Paul McGowan’s Post: More Than Enough

[awsmone]

It does depend on the sensitivity of the speakers, the size of the room, the listening distance, how loud you play and the dynamic range of the recording.

most of the time amplifiers are just cruising

However piano is often a amplifier crusher, given its power and transients, I have heard so many amplifiers clip or strain on dynamic piano recordings, no matter the quality

I remember going to hear Giya 1 at a store

The electronics were McIntosh 1201

I asked them to play Mahler 3

When it came to the hammer blows the needle jumped to 1200 watts!

The sales assistant didn’t believe it, and we replayed the section, and I filmed it on my iPhone
Sure enough when it came to that bit, the needle swept around to 1200 again....

 

[Al M.]

I personally like Pass' First Watt philosophy, where he believes the best sound comes from the simplest circuits with the fewest output devices and very careful parts selection and build execution. Pass has in inference stated that he likes the sound of lower power amps,

That is my impression too from his comments.

The enormous tubed units can sound majestic, but they don't seem to have the inner detailing of simpler tubed amps.

It depends. Madfloyd's CAT JL-7 monoblocks with eight KT150 tubes per channel (220 W) have incredible inner detail and resolution on his Magico M Project speakers.

So, as far as raw power and sound quality, I think that no generalization is the best generalization.

I can agree with that. See also my anecdote above about the auditioning of Rotel amps, where the least powerful one gave the best and most dynamic sound. On the other hand, I have heard high-power SS amps (Pass, Spectral) sound tremendously dynamic.

Some speakers really need big power. You can't drive everything with 20 W amps.

 

[caesar]

That is my impression too from his comments.



It depends. Madfloyd's CAT JL-7 monoblocks with eight KT150 tubes per channel (220 W) have incredible inner detail and resolution on his Magico M Project speakers.



I can agree with that. See also my anecdote above about the auditioning of Rotel amps, where the least powerful one gave the best and most dynamic sound. On the other hand, I have heard high-power SS amps (Pass, Spectral) sound tremendously dynamic.

Some speakers really need big power. You can't drive everything with 20 W amps.

Al, good points. Out of all the high-powered tube amps out there, CAT is an exception in bringing the magic of tubes to midrange, having extended and detailed highs, and that dynamic jump. Part of the reason for this is that Ken Stevens is a big fan of MBL, an extremely hard to drive speaker, and had to come up with a design that brought the speaker to life.

 

[Al M.]

Al, good points. Out of all the high-powered tube amps out there, CAT is an exception in bringing the magic of tubes to midrange, having extended and detailed highs, and that dynamic jump. Part of the reason for this is that Ken Stevens is a big fan of MBL, an extremely hard to drive speaker, and had to come up with a design that brought the speaker to life.

I suppose the Octave MRE 220 mono blocks (220 W/ch) are likely to do all that too, extrapolating from my Octave RE 320 stereo amp. However, I don't know how the extra complexity of 4 power tubes/channel vs. 2 power tubes/channel (my amp) would translate. And the Octave amps are class A/B, the CAT JL-7 is class A with 8 power tubes/channel, and I admire how these CAT amps can pull off the trick with so many tubes.

 

[Folsom]

Gentlemen, great discussion. Is there a definition of "headroom"? I also think we may need to distinguish between power and current...

Experientially, I think most modern speakers like the wilsons, magicos, mbls, as well as all electrostatics and magneplanars I have heard Suck with low powered ss amps and with most tube gear- they just sound dead and syrupy. Unless one is prioritizing midrange over dynamics.

High efficiency stuff is different, however. I think many trade off dynamics in favor of that rich, intoxicating tone. Interestingly, I have experienced 300b on high efficiency sound much more extended and NOT syrupy than , say , an arc, bat, vtl, etc., on the low efficiency, popular brands I mentioned above.

Not sure why people don't just go with 300b and high efficiency speakers instead of say, arc and Wilson, or say Martin Logan with bat, etc....

Current doesn't really change in audio gear. But what does happen is that inadequate pathways for it lead to sag/sat. Those however are often firstly a function of other attributes where current is being dumped across something, and there for you're getting that voltage flop. It isn't that there isn't enough current, it's how much of it is being wasted. The more current wasted the more voltage will drop because the power supply is limited (it's physics, not economics).

