Watts! How many do we need?

Current uber alles.

Looking at my own meters, it’s movies that are going to push the amps.

For the record, every meter I’ve seen is really a voltage meter in disguise. A low impedance speaker won’t push the needle further up. Perhaps someone has invented a smarter meter though. :)

On a slightly different angle than the OP has asked Nelson Pass has long advocated that the performance of an amp in the first few watts is what matters most.

How about this? Or this?

Or maybe that? Or that?

And who could forget this?

Search engines...I wuv you.

It is FAR easier to sell something on more-is-better because that is easy to understand. If the extra power does not come at an expense in terms of sound quality, that would mean no harm in having way too much power in reserve.  But, that is not really the case.  Of curse it is a matter of taste, but, I generally prefer the sound of low-powered tube amps over other types.  Often, my least favorite amps are high-powered tube amps running multiple KT88, KT150, KT170, etc. tubes (the sound is "hard" and harsh sounding to me).  These days, most good solid state amps are smooth and not harsh sounding at all, but, they tend to sound a bit lifeless and unengaging unless the volume is push up higher than one would play the speakers running good tube amps. 

I know Harbeth advocates for a lot of power and their speakers are somewhat inefficient, but, the best I have heard them sound was with medium powered amps.  A good 20-100 watt pushpull tube amp would be my choice.

I keep going down in watts. I do have high-efficiency Klipsch Cornwalls.

My amp voyage has been....

NAD M23 200 wpc

to 

Threshold CAS1 75 wpc

to

Manley Stringray switchable between 20 wpc (triode) or 40 wpc (ultra-linear) 

All have sounded great, but each has been a step up in quality; in my opinion.

I love the sound of the Manley in ultra-linear mode.

FYI

there are several Stringrays at a good price here in Audiogon.

as well as a great pair of Snappers; if you have a little extra and want 100wpc.

One. Million. Watts. (Max total peak)

From my experience with solid state, current was the primary determinant if the amp was up to the task, the more the better. From my tube amp experience, it is far, far less important. So I went from high current 350 wpc solid state to 70 wpc tube amplification and my system sounds so much better it is amazing. It is a high quality amp… I am sure there is a difference between 70 wpc with an inexpensive light weight tube amp and a good one.

I happened to be looking at the meters the other day and noticed they were peaking at 1.5 watts output. I realize analog meters are slow and will not capture the full extent of what is going on and perhaps the meters use a proxy parameter to measure watts… but still interesting. Seems there is some truth in the saying “it’s the first watt that is most important.”

I can switch my amp between 70 wpc and 140 wpc. I noticed the same average power output in either mode but the peaks in sound would send the meters quickly far to the right (higher output) in 70 wpc mode… which makes sense. Even at deafening volume my tube amp does not run out of power. While my past experiences with 75 - 150 wpc solid state either never had good solidity or loss it when pressed to higher volumes (not necessarily ear splitting). The other thing I notice is that my amp sounds much more natural and musical without loss of dynamics in the 70 wpc triode mode vs the 140 wpc ultra linear mode.

@ghdprentice 

I’m puzzled by your experience. I don’t doubt it but since current x volts equals watts. What you are describing is that lower current sounds better. So something else is going on when you switch to high watts. 

I have an original Schitt Aegir (20 w/ch 8 ohms) and a pair of Ohm 1000 speakers in a 15' X 24' room with a vaulted ceiling -- absolutely no problem reaching my desired listening level (85 dB average) with no hint of clipping or compressed dynamics.  Two caveats, however.  First I use a powered subwoofer and roll off the deep bass to the Ohms. Deep bass can be a power hog  Second, as many know, the relationship between watts and volume level is logarithmic, not linear. Doubling the power gets you 3 dB more volume. Doubling the perceived volume takes 10 times more power. So, if you are a headbanger, and have medium or low sensitive speakers, or have a large room, you're going to need a lot more power than I do.

But, as others have noted, lotsa watts are an attractive advertising carrot. And Class D amps have made it easier to offer that carrot these days.

Normal to loud-side for most speakers is around 1 - 2 watts.  Figure out what it is for your speakers and multiply by 100 to 200 to cover the peaks without capping out the amp's capacity.  And YES, momentary peaks can take that amount of power.

And my tube amp sounds better in every respect with my speakers in ultra linear mode, which is double the power. Also, with its 60 watt/ch it feels significantly more powerful than my previous 80 watt/ch ss amp at the same volume level. Well, you don't compare VAC to Burson, anyway.

