[Bill-P]

Quote:

Originally Posted by stv014 /img/forum/go_quote.gif

Well, it is your idea that more power, even when not actually used, improves sound quality, not mine.
 
How do you know that the right bass is not what you get when you drive the headphones with a near-zero impedance voltage source ? It is not like a DT990 is in the need of more V-shaped frequency boosting from the amplifier.

 
"More power" does not mean driving directly out of speaker output. That's called "overloading". You'll blow your headphones instantly.
 
And I know it's the "right bass" because I have heard it in a higher-end amp. It's completely different from "more bass".
 
Quote:

I do not recall mentioning IEMs at all. Those have quite different impedance vs. frequency characteristics, and typically do not resonate in the bass range.

 
There are dynamic IEMs as well, dear sir. Headphones are not the only audio devices to have dynamic transducers. And no, dynamic IEMs don't have different impedance vs frequency characteristics. They have different impedance vs frequency charts, but that's a completely different thing.
 
Seriously, it's getting ridiculous... so I'll just disregard this thread from this point on.
 
Good day, sir.

 

[stv014]

Quote:

"More power" does not mean driving directly out of speaker output. That's called "overloading". You'll blow your headphones instantly.

 
I actually tried it (with carefully set volume, and not too expensive headphones; the amplifier also has a fairly low DC offset on its outputs), and it did not damage the drivers. Of course, it did not improve the bass response either, not that I expected it to. It sounded like a lower powered source at exactly the same volume, except for some audible hiss and hum.
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
And I know it's the "right bass" because I have heard it in a higher-end amp. It's completely different from "more bass".

 
I do not doubt you heard what you heard, but there are good reasons to take subjective opinions with a grain of salt, unless the comparison was made under sufficiently well controlled conditions, which is rarely the case. "High end" audiophile products also sometimes have significantly colored and/or distorted sound.
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
Load impedance != output impedance
 
And don't just throw equations you don't understand in order to make yourself look smart...

 
I do not recall making any claims that load impedance and output impedance are the same. Your assumption that I do not understand the equations, and the other comments I did not quote are basically nothing but petty personal attacks that show desperation and running out of rational arguments. Do you have anything relevant to add to the discussion of the amplifier power and output impedance topic ?

 

[Bill-P]

I actually retracted that because it was not worth it, but if you need me to point it out...
 
[V load] = [V line] * [Z load] / ([Z load] + [Z output]) can be broken down as thus:
 
[V load] = voltage given to the load, or transferred to the headphone, which varies depending on the load
[V line] = constant voltage supplied to the amp
[Z load] = headphone impedance, which varies depending on frequency
[Z output] = output impedance of the headphone amp
 
In the case that [V load] is constant, it only means that [Z load] is constant since [V line] and [Z output] does not change. It doesn't mean anything else.
 
That equation basically calculates how much voltage the headphone amp needs from the power supply in order to deliver the required voltage to the headphone. Higher output impedance simply means more power needs to be drawn from the power supply (battery or wall socket). It doesn't really have anything to do with the impedance swing of the headphone or how output impedance affects volume or bass.
 
Or are you trying to show something that I'm not seeing here?
 
Edit: or in fact, allow me to show something to you using that same equation.
 
Based on that equation, if [Z load] is smaller than [Z out], it would cause [V load] to drop significantly depending on how small [Z load] is. That's why you don't want to try to plug low impedance headphones into amps with high output impedance, because it causes the voltage to drop, and the amplifier will have a harder time stabilizing that. But also conversely, a higher output impedance causes voltage [V load] to lower given that [V line] from the power supply stays the same. When [V load] lowers, then power output, which is [V load]^2 / [Z load] also lowers, and it causes volume to be tinier. BUT if you volume-match, the sound quality should still be the same. So then what effect do you propose that output impedance has on sound quality?

 

[stv014]

Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
In the case that [V load] is constant, it only means that [Z load] is constant since [V line] and [Z output] does not change. It doesn't mean anything else.

