Yet another "How much power do I really need?" thread

[sandeepsasi]

Hello everyone,

This topic has been covered in several posts both in this forum and elsewhere. There are too many parameters involved and it can be really hard to come up with a one-size-fits-all solution and often, analysis has to be done on a case by case basis. However, I'm looking for a reasonable starting point, based on the common parameters specified for amps and speakers. For example, if I would like to try Rega's -R line-up for a speaker of my choice, generally regarded as easy to drive, should I choose Brio-R, Elex-R or Elicit-R? Most of the times, the answer given by the audiophile community,
"Go for Elicit-R if you can afford it. It is a kind of future proofing, have enough reserves of power for the lows, will easily drive any low-efficiency speaker that you may choose to buy in the future, has a superior power supply and output topology, ..."

is a bit vague, inclines towards more is always good philosophy and will just add to the confusion of a beginner like me, who's always on a budget. The Elex-R is roughly twice as expensive as Brio-R and Elicit-R costs about thrice, so all those benefits of Elicit-R over Brio-R comes at a cost. Here, I'm trying make an engineering estimate of what is the minimum power needed, to short list the amps first. Later, I can factor in sound quality and then, go one level up, if budget permits. If I really have the budget, I can even consider a better quality amp, both in terms of build and sound, from a different manufacturer, rated at a slightly higher than the minimum estimated power, but lower that the power rating of Elicit-R. In other words, I'm trying to find out where we stop paying for watts and start focusing on sound quality instead.

This discussion assumes that the speakers have already been finalized or someone is trying to find a match for a pair of speakers from a manufacturer's amp line-up. I've also come across cases where someone (a non-critical listener) is trying to upgrade their speakers, but would like to know whether their choice of speakers will match their current amps, as far as power is concerned.

Based on facts like doubling of distance will drop SPL by 6dB and doubling of amp power raises SPL by 3dB, I tried to quantify the well known relationship among amplifier power, speaker sensitivity and listening distance, in my own way. I would like to get my approach as well as the results reviewed by the experts here and also, would like to get a few questions, that I have myself, answered. To begin with, I'll quote a more specific version of the problem statement here:

I have a speaker of sensitivity S dB/W/m and of 8 Ohms nominal impedance rating and my listening position is at a distance D meters (m) from the speakers, so how much amplifier power, P in watts do I need?

This discussion is going to be a bit technical (highly objective) and there will be a few equations containing logarithms. My apologies for that. To come up with an equation for P, we definitely need more inputs:

1. How loud do we listen? This varies significantly from person to person
2. What is the peak to average for the music we're playing? This will help us decide how much additional headroom we need for the peaks. To get the exact ratio, we really have to do a statistical analysis of our entire music collection
3. Do we need to account for additional headroom on top of this? As a rule of thumb, it is always better not to run entry level integrated amps at full volume setting
4. What if the rated nominal impedance of my speakers is 4 Ohms or there is a large impedance dip at certain frequencies? With watts getting cheaper, modern speakers take power for granted and a speaker having a nominal impedance rating of 8 Ohms can have it's impedance dip to 4 ohms or lower at certain frequencies
5. Are our speakers really as efficient as claimed by the manufacturer?
6. Do we need to consider additional effects like room absorption, reflection, etc?
7. ...

and, the list goes on. The equation I came up with, dials in reasonable values for points (1), (2) and (3) above. (4) is looked at as a secondary effect and is accounted for by modifying the equation. I'm ignoring rest of the effects in this calculation. Let's start building the equation by walking through points (1) to (4).

1. How loud do we listen?
The way I understand, this can be quantified by SPL, at our listening position. Only problems is, the preferred SPL varies vastly from person to person. For example, I prefer listening at volumes low enough to have a conversation with someone sitting across the table, when I'm not alone at home. At the same time, I don't mind cranking up the volume if only my furry friend and I are there at home. Many people listen very loud, and as I've read on Troels Gravesen's website, the less distorting the system, the more is the tendency to listen louder without even realizing it. If we google "Safe SPL for music listening", a number that pops up is 85dB, which also says, safe only for less than 8 hours a day. I'll take this safe value as reasonable level, though I myself listen at much lower levels. Also, please note that this value can be customized as per your preferred listening levels.

