"Pure Class A"

[mlsstl]

I was routinely pumping over 50w/ch into these speakers in routine listening...

If a person is using 50 watts to generate his true average listening level, then yes, you'll need 500 watts in order to have 10 dB headroom.

However, I owned a set of Maggie 1.6QRs for several years and drove them very successfully with a pair of 60 watt Dyanco Mk IIIs (modded, but not for more power.) I had no problem reaching an average listening level of 85 or even 90 dB (quite loud for home listening for most people) and never felt the need for more power.

If I recall correctly, the 1.6 QRs were about 86 dB sensitivity, so 1 or 2 watts was giving me an average listening level. The rest of the 60 watts was available for the peaks.

Of course, there are some people who listen at incredibly high levels. I've known a few people like that in my day. As one might say, more power to them. ;-)

If someone is cranking a constant average of 50 watts into a pair of MMGs, I can only imagine what type of volume level they prefer for listening.

It is one thing for a rock band to hit 120 dB in a big arena, or a symphony orchestra to hit something over 100 dB at full tilt in a 1,000 seat concert hall when you're dozens or even hundreds of feet from the players, but I find nothing realistic or attractive with those levels in a home living room.

However, that's just me and there are certainly plenty of big amps out there for those who want them.

[E-Stat]

And all that means to me is that you want a higher efficiency speaker even more in a larger room to counteract the db drop at distance.

To each his own. My choice of speaker is not a particularly efficient one.

But my contention is that I see no point in having a low efficient speaker and then putting a SS etchy high grain notch distortion ear fatiguing amplifier in the front.

Agreed. Which is why I use relatively high powered tube monoblocks each fed by a 20 amp dedicated circuit.

Granted you have Pass Labs right? A SE pure class A no feedback design? You must have decided on it for reasons other than looks no?

I use a pair of VTL MB-450s to drive the Sound Labs. While I also own a Nelson Pass designed Threshold amp for the vintage system, what I really lust for is a pair of VTL Siegfrieds. They sound positively magnificent in HP's Scaena system. Especially using Odins.

Still the best actual sound I have heard with even tough to drive panels and conventional lower efficiency speakers is with SE amplifiers.

While I listen at more moderate levels than many (typically in the 70 - 80 db range), a 10 watt SET would merely whisper with the U-1s.

There are bigger high efficiency speakers like the Tannoy Westminster that can play louder deeper and are even easier to drive!! That will beat the snot out of any 1000 watt amp and any panel or any speaker in the size and weight class of the Wilson Sophia.

Only if the utter coherence and purity of a single electrostatic driver is unimportant to you.

rw

[RGA]

Volume is a factor of the sensitivity of the speaker, the maximum watts the speaker can handle, and if provided the maximum decibel rating before compression, and impedance of the speaker, and the power supply quality of the amplifier (not quantity rating on the spec sheet).

Looking at a high power amplifier in isolation is a non issue.

If a typical speaker is 87db sensitive and an easy impedance and can take 120 watts then there is no point in buy an amplifier over 120 watts - Period - regardless of tube or SS.

1 w - 87db
2 w - 90db
4w - 93db
8w - 96db
16 w - 99db
32w - 102db
64w - 105db
128w - 108db
256w - 111db (This is the maximum level this speaker can play) You can talk about 3db drops at 2 meters all you want but that is still an apples to apples argument as all the other speakers will have the same issues). So you go out and buy your 256watt Bryston and you play to maximum you get 110db. (and then only in max peaks) Great. Whether you use 256watt amplifier or a 6million watt amplifier you are only ever getting ~110db.

And as E-Stat pointed out this level drops significantly at distance - more power won't help you though because the max you can get is 111db at 1 meter - so if you want more you need a HE loudspeaker capable of handling watts.

My Vanguard

1w - 95db
2w - 98db
4w - 101db
8w - 104db
16w - 107db
32w - 110 db. So I just matched your 256 watt Bryston power amp with an Antique Sound Labs or Jolida EL 34 tube amp. And I have had Bryston on these speakers. The ASL sounds better and costs less than 1/4 the price.
64w - 113db
128w - 116db
256w - 119db (granted the 256 watt Bryston will play up here on these speakers - cool man - but 110db is far too loud as it is so gaining 9db for and extra $5000 and getting worse sound is not a good trade just to play at levels you can no longer discern quality is IMO silly.

