What's the science behind more power leading to better sound?

[bravomail]

To topic originator - you still might be fine without buying fancy stuff if your headphones are low impedance and sensitive, say Audiotechnica M40x. Problem is, the headphones, which are considered truly HiFi (subjective, I know) have impedance of 50 Ohm or higher and sensitivities around 96dB or less. In other words - they r damn hard to properly drive by normal mobile phone or tablet. PC Mobo or laptop will give you audible hiss (mine does), even though they will have enough power. Solutions - phone/tablet with good built-in DAC (LG G6/G7/G8/V10/20/30/40/50 or International Samsung Galaxy S7...s10 with Exynos processor, ZTE Axon 7), for PC - external DAC and Amp (or combo).

 

[Idanonymous]

I also have this question. I don't really understand the science or math (not this major/field) and it does seem like there are a lot of myth out there that marketers of high end audio equipments sold to "audiophiles/audiofools". I want to preface by saying the idea of headroom makes sense. When there is a sudden loud sound (like an explosion that is much louder than something like dialogue), an amp needs to be able to handle that or it leads to clipping ("When an amplifier is pushed beyond its maximum limit, it goes into overdrive"). I used https://www.headphonesty.com/headphone-power-calculator/. I do see the power calculator showing that phones or laptops aren't recommended. Based on that, an o2 amp should be enough to drive a hifiman he-560 (7.27 VOLTS RMS) at 110 DBSPL. Would this actually lead to any clipping or uncontrolled bass? I also understand that better SNR means that when I increase the volume, there is less distortion so that plays a big factor. Would that play a factor with this headphone and amp combo solely based on the numbers? Just like the OP, I will like to know if it is myth that a more "powerful" amp (lets say 3.5 watts at 50 ohms with similar SNR) will make the sound "better". If i can play the less powerful o2 amp to the same volume with volume knob adjustment and human hearing cannot hear the increased SNR, will the bass become less controlled based on the numbers (doesn't clip but maybe distortion?)? I also don't understand why ppl say that more powerful amps make the midrange better (more detailed, fuller, etc). If headphone power calculator show that the o2 drives the inefficient he560 planar, then why would the midrange sound sonicly different from a more powerful amp? However, I do have to say that I do hear a difference between a more powerful amp compared to an o2. But that is from a limited listening and there might be subjective bias. That is why I am trying to get to the root of this objectively. I know there are a lot of factors that come into play. Like OP, I feel like a lot of people that have a much better technical understanding don't clearly come out to say whether "more power = better sound" is a myth. I am giving a clear test case (he-560 and o2 amp compared to he-560 with 3.5w at 50 ohms amp). Does more power in this case equal better sound? BTW, I know JDS atom is the better amp to buy than the o2 but the headphone-power-calculator site didn't have jds atom as an option. It seems like JDS atom is pretty much end game amp for precise reproduction of sound. What headphones will need more power than JDS atom can produce at normal listening volume?

 

[solderdude]

I also have this question.
I used https://www.headphonesty.com/headphone-power-calculator/. I do see the power calculator showing that phones or laptops aren't recommended. Based on that, an o2 amp should be enough to drive a hifiman he-560 (7.27 VOLTS RMS) at 110 DBSPL. Would this actually lead to any clipping or uncontrolled bass?

At 7.2V the HE560 reaches 116dB peak (about 100dB average).
Above that there will be clipping. No idea where the 'uncontrolled bass' comes from. I reckon its another way of saying they hear distortion (clipping)

I also understand that better SNR means that when I increase the volume, there is less distortion so that plays a big factor.

SNR and distortion levels are not really related. One can have a high S/N ratio and no distortion and have an amp with loads of distortion but a very low SNR.

Just like the OP, I will like to know if it is myth that a more "powerful" amp (lets say 3.5 watts at 50 ohms with similar SNR) will make the sound "better".

An amp does not provide or push 'power'. It provides a voltage and depending on the headphone's impedance a current s drawn (and thus power).
A powerful amp will not clip, a whimpy amp will. Whimpy amp may sound poor when maximum output voltage or current is reached.

If i can play the less powerful o2 amp to the same volume with volume knob adjustment and human hearing cannot hear the increased SNR, will the bass become less controlled based on the numbers (doesn't clip but maybe distortion?)?
I also don't understand why ppl say that more powerful amps make the midrange better (more detailed, fuller, etc). If headphone power calculator show that the o2 drives the inefficient he560 planar, then why would the midrange sound sonicly different from a more powerful amp?

The S/N ration doesn't have anything to do with this. As long as you do not clip the O2 it can not be distinguished from a more powerful amp at the exact same volume. Regardless what the opinion on the interwebs is.

However, I do have to say that I do hear a difference between a more powerful amp compared to an o2. But that is from a limited listening and there might be subjective bias. That is why I am trying to get to the root of this objectively.

Only controlled and level matched blind tests (properly conducted) will tell you.
Any sighted (knowing what you are listening to) will lead to incorrect assesment.