@ghdprentice  I assume it makes a great deal of difference on the individual speaker you are using.  I've always heard there is a difference between "tube" watts and SS watts.  On this very sight I have read that tube watts are roughly 3 times as powerful as SS watts.  To this end I've found this lengthy article penned by Roger Sanders to be quite interesting.  Especially as I am driving a set of SoundLab Majestic 745's.  Cheers.  

https://sanderssoundsystems.com/technical-white-papers/172-tubes-vs-transistors

@bpoletti 

That’s the issue I was addressing. The meters in my amp are set to freeze at the highest peak reached. So at the end of the night when I check them they never exceed 1 Watt. And that’s with me listening to full Classical orchestral recordings at  room filling volume that are very dynamic.  So even my large peaks  that are  musically dynamic do not move the meter past 1 watt. 

Maybe those meters do not capture those peaks.  I would not go less than 100 watts.  But that's just me.

It’s a new Accuphase amp and the meters are accurate. The amp is rated at 150 watts per channel into 8 ohms and 300 into 4 ohms. I think we are all using a lot fewer watts than we think even  in peaks. 

I would be curious to know what actual volume you are listening at? A crude app for your phone will tell you peaks at your listening position, and if you are feeling inquisitive, you can measure at 1m from the speakers. 
 

1 watt at 1m will give you 86db from your speakers. It’s math. As was stated above, every 3dB step  above that, will require a double amount of watts.

The other thing, is higher powered amps, have bigger power supplies. They will have more power in reserve at lower watt output.

Provided the transistors, and the topology of the amp is such that it is operating in a linear manner at low output, they will sound “better” at low volumes than a lower powered amp with an inadequate power supply.

Which brings us full circle back to Nelson Pass, and the relevance of the first watt.

Peaks are between 75 and 80db. I’m sitting and measuring about 8 feet from the speakers which are 7 feet apart. They are almost 3 feet from front wall and are 5 feet from side walls. Plenty of opportunities for sound absorption. The room is 17 x 15 and is carpeted and is not what I would call a live room as it has furniture as well. The peaks are as loud as I would want and doubling the loudness would apparently require 2 watts. When would I ever need 10 watts let alone 150? BTW these peaks I hear are produced effortlessly with a balanced presentation. No instruments or frequencies are emphasized. It’s a smooth delivery. 

The 1.5 watts of “audio magic” of the 45 tubes is plenty for me. My backloaded horns are 102 db efficient. Korneff integrated amp with Emission Labs 45 wire mesh tubes. Although I must admit that I cheat. With organ music I turn on a Velodyne 18GS subwoofer that has a built in 1500 watt solid state amp. Go figure…

I doubt those meters are able to measure actual peak power, it gets difficult to measure when a speaker is involved. 

The meters found on most amps are very rarely calibrated, and more often mislead rather inform.

Let’s assume the meters are not very accurate. How much off are they? By a factor of 10 so 10 watts peak? Or by a factor of 20 so 20 watts peak? And then we make the sound twice as loud (+3db). Now 20 or 40 watts.  Sill far below the amount of watts modern amps are offering. 

This is a portion of the text taken from Roger Sander's white paper on amp power requirements.  I found it to be very interesting and it appears to be pertinent to this discussion?  Take it away Roger:

"To see what is going on with an amp when playing music only requires an oscilloscope.  These are very fast (the slowest ones will show 20 MHz) and will clearly show amplifier peak clipping when music is playing.  A meter is too slow to do so.  A 'scope is cheap (you can get them for $100 on eBay all day long).  So you don't have to take my word for what I am about to explain.  Feel free to get your own 'scope and examine your system's performance.

You simply connect the 'scope across your speaker or amplifier terminals (which are electrically the same), adjust the horizontal sweep as slow as you can while still seeing a horizontal line on the screen.  Don't go so slowly that you see a moving dot.

Now play dynamic music at the normally loud levels you enjoy.  Adjust the vertical gain on the 'scope so that the trace stays on the screen.

As music plays, you will clearly see if clipping occurs.  The trace (which will just be a jumble of squiggly lines) will appear to hit an invisible brick wall.  It will appear as though somebody took a pair of scissors and clipped off the top of the trace.  That's where the term "clipping" comes from.