2. What is the peak to average for the music I'm playing?
The average corresponds to the RMS value of the output amplitude, that we hear. This value must translate to 85dB of SPL that I've assumed in (1). However, what makes a system less distorting and highly musical is its ability to reproduce the peaks effectively. The amp must have enough power in reserve for the peaks in musical passages. The peak to average ratio decides the headroom needed for the peaks. Higher the ratio, more is the headroom needed. When DRC is used extensively to minimize this ratio, then this value tends to 1 and we can get away with very less headroom. This particular ratio varies significantly from track to track. So, once again, we may have to make an educated guess here. I have analyzed about 100 tracks due to demands at work (otherwise, why would I?), and have seen this ratio to be around 3. This is in agreement with the values for this ratio I've seen in textbooks and various forums. Giving more headroom, if we round-up this value to 4, then this translates to additional SPL of 20 * log10(4) or 12dB and a peak of SPL = 85dB + 12dB = 97dB.

The effect of distance
==============
As we move farther away from speakers, they start to sound quieter. To be specific, if we double the distance (2x) from the speakers, SPL drops by one-fourth (1/4x) or by 6dB, in logarithmic scale. For example, if the SPL of a speaker is rated at 90dB/W/m and we supply 1W of power to the speaker, the SPL at a distance of 1m from the speaker will be 90dB, 2m will be 84dB, 4m will be 78dB, and so on. We can take into account of this relationship, in logarithmic scale, by introducing a term 20 log10(D), where log10(x) is logarithm to the base 10 of x, and D is the distance from the speakers in meters and the factor of 2 in 20 takes care of inverse square law of distance.

The effect of power
=============
As we start cranking up the volume, the speakers start to get louder. To be specific, if we double the power to the speakers, the SPL also doubles or increases by 3dB. We can take into account of this relationship, in logarithmic scale, by introducing a term 10 log10(P).

So, to play music in its fully glory, we must target 97dB of SPL for the peaks, at our listening position, D meters away from the speakers. SPL is deteriorated by distance, as we move away from the speakers, and the amplifier must deliver more power to compensate for this. Speaker's sensitivity, S works in the favor of the amplifier, as highly sensitive speakers demand less power from the amp. Putting it all together,
10 log10(P) + S = 20 log10(D) + 97, or
10 log10(P) = 20 log10(D) + 97 - S
This is a first-order approximation, based on what I have understood.

3. Do we need to give additional headroom on top?
This varies from amp to amp. A few amplifiers are over designed and can be run at full volume without distortion. Entry level amps may go for modest choices of power supplies and output devices, and may start distorting when their power supplies run dry or their output devices are pushed to the boundaries of their linear regions, while delivering the peaks at full power. I'm considering entry level amps here, and am giving an additional head room of 10%. This means that, the power calculated from the previous equation should not exceed 90% of the output power of the amp, rated at 8 Ohms. The previous equation can be modified to take into account of this,
10 log10(P / 1.11) = 20 log10(D) + 97 - S,
where 1.11 = 100/90, for 10% headroom.

An example. My speakers have an SPL rating of 90dB/W/m and have a nominal impedance of 8 Ohms. My listening distance is about 2.39m. Substituting these values in the above equation, we get:
10 log10(P / 1.11) = 20 log10(2.39) + 97 - 90 = 14.568
P = 1.11 * 10^(14.568 / 10) = 31.78W (The operation 10^y, means 10 to the power y, or antilog to the base 10 of y)
If I ever decide to purchase an amp from the the Rega -R series, then I will start with Brio-R and may even consider Elex-R, but Elicit-R will definitely be a stretch, as far as power is concerned. Once again, this comment is purely based on power, and how these amps sound and their synergy with my speakers, and rest of the chain, is a different topic altogether. Next, we'll try to address point (4).

4. What if the rated nominal impedance of my speakers is 4 Ohms or there is a large impedance dip at certain frequencies?
If the nominal impedance of the speakers is 4 Ohms, then the power calculated by the previous equation must also be doubled for an ideal amp, i.e., an ideal amp will double its power when the impedance is halved. Speakers don't present a constant impedance to the amp at all frequencies. To ensure that the amp does not run out of juice at these impedance dips, we have to make sure that the amp can supply sufficient power at the worst impedance value, which is the minimum impedance. A few manufacturers make our life simple by quoting the minimum impedance of speakers, at a certain frequency. We can cover both the cases of 4 Ohms speakers as well as impedance dips, by introducing a term 10 log10(8 / Rmin) in our equation, where Rmin is the minimum impedance of the speaker in Ohms. Thus, the modified equation is now,
10 log10(P / 1.11) = 20 log10(D) + 97 - S + 10 log10(8 / Rmin)
For speakers having a nominal impedance of 4 Ohms, put Rmin = 4, to calculate the minimum power at nominal impedance. Also, note that this term reduces to zero, if you put Rmin = 8. It can also be verified that the power at 4 Ohms is double that of the the power at 8 Ohms, using the above equation.