Tannoy Westminster

1 w 99db
2w 102db
4w 105db
8w 108db
16w 111db (So with an 8-16watt amp (tubes are ok going over their limits remember) you are now getting the same volume regardless of power amp used in system one. There are no other factors to consider - this is the level attained because you have efficient speakers. It's a much better sounding speaker, has far deeper better bass - takes advantage of the best sounding amplifiers - not just lousy sounding high power amps A/B amps

32 w - 114db (that ASL and Jolida will pound the snot out of system 1 even if you have a 6million watt amplifier.

64w - 117db
128w - 120db
256w - 123db (which is the maximum this speaker can handle.

So with the HE speaker you get better sound - deeper bass, endless amp choices and that is the reason this is one of best speakers ever built - granted you're into over 20 grand for it too but cost aside the argument for watts is a non issue if you have HE speakers - and headroom is included. These are the maximums and Tube amps typically can go over their maximums by signficant margins. http://stereophile.com/features/357/

[Ajani]

RGA - Yes, the maximum volume you can get is limited by the power handling of the speakers, so clearly there is no advantage to getting an amplifier in excess of what your speakers can handle...

Also, clearly a HE Speaker means you can use any amp from a 1 watt Sissy to a 2KW Savage...

All that is well and good... however, all that is IRRELEVANT if you don't like the sound of any HE speakers... You like Tannoy Westministers, etc... However, I'm sure E-Stat and Geoffcin chose their planars because they are the best sounding speakers they could afford to buy (in their opinions)...

The point being that it does not matter whether RGA thinks that HE Speakers and low powered amps sound better than anything else or whether Geoffcin/E-Stat think that LE Planars with High Powered amps sound best... All that matters is what each individual prefers... There are no right answers in audio... and we have all been on this forum and in this hobby long enough to know that this a never ending debate, that really just boils down to what sounds best to each of you...

[RGA]

Estat I think you'll agree that your system and budget is quite an exception to the average person my comments are for.

With the typical budget and typical midsize room with people not spending over $70,000 on systems - My comments are directed at the lower end of the higher end market. And generally with the exclusion of panels which are in a serious minority of buyers.

[Feanor]

Volume is a factor of the sensitivity of the speaker, the maximum watts the speaker can handle, and if provided the maximum decibel rating before compression, and impedance of the speaker, and the power supply quality of the amplifier (not quantity rating on the spec sheet).

Looking at a high power amplifier in isolation is a non issue.

If a typical speaker is 87db sensitive and an easy impedance and can take 120 watts then there is no point in buy an amplifier over 120 watts - Period - regardless of tube or SS.

1 w - 87db
2 w - 90db
4w - 93db
8w - 96db
16 w - 99db
32w - 102db
64w - 105db
128w - 108db
256w - 111db (This is the maximum level this speaker can play) You can talk about 3db drops at 2 meters all you want but that is still an apples to apples argument as all the other speakers will have the same issues). So you go out and buy your 256watt Bryston and you play to maximum you get 110db. (and then only in max peaks) Great. Whether you use 256watt amplifier or a 6million watt amplifier you are only ever getting ~110db.

...

Pretty close to the mark. My Magneplanar MG 1.6 are about 86 dB sensitivity; my Monarchy amps are 120 wpc into 4 ohms. I never played anything, even orchestral crescendos, above 100 dB, ergo I have plenty of power as is born out by my experience.

Yes, there are plenty of tube amps that would work fine for me. I fancy a pair of AES Six Pacs, not that I'm anything but please with the Monarchys.

Six Pacs ...

[E-Stat]

Estat I think you'll agree that your system and budget is quite an exception to the average person my comments are for.

True, but that really doesn't change the nature of my speaker choice which is driven by its sound, not efficiency. I could very easily live with a pair of nicely driven Maggie 1.6s or Peter Gunn modified SMGs.

And generally with the exclusion of panels which are in a serious minority of buyers.

What a shame to deny the realism I find they provide.

rw

[Geoffcin]

If a person is using 50 watts to generate his true average listening level, then yes, you'll need 500 watts in order to have 10 dB headroom.

However, I owned a set of Maggie 1.6QRs for several years and drove them very successfully with a pair of 60 watt Dyanco Mk IIIs (modded, but not for more power.) I had no problem reaching an average listening level of 85 or even 90 dB (quite loud for home listening for most people) and never felt the need for more power.

If I recall correctly, the 1.6 QRs were about 86 dB sensitivity, so 1 or 2 watts was giving me an average listening level. The rest of the 60 watts was available for the peaks.

Of course, there are some people who listen at incredibly high levels. I've known a few people like that in my day. As one might say, more power to them. ;-)

If someone is cranking a constant average of 50 watts into a pair of MMGs, I can only imagine what type of volume level they prefer for listening.