If you see clipping at the levels you like to listen, then you are not using a sufficiently powerful amplifier to play your music cleanly.  Your system is compromised because your amplifier will have compressed dynamics, sound strained, lose its detail, and have high levels of distortion.

The 'scope will be calibrated so that you will know the voltage at which clipping occurs by observing the grid lines.  If you know the voltage and the impedance of your speakers, you can easily calculate the power.

Power is the voltage squared, divided by the impedance.  So if the 'scope measures 40 volts at clipping, and you are driving 8 ohm speakers, you know that 200 watts are being produced at clipping -- and this is insufficient power for your particular system because it is clipping.

You will find that conventional, direct-radiator (not horn-loaded), magnetic speaker systems of around 90 dB sensitivity, require around 500 watts/channel to avoid clipping.  More power is needed in larger rooms or if you like to play your music more loudly than most.

The key point I'm trying to make is that audiophiles usually are using under powered amplifiers and are therefore listening to clipping amplifiers most of the time.  When an amplifier is clipping, it is behaving (and sounding) grossly differently than its measured performance would suggest. This is because we always measure amplifiers when they are operating within their design parameters -- never when clipping.  A clipping amp has horrible performance, so attempting to measure it is a waste of time.

In other words, we usually listen to an amplifier when it is clipping and we measure it when it is not.  This is why amplifiers sound so different than their measurements would imply.  It is not that measurements are wrong, it is simply that we are listening and measuring different conditions.

It is essential to understand that when an amp is clipping, it will sound quite different than when it is not clipping.  It is also important to realise that different types of output devices (tubes vs. transistors) clip in very different ways, so sound quite different when they are clipping.

Finally, it is important to realise that an amp does not instantly recover from clipping.  It takes several milliseconds for its power supply voltage to recover, for it to recharge its power supply capacitors, and for its internal circuitry to settle down and operate properly again.  Therefore, even though an amp may only be clipping on the musical peaks, it will not immediately operate properly at average music levels where it is not clipping."

I suspect that most of us are not aware of how our amps are truly performing?  I for one have never attached an oscilloscope to my equipment.  

The full article can be found here:

https://sanderssoundsystems.com/technical-white-papers/172-tubes-vs-transistors

OP,

‘The difference in sound quality between my ARC Ref 160s with 70 wpc and 140 wpc is topology… one configuration is ultra linear and one is triode (70 wpc). What sounds better to me is the more natural - musical sound of the triode mode… what I am not hearing is any difference in dynamics caused by the drop in power. 

@jfmusic, I’d often seen stereo amps displaying wildly different readings when fed mono signals. While it’s very true that the fault could lobe elsewhere, it’s doubt it is not more often than not that the meters are at fault. When queried about the meters, and when not being evasive just about every manufacturer has admitted that inclusion of meters was because they looked cool, and shouldn’t be relied on for critical use.

What Harbeth's do you have?

@macg19 

I have the C7ES-XD Harbeths 

That's a very nice amp for those speakers, congrats. Fellow Harbeth owner - I'm seeing/hearing similar results. I think the "low" sensitivity of Harbeth's isn't that relevant with well-matched, high-current SS power which is exactly what you have based on Paul's video (able to double power from 8 to 4 ohms).  

Even if our meters are not perfectly calibrated, there is no doubt in my mind that they are a very good indication that the amp is barely breathing, and having virtually unlimited headroom is a really good thing. Enjoy it! 

@bigtwin 

Thanks that was an interesting article. 
 

@macg19 

Yes exactly how I feel  The amp reproduces the peaks effortlessly. No sounds of straining or harshness.  The entire frequency range just gets louder with no over emphasis of upper frequencies A very even response  

@bigtwin Very interesting, thank you.

@bigtwin -- the one thing missing in the info you posted is a reference to the listening volume in the room. Not everyone wants live rock concert volume in their home.

The other common mistake concerns the level of peaks above the average playback volume. People often think that recordings have more dramatic peaks than they actually do.  Digital recordings in particular have a max level that can't be exceeded. I've used Adobe Audition for years to edit all manner of recordings -- rock, folk, jazz, classical and others.  One thing I've noticed, in particular in reference to albums recorded since the advent of digital in studios, is the volume uniformity of drum strikes -- a common source of audible peaks in music. Viewed digitally, one can easily see that dynamic peaks have been limited during the recording process. The uniformity is is sharp contract to the variable peaks one sees from old LPs recorded on analog tape. Even with the latter there was a max possible volume that could recorded on the tape so limiters were still used so the average playback level wouldn't be too soft. No one wants the average level so low that background noise gets in the way.

