From the Roger Sanders interview on Dagogo:
Below is Roger Sanders opinion on power previously quoted here and in his whitepaper:
As an aside, I continue to be amazed by the fact that most audiophiles are not aware that they are using underpowered amplifiers that are clipping and distorting most of the time. It is easy to show that most speakers need around 400-500 W/channel to play dynamic music at the loud levels audiophiles enjoy. If you doubt this, just connect an oscilloscope to your speakers and watch it. You will see the trace run into an invisible "brick wall" when the amplifier runs out of power and clips. In most systems, you will see the amplifier clipping constantly on musical peaks.I routinely use an oscilloscope to look at power in my system. I have permanently installed a classic Textronics rack mount scope in my rack so I dont have to fetch one off the bench when I am curious.
I set the brick wall at the top of the screen and indeed watch how close I come to it. The headroom I see (the distance from the peak to the brick wall) is often 10 db on peaks. Using a 100 watt amp that is a peak of 10 watts with 90 watts to spare. I know Roger Sanders work and I know he makes a very power amplifier for his speakers and I have no doubt he uses its full power listening at very high levels (which I don't). The big difference is not in the speaker and amplifier but in the listener and his desired levels.
I do agree with Sanders that you can see your headroom with the most basic oscilloscope which can be had on eBay for under $100, (just make sure the seller promises that it works, is reasonably accurate and will take it back if not). I do not agree
"that most audiophiles are not aware that they are using underpowered amplifiers that are clipping and distorting most of the time".
There are several easy ways to check accuracy which I will be happy to write about if asked. Most scopes have a built in calibrator and even if the accuracy is 10% off, its plenty good enough to determine peak power and clipping and further determine if your amp is putting out what it should or if it is being limited by speaker impedance dips, cable losses, low line voltage or any number of things.
All this being said, with a 90 db speaker I do not see clipping at my average listening levels with even the 15 watt EL-84 Class A version of the RM-10 that I am currently working on. Furthermore the standard 35 watt per channel RM-10 satisfies most listeners and they often comment on how loud it will play their speakers and how much bass impact this little, light weight amplifier has.
Sanders further states:
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.I would say from the above that Mr Sanders
likes to play his music more loudly than most. Due to what I cannot say, though I hope he has not severely damaged his hearing. I routinely check mine, often in a group of several people at once, it's not hard to do. I connect an oscillator to a speaker, set a moderate level like 70-80 dB start at 400 Hz, everybody raises their finger and drops it when they can no longer hear the tone. You could do the same at several frequencies and lower the level and watch fingers drop. It's a very interesting test and you can use CD with a sweep and tones or get them off your computer with free software. Get a group together on a Saturday and see where everyone is at.
Another thing I do when playing my system for others is to routinely ask if the volume I set is higher or lower than than they usually listen to. I find a fairly equal number who like higher and lower and the same as I listen to.
The room size matters little as most of our rooms are relatively dead (by choice) and the reflected energy is thus low and not material. What does matter is how far you listen from the speaker as most speakers fall off 6 dB as you double the distance. At 4 meters (13 feet) the 90 db speaker is now 78 dB per watt but two of them are 84 dB which is a level most people listen at, which means one watt per channel. We can argue for large peaks in live music but with music as compressed as it is these days on CDs 10 dB peaks are about all you can expect unless it is a very quiet CD on the average. Ten dB above 1 watt is 10 watts and you still have headroom. In my experience occasional clipping on the scope is not heard by even the most golden eared listeners. However if you do see clipping on the scope, just turn it down a bit.
Another thing you may see on the scope even with well recorded vinyl like Willie Nelson's Stardust is clipping in the recording on rim shots on the ALL of Me track. We used that record extensively at Beveridge and I never noticed it till I saw it on the scope. So in many cases, the source material is going to clip before your amp.
One thing that Sanders does not address is the "First Watt" principle that Nelson Pass, myself and many others feel is more important than unused headroom. Granted some headroom is a good thing but for a listener who is using one watt, 99 watts of headroom are not needed. People listening at 70 dB (and there are many) are using far less than that.