 

[NorthStar]

It does depend on the sensitivity of the speakers, the size of the room, the listening distance, how loud you play and the dynamic range of the recording.

most of the time amplifiers are just cruising

However piano is often a amplifier crusher, given its power and transients, I have heard so many amplifiers clip or strain on dynamic piano recordings, no matter the quality

I remember going to hear Giya 1 at a store

The electronics were McIntosh 1201

I asked them to play Mahler 3

When it came to the hammer blows the needle jumped to 1200 watts!

The sales assistant didn’t believe it, and we replayed the section, and I filmed it on my iPhone
Sure enough when it came to that bit, the needle swept around to 1200 again....

Music on peaks can hit the four digits, even five! And you need loudspeakers with drivers that can handle those peaks, or it's going to be expensive in driver's replacements.
____

• https://www.rollingstone.com/music/music-news/hi-fi-how-much-power-is-enough-39088/
____

For fun: http://www.audiomasterclass.com/newsletter/how-much-power-do-you-need-to-fill-a-venue-with-sound

• Home stereo: 150 W for 85 dB SPL average (with 15 dB peaks),
1,500 W for 95 dB SPL average (with 15 dB peaks)

 

[cjf]

I used to be a big fan of having way more power then I would ever need for a particular speaker sometimes even going for having double the RMS rating of the speaker. I never had any complaints by doing so in terms of SQ but at the same time, especially in this hobby, going this route means spending big money that in reality you didn't really need to spend.

With my current setup which takes the opposite approach I again cant find anything to complain about in terms of SQ. In fact, I'm continuously impressed with how effortless sounding my "baby" Pass XA60.8 amps sound despite what their low power output rating would imply. I've not yet found a volume level where I can get the needles to move more than just a slight wiggle while powering my somewhat difficult to drive Magico S3's and I tend to listen rather loudly most of the time (90-105db range). The speakers are rated at 500wrms so I'm not even touching that recommendation. If I listened to Magico and bought an amp that could do 500wrms into 4ohms I should be using a Pass XA200.8 @$42K a pair instead of my meager $14K a pair XA60.8's.

Even the most well healed folks might even blink at wasting that kind of coinage for "JIK I need it" purposes.

 

[DaveC]

I'd agree with folks saying this needs qualification. It depends... MikeL's speakers are a great example of speakers that are efficient, but really need a big amp to perform as intended. OTOH, take a single driver/no crossover speaker with the exact same sensitivity specs and you won't see the same benefits to using a larger amp, in fact it might be worse because the amp will be more complicated.

Another case of specs in audio not telling the whole story, who woulda thunk?

 

[PeterA]

I used to be a big fan of having way more power then I would ever need for a particular speaker sometimes even going for having double the RMS rating of the speaker. I never had any complaints by doing so in terms of SQ but at the same time, especially in this hobby, going this route means spending big money that in reality you didn't really need to spend.

With my current setup which takes the opposite approach I again cant find anything to complain about in terms of SQ. In fact, I'm continuously impressed with how effortless sounding my "baby" Pass XA60.8 amps sound despite what their low power output rating would imply. I've not yet found a volume level where I can get the needles to move more than just a slight wiggle while powering my somewhat difficult to drive Magico S3's and I tend to listen rather loudly most of the time (90-105db range). The speakers are rated at 500wrms so I'm not even touching that recommendation. If I listened to Magico and bought an amp that could do 500wrms into 4ohms I should be using a Pass XA200.8 @$42K a pair instead of my meager $14K a pair XA60.8's.

Even the most well healed folks might even blink at wasting that kind of coinage for "JIK I need it" purposes.

Another happy Magico/Pass owner. Congratulations. I had the XA100.5 with my Magico Mini IIs. The needle barely moved, but I upgraded to the XA160.5s and the sound became more effortless. There was a new sense of ease and control that the 100.5s just could not do. Now I have the Q3s which are a bit easier load, and the sense of ease is even better. I think this is a case of being happy with the sound and not realizing there are any minor issues until you hear something better in your room. I agree that the prices of these amps do not make one feel eager to step up to the next level, especially if he is happy with what he has.

 

[RogerD]

I use 5 amplifiers that total about 4500w powering 23 drivers. All speakers are at least 90db efficient. Could 800w be sufficient,maybe, but I am so addicted to a soaring portrayal of a full orchestra,it has made me a firm believer in high power.

