I found this equation in article about perceived loudness:
L[%] = k^(1/3.5) where k is ratio of power.
According to this 200W amp will sound 21.9% louder than 100W amp.
150W/100W ratio will make it 12.3% louder.
Somehow it is easier for me to comprehend it in percentage.
Perception and Watts: Doubling of power
There's a curious rule of thumb, which to my ears seems mostly true:
- To double the perceived volume, you must output 10x more power.
10x power = 10 dB by the way. We've read this as we were buying amps and trying to decide between 100w/channel and 150w/channel. We are told, repeatedly that 50 W difference isn't really that much.
On more than one occasion I've tested this and found it's pretty much spot on. Here's my question:
How can any of us really tell what half as loud, or twice as loud is?
I mean, think about this for a bit. I cannot tell half as bright, or twice as bright, but it seems I actually CAN tell what half as loud is. How does this even begin to work in the ear/brain mechanism?? 😁
56 responses Add your response
Different frequencies consume different power so "twice as loud" would be frequency dependent in many cases. Paying for an additional 50 watts will improve low frequency performance -- you may not hear it louder but you will hear it cleaner with respect to transients and lower distortion from better controlling the woofer. It gets integrated into the music a little better. |
@gs5556 This is my thought as well. Even at the same overall volume, the music may sound better and more controlled, with better impulse response. And, some amps distort closer to full power if that is where I have to run it it get the volume I like. I have found there to be better bass response in systems with more power, even if the average overall volume is about the same. |
Hey @fuzztone , based on you asking questions like "Why do I care" on another thread, it may be helpful to remind you that this is a community driven site and not one you pay for. If you find threads you are not interested in feel free to ignore them. You might also want to take a look at the number of others who have participated in these threads and think "well, others are having fun so I should let them." Just a thought. |
We can easily distinguish changes in loudness but human beings are not SPL meters - our sensory system doesn't work that way in the sense of being able to calibrate what we hear in dB - cue the eleven is one louder sketch in Spinal Tap. As regards the reference to amplifier power in the OP. I don't think many people look for more powerful amplifiers from the perspective of playing louder. It's more about dynamic range and playing at the desired volume level without distortion. Speaking of distortion, a low wattage amplifier driven into clipping will sound subjectively louder than a high power amplifier playing at equivalent or lower SPLs. As another example, modern popular music which has been compressed to hell sounds superficially louder - hence the loudness wars in modern mastering. So loudness is a bit of an ephemeral concept. |
Playing loud and watts seldom comes into the picture in… well non-low end systems (ok, maybe for college parties). Unless there is a gross mismatch in sensitivity and power. More power tends to add weight and realism to the sound you get… it fills it out more than making it louder. I have Audio Research Ref160 monoblocks, they can be run in linear (140wpc) or in triode mode (70 wpc)… there is no obvious difference in volume what so ever. I was actually surprised… I was listening for solidity, but thought I would hear a difference. The only difference I heard was in the ultra linear mode the treble became a bit hard and some of the midrange bloom dried up. I operate them in the triode mode, and they can play at deafening sound levels… (a dealer came over to listen and cranked them… wow, that was loud). In solid state amps the power directly transforms itself into slam… that is one of the reasons high power ss amps are prized. I only feel like I have scratched the surface. We’ll see if anyone wants to continue. |
First you have to understand there are quality 100 watts of power and there are garbage 100 watts of power. If all you are concerned about is the decibels or loudness of the music come over to my house and I will hook my Sansui 7070 and the Bose 901’s sitting in their original boxes. It will play so loud your ears will ring for days and it will sound like garbage. Then I will play my 80 watt monos into my Vandersteen’s and not only will they play loud but it will sound beautiful to boot. They did not name the company first watt as a gimmick. |
I can’t count how many times I’ve read or heard people erroneously claim that a more powerful amp sounds better. If that were true, a 300b "flea watt" amp would sound terrible. |