I know there are a lot of factors that come into play. Like OP, I feel like a lot of people that have a much better technical understanding don't clearly come out to say whether "more power = better sound" is a myth.

It is not a myth when the less powerful amp is clipping or when there are other factors at play.

I am giving a clear test case (he-560 and o2 amp compared to he-560 with 3.5w at 50 ohms amp). Does more power in this case equal better sound?

Only when you reach clipping levels on the O2.

BTW, I know JDS atom is the better amp to buy than the o2 but the headphone-power-calculator site didn't have jds atom as an option. It seems like JDS atom is pretty much end game amp for precise reproduction of sound. What headphones will need more power than JDS atom can produce at normal listening volume?

Atom can play 1.5dB louder than O2 (on mains)
In normal to loud circumstances they are indistinguishable.
O2 can be used on batteries and is metal, Atom only on mains and is plastic. Pick your poison

 

[koro]

At 7.2V the HE560 reaches 116dB peak (about 100dB average).
Above that there will be clipping. No idea where the 'uncontrolled bass' comes from. I reckon its another way of saying they hear distortion (clipping)



SNR and distortion levels are not really related. One can have a high S/N ratio and no distortion and have an amp with loads of distortion but a very low SNR.



An amp does not provide or push 'power'. It provides a voltage and depending on the headphone's impedance a current s drawn (and thus power).
A powerful amp will not clip, a whimpy amp will. Whimpy amp may sound poor when maximum output voltage or current is reached.



The S/N ration doesn't have anything to do with this. As long as you do not clip the O2 it can not be distinguished from a more powerful amp at the exact same volume. Regardless what the opinion on the interwebs is.



Only controlled and level matched blind tests (properly conducted) will tell you.
Any sighted (knowing what you are listening to) will lead to incorrect assesment.



It is not a myth when the less powerful amp is clipping or when there are other factors at play.



Only when you reach clipping levels on the O2.




Atom can play 1.5dB louder than O2 (on mains)
In normal to loud circumstances they are indistinguishable.
O2 can be used on batteries and is metal, Atom only on mains and is plastic. Pick your poison

Thank you solderdude, I’ve always wondered the same as OP.

So, if I’m getting this correctly, assuming I have two transparent amps with different power output capabilities but with low enough output impedance to not alter FR and I’m listening at a level in which neither are clipping, we shouldn’t expect differences in, say, bass level right? In this case, the output from a phone with low but clean output and a dedicated amp should sound the same?

I put the bass in the example because I’ve read many times the Hifiman 400i/4xx I own need an amp even at low volume to “properly“ hear “controlled” bass, is there any logic to those claims?

I’m just guessing here, but could it be that what some (subjective) reviewers consider low volume is higher than mine, and what they’re hearing is clipping and thus “uncontrolled”, “weak” bass?

 

[Wes]

another question is what music has the greatest dynamic range and in the lowest bass...

my bet is 1812

 

[DonH56]

The Telarc 1812 had 6 Hz fundamentals during the cannon blasts. They are obvious on the LP and would tend to make most cartridges hide in fear. That said, although I do not have statistics in front of me, the actual peak to average and "quiet" levels for a number of CDs well exceed that of records. The problem is the noise floor of the LPs themselves usually only provides about 50~60 dB dynamic range, and tape can hit 70+ dB, so the recordings have to fit the medium. We can hear below the noise floor, of course, so can pick up a few dB.

Hopefully someone can find some of the early AES papers that looked at frequency content; it was rare, but single-Hz content was found, sometimes due to beats from multiple instruments (or pipes) modulating, though typically at low levels.

 

[solderdude]

So, if I’m getting this correctly, assuming I have two transparent amps with different power output capabilities but with low enough output impedance to not alter FR and I’m listening at a level in which neither are clipping, we shouldn’t expect differences in, say, bass level right? In this case, the output from a phone with low but clean output and a dedicated amp should sound the same?

Could sound the same....
The frequency response, especially in the low frequencies and depending on the headphone impedance can still differ.
The phone could be rolled off when coupling caps are too low in value where it most likely won't be the case with an external amp.
Measurements can easily show this.

I put the bass in the example because I’ve read many times the Hifiman 400i/4xx I own need an amp even at low volume to “properly“ hear “controlled” bass, is there any logic to those claims?

I’m just guessing here, but could it be that what some (subjective) reviewers consider low volume is higher than mine, and what they’re hearing is clipping and thus “uncontrolled”, “weak” bass?

I often read that certain amps have a better 'grip' on the drivers (supposedly due to 'synergy' or 'moar powerrr') and that bass is tighter.
Let's just say that in such case the only way to really find out is to AB blind.
It's a myth that is continued by sighted comparisons. They know what is playing and also often turn up the volume to hear it better and with it come in clipping territory.

 

[BolusOfDoom]

Here's what I think the OP is getting at. Two excerpts from ASR product reviews:

Fiio K3:

Switching to Hifiman HE-400i to represent low impedance headphones, performance improved a bit with higher output level but again, bass notes were on vacation and nowhere to be found.