By bringing Rmin to the equation, we'll have more things to worry about. Not all amplifiers can double their power when impedance is halved. In fact, many amps (Rega -R series, Marantz PM8006, Luxman-505uX II, and a few others on hifimart.com for which this data is available), can only increase their output power by 1.5x, when the impedance is halved from 8 Ohms to 4 Ohms. If the minimum impedance, Rmin, is in the ballpark of 4 Ohms, then we will have to consider an amplifier that can drive 4 Ohm loads and for which this data is explicitly specified by the manufacturer. If no data is available, it's quite possible that the amp is not designed to handle 4 Ohms speakers. Also, in these cases, we will have to compare the power estimated by the equation against the power rating of the amplifier at 4 Ohms, rather than 8 Ohms.

Notes:
====
1. I found this equation as a way of checking whether a speaker is too big for our room. A few manufacturers provide a spec called recommended minimum amplifier power. This is said to be the minimum power at which the speakers start to sound their best. If this power turns out to be much higher than what is estimated by the above equation, then it could be a sign that the speakers could be too big for our room. For eg, Monitor Audio Silver 500 speakers have a recommended minimum amp power rating of 80W. Given, a sensitivity of 90dB/W/m and nominal impedance of 8 Ohms, the minimum power required to produce 97dB of SPL at my listening position is only 32W. If I ever attempt to pump 80W at 8 Ohms to these speakers at home, I expect somebody from hifivision.com to bail me out of "domestic violence by noise pollution" charges. This speaker is also an interesting speaker in that it's minimum impedance dips to 3.1 Ohms at 2.45kHz, which means we'll be needing to pair an amp capable of driving 4 Ohm loads, with these speakers and the amp should be rated at least 82W at 4 Ohms.

Questions:
=======
1. Is there anything like a stereo reinforcement? When we play stereo, the SPL at the listening position is produced by two speakers, combined. Do we have to modify the equation to take account of this?
2. How do we exactly measure the listening distance?



Is it x or y, in the above figure? It is 7.83 feet v/s 7 feet in my listening room; so, irrespective of which value I choose, the calculated power doesn't vary a lot.

With regards,
Sandeep Sasi

 

[arj]

I will confess I did not read all the text , but the distance you can look at is Y and as you said it will not vary much even if you take x.

In the end you need the speaker to pressurize the entire room and thats independent of the seating distance. I would suggest you go by the recommended minimum power rating that the speaker is recommended for rather than calculating based on the sensitivity. Any power over that minimum is based on the flex of your budget ( anyway doubling the power will only give you 3db more)

As you go up an amp chain it is not necessarily only power but very often also better components, power regulation etc which can have a bigger impact on sound quality than just power and for most speakers the quality of the first few watts is far more important than the others.

 

[Surrealistix]

I like how you have gone about calculating power needs. Excellent work. You could take this further by considering phase angle in addition to the speaker's impedance. This would require measuring the speakers phase and impedance over frequency or use published data from a review.

 

[Deleted member 15865]

In the end you need the speaker to pressurize the entire room and thats independent of the seating distance.

Arjun, can you please help understand this statement? Why should the speaker pressurise the entire room? (And what exactly is this pressurisig? The sound wave pressure?) Is it a function of the kind of music one listens to?

What I am saying is that I have sensed it’s a common objective of all audiophiles to get an exact feel of the actual performance, whether studio or live (more precisely, post recording, mixing and mastering) even if they may not agree on a lot of other things. In that case, I can understand the need to pressurise the entire room to replicate the stadium/arena feel of a live rock concert which has that kind of pressure. But what about other kinds of, esp softer music? Like a country singer strumming and singing to a small audience in a studio? Or a classical artist playing her sitar in an auditorium? The strains of music keep floating towards you at a good enough volume, but the place doesn’t feel pressurised - if it does, would it even sound good? Or is that an essential part of a sound being hi-fi?

Or am I understanding the ‘sound pressurising if a room’ wrongly? I’d like to understand this. Because I sometimes wonder about it too.

@sandeepsasi, I tried reading the entire post, but had to give up midway due to my limited technical apprehension. However I was still left admiring how well you have gone on postulating the problem - I’d hazard a guess that you are a PhD. Also your hold of the language is evident in its elegant simplicity.