Lots of interesting discussion here, of course, but I really did just want to ask, how can we tell what twice as loud is? I mean, I think I can, and it happens to match 10 dB. WHY? Our visual systems however seem very good at seeing twice the length or half the length, or finding the mid point on a line. Kind of weird that we also have a sense for half or twice as loud. |
I also never really understood what was meant by “perceived” as sounding twice as loud…I figured it was too subjective, but apparently there are some safe generalizations too be made with the way human ears work. @mrskeptic I agree |
Here's a question, @johnnycamp5 - Lets say you play some music. Can you turn it down by half? Or can you turn it up to be twice as loud? How do you know?? I mean, I can, and I think you can too! but... how do we know what half is? |
I have recently been playing with speaker cables on my tube amp system (PM EVO 400 pre and power feeding Klipsch Cornwall 4's) and have found that the choice of speaker wire gauge has large effects on perceived volume and "control." There is nothing new here,of course, but it is fun to experiment with the so called "Western Electric" multi-stranded tinned copper wire. From 16 gauge to 10 gauge there is room for lots of slants on the sound emerging from the speakers. Since the Cornwalls are bi-wireable there is also room to play with mixing wire gauges between the bass taps and the treble taps too. Have also been experimenting with ziploc baggies filled with powdered Rochelle salt wrapped around speaker cables. This is based on a comment Caelin Gabriel of Shunyata once made about these salts being good at attenuating ambient electromagnetic noise,
|
Hi Erik, twice the volume, i.e. twice as much sound pressure level, is 6dB(spl). And this requires 4x the power in watts. It seems that 10dB is what most of us perceive as double sound intensity, i.e. from a psychoacoustic point of view. And even then, our perception is affected by the frequency. The lower the frequency, the lower our sensitivity, so +10dB may not be enough in the bass, for example... Interesting stuff! |
@iseland I think you mean an increase by 3db requires double the power (or watts) Certainly this is not a tripling of average SPL |
Back in the day we found that more speakers (specicially: tweeters) got killed by too SMALL amp that were clipping like crazy when playing louder, and the clipping sent huge amounts of watts to the tweeter. Most tweeters blow at 10W (just throwing a number at it), so a 20watt amp is capable of killing a tweeter (if sending a lot of high frequency clipping noise to the tweeter). |
Couple of things come to mind. When you buy an amp w twice the rated power, the power supply in that amp is generally much bigger than in the smaller one. As was mentioned in a post above, a bigger amp playing at the same volume as a smaller amp will sound more dynamic - provided they are built to the same quality level. I tend to believe this difference is connected to the size of the power supply, more so than the power (watt) capability of the amp. The Hiraga Super 30 watt class A amp I built, has 192K uF in the PS, with a 500VA transformer. People build these amps with insane amounts of capacitance, but for this built, I went with this. Next time I build this amp, I’ll be building it as a dual mono design with greater capacitance in the storage so I can hear first hand how that translates into real world listening experience. I recently built a single ended EL84 based amp. It’s about 3 watts. Love the sound, but want a bit more usable volume. So I’m about to built a single ended EL34 based amp that will get me approximately 6 watts. Double the power. Different tubes and circuit design, I know, but I’m just curious how that 6 watts will present compared to 3 watts. 3 watts don’t mean a hill of beans if your amp is pushing 200watts into the speakers, but going from 3 to 6 is doubling the power. Thankfully, going from 3 to 6 watts is a heck of a lot less expensive than going from 150 to 300 watts :) |
Stated a different way, a 3dB increase in loudness requires twice the power. This is the reason speaker sensitivity is so important when matching to the power amp. If your speakers are inefficient and maxxing a 300 watt amp, 600 watts gets you just 3db. Re. the original post, humans are remarkably good at judging relative sound pressure differences. I read about this long ago so I can’t cite the research. But for grins I tried this with my sound pressure meter. I was pretty darned close doubling or halving loudness by ear.
|
Erik: Once I get past the first 2 "steps" on my dual mono volume pots each step offers an approx. 3 dB increase. 3 clicks is then approx. twice as loud. When I hold Alexa in my hand, while she is screaming @ me, the SPL is approx. 80 dB. When I then throw her across the room the SPL diminishes considerably.