@ghdprentice 

I’ve been looking at SF speakers and the dealer I’ve been talking to has mentioned several times that “These speakers need a lot of watts to sound good!”  One of the speakers I am looking at is the Amity G5 and he said that it needed 600 watts to really sing.  And I immediately thought of you and had to stifle a laugh.  He’s talking and I’m imagining you with 600 X 2 watts of ARC tube equipment.

@mlsstl  I'll be the first to admit how little I know about much of this hobby.  I believe what Sanders is trying to get across is the fact that many people are oblivious to the fact their amps may be clipping and are under powered, and that their system could sound even better with a higher powered amp?.  I suspect you don't need to be playing at excessive volume to be stressing an amp.  I think, and correct me if I'm wrong, that the load the speaker presents to the amps has a lot to do with it.  I owned a pair of Acoustic Zen Crescendo Mark ll.  When you look at the impedance graph, that speaker is almost a flat line at 6 ohms.  Extremely easy on your amp.  My current speakers are the opposite.  Dipping as low as 1 ohm which is a punishing load.  Luckily my amps are stable at 1 ohm.  In summary, I found the Sanders article very interesting and will assume he knows a lot more about this than I do. I don't think it ever hurts to read the lengthy papers written by the designers and builders.   Cheers, 

As I think I remember @ghdprentice  typing, I have also read that maybe not all watts are created equal.

👍 @immatthewj I did mention that and I think @curiousjim ’s dealer is trying to make Sonus Faber sound more like Wilson. I think he is trying to put so much fast current to them (and probably a lean midrange) to increase slam. These speakers are made to reproduce natural sounding music not to do shock and awe. I know it is hard to attribute intent to others. But I have heard SF with very high powered solid state amps, they do make them sound better to me.

@immatthewj   Actually, all watts are the same.  There's a scientific calculation that defines a watt.  It's all the other stuff that makes the difference.   

@bigtwin -- I appreciate your points, but high wattage alone doesn't tell you anything about how an amp behaves when presented with a low impedance load of 4 ohms or under.  And wattage also says nothing about the other elements of sound quality. 

Especially since I don't listen at high levels (85 dB tops) I happen to enjoy my 20 w/ch (8 ohms) Aegir more than the 150 w/ch VTV amp I had immediately prior.

To use a car analogy, it is like focusing only on horsepower while ignoring torque, gearing, suspension, handling, braking and so on. (Of course, the same as with audio buffs, there are some car enthusiasts that will focus only on one trait.)

@bigtwin  , is an unltralinear watt identical to a triode watt?  That's a serious question, I am not intending to be sarcastic.

Beyond my paygrade.  We should all buy and listen to what we like.  There is no one right answer when it comes to amp power.  As pointed out in several comments, which I agree with, there are many factors that determine the proper combination of equipment.  I can only say that with my speakers, a low powered amp is not that correct equipment.  Cheers.

@immatthewj   I think that Sanders covered this question fully in his paper?  

@immatthewj   I think that Sanders covered this question fully in his paper? 

@bigtwin  , I am honestly not familiar with Sanders or the paper you referenced, but is it accessible on this site?  What would I need to use (besides "Sanders") as a search engine to find it?

@immatthewj  Scroll up about 15 posts and you will see I pasted a potion of the paper and included a link to the full paper.  Sanders does a good job explaining the perceived difference in tubes and SS, and why many systems may be underpowered.   Once again, I am assuming he knows what he is talking about after designing and building amps and speakers for decades.   For me anyway, he calls into question all the members who claim only fhe first watt is important and that 50 watt amps are all you need.   Read the whole paper and tell me what you think.  Cheers. 

The Sanders white paper is excellent. The two big takeaways are 1. that transient power demand is far higher than we think or that anything short of an oscilloscope can measure and, 2. the recovery time of the power supply in responding to transient power demands is critical.

"is an unltralinear watt identical to a triode watt? That’s a serious question, I am not intending to be sarcastic."