 

[cjfrbw]

I guess having all that power is like having a back forty. You never go there, but you like to know it's available.

 

[morricab]

. . .

What we want in a power product is called headroom, the ability of a device to exceed demand by an appreciable amount.

Headroom is important on a number of levels: lowering parts stress, relaxing audio presentation, removing strain from both the equipment and the music. If you think you need 100 watts, go for 200 to 300 instead.

It’s easy to understand too little strangles performance. The difficult argument is that bigger is better than enough. Taking your equipment right up to the edge, or anywhere even close to shore, isn’t worth the initial savings on equipment.

When it comes to deciding how big to go, more than enough should be your guiding light.
​

I post this because I agree with it. More importantly MikeL’s long experience with amplifiers and headroom has lead him to this same conclusion as Paul.

I think this is a somewhat shamefully simplistic statement from Paul. Obviously, anyone with even a little inclination for math understands that dynamic peaks on relatively uncompressed recordings can demand a high instantaneous power and all things being equal more power would cover these rare occurrenes. However, NOTHING is equal about going to more power and do you need an amplifier with a high continuous power or one that can generate 2 or 3x the rated power for short burst to cover some of these music peaks?

Amplifiers of the same topology will not scale equally. As you start adding more transistors and more parts required to make the higher powered model, you are introducing more distortion and noise making devices into the mix and often require more, not less, negative feedback to keep the whole thing on track. The result? An amp that MIGHT be able to track that music peak you wanted to have the headroom for but for 99% of the rest of the music you have worse sound.

Also, for a given power, the topology and power supply design matter...a lot. A good example of this was the CAT JL2 Signature vs. NAT SE2SE (older bigger version with 211s). On paper the CAT was a 100 watt Class A PP amplifier (and a good sounding one too) and the NATs were 75 watt SET monos (in actual test they were 126watt monos in CLass A2). So, very similar power and both Class A but one with multiple output tubes in PP and the other with two big tubes in parallel. The NAT was the better sounding of the two by a significant margin...

I have never heard this removal of strain simply by going to more power because what I normally hear by going in that direction is a more sat on sound with less dyanmics. Some of the wost sounding amps I have heard from a dynamics standpoint were also the most powerful on paper.

Another point: The most dynamic amp we ever tried on the Thiel CS3.7 was the Aries Cerat Diana Forte, followed cloesly by the Diana Integrated at 60 and 25 watts respectively. Both of these amps sounded totally at ease with the speaker even at higher levels...there was arguably no need for the extra 35 watts from the Forte. Both of these amps though were clearly well ahead of the rest that we tried, up to and including McIntosh MC-501 monos (so approx. 10-20x power), Octave MRE 130 with SBBs (2-4x power), Lamm M1.1 (2-4x power), NAT Symbiosis SE (2-4x) power, Cayin SET (same power as Diana Int), VAC 30/30 (same as Diana Int), KR VA350 (same as Diana Int), Brinkmann monos (2-4x power), KR VA990 (2-4x power), BAT VK200 (2-4x power).

Amps of a given power will not be able to deliver that power equally and a lot depends on how the power supply is designed and if it can really deliver on demand what is requested by the signal. Then the topology matters. It seems also that many smaller amps have higher dynamic power capability (as a ratio of their rated power), which allows a better sounding simpler amp that can still handle many of the peaks just fine. Also, the heavy use of negative feedback will often result in a powerful amp that sounds small and "constipated" sounding...like you always feel the need to turn it up because the sound just can't get out.

Finally, I would direct attention to the articles by Peter Van Willenswaard where he demonstrated that tubes can do much bigger voltage swings than SS amps of similar power ratings. Being able to track the full signal envelope for longer will matter in the perception of dynamics.

I think most people would be far happier with the sound of an amp that was 10 watts RMSbut co uld generate 100 watts for say, 100ms burst, than an amp that is 100 watts RMS but with little to no dynamic headroom. Having really high powered RMS amps compromises too many other things, like purity of the signal, noise etc.


