Watt nonsense!

[dcmfan]

The research: so I started to read up on the relationship between power output, and decibels, and decibels and sound pressure. I got confused. But my take away from that was this: a 40w amp is certainly not twice as loud as a 20w amp. It's really only a little bit louder.

Several folks in this thread have given good input on this already. Maybe a short "cheat sheet" will help as you read their responses.

First, here are two key things to keep in mind:

• A change of 3db is a "noticeable change", not a huge change.
• A 3 dB increase requires double the power from the amp.

And- your Celestions have a sensitivity rating of 90 db, which means that they produce 90db of " loud" from 1 watt, measured one meter from the speaker.

With that, we can make a table like this to see how loud things can go at various wattage:

1 watt = 90 db
2 watts = 93
4 watts = 96
8 watts = 99
16 watts = 102
32 watts = 105
64 watts = 108
128 watts = 111
and so on...

At the levels I normally listen to, my amp loafs along at less than 1 watt. But there can be sudden peaks from drum strikes or orchestral crescendos that ask the amp to quickly deliver WAY more than one watt. When that happens, the extra watts (aka headroom) from a larger amp come into play.

Hope that's not a re-hash of things you already knew! When I was trying to figure out how large an amp I needed, basic stuff like this helped me figure out roughly where I needed to be as a starting point.

 

[idf.bts]

Interesting thread. It's why I argue that there is no replacement for straight up power. The first watt discussion is an esoteric discussion, and I am sure Pass himself would probably say that.

First and last is dynamic range for critical listening. If you have a passage with a 103db peak and your little puffer can only spit out 100db... the party is over. That peak is half as loud as it should be, and will not, I don't care how many first watt threads there are, reproduce the passage at the correct SPL.

I like to listen at live symphony concert levels. Based on my own measurements that is c. 87dBSPL average and 102dB peak where I sit in my usual concert hall. My speakers produce 91dBSPL @1m for 1W equivalent at "8 ohms" 1m distance, that's 11dB less than the max of 102dB, but I listen at 3m from the speakers, so that requires c. 10dB more power, so we're talking about 21dB more power than 1W. if 3dB is a doubling this is about 128W (7 fold doublings), and that assumes an 8ohm speaker. If it's 4 ohms- which it nominally is-, then the power is doubled to 256W, and if the impedance/phase angle drops at any point it can be significantly more.
My amps can deliver c.360W into 4 ohms per channel, and they never clip. But there are times it gets close, and when I just had a single unit rather than the dual mono it was not uncommon for the clip indicator to light up.

Well argued. In an auditorium, the SPL and wattage are way more correlated. You spend all that money on power to reproduce the dynamic range on the recording at home.

The analogy that I like to think about happened to me in the 90s at a Formula 1 weekend in Canada. I was watching the Williams Renault garage when they fired up the engine. They had the laptops plugged into the car and after the engine idled for a moment they sent a few peak max rev signals in and back to idle. With all the inertia in that spinning engine block, the speed at which that engine went from idle to peak left me in disbelief. It was instantaneous up and just as fast back down. Racing digression aside, the point is that engine's power is what made that piece of "dynamic range" I witnessed possible, same physics with music.

 

[62caddy]

TL;DR - Is stated power output highly overvalued?

I don't suppose I'm the first person to comment on this, but as someone still in the honeymoon period of vintage audiophilia, I'd be interested to hear some community wisdom on the question of power output. Or more specifically, its relationship to "how loud the thing goes", if that's not too crude a term.

The context: my love affair with hifi re-ignited over the 2020 lockdown and has been focussed on 1970's kit, and specifically, Japanese-made receivers. In a relatively short space of time (they're not making any more, quick, you might never see that one again!) I acquired a number of units from Pioneer, Yamaha, Rotel and Luxman, all rated between 25 and 50 watts.