Basically, yes. But ultralinear operation has plusses and minuses. and a lot of those depend on the quality of the output transformer, the percentage of the screen taps relative to the plate taps, etc. You can get essentialy twice the power of triode operation (or more) but you also introduce stability and distortion factors that must be addressed. In a typical feedback amplifier, switching from triode to ultralinear operation requires adjustments in the feedback components to optimize the amp’s performance and stability. So a big question when buying a switchable amp is, does switching from one mode to another involve correctly adjusting the feedback components to allow for optimal performance? An amp oprimized for triode operation can encounter some serious issues when switched to UL mode, and vice versa. If no such arrangements are made, then one mode or the other is going to suffer. Often times, people with switchable amps will report that they tried the triode switch but didn’t like it because the sound quality was reduced. Or, they’ll report that triode mode was more "open" but the bass control was worse, etc. Usually these negative impressions are the result of the amp not being properly designed to optimize each mode of operation.

@bigtwin thanks for the Sanders white paper! An excellent read.

The Sanders paper is very interesting.  But one should remember a couple of things:

1) He's talking about ESL speakers in the main;

2) I think he exaggerates the frequency and extent of clipping in an appropriately matched system.

I would venture to suggest, at the risk of possible ridicule, that an audiophile in 1950 with a pair of Klipsch Cornwalls and a 15 wpc Wliiamson amp would rarely have encountered clipping and was probably enjoying a better musical experience than a lot of the power-eating systems available today.  In modern terms, I'll also stick my neck out and suggest that an audiophile with a pair of Vandersteen 1Cs and a well-restored Eico HF-87 will find a good deal of musical contentment with that arrangement, with far fewer headaches and travails than an amateur attempting to joust with walls of drivers and 600 wpc. ;-)

As a counter-argument to Sanders, I enjoyed this interview with amp designer Justin Weber, who essentially maintains that most audiophiles need *less* power than they think.  He also offers a very amusing and (IMO) very pertinent discussion about single-ended amplifiers:

@dogearedaudio   You are correct that Sanders is speaking about ESL for the  most part.  I listened to your attached file.  Very interesting.  I would suggest you could start at the 25 min mark as he doesn't really get into the meat of the subject before that.  Incredibly knowledgeable guy.  Kind of funny how many of the amps his own company produces are in the 50 - 100 watt range when 5 watts is sufficient for 95% of speakers and 99% of listening rooms (I think I got that right).  Anyway, a good listen for sure.  I liked his comment that we should all start by choosing our speakers first and then work backwards to the right amp.  Maybe that's why I liked the Sanders paper as I am running very large panel speakers.  Cheers. 

Also consider clipping for tubes and SS is very different. (I'm running SS now but owned 3 tube amps and use a tube DAC now so I'm not trying to pick a side here)

Part of the reason we like tubes, is the distortion, which increases as they get overdriven, and is obviously part of the design/intent in tube guitar amps. 

It's Jazz Sunday in our house today, and something else I thought of as I play both analogue and digital albums, is the wide variation of recording levels. It's nice to have (virtually) unlimited reserves of clean power for those lower level recordings.

Lastly, I don't understand this.

with far fewer headaches and travails than an amateur attempting to joust with walls of drivers and 600 wpc

You don't need walls of drivers and the only jousting with 600wpc is financial/spouse approval.

I don't think you need that many watts to drive most speakers.  I was using a 10 watt class A amp to drive a pair of Super Tablettes (86 db I think) in a 13x14 listening room with zero issues.  So then I thought, "What the hell!" and put in my 2A3 integrated amp.  Still no issues and the speakers sound great.

Maybe my musical choice isn't demanding?  I normally don't listen above 85db or so in volume.

In the past I would have been afraid to try this combo in fear I might damage something.  But I guess as I get older I take all the experts' advice as they really are...an opinion.  And you know what they say about opinions.

This combo works for me.

"You don't need walls of drivers and the only jousting with 600wpc is financial/spouse approval."

My point was that I have seen systems so expensive, so complicated and striving for so much power and scale that they become unmanageable for the owner.  They're often difficult to listen to as well.

"I was using a 10 watt class A amp to drive a pair of Super Tablettes (86 db I think) in a 13x14 listening room with zero issues. So then I thought, "What the hell!" and put in my 2A3 integrated amp. Still no issues and the speakers sound great."

ProAcs are rather famously amenable to lower-powered tube amps. My first audio heartthrob was a pair of Studios with the little Cary stereo 300B amp I heard in a shop. I was thrilled a few years later when a friend sold me his Response 2’s. Thirty years on, with one set of replaced drivers and two sets in storage, I’m still enjoying them. ;-) I used 300B amps with them for years but the quest for better bandwidth and slightly more power led me to my current 20 wpc Williamson amps.