 

[sandeepsasi]

I will confess I did not read all the text , but the distance you can look at is Y and as you said it will not vary much even if you take x.

Yes, it doesn't vary much in this case. The speakers and the listening position happen to form an equilateral triangle in my room, and the variation happens to be just 20%. Because the room dimensions are small, the power is also small and the delta in this case is also small. However, if the room dimensions are large, this may matter as the choice eventually may boil down to two amps rated at 200W v/s 250W, for example. Also, X and Y variation can be larger, if the speakers and the listening position form an isosceles triangle with the angle at the listening position is slightly greater than 60 degrees. I think this kind of placement is common when the listener sits facing the longer walls of a rectangular room, in order to widen the sound stage.

In the end you need the speaker to pressurize the entire room and thats independent of the seating distance. I would suggest you go by the recommended minimum power rating that the speaker is recommended for rather than calculating based on the sensitivity. Any power over that minimum is based on the flex of your budget ( anyway doubling the power will only give you 3db more)

Thanks for the inputs.

With regards,
Sandeep Sasi

 

[arj]

Too much of theory for me to remember and summarize unfortunately.. additionally my knowledge is also limited !

But Yes I do mean the SPL since sound can also be looked at as Waves of pressure which is a function of the power used to generate it and how it interacts with space boundaries.

I believe it has to do with how reflected waves work with the primary waves, hence the total SPL at any point would be a summation of the primary wave + the pressure caused by the radiated waves..the higher the amplitude of the reflected wave, the higher the pressure. and the closer a wall the higher the amplitude and hence the "pressure" of the wave.

Depending on the wall dimensions you can have a phase lag and hence there will be points where pressure actually negates and reduces. And you are right on the impact of this Pressure. a guitar sounds so much better when you can feel the pressure of the strings

Yes, it doesn't vary much in this case. The speakers and the listening position happen to form an equilateral triangle in my room, and the variation happens to be just 20%. Because the room dimensions are small, the power is also small and the delta in this case is also small. However, if the room dimensions are large, this may matter as the choice eventually may boil down to two amps rated at 200W v/s 250W, for example. Also, X and Y variation can be larger, if the speakers and the listening position form an isosceles triangle with the angle at the listening position is slightly greater than 60 degrees. I think this kind of placement is common when the listener sits facing the longer walls of a rectangular room, in order to widen the sound stage.


Thanks for the inputs.

With regards,
Sandeep Sasi

The choice between a 200 or 250 W should be based on how big your room is and how much better the quality of sound is is rather than the quantity of power. I would guess that all things being same, a 200W or 250W should not matter much. in fact for most speakers even 100W and 200 might not make much of a difference

The actual SPL is dependent on so many things - the efficiency of the speaker ( not the sensitivity), the speaker sensitivity, speakers dispersion, the distance for measuring the absorption/reflectivity of the walls and other objects including people ! To add complication many of these absorbtion/dispersion are dependent on the frequency of sound hence if we look at music it could affect the tone of sound as well

you will find a good calculator here . I had it saved some years back and just check if it works and this calculation is what is used for Pro Audio

 

[sandeepsasi]

I like how you have gone about calculating power needs. Excellent work.

Thank you!

You could take this further by considering phase angle in addition to the speaker's impedance. This would require measuring the speakers phase and impedance over frequency or use published data from a review.

Thanks for this suggestion and I spent some time reading more about magnitude and phase plots of speaker's impedance. The way I understand, the magnitude plot of the impedance already takes care of the apparent power (vector sum of real and reactive powers) the amp must deliver to the speaker, at various frequencies. Phase angle or power factor (cosine of phase angle) decides how much of this power is burnt as heat, in the amp itself. So, the worst case scenario can be an impedance dip combined with a large phase angle (or low power factor). This is when the amp must supply large amount of current to meet the high power demand (due to impedance dip) and at the same time dissipate most of this power as heat in the amp itself (due to the low power factor). The heat generated will be much higher than that when the amp is driving a purely resistive load of the same value and therefore, appears as though the impedance of the speaker is even lower than that at the dip.

Usually, if the amp is well designed, both in terms of power delivery and proper heat sinking and cooling, then this may not be a problem. Unfortunately, this may not always be the case for many integrated amps. So, instead of relying on the amp designers to do their job, we can take things into our own hands and pick an amp with a higher power rating, which we can hope is capable of dissipating the heat generated during the deadly combination of low impedance and large phase angle. How high? A few threads on avsforum.com suggest that 2x the power is a safe bet. Interestingly, in this case, we are not interested in the extra 100% power reserve in the amp but in the power dissipation capabilities that, we hope, come with it. Once again, this is the way I have understood it and I could be overlooking several things.