The confusion: The two more powerful units (sx780 and sx750) did sound louder than the 25/30w units, and so with some excitement, I sold my old Saab (wasn't getting used) and increased my budget to some upper-mid level machines, notably a Marantz 2285 and a Luxman R-1120. With 85w and 120w respectively, I assumed these were going to offer intoxicating, near limitless reserves of power.
I was wrong.
They sounded maybe a "bit" louder than the two Pioneers, but not my much. In fact it may have been wishful thinking as I'm not completely convinced the Sx780 doesn't still have them both beat for sheer grunt.

The research: so I started to read up on the relationship between power output, and decibels, and decibels and sound pressure. I got confused. But my take away from that was this: a 40w amp is certainly not twice as loud as a 20w amp. It's really only a little bit louder. If you want to get an amp that sounds twice as loud - then you'll need a 200w amp.

The questions:
1. That may be over simplistic, but my experience seems to suggest it may not be far off the mark. Can any of AK's more technical members corroborate that ballpark 10:1 power output ratio required to double perceived volume?

2. Also, if that is the case - and apologies if my post starts to trespass here into values territory - why the huge premiums paid for high power machines if the difference is volume is really quite modest? I realise some of the bigger receivers have better tuners and more speaker outputs, but it seems equally true, especially for, say, Marantz, that the lower power machines sounds just as, if not, sweeter than their bigger brothers.

3. I have also noticed that the CD out signal (from a Marantz CD73 and Philips CD160) sounds noticeably louder than other sources such as the built-in tuner or playing my iphone via iFi Zen Blue DAC into aux. Why is that? Because it's a higher voltage? Could you increase the voltage on any source (eg a DAC) and reap similar "gains" in volume? Is there any risk to amp/speakers from "louder" sources?

4. I also read - on AK I think - that Pioneer tended to calibrate the volume control so that most of the welly comes before the midday position, and after that, there's not much to give. Is there any truth in that? It's hard for me to test as most of the units I have are already pretty damn loud by that stage and I wouldn't want to risk blowing up anything by going into the afternoon, as it were. How many on here regularly crank up the volume past the half way mark?

5. I read people talking about "headroom" and how good it is to have power in reserve, but is there any science in that? I mean it's not like we're driving and we need power in reserve to overtake. If you had a blind test with an sx450 playing at 50% and an sx950 playing at say 30%, so that the sound level seemed the same, would you be able to tell the difference?

Thanks in advance for anyone with the patience to read this far and consider any of these rather ill-informed queries. It would help me allay current suspicion that, for the great majority of speakers, especially if only one pair is connected, having anything over 50 watts is probably overkill.

Craig

ps 1. I appreciate this is not a very sophisticated line of questioning, and that volume/power output is only one part of a system's ability to entertain, but who doesn't enjoy that moment when the volume dial is only at 2.5 and you can feel the floor vibrating already ; )
ps 2. Receivers normally tested with Celestion Ditton 25s, but also Beovox M70 and S45, and a modern pair of Monitor Audio (Monitor 50). The woofers on the Dittons are nearly 50 years old and positively heroic in their ability to move air

You're entirely fixated too much on raw wattage numbers which is only one side of the SPL (sound pressure level) equation. Just as important, if not more so, is loudspeaker sensitivity. For example a loudspeaker with 110 dB sensitivity will play just as loudly (at maximum) driven by 1 watt amplifier as an 80 dB sensitivity speaker driven by a 1,000 watt amplifier. Now that doesn't take into account the very different sonic characteristics that are likely to exist between two such speaker designs, ie: the lower sensitivity model will probably have superior low frequency response while the mids on the high sens model will be more pronounced, in essence more "volume" in that part of the frequency spectrum.

Secondly, when amplifiers are operated within their output limits for clean power (ie at low distortion), there is no reason one should sound any "louder" than the other assuming they're both operating properly. The loudspeaker has no way of knowing, nor responding any differently whether, say a 10 watt signal is being supplied by a 20 watt amplifier than 10 watts being supplied by a 1,000 watt model. It is when you feed the speaker more power than the lower power amplifier is capable of - only then would the higher powered unit make any difference in loudness.

