[STC]

27 minutes ago, esldude said:

The second best at driving them with aplomb was a bridged pair of Classe 25's. 

 

I was using Classe 25. The highs were missing. I thought I needed more power and all I could barely afford was the CA-2300. I am a Classe fan and reluctant to go for other brand. And I prefer new. 

 

I got the Crown XLS 2000 ( although the current model is XLS2002) to check whether the 300W was enough. 

 

Yes, I had all the initial thin sounding feel with the Crown but the longer I listened I noticed the highs were correct.  

 

I couldn't do blind test because I know the difference in the highs so I have to rely on others who are not familiar with the brands. 

 

I think I also posted the C Vs C video here. Sound Lab informed me the reason for the missing highs were due to the output impedance. Apparently, Crown is lower. 

[Jud]

On ‎5‎/‎1‎/‎2017 at 1:12 PM, GUTB said:

But the problem with class D is that it sucks by nature. That's not a troll, but a hard fact.

 

It ain't so much what you don't know, but what you know that just ain't so. 

[Jud]

10 hours ago, cjf said:

I've written at length about my various experiences and experiments here and elsewhere while trying to find a happy combo using the Class D amps I've owned. I eventually had success but the road was long and expensive. I've since moved on to a seemingly immune to anything bad set of Pass Labs XA Mono's and intend to take them to my grave.

 

I like your taste. 

[Jud]

17 hours ago, gmgraves said:

First of all, your post doesn't tell us the frequency of the square wave.

 

There is an important sense in which a square wave has no frequency - those infinitely fast rise and fall times.  Thus square waves at some "audio frequency" are often used in misleading demonstrations that DSD is superior to PCM, though no audio band wave is going to have such rise and fall times.  So PCM looks bad because illegal frequencies have been brought into the demonstration by sleight-of-hand. 

 

This is very similar to what is being done in many of these Class D amp demonstrations - rise and fall times characteristic of sine frequencies far above the audio band that the amp is not spec'd to deliver and will never encounter in playing music are used by arbitrarily having the square waves occur at particular time intervals that are misleadingly (for the purpose for which they're being used) characterized as audio band frequencies.  In essence these square waves are something like impulse tests.

[Nikhil]

10 hours ago, cjf said:

I've written at length about my various experiences and experiments here and elsewhere while trying to find a happy combo using the Class D amps I've owned. I eventually had success but the road was long and expensive. I've since moved on to a seemingly immune to anything bad set of Pass Labs XA Mono's and intend to take them to my grave.

 

So the Mola Molas are out?  It's been a while since we last exchanged notes on the Components thread.

Congratulations on those Pass Labs monos ... great amplifiers.

 

[pawsman]

4 hours ago, STC said:

 

IME, I find Class-D CrownXLS is better at driving the difficult ESL Sound Lab than my other audiophile approved Amp. I am not naming the amp because mine was rather outdated and the newer ones probably are probably better. Having said that, I was told that Sound Lab uses Crown as their in-house Amp, a fact I came to know after I acquired the Crown amp which I bought to check whether a 300 Watter would be a better replacement to my 250W Amp.

I picked up a Crown XLS 2500 on closeout for $299; it's the best buy I've ever made in my many years in this Hobby.

I drive a pair of Selah Audio ported Veritas (5 ohm/83db sensitivity) and have no problem achieving SPLs as loud and dynamic as I'd ever want to hear (with mainly classical music). At a weight of 10.5 Lbs., it puts out  440/775/1200 wpc into 8/4/2 ohms; it's has almost unbelievable power for such a small amplifier. The highs ARE slightly different than other (Class A/B) amps that I've owned; I'd describe the highs as better damped and more accurate that the other amps, but still with plenty of extension and air. Many Class A/B amplifiers I've heard have a slight sheen/resonance/brightness  in the highs (crossover distortion?) that I don't hear with the Crown-

 

pawsman

[mansr]

28 minutes ago, Jud said:

There is an important sense in which a square wave has no frequency - those infinitely fast rise and fall times.  Thus square waves at some "audio frequency" are often used in misleading demonstrations that DSD is superior to PCM, though no audio band wave is going to have such rise and fall times.  So PCM looks bad because illegal frequencies have been brought into the demonstration by sleight-of-hand. 

 

This is very similar to what is being done in many of these Class D amp demonstrations - rise and fall times characteristic of sine frequencies far above the audio band that the amp is not spec'd to deliver and will never encounter in playing music are used by arbitrarily having the square waves occur at particular time intervals that are misleadingly (for the purpose for which they're being used) characterized as audio band frequencies.  In essence these square waves are something like impulse tests.

Square waves, like any periodic signal, have a frequency. What you want to say is that they have infinite bandwidth. The rest of your post is spot on.

[Jud]

7 minutes ago, mansr said:

What you want to say is that they have infinite bandwidth.

 

Thank you.

[Panelhead]

2 hours ago, pawsman said:

I picked up a Crown XLS 2500 on closeout for $299; it's the best buy I've ever made in my many years in this Hobby.

I drive a pair of Selah Audio ported Veritas (5 ohm/83db sensitivity) and have no problem achieving SPLs as loud and dynamic as I'd ever want to hear (with mainly classical music). At a weight of 10.5 Lbs., it puts out  440/775/1200 wpc into 8/4/2 ohms; it's has almost unbelievable power for such a small amplifier. The highs ARE slightly different than other (Class A/B) amps that I've owned; I'd describe the highs as better damped and more accurate that the other amps, but still with plenty of extension and air. Many Class A/B amplifiers I've heard have a slight sheen/resonance/brightness  in the highs (crossover distortion?) that I don't hear with the Crown-

 

pawsman

  I read several posts where people bragged on the XLS. When I found a XLS-1500 for less than 200.00 with shipping I pounced on it. 