 

[sandeepsasi]

@sandeepsasi, I tried reading the entire post, but had to give up midway due to my limited technical apprehension. However I was still left admiring how well you have gone on postulating the problem - I’d hazard a guess that you are a PhD. Also your hold of the language is evident in its elegant simplicity.

Thank you! Very kind of you to think that I'm a PhD, but that would be far fetched. I have a masters in electronics engineering and am also a budding audiophile.

I've come across the topic of amp power and its relationship with speaker sensitivity and listening distance, many times, in many places. For example, the online calculator suggested by @arj already does this. Only problem I've seen with such calculators is that they leave, desired SPL at the listening position and headroom, as a burden for users to decide. Many newcomers like me don't have a feel for SPL in dB and how much headroom is required. So, my thought was to dial-in the safe SPL for music listening as the desired SPL at the listening position. I came across the relationship between headroom and reasonable values for peak to average while generating white noise to test digital audio signal paths. It will be great if online calculators present these values as defaults, for the benefit of newcomers, and senior folks in audio can fine tune these values as per their preferences. 10% back-off from full volume setting is commonly discussed in forums. Adding additional 3dB during the power calculation for 4 Ohms loads is explained in the book The Complete Guide to High-End Audio by Robert Harley, the editor-in-chief of Absolute Sound, though I don't remember coming across a complete equation like this. I thought I can generalize this idea to take care of impedance dips as well.

 

[jls001]

@sandeepsasi the two speakers "reinforce" each other in the sense that the two of them are needed to create a single stereo image from the left and right channels. They would be reinforcing in the x dB sense of the word if they were playing dual mono material, meaning left and right channel are exactly same content. This is not the case in true stereo recordings as left and right carry different signals. And even if the program material played is dual mono the reinforcement would be <3 dB because the two speakers are not coincident, or at least nearly so.

For my 86 dB efficient speakers in my 19' x 10' room I have tried quite a few power amps rated at - 50W, 50W, 50W, 60W, 60W, 60W, 85W, 200W all at 8 Ohms, and each and each one of them could play very loud and didn't falter during high dynamic swings. And they were class AB discrete solid state, chipamp AB, class A solid state, tubed class A, and class D chipamp. All of them have more than decent power supply sections. My conclusion is 50W is more than enough power for most stereo setups provided the power supply of the amp is stiff enough.

 

[sandeepsasi]

@sandeepsasi the two speakers "reinforce" each other in the sense that the two of them are needed to create a single stereo image from the left and right channels. They would be reinforcing in the x dB sense of the word if they were playing dual mono material, meaning left and right channel are exactly same content. This is not the case in true stereo recordings as left and right carry different signals. And even if the program material played is dual mono the reinforcement would be <3 dB because the two speakers are not coincident, or at least nearly so.

What you said is a very good point! Thanks for the input. Even for stereo, the instantaneous level can even go negative, if the sound from both speakers are meant to cancel each other to from the stereo image. Often, I've observed that if one speaker is off, or goes off all of a sudden, I feel a drop in loudness and an amateur listener like me will catch this even before I realize that the stereo image has collapsed. This means that there is also a net increase in SPL, when two speakers play stereo, even though the waves are not in phase instantaneously, and this must of course be < 3dB, as you just pointed out.

For my 86 dB efficient speakers in my 19' x 10' room I have tried quite a few power amps rated at - 50W, 50W, 50W, 60W, 60W, 60W, 85W, 200W all at 8 Ohms, and each and each one of them could play very loud and didn't falter during high dynamic swings. And they were class AB discrete solid state, chipamp AB, class A solid state, tubed class A, and class D chipamp. All of them have more than decent power supply sections. My conclusion is 50W is more than enough power for most stereo setups provided the power supply of the amp is stiff enough.

You are blessed with a very large room and can experiment with speakers of different sizes and can try different placement options, unlike the case with me. In theory, for the largest triangle that can be placed in your room will have y ~= 6m (from the figure that is attached in OP). Out of curiosity, what is the listening distance in your case?