 

[wyn palmer]

Interesting thread. It's why I argue that there is no replacement for straight up power. The first watt discussion is an esoteric discussion, and I am sure Pass himself would probably say that.

First and last is dynamic range for critical listening. If you have a passage with a 103db peak and your little puffer can only spit out 100db... the party is over. That peak is half as loud as it should be, and will not, I don't care how many first watt threads there are, reproduce the passage at the correct SPL.



Well argued. In an auditorium, the SPL and wattage are way more correlated. You spend all that money on power to reproduce the dynamic range on the recording at home.

The analogy that I like to think about happened to me in the 90s at a Formula 1 weekend in Canada. I was watching the Williams Renault garage when they fired up the engine. They had the laptops plugged into the car and after the engine idled for a moment they sent a few peak max rev signals in and back to idle. With all the inertia in that spinning engine block, the speed at which that engine went from idle to peak left me in disbelief. It was instantaneous up and just as fast back down. Racing digression aside, the point is that engine's power is what made that piece of "dynamic range" I witnessed possible, same physics with music.

Even at home, unless you are past the critical distance from the speakers- i.e. you're not listening in the region where the reverberation sound field exceeds the direct field- the calculations are very simple if you know the sensitivity of your speakers.
amroc - THE Room Mode Calculator (amcoustics.com)
Everyone should use the above calculator at least to get some idea of the basic state of their listening environment.

By the way, the benefit of the "first watt" concept is that it recognizes a fundamental aspect of hearing, that at low acoustic volumes the ear/brain interface is most sensitive to extremely low levels of high order distortion, and that is where many amplifiers are in fact worst due to aspects like crossover distortion artifacts.

 

[idf.bts]

Even at home, once you are past the critical distance from the speakers- i.e. you're not listening in the region where the reverberation sound field exceeds the direct field- the calculations are very simple if you know the sensitivity of your speakers.

This is the physics driving the near-field listening advocates.

By the way, the benefit of the "first watt" concept is that it recognizes a fundamental aspect of hearing, that at low acoustic volumes the ear/brain interface is most sensitive to extremely low levels of high order distortion, and that is where many amplifiers are in fact worst due to aspects like crossover distortion artifacts.

Well summarized. Pass has a highly refined set of ideas not suited for sound bites so I once heard first watt morph into "you only need 1 watt" or normally "most important watt" becomes mantra-tization without knowing what was intended by it.

 

[wyn palmer]

This is the physics driving the near-field listening advocates.




Well summarized. Pass has a highly refined set of ideas not suited for sound bites so I once heard first watt morph into "you only need 1 watt" or normally "most important watt" becomes mantra-tization without knowing what was intended by it.

The difference is that I'm not a Pass advocate, per se. Just an advocate for well-designed ultra-low distortion at all listening levels amplifiers such as the AHB2 and Halcro amps. Probably also the purifi class D amps, but I've not had the opportunity to test/listen to any of those.

 

[daveshel]

A read of DaveG's analysys of the shortcomings of the Dynaco ST-35 (and ST-70) shows how often an amplifier's power supply is the weak link - many amplifiers' power supplies will sag under the demands of having to supply power to both channels at the same time. His EFB/power supply boards go a long way toward improving the situation while maintaining stereo operation, but another (albeit more expensive) way to achieve the same end is to duplicate the entire power supply in the form of monoblocks. The idea is simple: a dedicated power supply for each amplifier allows for more current in the amplifier circuits, which means more amplification. Transistors came along and amp designers created massive power supplied with beer-can-sized capacitors and built the monster solid-state amps of the 70s - the Dynaco 4xx series, Phase Linear, and eventually Soundcraftsmen. Look inside these and you will see lots of bug caps. Looking inside receivers, you will occasionally see impressive caps, such as on the SX-1980, but even there, these power supplies have to run the preamp and the tuner and the phono preamp in addition to the amplifiers. (And the front panel lights, which were still incandescent and relatively power-hungry, thus the dimming of panel lights with peaks mentioned above.)