It is one thing for a rock band to hit 120 dB in a big arena, or a symphony orchestra to hit something over 100 dB at full tilt in a 1,000 seat concert hall when you're dozens or even hundreds of feet from the players, but I find nothing realistic or attractive with those levels in a home living room.

However, that's just me and there are certainly plenty of big amps out there for those who want them.

Magnepan rates their speakers into 4 ohms, so if they really claim 86dB per watt, it's more like 2 watts. Also very few people listen at 1 meter, especially maggies owners as you would be wearing them like headphones at that distance! Mine are just past 3 meters from my seating and at that distance to create 85dB we're at more like 95dB at one meter. Crank it up to 95dB at my listening distance and your at 105dB at 1meter. With a speaker that has 85dB sensitivity that your asking to produce 105dB and also have headroom for peaks you can see where you would run out of watts fast. With that being said, my 75wpc tube amp has more than enough power to make them sing with most program material.

[E-Stat]

It is one thing for a rock band to hit 120 dB in a big arena, or a symphony orchestra to hit something over 100 dB at full tilt in a 1,000 seat concert hall when you're dozens or even hundreds of feet from the players, but I find nothing realistic or attractive with those levels in a home living room.

Amen. High resolution systems don't need to be played at ear bleeding levels in order to reveal their beauty.

rw

[JoeE SP9]

Speaking of Pass Labs and Class A, I'm thinking I'd really covet a pair of XA200.A's, (400 wpc @ 4 ohms), to power my Maggies.




A pair of these beasts is in the range of $35,000, weighs 360 lbs, and sucks 1400 watts at idle.

Yeah, but they don't suck much more at full power. Full Class A amps act like they are producing full power 100% of the time.

Most amps are of the push pull variety. One half of the sine wave is amplified by one set of transistors the other half is amplified by the other set of transistors. Most amps operate in class A up to a certain point. This is usually only a couple of Watts. While in Class A all the output transistors are on even when the sine wave is on the other side of the zero volt reference and they are sitting idle. When an amp switches to Class B the set of transistors needed are switched on as the sine wave goes from negative to positive. The unused transistors are turned off. Any switching distortion generated is masked by the overall signal level.
This can cause switching distortion. It is the distortion generated when a bank of transistors are turned on or off. Class A operation does not have this because all the transistors are always on.

In general the more Watts an amp produces before switching to Class B the larger, more expensive and bulky the power supply becomes. The large power supplies in full Class A amps are necessary because the amp acts like it's producing full power all the time. The heat production is a byproduct of dissipating heat at 100% current draw 100% of the time.

[E-Stat]

Yeah, but they don't suck much more at full power. Full Class A amps act like they are producing full power 100% of the time.

Ironically, they produce less heat when run at full power. When idling, the extra power consumed is dissipated as heat.

rw

[mlsstl]

Geoffcin, I believe you're overstating the volume drop-off past 1 meter. The volume decrease you note is for anechoic conditions or a speaker used outdoors. When a speaker is in a typical residential room the distance drop-off isn't nearly as great.

I'm listening to my second system tonight, so got my sound level meter out. The one-meter pink noise level was about 69 dB. At my listening seat, about 2 1/2 meters, the meter reading varied from about 66.5 to 67 dB. Far less than the anechoic measurements would indicate.

I'm also fortunate enough to have acquired quite a bit of test equipment in support of my hobby over the years (Tek scope, distortion analyzers, power meters, frequency generators, counters and the like) so I'm quite confident in the approximate power figures I gave in the earlier email.

Yes, extra power can make a difference in some situations, but it also brings its own problems. All engineering is a compromise that balances numerous different variables. For me and the way I listen, gobs of extra power just isn't an issue. It is kind of like E-Stat and RGA arguing over their speaker preferences. I've found what works for me and have two systems that give me great pleasure. However, I can easily see that others might prefer something else.

However, the point I first made earlier in this thread is that I know of no recordings that contain the type of dynamic range that would cause a listener who was cruising along at an average listening volume generated by 1, 2 or 5 watts to suddenly need 200 or 500 watts for a peak.

Now if your speakers truly need 50 watts to hum along at average volume, then the hundreds of extra watts start making a bit more sense. But that person isn't a type of listener I can't relate to very much.

[RGA]

Edited due to being argumentative - I apologize if you read it before I deleted it.

[Geoffcin]

However, the point I first made earlier in this thread is that I know of no recordings that contain the type of dynamic range that would cause a listener who was cruising along at an average listening volume generated by 1, 2 or 5 watts to suddenly need 200 or 500 watts for a peak.