One more thing, the whole purpose of this exercise is to do a quick, back of the envelope calculation to see whether our choices are correct, especially while choosing modest, entry-level components like Marantz PM6006, Rega Brio-R, etc. Instead, if you consider amps like Bryston B60R or even power amps and monblocks, I wouldn't even worry about the 10% back-off and additional 12dB headroom. I once had a case of a salesman of a boutique brand trying to sell a 200W Class A/B SS integrated amp along with his speakers rated at 98dB/W/m (had horn loaded woofers and CD) to a friend of mine who has a non-engineering background. I couldn't listen to the set-up myself, as he was in a different country at that time. In these kind of cases, this formula has raised a bright red flag, saying something is seriously wrong with the recommendation. Today he runs those speakers from a modest 20W/channel amp.

 

[keith_correa]

My conclusion is 50W is more than enough power for most stereo setups provided the power supply of the amp is stiff enough.

My exerience has been the same - for a room sized 150-200 sq. ft and speakers that are ~85 db efficient, 40W-50W of clean power is more than adequate for listening at uncomfortable levels if one chooses to. More power than that gets one just bragging rights.

 

[sandeepsasi]

Hi @jls001, @keith_correa,

How do you know the efficiency of your speakers? Usually the common spec is sensitivity in dB/W/m, which what is needed to compute power and they are not definitely the same. For most of the moderate sized urban living spaces, the number is in the ballpark of 30-60W at 8 Ohms, unless you are talking about horns or power hungry speakers.

My exerience has been the same - for a room sized 150-200 sq. ft and speakers that are ~85 db efficient, 40W-50W of clean power is more than adequate for listening at uncomfortable levels if one chooses to. More power than that gets one just bragging rights.

My listening space is even smaller ~ 120 sq. ft and my amp is just rated at 30W at 8 Ohms. It does linear power delivery though, all the way to below 4 Ohms, which also a kind of gives me bragging rights, but only while talking to the right people

 

[keith_correa]

Hi @jls001, @keith_correa,

How do you know the efficiency of your speakers?

Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Sengpiel Berlin

Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter dB/W/m loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Eberhard Sengpiel

www.sengpielaudio.com

Usually the common spec is sensitivity in dB/W/m, which what is needed to compute power and they are not definitely the same.

You're absolutely right. I misspoke. I meant sensitivity and not efficiency!

 

[jls001]

I once had a case of a salesman of a boutique brand trying to sell a 200W Class A/B SS integrated amp along with his speakers rated at 98dB/W/m (had horn loaded woofers and CD) to a friend of mine who has a non-engineering background. I couldn't listen to the set-up myself, as he was in a different country at that time. In these kind of cases, this formula has raised a bright red flag, saying something is seriously wrong with the recommendation. Today he runs those speakers from a modest 20W/channel amp.

This should not necessarily raise a red flag. Many highly efficient speakers work very well with high-powered amps. Case in point: Tekton Double Impacts which are rated at 99+ dB/W/m work very well with 200W Pass monos, as well as with 2W microZOTL tube amps. The point being there seems to be more to amp-speaker pairing than simple power-sensitivity relation. I'm not sure what are the relevant metrics are but damping factor could be one.

 

[arj]

Hi @jls001, @keith_correa,

How do you know the efficiency of your speakers? Usually the common spec is sensitivity in dB/W/m, which what is needed to compute power and they are not definitely the same. For most of the moderate sized urban living spaces, the number is in the ballpark of 30-60W at 8 Ohms, unless you are talking about horns or power hungry speakers.


My listening space is even smaller ~ 120 sq. ft and my amp is just rated at 30W at 8 Ohms. It does linear power delivery though, all the way to below 4 Ohms, which also a kind of gives me bragging rights, but only while talking to the right people

A speakers efficiency is usually very low from 0.5-3% only

Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Sengpiel Berlin

Efficiency and sensitivity conversion - loudspeaker percent and dB per watt and meter dB/W/m loudspeaker efficiency versus sensitivity vs speaker sensitivity 1 watt = 2,83 volt box chart - sengpielaudio Eberhard Sengpiel

www.sengpielaudio.com

You're absolutely right. I misspoke. I meant sensitivity and not efficiency!

Nice link..thanks !

 

[Dr.Bass]

In the end you need the speaker to pressurize the entire room and thats independent of the seating distance.

This one sentence sums it all.
The key is to differentiate between filling the room and pressurizing the room.

 

[mpw]

This one sentence sums it all.
The key is to differentiate between filling the room and pressurizing the room.

Pressurizing the entire room will mean a super wide sweet spot ?

One can hear everything and more anywhere in the room ?

 

[Dr.Bass]

Pressurizing the entire room will mean a super wide sweet spot ?

One can hear everything and more anywhere in the room ?

Not every detail but the intensity would be very well reproduced everywhere.