"I then connected my 300B amp and repeated the tests, with the 'scope's scale set to 10V/div. Fig.3 shows that this amplifier went into mild clipping (estimated THD 3%) into 8 ohms at 14Vp for the positive, least-clipping side, and 11Vp negative. This suggests a maximum output power of 11W RMS. Doubling the amp's input voltage produced heavy clipping at 17Vp positive.
Read more at https://www.stereophile.com/content/tubes-do-something-special-page-2#0k0dCtehfVGhy4GL.99"

"I replaced the 8 ohm load with the speaker and tried to see how far I could crank up the volume with this passage on the CD until no further increase in output occurred. I got fig.4: certainly distorted in comparison to fig.2, although I could hear nothing at all problematic. But look at that 36Vp in the negative half of the picture—it would take an 80W class-A transistor amp to allow such a voltage excursion! Fig.4 also suggests that if the 300B output stage were dimensioned differently and optimized for these transient conditions instead of the usual steady-state sinewave condition, the heavy positive clipping could have been avoided. This deserves investigation, but that means a whole new project...
Read more at https://www.stereophile.com/content/tubes-do-something-special-page-2#0k0dCtehfVGhy4GL.99"

Even unoptimized the 300B amp made signal like an 80 watt amp! Optimizing it would possibly take that dynamic performance even higher. One could expect similar results from 20 or 30 watts SETs as well and them behaving more like 200 or 300 watt SS amps in terms of voltage swing but a lot better sounding for a host of other reasons.

 

[morricab]

Paul’s view is apparently based on the following premise: more (and more and even more) watts do not deteriote the quality of these watts. However, based on my listening experiences with a lot of powerful poweramps, in particular many solid state poweramps, this premise is in many cases incorrect in our real audio world (and really transparent audio systems casu quo hifi systems with a (very) low noisefloor will demonstrate this). I am not willing to substitute quality for quantity.

Agreed

 

[morricab]

Gentlemen, great discussion. Is there a definition of "headroom"? I also think we may need to distinguish between power and current...

Experientially, I think most modern speakers like the wilsons, magicos, mbls, as well as all electrostatics and magneplanars I have heard Suck with low powered ss amps and with most tube gear- they just sound dead and syrupy. Unless one is prioritizing midrange over dynamics.

High efficiency stuff is different, however. I think many trade off dynamics in favor of that rich, intoxicating tone. Interestingly, I have experienced 300b on high efficiency sound much more extended and NOT syrupy than , say , an arc, bat, vtl, etc., on the low efficiency, popular brands I mentioned above.

Not sure why people don't just go with 300b and high efficiency speakers instead of say, arc and Wilson, or say Martin Logan with bat, etc....

A lot of people in the last 20 years have come around to this idea...thus the surge in SETs on the market as well as a general trend to higher sensitivity speakers (both horn and non-horn). A lot of people I know who were heavily down the first path (myself included) are now firmly on the latter path.

 

[morricab]

Another point I meant to make but overlooked in my previous long post is that Paul's logic can be exposed as a fallacy with a thought experiment that is borne out in the real world.

If we follow the headroom/power-on-tap logic to the extreme then we would want to have both high power AND high sensitivity loudspeakers. This would give the maximum potential headroom (assuming the speaker can absorb that kind of power).

This would mean match a high sensitivity horn system with a powerful SS amplifier (or perhaps a very powerful tube amplifier like the VTL Wotan or whatever they call it these days).

If power is the main criteria then putting some large MC-1201s or even their 3KW model or big Boulder monos or Krell monos or whatever. In the real world, almost no one does this. Have you guys ever heard an SS amp on a Klipschorn?

Why does (almost) no one do this in the real world? Probably because the small tube amps, which as was demonstrated by Mr. Van Willenswaard are not so puny as one thinks, SOUND much better on these speakers. The other amps will have you running for the door.

Ever heard Avantgarde's with their own amps, which are SS? Terrible, really foward and hard. Avantgarde speakers with a good tube amp? Can be very nice indeed (I am reminded of Munich 2015 with Duo Omegas and Audiopax amps or earlier times with Thomas Mayer amps).

The simple fact that, when given a choice of amp for a high sensitivity speaker (and they do allow for the widest possible choice of amp being high sensitivity and usually easy load as well) 99% of owners of such speakers will not choose a high powered SS amp. They MIGHT choose a low powered SS amp, like a First Watt SIT or some other Class A little guy. One reviewer went to the Pass XA30.8, but we can all agree this was an exception and still a low powered amp. They might go with something like a 100watt Gamut amp, which uses minimal active parts (these don't sound very good IMO).