 
No, that is not what I meant. It is not Vload that is constant, but Vout. Vload is the voltage that actually falls on the headphone drivers. On the pictures below, replace V1 with Vout (an ideal amplifier with zero output impedance; I did not refer to the power supply voltage, I do not know why you changed it to Vline), R3 with Zout, and the R1/L2/R2/L1/C1 network (which is a simplified simulation of a headphone driver) as a whole with Zload, and then the voltage on Zload is Vload, as plotted on the right, and according to my formula, where all voltages and impedances are complex numbers. The lower Zout (R3 on the schematic) is, the flatter the response becomes.
 
   
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
That equation basically calculates how much voltage the headphone amp needs from the power supply in order to deliver the required voltage to the headphone. Higher output impedance simply means more power needs to be drawn from the power supply (battery or wall socket). It doesn't really have anything to do with the impedance swing of the headphone or how output impedance affects volume or bass.

 
That could be true if Zout was inside a negative feedback loop (assuming there is a high loop gain, and the system is stable), but I meant the actual closed loop output impedance, which does affect the frequency response when driving a reactive load, and making it high is how you typically get the alleged "proper driving" behavior shown in your earlier post with the HeadRoom graphs. However, with a near-zero closed loop output impedance, flat frequency response, low distortion, and no stability issues - these are the properties one would normally expect from an objectively "good" amplifier - impedance peaks of the load do not result in voltage peaks on the output.

 

[stv014]

Quote:

So then what effect do you propose that output impedance has on sound quality?

 
Well, you have already explained it, maybe without knowing: "Based on that equation, if [Z load] is smaller than [Z out], it would cause [V load] to drop significantly depending on how small [Z load] is.". So, if Zload varies with frequency (like it very often does), then Vload will vary too, resulting in frequency response variations. Furthermore, if Zload is non-linear, then Vload will have non-linear distortion (see the Benchmark article linked earlier).
 
You may want to create a thread on the Sound Science forum for further discussion, or post in this already existing one. You could get responses from others as well, who probably do not read this thread.

 

[Bill-P]

Yeah, but that has nothing to do with output impedance.
 
And the headphone amp's worth is actually in how it controls [V load], and not in how much [V line] it can draw. When I wrote "more voltage", I really meant "more [V load]".
 
But [V load] is not always linearly proportionate to [V line] due to instability of voltage line in a circuit... either because of the opamp (which in reality is just a very small circuit) or the circuit design. As a result, different amps sound... differently, because they distribute [V load] differently. If all headphone amps distribute [V load] the same way, then we wouldn't have this absurd race to "flat frequency response", or "perfectly proportionate [V load] and [V line]". But bottom line is... it's apparent that not all headphone amps distribute [V load] the same.
 
Now, if you take [V line] into account, portable headphone amps like the Fiio E11 runs mostly on battery. The battery in the Fiio E11 is rated for 3.7v. Provided that Fiio E11's output impedance is negligibly small (around 0.5), I'd say for headphones with sufficiently high impedance, [V line] = [V load] for E11.
 
However, with a max [V load] of 3.7v, Fiio E11 can only output about [power] = [V load]^2 / [Z load] = [3.7]^2 / [250] = 0.055W, or 55mW to a 250 Ohm load, and that's considering impedance is constant at 250 Ohm. If impedance swings to, say, 350 Ohm, then power output drops to [3.7]^2 / [350] = 0.039W, or 39mW to a 350 Ohm load.
 
Now, sound level difference is generally measured by the equation [Loudness in dB] = 10 * log ([power after] / [power initial]) so the difference for 39mW and 55mW is: 10 * log ([39mW] / [55mW]) = -1.49 dB. And a difference of 1.5 dB is very audible.
 
You can take a look at that loudness equation here:
http://www.musiccenters.com/vol.html
 
And that's what I meant when I wrote that lack of power would generally decrease bass. That's not even a subjective listening experience. It's hard-coded data.
 
And by the way, that's assuming a lot of things... like that the Fiio E11 doesn't have undue colorations due to unbalances in its design (as in it's a perfect circuit with perfectly crafted components), and that [V line] is constantly at 3.7v (this is not true because it's drawing from a battery)... among other things. But theoretically, even with a perfect design, I don't see how the Fiio E11 can sufficiently satisfy the DT990.
 