JDS Atom:

As shown in measurements, there is no rise in distortion up to max volume at 5:00 at which time, you are hearing thundering bass and control of the drivers.

Is there a way to quantify this difference, an objective way to measure if bass notes are "on vacation" or "thundering and well-controlled"?

 

[koro]

Here's what I think the OP is getting at. Two excerpts from ASR product reviews:

Fiio K3:


JDS Atom:


Is there a way to quantify this difference, an objective way to measure if bass notes are "on vacation" or "thundering and well-controlled"?

Hehe, I must admit I had these in the back of my head as well...

 

[mhardy6647]

The Telarc 1812 had 6 Hz fundamentals during the cannon blasts. They are obvious on the LP and would tend to make most cartridges hide in fear...

source: https://www.vinylengine.com/turntable_forum/viewtopic.php?t=94805

 

[infinitesymphony]

View attachment 59945
source: https://www.vinylengine.com/turntable_forum/viewtopic.php?t=94805

This poor guy's enduring a heavy shelling:

 

[tdastrup]

"Is there a way to quantify this difference, an objective way to measure if bass notes are "on vacation" or "thundering and well-controlled"?

I'm looking into inefficient 50 ohm headphones. "Subjective" reviewers have said they should be driven with upwards of 8 Wpc. This is exactly what I'm hoping this thread eventually answers.

 

[Incursio]

I'm looking into inefficient 50 ohm headphones. "Subjective" reviewers have said they should be driven with upwards of 8 Wpc. This is exactly what I'm hoping this thread eventually answers.

I know exactly what you're referring to.

T60RPs or similar can be driven perfectly with an SMSL SP200 (or any similarly measuring amp), with headroom to spare. It delivers around 2.1 watts into 50 Ohms, single ended.

Unless you're referring to the HE6se, in which case, I can't really say (I don't own one).

 

[solderdude]

Here's what I think the OP is getting at. Two excerpts from ASR product reviews:

Fiio K3:


JDS Atom:


Is there a way to quantify this difference, an objective way to measure if bass notes are "on vacation" or "thundering and well-controlled"?

The secret here is output power, headphone efficiency and 'equal loudness contour curves. There is nothing more to it.
It is very simple... when the frequency response is flat there is no such thing as more bass or anemic sound. There can't be. It is easy to hear this though when levels are different. The louder one plays the more power there is in the lows.
When you want impressive bass you have to increase output levels. That's simply how it works.

One should only compare the sound of amps side by side, level matched and blind. This wasn't the case here. There could be many reasons as to how Amir perceived the sound as he did at that moment not having them level matched side-by-side.

 

[mhardy6647]

This poor guy's enduring a heavy shelling:

Most impressive.
That disk was (in)famous for sending tonearms flying in its time.

 

[A800]

More power <> better sound.

 

[BolusOfDoom]

One should only compare the sound of amps side by side, level matched and blind. This wasn't the case here. There could be many reasons as to how Amir perceived the sound as he did at that moment not having them level matched side-by-side.

Right. I didn't put these two quotes together as a "gotcha", I've just been wondering if there was something significant that wasn't yet being measured. At the time I read both reviews, I had a Fiio K3 and HE-400i, and it sounded great to me. Plus being so small and USB-powered, I can just wear the K3 on a lanyard and control the volume.

Since the Atom got so much praise, and it's wattage can get so much higher, and it's inexpensive, I bought one. It sounded even better than the K3 and the bass was amazing, but what I was doing was playing a lot of bass-heavy tracks with a lot of percussion which isn't what I normally listen to. So of course the bass was thundering and controlled. I was listening to different stuff on the Atom since it was supposed to be so good at it.

Since then, I've had a hard time telling the K3 and the Atom apart without a way to do an AB test. But I've always wondered if there was a way to measure this subjective perception of bass response that supposedly improves with extra power or degrades with insufficient power.

 

[tdastrup]

I know exactly what you're referring to.

T60RPs or similar can be driven perfectly with an SMSL SP200 (or any similarly measuring amp)...

I finally got the go-ahead from the finance department (the wife) to grab some Argon MK3s. They'll be driven by a DX3 Pro which only does 700mW x2@32Ω and 125mW x2@300Ω. Since the Argons are 50Ω it'll be somewhere between 125-700mW but on the upper end, I presume.

Upgrade path to end-game will be adding a THX AAA 887, then a D90. Those are still a year or two out, so hopefully my DX3 Pro can soldier through until then.

 

[solderdude]

750mW to be exact.
You can reach 120dB peak levels ( is not the same as average levels !).

 

[Wes]

More power <> better sound.

More power can give better sound under a few conditions:

1. avoiding amp clipping
2. enable the amp to drive the speaker ("assuming a cone-head speaker") better say at a very low impedance
3. allow the amp to stay away from the A-B class transition (sounds sociological...) and avoid xover distortion