Where are the guys driving horns or single driver 99db speakers like Voxativ with 500 watt SS amps? If you are out there, speak up and tell us all how having that that power and headroom has benefitted your listening experience.

What about high powered tubes? Well, not many seem to go that route either... any guess why?

 

[Folsom]

Those are some long posts to repeat that it's about efficiency.

 

[cjfrbw]

Somewhere in the lexicon and the mythologies, likely generated by various audio critics and their poetics, the idea of 'ease' of high powered amps crept in, as if you can 'hear' all those empty watts on routine program material.

I think the idea comes from a false comparison between a car with lots of horsepower and and a car with less horsepower, that you automatically get a more instantaneous acceleration that can be felt. A more powerful amplifier has extra horsepower that can be 'heard'. Of course, a lot of 'guys' in a phallo-centric hobby will always go for horsepower for it's own sake.

There is no reason to believe that two amps operating without compression or clipping, delivering adequate current within the parameters of power supply voltage, acting on a music signal would sound different merely on the basis of power rating, whether or not one is 10 watts and the other is 1000 watts .

Clearly, there would be lots of other topology and construction differences that could make them sound different, but power superiority isn't really the relevant feature. Are there still audiophiles that believe that mere power superiority creates sonic superiority?

You hear designers bemoan about the current sharing problems with very powerful solid state devices used in multiples in large banks.

Another notion from years ago is the 'jangling keychain' idea, that a jangling keychain has 120db micro transients. In order to adequately 'capture' these transients, your amplifier needs all those extra watts for the micro transients.

Whether or not these micro transients exist in nature, were they captured by the microphone, recorded without alteration, survived the mastering suite, and conveyed to the final recording without compression or compromise? Do you even have speakers that can adequately reproduce these hypothetical micro transients? You can pretty much see the dynamic peaks of a recording on a time based voltage diagram of a music signal. Oops, where are the micro transients?

So I am not really thrilled with either the notions of 'micro transients' or 'ease' being anything but poetry bogeys or conceits.

Members have already discussed the 'problems' of high power for it's own sake, potentially compromising sound quality on standard program material in order to get those less common 'peaks' right.

Is your lovely pianissimo really going to sound better with twenty devices dribbling current in tandem rather than one or two?

 

[cjfrbw]

https://www.youtube.com/watch?v=R91HAP0t3AA

The ever entertaining Steve Guttenberg from one of his you tube videos.

 

[Alrainbow]

What we really need and what are told can be very wrong . Speaker imp matters and confuses most coz some amps double in power output. But this is not a measure of dynamics in headroom. In fact an amp at 8 ohms has more headroom then the same amp at 4 ohms. Most Amps made are for 8 ohms but can play 4 but there is sacrifice. I bought the amps I have based on a small group that can play them right. An amp playing music at a good level maybe just a few watts , but a snare drum hit or even worse a female vocal needs hundreds of times more.
My last amp I was using was a Heavily modded Aragon 4004 Mkii . New caps more than double size caps , dual bypass caps . Gave me a good sound on them. A low dynamic song is easy to play loud it has little increase. Now play some classical like beethtovin no 9 chorals 5/6/7 well stuff happens. But the same can be heard on simple female vocals. I think most don't know what clipping sounds like . Tubes smear but SS jumps at you. And in some cases both tubes and SS the music just gets flat as the amp runs out of headroom totally.
Paul has my speakers and I cannot imagine his amps with there tiny caps not clipping but I have not been to his room.
I can name some songs to,try and can tell you Just what word will jump at you. If SS. but be for warned once you hear it there is no going back. Only a handful can play my speakers well. I do play loud and my room is a good size. Heck my seat is 16 feet back.

 

[morricab]

https://www.youtube.com/watch?v=R91HAP0t3AA

The ever entertaining Steve Guttenberg from one of his you tube videos.

We were doing this frequently when my colleague had McIntosh MC-501 monos on his Thiel CS3.7s. Even listening quite loud we were never driving the needles past 5 watts. i think one orchestra piece got to 50 watts at the very climax but the rest was cruising well under 5 watts... often 1 watt or so. God, were they boring sounding amps though...