Too much power (voltage) is not necessary, I agree. But when you bump into voltage walls like with the Fiio E11, it's... disheartening.
 
And I wouldn't create a new thread in Sound Science... since I'm sure this topic has been discussed to death and back already.

 

[Yero]

Well, let's see...
 
1) You will BARELY hear anything out of an iPhone 4 or iPod if you try to drive a 250 Ohm headphone. It gets to about 20-25% the max volume that the headphone can make from my experience.
 
2) While power has to do with volume, it's usually more about supplying the headphones with enough voltage for the drivers to pump out music as they should. Lack of power will cause quality degradation. Typical symptoms include lacking bass, closed-in soundstage, less midrange details, etc... And yes, it is very noticeable when there is a lack of power... provided you have heard the headphones at their best.
 
3) If the laptop has a bad headphone out, then yes, DAC does improve a great deal. For MacBook specifically, the quality of the headphone out varies. But in general, I've found that a DAC will improve the sound quality... though whether your setup can "hear" the improvement is another matter altogether.
 
4) The Digizoid ZO has its own... "interesting" way to process music. It's more of a portable EQ than an amp. An amp should only increase volume and power output to headphones. It should not alter the sound of the source (DAC).
 
5) It depends on your ears, too. If you can't "hear" the difference, then leave it at that.
 
6) The DT880 Pro is a significant loss on bass and isolation compared to DT770 Pro. On the other hand, DT770 Pro is a significant loss in soundstage and instrument separation compared to DT880 Pro.
 
If you don't plan to lug that portable amp around too often, then I'd actually suggest that you look at other headphones. In fact, if you plan on just plugging the headphone directly into your iPhone 4, then... I think a pair of IEMs would give you better sound quality.
 

Thanks for a great reply and everyone else as well!
 
I am not sure how conversation went into DT990, I was only looking at DT770 and DT880, but anyways =)
 
I am kind of leaning towards this variations:
 
a) DT 770 250 Ohms + FiiO E11
b) DT 770 80 Ohms + DigiZoid ZO2
c) DT 770 80 Ohms
 
Now if different people have different opinions on driving DT 770/880 250 Ohms straight from iPhone, then 80 Ohms would need amps at all I would imagine?
 
I kind of went for the better bases and isolation with DT 770 Pro
 
What other headphones in your opinion, would compete with the looks, quality, and comfort of Beyers? I listen to everything, but a lot of it bass heavy music/ or supposed to be. Just trying to see some options before purchasing.

 

[Bill-P]

Sorry, it kind of went off a tangent there because I made a statement without backing it up...
 
I find that 80 Ohm still kind of needs amping, but if you want to drive it straight out of your iPhone, it'll at least sound better than 250 Ohm.
 
In the same pricing tier, I think the DT770 (Pro or Premium) is very hard to beat. The other alternatives have trade-offs of their own, and you likely won't get "better", just "different" sound.
 
Bass won't be a concern at all with the DT770. I think what you'll want to consider is how much treble you want to hear in your music. The 250 Ohm version has a tad more treble than the 80 Ohm version, and it's quite a noticeable difference, unlike the usual 5% difference in most Ohm versions of the Beyer line.

 

[stv014]

Quote:

Yeah, but that has nothing to do with output impedance.

 
Yes, it does. The higher the output impedance is, the more the voltage on a reactive load will vary with frequency. This can easily be proven in theory, and in practice with measurements.
 

Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
However, with a max [V load] of 3.7v, Fiio E11 can only output about [power] = [V load]^2 / [Z load] = [3.7]^2 / [250] = 0.055W, or 55mW to a 250 Ohm load, and that's considering impedance is constant at 250 Ohm. If impedance swings to, say, 350 Ohm, then power output drops to [3.7]^2 / [350] = 0.039W, or 39mW to a 350 Ohm load.

 
The E11 has already been measured to be able to output a maximum of about 2.7-2.75 Vrms into high impedance loads before it starts to clip (posting a link to the source is not allowed on this forum, but it should not be hard to find it). That translates to ~30 mW into 250 Ω.
 
