Any truth to speakers being better with a more powerful amp?

[Sokel]

Also, I tested it first at 4m from my speakers, than got closer all the way to 1m. I expected the dB reading to increase by 12db, it did not (barely moved). Could it be that this is due to me seating inside a "box" - speakers 30cm from the wall on one end, seating position 4m away against the opposite wall - and the sound bounces off each sides and gets diffused (if that's the right word) more evenly within the room?

Your SPL meter has to be calibrated and it matters a lot what kind of weighing you're using.
It must also have the abilititty to measure average -max and peaks (most can't measure peaks) .

The good news is that REW has a real-time logger than can measure all that and much more,you'll only have to calibrate your mic:



Edit:About weighting,here's how different the same play measures with different weighing at the exact same level:

How loud do you like your music?

I voted under 50db. I'm listening right now about 16 feet away from the speakers and measured the SPL using my phone. It was mostly under 40db and I wouldn't want it much louder. When I crank it up to get even 60db at listening position, I don't want to listen for long. I'm the guy who hates...

www.audiosciencereview.com

 

[fpitas]

You may also have to time-gate to eliminate the reflections.

 

[DanielT]

You may also have to time-gate to eliminate the reflections.

From another forum:

"It has been shown in several threads that many people have questions about measurement and equalization of speakers.

There are lots of strange myths in the field, e.g. that you should not equalize at all, that you should only equalize with small interventions or that you should possibly only equalize for the very lowest frequencies. However, the answer is that you can and should equalize in the entire audible range and that you only need to keep an eye on limitations for power and cone displacement. For meaningful equalization, however, you need to perform some relatively simple measurements.

It is best to divide the measurements into two different frequency ranges, which are treated separately. Never use LTAS measurements (Long Term Average Spectrum), i.e. measurements with a long time window for equalization above about 1 kHz.
So-called gating (or windowing) becomes a decisive issue here. Windowing is not as advanced as it sounds, at least not to the user - the signal analytic background and implementation is not entirely straightforward. By time-limiting sound waves arriving at the microphone during post-processing, reflected sound can be excluded with an arbitrary time limit.

For mathematical reasons, the length of the time window is inversely proportional to the frequency resolution (you must have at least one full period within the time window to be able to determine the frequency), so 500 ms (0.5 s) corresponds to 1/0.5 = 2 Hz frequency resolution. In this case, the frequency bands are thus so wide that everything within 2 Hz is presented as an average value within the same frequency band."

faktiskt.io • Visa tråd - Mätningar för ekvalisering

 

[Nicolas_]

I just tried very quickly and not seriously at all, with almost the first app that popped up on the store - it was just to get a rough idea of what 60db/70db/80db sounded like in the real world, so it's done with the phone mic. I will do some more serious measurements in the future, when I get hold of a proper mic, and I'll also do room measurements with REW (or Housecurve).
I did get peak readings though, I think ~65db was the average I was happy and comfortable listening at, with peaks of ~75-80dB.

 

[Sokel]

You may also have to time-gate to eliminate the reflections.

Why?We are interested for the MLP,room gain and reflections is at play so we want the overall (which might help)

 

[Sokel]

I just tried very quickly and not seriously at all, with almost the first app that popped up on the store - it was just to get a rough idea of what 60db/70db/80db sounded like in the real world, so it's done with the phone mic. I will do some more serious measurements in the future, when I get hold of a proper mic, and I'll also do room measurements with REW (or Housecurve).
I did get peak readings though, I think ~65db was the average I was happy and comfortable listening at, with peaks of ~75-80dB.

What kind of music is that with only 10db peaks,that's normally the max.
Peaks get at least 15-20db more than average,with classical a lot more.
What weighting does it use?

 

[DanielT]

What kind of music is that with only 10db peaks,that's normally the max.
Peaks get at least 15-20db more than average,with classical a lot more.
What weighting does it use?

Measured with a free dB app, take it with a grain of salt.

Also, with some modern recorded music and 10 dB peaks, isn't that even considered good?

Edit:
It's sad and even sadder that HiFi geeks like me completely clear out all flea markets on good old dynamically powerful CDs. Nowadays I can't find any decent CDs at flea markets.

 

[fpitas]

Why?We are interested for the MLP,room gain and reflections is at play so we want the overall (which might help)

I thought the idea was to observe the fall off in direct sound with distance.

 

[Sokel]

Measured with a free dB app, take it with a grain of salt.

Also, with some modern recorded music and 10 dB peaks, isn't that even considered good?

Yes,I took a look,most measure with A weighting and they don't even have Z weighting for the peaks.
If there is such music I don't want to know

 

[Sokel]

I thought the idea was to observe the fall off in direct sound with distance.

Oh,then you're right.Different subject.

 

[fpitas]

Oh,then you're right.Different subject.

Well, I'm not entirely sure of the exact subject either.

 

[Sokel]

Well, I'm not entirely sure of the exact subject either.

The one I'm talking about is what power does it needs,so one measuring low can extrapolate what it needs to go higher and also where the limits are (distance between max-peak will be reduced).

 

[DanielT]

Yes,I took a look,most measure with A weighting and they don't even have Z weighting for the peaks.
If there is such music I don't want to know

Sure, but they're free. Just to get an estimate. But I wonder how much they deviate from more professional ditto?

Dynamic peaks in music can be investigated through theoretical rough estimation based on the music you listen to.