This is why I have never owned a receiver.

 

[ejman]

Even at home, unless you are past the critical distance from the speakers- i.e. you're not listening in the region where the reverberation sound field exceeds the direct field- the calculations are very simple if you know the sensitivity of your speakers.
amroc - THE Room Mode Calculator (amcoustics.com)
Everyone should use the above calculator at least to get some idea of the basic state of their listening environment.

By the way, the benefit of the "first watt" concept is that it recognizes a fundamental aspect of hearing, that at low acoustic volumes the ear/brain interface is most sensitive to extremely low levels of high order distortion, and that is where many amplifiers are in fact worst due to aspects like crossover distortion artifacts.

Very good point. I noticed with a high power amp I have that it sounded better, less "edgy" when I switched from short 6 ft regular 12 gauge speaker wires to long 18 ft 18 gauge solid core wire. I noticed that I had to set the volume control higher for equivalent listening level. This told me that the wire resistance went up so that the amp had to work harder and probably get out of the zone where crossover distortion products were audible. (The amp runs warmer too). This was at 60-70 db listening levels according to my phone app.

 

[goodolpg]

For about 40 years I did just fine with 2 receivers that had 90 wpc.
When nostalgia hit me around 2013 I proceeded to buy 1 120 wpc receiver and 2 125 wpc receivers. No big gain in power, just receivers I would have liked late 70's early 80's and couldn't afford then.
At some point I went to separates, first amp might have been because I found a 300wpc Rotel amp for about $200 on ebay. Then a while later I saw a half price discontinued sale on a Crown XLS2000. 650 wpc into 4 ohms.
Now I had the power to run some difficult to drive speakers.
I think the increase in power might have been audible, but I don't give my system hell like I did 45 years ago. Still, the extra power is not a bad thing to have and allows me to consider some "tough load" speakers that I'd never consider if I was still using a receiver.
More power, used wisely, is almost always a good thing.

 

[wyn palmer]

Very good point. I noticed with a high power amp I have that it sounded better, less "edgy" when I switched from short 6 ft regular 12 gauge speaker wires to long 18 ft 18 gauge solid core wire. I noticed that I had to set the volume control higher for equivalent listening level. This told me that the wire resistance went up so that the amp had to work harder and probably get out of the zone where crossover distortion products were audible. (The amp runs warmer too). This was at 60-70 db listening levels according to my phone app.

Depending on your amp and speaker, the change in wire could have altered the frequency response in ways that you preferred.
Yes. Speaker cables, amp output impedances and speaker complex impedances do matter- and it is demonstrable and provable.
However, that's a subject for another thread.
Speaker wire | Audiokarma Home Audio Stereo Discussion Forums

 

[Snow]

Not all of us can ever hear the difference a powerful amp can produce at low volumes and some are perfectly happy with a receiver or lower powered amp. I know I was satisfied with my first new receiver it was a Sony ES model that I paid $1500.00 for from a local audio store but after I borrowed a friends amplifier I instantly knew that a separate amp was the way to go for me and after I bought my first powerful amp an Altec 9440A I knew then the difference between that and other amps that I owned was profound even at lower volumes the music sounded different. There is nothing wrong with whatever you choose to power your speakers but if you want to listen to music and hear it as it was designed to be heard then I don't know of any other way of hearing it with SS equipment without using a powerful amp. I typically listen to music at low levels but it sounds better with a powerful amp even at lower levels to my ears then a lower powered amp does, perhaps it is because of the extra headroom larger caps that can replicate the intended sound when a jump from a few watts to a few hundred watts needed happens I dunno I just know that it sounds better to my ears.

 

[rothwellaudio]

...it was a Sony ES model that I paid $1500.00 for from a local audio store but after I borrowed a friends amplifier I instantly knew that a separate amp was the way to go for me and after I bought my first powerful amp an Altec 9440A I knew then the difference between that and other amps that I owned was profound even at lower volumes the music sounded different.