You cannot calculate the maximum output voltage directly from the battery voltage, because it depends on a number of other factors. First, the E11 boosts the power supply voltage to much higher than 3.7 V with a DC-DC converter. Second, there is always some loss of voltage in the amplifier, even in a "rail to rail" design like the AD8397 used in the E11. Your calculation is wrong anyway, since without any tricks to increase the power supply voltage, 3.7 V would be the maximum peak to peak voltage (somwhat less in practice with a real amplifier), rather than RMS. A 3.7 Vp-p (1.308 Vrms) sine wave would translate to a power of 6.845 mW when driving a 250 Ω load. Of course, the E11 is capable of more because of the above mentioned DC-DC converter.
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
Now, sound level difference is generally measured by the equation [Loudness in dB] = 10 * log ([power after] / [power initial]) so the difference for 39mW and 55mW is: 10 * log ([39mW] / [55mW]) = -1.49 dB. And a difference of 1.5 dB is very audible.

 
You are still assuming that the output voltage has to increase at a frequency where the load impedance increases. That is simply wrong, unless you want an intentionally colored sound. The frequency response graphs you see at HeadRoom and InnerFidelity assume the same voltage (not power, nor current) at all frequencies. Yes, the power is reduced at impedance peaks, but that is already factored into the "standard" frequency response of the headphones. In other words, at the resonant frequency, the driver is more efficient, and needs less power for the same output SPL.
 
Did you read the Benchmark article I linked earlier ? As you can see on the graphs, the $1000+ Benchmark DAC1, which has very low output impedance, does not output a higher voltage at the resonance frequency of the headphone drivers than it does at 1 kHz. Do you think it cannot drive headphones properly ?
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
And that's what I meant when I wrote that lack of power would generally decrease bass. That's not even a subjective listening experience. It's hard-coded data.

 
No, it is simply incorrect information that stems from not understanding some basic concepts. Even if your assumptions were not wrong, they fail to explain how "lack of power" would decrease bass at low volume where the higher power is not needed.
 
Quote:

Originally Posted by Bill-P /img/forum/go_quote.gif
 
Too much power (voltage) is not necessary, I agree. But when you bump into voltage walls like with the Fiio E11, it's... disheartening.

 
So, how do you define enough voltage, especially if you want to set an absolute limit that does not even depend on volume ? Let's say I want to listen at a volume where the voltage of a 0 dBFS sine wave would be 0.5 Vrms, maybe 1.0-1.2 with more dynamic music, or 0.35 with heavily compressed "loudness war" recordings (these are a realistic figures from my experience with the DT770-250, but of course everyone's tastes are different). You think the E11's 2.7 Vrms maximum output voltage is not enough for that, and will result in "lack of bass" for whatever reason. Exactly how much more do I need then ?
 
Quote:

And I wouldn't create a new thread in Sound Science... since I'm sure this topic has been discussed to death and back already.

 
You do not want to discuss the topic in Sound Science, because you know others would agree with me there.

 

[MEDO]

To the OP, I've had the DT990\880\770 32\250\600ohm and I assure you that the DT990 will suit you best IF you don't need a closed headphone, they are MARVELOUS headphones for the value. of course the DT880 is nice too but the other one has more bass and exciting sound without losing detail, and NEVER go with under 250ohm from beyerdynamic headphones. i can see you're so concerned about the bass so the 990 is the best, they have even more soundstage than 880 and more bass and overall fun and exciting sound.

 

[Lorspeaker]

This is all too confusing...ohmspowerimpedencevoltage...
i went ahead n bought a pair of this :
14bucks refurbished.
http://www.amazon.com/gp/product/B001UE6HZM/ref=olp_product_details?ie=UTF8&me=&seller=
 

 

[planx]

Quote:

This is all too confusing...ohmspowerimpedencevoltage...
i went ahead n bought a pair of this :
14bucks refurbished.
http://www.amazon.com/gp/product/B001UE6HZM/ref=olp_product_details?ie=UTF8&me=&seller=
 

 
Haha exactly my thoughts. Trial and error people! Whatever sounds good to your ears is good enough! 