Now if your speakers truly need 50 watts to hum along at average volume, then the hundreds of extra watts start making a bit more sense. But that person isn't a type of listener I can't relate to very much.

For sure most high power amps never get to use their reserves, and most people don't listen at high voume on any regular basis, but I would rather have a few dB's headroom than run the risk of hard clipping especially with a SS amp. I too know how to do engineering math and for what I was listening at last night, at my listening distance required power maxed out at about 10 watts. However at the end of Ravel's Bolero I'm sure a few extra dB's of headroom was needed. With a 40 watt amp I would have been very close to clipping, thankfully the amp I was using at the time (I have several) produces double that output.

[E-Stat]

However, the point I first made earlier in this thread is that I know of no recordings that contain the type of dynamic range that would cause a listener who was cruising along at an average listening volume generated by 1, 2 or 5 watts to suddenly need 200 or 500 watts for a peak.

Hmmm. I have a handful that have more than 40 db range. Le Sacre du Printemps places pretty heavy demands on my system.

Dynamics

rw

[mlsstl]

Hmmm. I have a handful that have more than 40 db range.

I have a lot of music that has a range of more than 40 dB from the softest parts to the loudest, but as pointed out before, the soft passages aren't meant to be played at an average volume level of 80 or 90 dB!

A good example is an album I recently converted from LP - the 1965 album of Raymond Lowenthal's recordings of Charles Valentin Alkan. This recording of solo piano has great dynamic range from the very soft rendering of Barcarolle to the much louder Finale Presto of Alkan's Symphonie.

If you set the playback level for the Symphonie at a reasonable average of 85 dB in the room, the average level of Barcarolle is much softer, perhaps 70 dB. This does not include the extremely soft passages within the music itself which are easily down in excess of 40 dB from the loud passages on the recording.



You didn't state which Rite of Spring performance you have, but I would be very surprised if the parts you mention as 40 dB down in volume were meant to be played at a loud average level. I suspect if 80, 90 or even 100 dB is the average volume for the louder parts, that your soft parts were meant to be heard in the room at 40, 50 or 60 dB SPL. At least that's the way I've always heard the performance - the quiet passages are precisely that - they aren't played at a higher volume.

[E-Stat]

You didn't state which Rite of Spring performance you have, but I would be very surprised if the parts you mention as 40 dB down in volume were meant to be played at a loud average level.

That's certainly true, but one must set the output level such that it can handle the loudest passage(s) without clipping - even if they only occur for a fractional amount of the playing time which may be measured in seconds. That particular version is the ASO on Telarc.

rw

[frenchmon]

Any one heard of these? http://www.jamnaudio.com/claytonaudio.html

Frenchmon, care to guess why they are called "Clayton" Audio? They are pretty expensive as well. Not like a Pass monoblock but several thousand. I heard a pair of the CA monoblocks drive the Dynaudio Sapphires and it was amazing.

That KSA would be a sweet amp.

He Mr.Peabody. I think that's a local company...am I correct? If so do they have a show-room? Theirs a few audio company's located here....ADD speakers, Herron Audio, Dynavector, Vibrapods and a few others. I wonder if these places have a show room?

frenchmon

[mlsstl]

...but one must set the output level such that it can handle the loudest passage(s) without clipping

Granted, but that's not a particularly complicated situation. Digital material has a pretty absolute maximum volume for its output signal. The loudest possible signal from your over-compressed, peak-limited pop music CD is at the same level as for your most refined classical recording.

LPs and tapes are not quite so precise in their output limitation but still, there are no giant surprises lurking. I noted before that I've transferred a lot of LPs and open reels to digital so I get a very exact visual indication of what's on the analog source.

Without fail, the only giant dynamic peaks I ever see well beyond the average loudness level are invariably a click or pop in a vinyl record. This is for everything from the ordinary pop/rock pressings to classical to Sheffield type recordings.

Open reel recordings are limited by tape saturation issues once out of their linear range. Similarly, LPs have a set of limitations that revolve around playing time and the amount of low bass content. A record producer is also aware he has no control over the playback systems of the end users so needs to keep modulation levels within generally acceptable parameters for stylus tracking.

The only recordings that I know of as exceptions are specialty ones. Tom Danley once did a recording of fireworks that will challenge a system beyond any available commercial recording, but for me that is a novelty. It is certainly not something I'll elect to build my system around. (Note, you can download the wav file from his web site at http://www.danleysoundlabs.com/techn...downloads.html.)