So, the Altec sounded better than the Sony, but can you attribute that to the higher power? Or is the Altec just a better designed amplifier with a better circuit? The power output is a very easy to latch onto spec, but the circuits employed in the two amps could be very different, though not apparent to the average consumer who doesn't study circuit diagrams.
I think it's entirely possible that the extra power of the good-sounding amp is purely coincidental. After all, you do say the improvement is apparent at low listening levels where both amps must surely be operating well below their maximum output.

 

[toddalin]

So, the Altec sounded better than the Sony, but can you attribute that to the higher power? Or is the Altec just a better designed amplifier with a better circuit? The power output is a very easy to latch onto spec, but the circuits employed in the two amps could be very different, though not apparent to the average consumer who doesn't study circuit diagrams.
I think it's entirely possible that the extra power of the good-sounding amp is purely coincidental. After all, you do say the improvement is apparent at low listening levels where both amps must surely be operating well below their maximum output.

Agreed, the design of the amp has a he-- of a lot to do with the sound you hear, probably moreso than the power level when operated within the designed range.

And here's examples of a high current, solid state RECEIVER (Yamaha RX-Z9, 170 WPC, 540 WPC dynamic power @ 2 ohms) vs a p/p KT-88 (80 wpc) tube amp. The speakers have an efficiency of over 96 dB @1 watt @ 1 meter, so power is not really a factor. But the Heil AMT certainly seems to think that current is. Both tracks are matched to within 0.1 dB using pink noise.

Vanessa on tubes:

Vanessa on high current solid state:

SRV on tubes:

SRV on high current solid state:

BTW, anyone who thinks that 3 dB is the minimum for detectable change is sadly mistaken. My Yamaha is calibrated in 0.5 dB steps and I can hear changes of this magnitude in the sensitive area and changes of 1 dB in the less sensitive areas of hearing. Studies have actually shown that we can hear changes of 0.2 dB if it shifts the frequency balance.

 

[ejman]

Agreed, the design of the amp has a he-- of a lot to do with the sound you hear, probably moreso than the power level when operated within the designed range.

And here's examples of a high current, solid state RECEIVER (Yamaha RX-Z9, 170 WPC, 540 WPC dynamic power @ 2 ohms) vs a p/p KT-88 (80 wpc) tube amp. The speakers have an efficiency of over 96 dB @1 watt @ 1 meter, so power is not really a factor. But the Heil AMT certainly seems to think that current is. Both tracks are matched to within 0.1 dB using pink noise.

Vanessa on tubes:

Vanessa on high current solid state:

SRV on tubes:

SRV on high current solid state:

BTW, anyone who thinks that 3 dB is the minimum for detectable change is sadly mistaken. My Yamaha is calibrated in 0.5 dB steps and I can hear changes of this magnitude in the sensitive area and changes of 1 dB in the less sensitive areas of hearing. Studies have actually shown that we can hear changes of 0.2 dB if it shifts the frequency balance.

wow remarkable difference between the 2 amps. What's the kt88 amp?

 

[Powertech]

The issues you have when an amplifier goes into clipping -

For the uninitiated this is where in order to achieve the required output voltage to follow the signal correctly the amplifier output approaches or exceeds it's supply voltage. This usually occurs when the supply voltage falls because the power supply can no longer provide sufficient current to support the amplifier's demand on it.

The usual candidates are load excursions in the bass region as that is where the highest power demand lies in most music signals. When this occurs the amplifier clips, but most of the distortion will in fact be heard in the midrange and treble regions because the human ear is less sensitive to low frequency distortion, but very sensitive to higher frequency distortion. As the signal effectively has the tops and bottoms of the waveform chopped off, this is effectively a square wave. It's this distortion that is most audible and it is this squarewave like distortion that destroys tweeters. Remember that with multiway passive speaker systems (most loudspeakers) the amplifier is handling the entire audio frequency spectrum. The tweeter is the most fragile driver in a loudspeaker system so it tends to take the hit.