 

[anmpr1]

I somehow don't recall the High Fidelity review. I owned and was thinking of the Acoustat Two. I measured it outdoors and came up with my approximately 80 db number. The actually very similar Soundlabs I came up with about 82 db. Friends owned 3s, 2+2s, and 1+1s. I thought the Model 3 was their best and thought the 1+1 was a mistake. I modded those interfaces a few ways with/for other people. As far as I know the change Acoustat made was essentially altering the crossover point between the high and low frequency transformers. It should help a little, but didn't fix the low 1 ohm or so impedance in the treble. One friend had an amp that didn't like that we ended up putting a half ohm power resistor in series with the feed to the high frequency transformer which fixed his problem. The very top end rolled off some more of course which he liked anyway. I'm surprised more amps didn't 'complain' about 1 ohm impedance with about 75 degrees phase shift in the treble.

I always wanted to come across those Acoustat X's with direct coupled amp.

From my very limited experience with the SL, they were a better overall loudspeaker, but larger and more expensive. As with any ESL, you just couldn't stick them anywhere in your living room and expect decent results. Certainly not like shoe-box monitor style loudspeakers.

I enjoyed the Acoustats, but eventually all good things end.

Anecdote: Harvey 'PT Barnum' Rosenberg claimed that a lot of his NYAL Futterman warranty problems were from customers using the amps with Acoustats. The Futterman would play happily with the high impedance Quad, but self-destructed with the Acoustat.

Of all the loudspeaker brands still around, I am somewhat surprised that Acoustat is not. For the price they offered good value, for that sort of thing. And they were generally reliable, for that sort of thing. Last I heard, Rockford sold the IP to an Italian firm, which sold it to a Chinese outfit. A quick search doesn't come up with anything now.

 

[Blumlein 88]

Also, I tested it first at 4m from my speakers, than got closer all the way to 1m. I expected the dB reading to increase by 12db, it did not (barely moved). Could it be that this is due to me seating inside a "box" - speakers 30cm from the wall on one end, seating position 4m away against the opposite wall - and the sound bounces off each sides and gets diffused (if that's the right word) more evenly within the room?

There is a thing called critical distance with loudspeakers. It is the point where reflected sound equals direct sound from the speaker. Inside critical distance you will get about that 6 db drop per doubling. As you move away further it lowers and critical distance is where doubling distance results in only a 3 db drop in SPL. A rule of thumb which is generally not far off is a speaker's sensitivity rating is close to what you'll get in room at the listening position with the same input. An 87 db SPL @ 1 meter rating will often give about that 87 db at the listening position in a room. Now like all rules of thumb this is close, but precise.

 

[DanielT]

There is a thing called critical distance with loudspeakers. It is the point where reflected sound equals direct sound from the speaker. Inside critical distance you will get about that 6 db drop per doubling. As you move away further it lowers and critical distance is where doubling distance results in only a 3 db drop in SPL. A rule a thumb which is generally not far off is a speakers sensitivity rating is close to what you'll get in room at the listening position with the same input. An 87 db SPL @ 1 meter rating will often give about that 87 db at the listening position in a room. Now like all rules of thumb this is close, but precise.

And you can easily measure that with a free dB app to see how it is at different distances. ..#79 in the thread about it. Up to two meters I said but well that's probably too much.

 

[Barrelhouse Solly]

Depends on speaker sensitivity. It's possible to have an amp that's under powered for a given speaker if it can't reproduce the full dynamic range of the program material using your speaker. In your case I doubt it. As others have mentioned, 2X power = 3 Db volume increase.

 

[Duke]

"What you really need is a power amp. That Denon has soft power. If you pump 200 watts into those things they'll really come alive" is what my local dealer told me today.
The speakers are Monitor Audio Silver 500 driven by a Denon Avr rated at 105 watts.
Considering I only need about 25 watts to listen at around 95db how can having an extra 100 watts over what I already have make any difference?

I did not read the entire thread so apologies if this has already been covered. AND I am not an amplifier guy so I welcome correction from anyone who is, anyway the following will not be strictly accurate but I think the trends described are in the ballpark.

When it comes to the amplification section of AVRs, the fine print matters. What exactly does the Denon's spec sheet say? The following will be a hypotherical "worst case scenario" - I am NOT accusing Denon of being this optimistic with their specs! But this will be something that imo is worth being aware of.

IF the spec sheet of an over-optimistically-rated AVR says something like "100 watts per channel at 1 kHz with one channel driven", and then they multiply that 100 watts by the number of channels, say six, they might claim that it's a "600 watt" receiver. If the limiting factor is the power supply, which is quite likely, then the actual output with all six channels driven equally might only be 100/6 = 16.7 watts. This is an extreme example, for illustration purposes.

If you are only using the front three channels, we'd be looking at 33.3 watts per channel, assuming our hypothetical wimpy power supply only delivers enough power for the amplifier modules to put out a combined total of 100 watts.

But it gets worse.

That 33.3 watts/channel with only the front three channels driven is with a 1 kHz sine wave. Real program material is much more broadband,. and amplifiers generally deliver far less wattage broadband than they do into a sine wave. Like maybe ballpark 20% of their sinewave power output. So now our 33.3 watts sine-wave is down to about 7 watts broad-band.

Something like the above might be why your dealer used the term "soft power".

(The substantial difference between an amplifier's sine-wave power output and broadband power output was described to me by the late, great Henry Wolcott, who designed an outstanding tube amp. It was based on a tube amp that he designed for powering a bank of computer hard drives for a large government contractor; unfortunately I don't recall why tubes were preferred for that application. Henry's day job also included designing and building unique test equipment for Defense Department contracts. So if Northrop Grumman won the contract to build the new radar for the F-35, Northrop's claimed performance would have to be vertified by testing. And if sufficiently capable test equipment did not yet exist, Henry's company Metrology Instruments would be the one to design and build it. I think that actually happened, though he never said so to me in so many words.)

 

[pogo]

 

[Holmz]

Probably not. Ask the dealer to define soft watts and how watt hardness is rated.

Moh hardness?