The net result is that, during the average loud passages of the vast majority of musical recordings, there is simply little chance that some hidden 15 dB or 20 dB peak is lurking, just begging to be clipped by our system. Even in classical music, most of the unusually large peaks were lost during the recording process due to the natural compression imposed by mikes, recording consoles and other devices once pushed out of their linear range. And, as we know from pop/rock recordings, there is a lot of artificial processing that can be further lumped on top of this.

[E-Stat]

that's not a particularly complicated situation.

It's not complicated at all - just requires more power to maintain equally realistic average levels with wider dynamic range content.

Tom Danley once did a recording of fireworks that will challenge a system beyond any available commercial recording, but for me that is a novelty.

I downloaded that some time back when he first posted it over at AA.

The net result is that, during the average loud passages of the vast majority of musical recordings, there is simply little chance that some hidden 15 dB or 20 dB peak is lurking, just begging to be clipped by our system.

It is always the exceptions that drive ultimate requirements.

rw

[mlsstl]

It is always the exceptions that drive ultimate requirements.

And, in all my years in this hobby with well over 40,000 songs in my collection, I've still never seen or heard one of those exceptions on a music recording.

I suppose I could spend a lot of time completely revising my system in preparation for the fraction of a second that may or may not ever occur, but that doesn't make a whole lot of sense to me.

If it works for someone else, well, once again, more power to 'em. ;-)

[E-Stat]

I suppose I could spend a lot of time completely revising my system in preparation for the fraction of a second that may or may not ever occur, but that doesn't make a whole lot of sense to me.

May not ever occur? I have no idea what you are saying. I have perhaps two dozen or so favorite recordings of very high quality and wide dynamic range that determine for me how much power I need. Then again, with a higher budget I wouldn't even think about it. YMMV.

rw

[mlsstl]

May not ever occur? I have no idea what you are saying. I have perhaps two dozen or so favorite recordings of very high quality and wide dynamic range that determine for me how much power I need.

I've been very specific with my examples and explanations in my prior posts. This is based on the more than 2,000 recordings I've transferred to digital. Adobe Audition gives a very clear view of the dynamic range involved of everything I run through it.

You earlier gave an example with a 40 dB dynamic range, but I pointed out that that the soft parts of a music piece should be played softly. Take a medium efficiency 85 dB/watt speaker and those quiet sections of a recording down at 60 dB SPL are only consuming four one-hundredths of a watt. That is why some music listeners are more concerned about the low-wattage end of things more than the high end.

The question is what peaks are still above the loud parts. The answer, as is clearly shown in my prior post concerning the maximum peak recording level that applies to all formats, is not much.

I've got a meeting to attend this evening, but afterward I'll get one of my fancy classical recordings with good dynamic range out and post a screen shot from Audition. Perhaps a picture will better illustrate things.

[E-Stat]

You earlier gave an example with a 40 dB dynamic range, but I pointed out that that the soft parts of a music piece should be played softly. .. Perhaps a picture will better illustrate things.

What I said was "more than 40 db". There is no reason to post pictures, because I'm not talking theory or hypothetical efficiency ratings, but direct experience with my system. I'll repeat: it is only for a dozen or so recordings that I need the power I have (actually about 3 db would be better). I know exactly what the live levels would be for several ASO Telarcs since I've been in that hall countless times and have participated in one of the recordings. Your system is likely very different from the one to which I refer. The two other ones require far less power. With the garage system, I rarely need more than 4 watts unless I want it VERY LOUD.

rw

[mlsstl]

Here's an image from the finale of Wagner's "Ride of the Valkyries"

Note the orchestra is cranking along with a lot of fervor at full tilt. You can see the volume bar at the bottom of the screen shot shows right at - 6 dB. That means there is 6 dB of headroom left for peaks. Nothing louder than that will be reproduced. There is simply nothing available for 20, 15 or even 10 dB peaks.

This is pretty typical of classical recordings. The start of the screenshot is about -24 dB. The decaying sound level right before electronic fadeout is right at -40 dB.

You can see we have about 5 or 6 peaks that make use of some or all of the remaining 6 dB of headroom. As noted earlier, there simply isn't a massive amount of headroom on recordings above the average loud portions of a composition.

I'll certainly agree that: 1) some people listen at much louder average levels than I do. The guy that uses 50 watts on average will probably benefit from those hundreds of watts. They are certainly free to pursue that, though it simply doesn't interest me and I don't require it for my system.

2) Some people simply prefer having an amp with enormous power reserves at their disposal. That may well influence how their system sounds to them.

My primary point all along has been is that if you are getting the volume you want on the average loud portions of a record from just a couple of watts, there is nothing on a recording that is going to produce a sudden demand for 200 or 300 watts. If this person prefers the more powerful amp it is likely due to some other factor.