I run a fully active system. In this case, any clipping initiated distortion which occurs in the bass region will not be appear in the midrange or treble as these ranges use separate amplifiers. My amps are conservatively rated at 60watts per channel (all 6 of them), but probably about 70 - 75 watts into the actual speaker loads in use. When the average signal balance between them is taken into account there is probably at least 170watts per stereo channel available for a system that normally averages around 10 watts per channel at normal listening levels. So you can see the amount of headroom that is available to take musical peaks into account.

I should add that I have never heard this system go into clipping.

What has been iterated above is quite true. A struggling underpowered amplifier is the one that is going to damage tweeters. A good quality high powered amplifier doesn't generally damage loudspeakers with loud short term transients even when they are effectively being overdriven because the signal doesn't contain squarewave like distortion.

 

[ejman]

Sorry I just realized the amps are identied on the video

 

[Snow]

So, the Altec sounded better than the Sony, but can you attribute that to the higher power? Or is the Altec just a better designed amplifier with a better circuit? The power output is a very easy to latch onto spec, but the circuits employed in the two amps could be very different, though not apparent to the average consumer who doesn't study circuit diagrams.
I think it's entirely possible that the extra power of the good-sounding amp is purely coincidental. After all, you do say the improvement is apparent at low listening levels where both amps must surely be operating well below their maximum output.

Anything is possible I suppose and I have certainly heard some great sounding receivers the Pioneer SX-1010 and Sansui G9000DB come to mind but to my ears the best sounding were the higher powered amps Like the Altec 9440A Sunfire 300 X 2 Bedini BA-803 Bose 1801 ETC the speakers I was using were Polk SDA 1.2TL's these all sounded better to my ears at both low and higher volume levels even though there were some nice sounding amps that had less power Like the Kenwood Supreme 600 The higher powered ones sounded better to my ears. Now I am not a technical person so I cant say with any real knowledge as to why they did other then perhaps better deigns and or parts or more headroom bigger caps.

Of course this is just my findings after listening to virtually every amp receiver int amp I could get my ears on your findings may be different everyone ears and likes are different. I no longer have the ability to hear like I used to, the last test I did I could barely hear 21 khz as a young man I could clearly hear 25 khz

 

[Snow]

Agreed, the design of the amp has a he-- of a lot to do with the sound you hear, probably moreso than the power level when operated within the designed range.

And here's examples of a high current, solid state RECEIVER (Yamaha RX-Z9, 170 WPC, 540 WPC dynamic power @ 2 ohms) vs a p/p KT-88 (80 wpc) tube amp. The speakers have an efficiency of over 96 dB @1 watt @ 1 meter, so power is not really a factor. But the Heil AMT certainly seems to think that current is. Both tracks are matched to within 0.1 dB using pink noise.

Vanessa on tubes:

Vanessa on high current solid state:

SRV on tubes:

SRV on high current solid state:

BTW, anyone who thinks that 3 dB is the minimum for detectable change is sadly mistaken. My Yamaha is calibrated in 0.5 dB steps and I can hear changes of this magnitude in the sensitive area and changes of 1 dB in the less sensitive areas of hearing. Studies have actually shown that we can hear changes of 0.2 dB if it shifts the frequency balance.

I don't think it is fair to compare a tube amp with a SS amp if for no other reasons then you can change the sound of the tube amp by putting in different tubes and a tube amp when it distorts sounds way different then a SS amp does and the fact that 10 watts of tube power sounds much louder then 10 wats of SS power and as a rule people tend to initially like a louder sounding amp.

 

[toddalin]

wow remarkable difference between the 2 amps. What's the kt88 amp?

http://meltonaudio.com/cart/amplifier/integrated-amplifier-mkt88p-80w/

 

[62caddy]

I place zero stock in subjective listening comparisons in any situation in which the listener is aware of which amplifier is playing - or for that matter, any audio component.

The negative effects of clipping are generally accepted and are not in dispute. This discussion should only be concerned with sonic differences between amplifiers when driven within their operational limits.