  Did a little more reading and it seemed the sweet spot in the product line was considered to be the XLS-2000. Paralleled output devices, twice the filter capacitance. Like the 2500, just lower voltage rails. 

  Was using either a MP Evolution Full or MP Evolution Rev A to drive anything from a B&W 801 to Lowthers. 

  The sound was a little thin with the 1500, better with the 2000. Did a little work on the linear power supply in both. This cleaned up the presentation some. 

  What have noticed is the lack of sonic signature. Poor recordings sound poor. Good recordings sound better. 

  Small changes in playback, whether upsampling red book to 88.2, 96, 176.4, or 192k are clearly audible now. Same with converting ISO files to 88.2 or 96k. 

  This should not occur, I think the DriveCore operates at 48k. But it is transparent enough to hear differences in higher sampling and the filtering used.

  I am very happy with my two XLS amps. One tip, only use the balanced inputs. The complaints about noise level and turning the input attenuators to 11 - 1 o'clock to reduce noise all come from users with single ended sources. 

  I think the amp puts out a lot of noise that single ended cabling picks up. Using balanced cabling rejects this noise.

  I purchased the first one because I had sold all my amps with balanced inputs. Every Dac/pre I own has balanced outputs.

  Are these the best amps I have heard, no. Best amps I own, no. Best two amps purchased for 500.00 delivered with warranties. YES. Better than I can build myself with 500.00 in parts, easily.

  Being a switching amp with a linear supply does not limit the sonics. Class D does not suck, just different. Tripath does suck.

  I think Crown could do better if they built at a higher price point.

[esldude]

4 hours ago, STC said:

 

I was using Classe 25. The highs were missing. I thought I needed more power and all I could barely afford was the CA-2300. I am a Classe fan and reluctant to go for other brand. And I prefer new. 

 

I got the Crown XLS 2000 ( although the current model is XLS2002) to check whether the 300W was enough. 

 

Yes, I had all the initial thin sounding feel with the Crown but the longer I listened I noticed the highs were correct.  

 

I couldn't do blind test because I know the difference in the highs so I have to rely on others who are not familiar with the brands. 

 

I think I also posted the C Vs C video here. Sound Lab informed me the reason for the missing highs were due to the output impedance. Apparently, Crown is lower. 

Yes the Soundlabs are very low impedance in the treble and the Crown's much lower output impedance.

[wgscott]

19 hours ago, esldude said:

They didn't blow much of the budget on beautiful build quality either. Kept it where it counts....producing the kind of special distortion people crave. 

 

 

 

 

That really is horrifically ugly.

 

 

 

 

 

 

 

[monteverdi]

The title of this thread is just an opinion enforced by class A/AB designers and if one would choose more class D designers it would be contradicted. Besides there are a lot of implementations of class A which suck as much as some class D.

I have a Devialet 250Pro and it sounds pretty perfect too me. I could find as good or better sounding class A/AB amps but for even higher costs and much less functionality!

I used different pure class D amps before. The first experience was with early Hypex modules and they fulfilled all negative impressions of class D. I then got some Kharma MP150 (also a Putzey design) and they were quite good but somewhat lame in the high frequency. Now I am built some amps from Anaview AMS-100 modules (bridged) for my woofers and I like them for that purpose.

Before I was using class A amps first a Krell KSA which had was just too noisy fans to keep in the listening room. It got replaced by the first generation PassLabs. These were quite good but deteriorated over time as the high internal temperature caused fast aging of components, especially caps. So one thing I have not seen discussed here is the environmental impact of different amp designs and Class A really sucks in that respect (except you need space heaters)!

[semente]

5 minutes ago, wgscott said:

 

 

That really is horrifically ugly.

 

 

 

 

 

 

 

 

It's missing the side panels.

 

[Forehaven]

2 hours ago, Panelhead said:

  One tip, only use the balanced inputs. The complaints about noise level and turning the input attenuators to 11 - 1 o'clock to reduce noise all come from users with single ended sources. 

  I think the amp puts out a lot of noise that single ended cabling picks up. Using balanced cabling rejects this noise.

  I purchased the first one because I had sold all my amps with balanced inputs. Every Dac/pre I own has balanced outputs.

 

I replaced Bryston 7BST monoblocks for the Crown XLS 2000 which made the Brystons sound muddy in comparison.

I have on order though a SE preamp, i've been using balanced so far.  And, the amp interconnects are 15' long.  Humm...

[wgscott]

31 minutes ago, monteverdi said:

one thing I have not seen discussed here is the environmental impact of different amp designs

 

That was one of my primary motivations for trying Class D.  I am trying to produce more electricity that we consume. 

 

We also have a serious ecological catastrophe unfolding ...

 

Even if the inferior sound quality allegation amounts to anything more than snobbery, I think I could live with the very minor difference, given the difference in power consumption.

[gmgraves]

21 hours ago, esldude said:

As long as we are posting deceptive squarewave graphs out of context, I offer this small signal 10 khz square wave:

 

 

Thoress 300B monoblocks which according to JA:

 

Overall, the Thöress 300B monoblock measures well for a SET amp.

 

It manages almost (not quite) 10 watts at 3% THD.  So I am not sure it could do anything other than a small signal square wave.  Price is only $12,995 per pair. I guess these don't suck like some stupid class D designs. 

 

They didn't blow much of the budget on beautiful build quality either. Kept it where it counts....producing the kind of special distortion people crave. 

 

 

 

 

So once again, we see that people will always use square waves to back-up their pre-conceived ideas about a component. The  class D amps' square waves look much better than this SET at 10 KHz, but because GUTB likes SETs and has decided that all class D amps suck, this square wave showing poor high frequency response and poor power bandwidth is O.K. ????!!!!