 

[Bill-P]

Quote:

 
No, it is simply incorrect information that stems from not understanding some basic concepts. Even if your assumptions were not wrong, they fail to explain how "lack of power" would decrease bass at low volume where the higher power is not needed.
 
 
So, how do you define enough voltage, especially if you want to set an absolute limit that does not even depend on volume ? Let's say I want to listen at a volume where the voltage of a 0 dBFS sine wave would be 0.5 Vrms, maybe 1.0-1.2 with more dynamic music, or 0.35 with heavily compressed "loudness war" recordings (these are a realistic figures from my experience with the DT770-250, but of course everyone's tastes are different). You think the E11's 2.7 Vrms maximum output voltage is not enough for that, and will result in "lack of bass" for whatever reason. Exactly how much more do I need then ?
 
 
You do not want to discuss the topic in Sound Science, because you know others would agree with me there.

 
God... I'm done. Explaining this to you has been a really hard task because you keep going off on your own assumptions and you keep disregarding what I wrote.
 
If you don't believe me, well, then please enjoy your Fiio E11 with 250 Ohm headphones, sir. Maybe one day, when you have scaled up in amp, you'll see what I'm talking about.

 

[stv014]

Quote:

God... I'm done. Explaining this to you has been a really hard task because you keep going off on your own assumptions and you keep disregarding what I wrote.

 
That is exactly what you do. You did not even respond to my last post in any meaningful way. I explained, more than once, why your claims are wrong, but you either ignored or misrepresented it (not sure if deliberately or because of the lack of understanding), or responded by obfuscation and veering off the original subject, or just resorted to personal attacks. Very typical of audiophiles when their beliefs are questioned.

To reiterate, here is a summary of your original statements (both are popular myths, by the way):
 
- higher maximum power output improves sound quality, even if never actually used, especially in relation to high impedance headphones (the standard "it will be loud enough, but not driven properly" FUD). Let's say I monitor the voltage on a headphone driver with an oscilloscope or audio analyzer, and, at the desired volume, it never exceeds a peak level of +/- 1.5 V; having two amplifiers, A and B, where A can output a maximum of +/- 3 V, and B can output a maximum of +/- 6 V, B will - according to you - sound inherently better, particularly by having an improved bass response
- when driving a transducer that is not a pure resistive load, the amplifier needs to provide constant power from a constant input voltage at any frequency for a "correct" response, rather than a constant voltage; that is, if the load impedance is 100 Ω at 1 kHz, and 1000 Ω at 100 Hz, then the voltage output at 100 Hz needs to be increased by 10 dB relative to the level at 1 kHz, otherwise the bass response will be lacking (note that this is not related to the absolute level, and thus the maximum power output, but you obfuscate the two issues)
 
Can you back these claims up, or would you prefer to change them ?
 
Quote:

If you don't believe me, well, then please enjoy your Fiio E11 with 250 Ohm headphones, sir. Maybe one day, when you have scaled up in amp, you'll see what I'm talking about.

 
I do not use my 250 Ω headphones with the FiiO E11 (as I do not have an E11, in case you wonder why), but even if I did, I would need more convincing arguments to "scale up" than condescending and snobbish remarks like what is quoted above. Would you be willing to prove the inadequacy of the E11 to drive such headphones in a level matched blind test against a more powerful amplifier that meets all objective criteria of being transparent ?
By the way, I am aware, as indicated right from the beginning, that it might not be enough for everyone and with every type of music, but when it is not, it is simply an issue of not being able to reach a sufficient level of loudness without audible distortion, rather than an inherent lack of bass, sound stage, or whatever other nonsense. Maybe the general sound quality of the E11 could be better, but that is not something that specifically affects high impedance headphones.
 
I see you still avoid discussing the topic in Sound Science. Does that indicate a lack of confidence in your argumentation ?

 

[Lorspeaker]

Once upon a time when i was toying with home speakers,
i had a 100W amp n switched to a 250w amp...
the difference in bass control n soundstage was significant enough for me to dump the 100watter...
Think the salesman told me more power gave me "better transicent control", better "startstop" grip on the diaphragm..
something like that..its not about the volume.
 
Anyway, i am no techqie...as far as my ears goes, its all good, more power, more joy.