DeKay |
10db should be perceived as double loud by most people, so to understand double loud you would need a 500w amp instead of a 50w one on same speaker. On the other hand our ears are not measuring equipment, take some time to adjust and have self protection that would change how some things sound in loud volumes. My preamp increases +3db on every 6 steps (0,5 a step), ha!
|
@johnnycamp5 you are absolutely correct, was a bit tiered at that moment.. |
@anotherbob Don't get taken in by this pseudo science. It is a con. There is no system that can instantaneously increase power. Musical transients are gone by the time the power is produced. Krell started this when they wanted to appear greener. They stopped making pure Class A amps for which they had been rightly applauded and said their new amps could increase power instantly so that transients could still be correctly reproduced. Further, such a system might not reproduce music consistently because the watt or two that suffices most of the time would not sound the same as 100 or 200 watts suddenly introduced by the amp. I don't have such an amp but I wonder if anyone has noted this. |
@anotherbob - Marketing hype for a couple of different things some amps do. There were some like NAD / Proton (back in the day) which famously used 2 voltage rails in the amps. The high voltage had no staying power, but for short transients could deliver more than the low voltage rails could do. I think this was Class H. Bob Carver’s amps did something like this, using a linear amp which would float among multiple voltage rails. The NAD/Hypex hybrids do this also. A lot of this has to do with federal regulations about how you rate a power amplifier. To combat outrageously useless amplifier specs of the day the FTS imposed regulations about not only stating distortion and power together but also your amp must be preheated. The preheating required a lot more heat sinks to achieve the same wattage rating. As a result, some amplifiers may have significant more headroom than they may show in the power spec alone. To be honest, music is never steady state and a lot of audiophiles with 300 Watt amps never go beyond 30, so there’s a lot to be said about right sizing an amp to your needs. . |
Erik: The early NAD PE's used class G. I demo'd a few of those integrated amps and SQ improved, IMO, when the Power Envelope circuitry was defeated. Do the meters on your Luxman register "fast" peaks? I had a Carver MXR-130 hooked up to single driver speakers (Stephens 80FR) and the meters showed them to be pulling quite a bit of power on musical peaks. There are two meter settings (low/high) and it only showed up on one of the settings (forget which). This said, I do not notice clipping/compression when running them with my 3 watt 2A3 amps, but I doubt that I ever listen louder than 80 dB (probably much lower, but with unknown peak levels).
DeKay
|
Thanks for the correction, @dekay
Not a tall, I don't think of them as very twitchy. |
Here's a blurb on meters. I recall a few vintage pieces with fast/slow meter settings, but the Carver switch is marked low/high. The Carver has not been used for @ least 10 years (afraid to power it on). http://www.cordellaudio.com/instrumentation/power_level_meter.shtml DeKay |
How can any of us really tell what half as loud, or twice as loud is? Using music as a test signal for determining half or twice loudness is indeed difficult. Music varies in pitch and loudness and so is a moving target. The bel scale was developed by Bell Labs using steady tones on multiple test subjects and determining an average estimation of half or twice as loud. The difference in acoustic pressure called a bel. Decibel, of course being 1/10 of a bel. So a bel is what the average human perceives as twice of half as loud. The actual acoustic pressure between the threshold of hearing and the threshold of pain is more than a million to one. The human ear/brain compresses this range in a logarithmic manor so as to be manageable. This is why the objective acoustic power must be increased by 10x to be perceived as twice as loud. To answer the original question: there is generally no reason for us to worry about what is half of twice as loud. But, it is good to understand that a 200 W amp will only play 3db (21.9%) louder before clipping than a 100 W amp. |
