[colonelkernel8]

Where did you get the "of course" part? The measurements are tapped at the input to the headphone. Where do you think that signal is going if not into the headphone?

There are two white papers on this, one from Benchmark and another from Texas Instruments. You can start reading from here: https://www.audiosciencereview.com/...ments-of-rme-adi-2-dac.2582/page-5#post-74064

Ah you're right, I'm not sure how I was twisting that in my head. I'd be interested in seeing the THD components at one of the low frequencies, say like a 50 Hz sine wave with the Sony MDR-V6s or the Grado SR60es you tested. Do you have that perchance?

Apparently I need to buy a Benchmark DAC...it's dead flat. What gives?

 

[pwjazz]

@amirm It would be really interesting to see the same graph from a different amplifier (perhaps a drastically different one like an OTL tube amp or something).

 

[amirm]

@amirm It would be really interesting to see the same graph from a different amplifier (perhaps a drastically different one like an OTL tube amp or something).

If you follow the link I provided, in a page or two you see more test results from other amps. It includes a Schiit Lyr tube amp.

I plan to run this test from here on so there will be more results coming.

 

[colonelkernel8]

If you follow the link I provided, in a page or two you see more test results from other amps. It includes a Schiit Lyr tube amp.

I plan to run this test from here on so there will be more results coming.

It does make me a little sad

If the RME is all over the place, no doubt my iFi gear is just as bad if not worse. I HAVE BEEN DECEIVED!

Edit: maybe not, after reading the rest of that thread @amirm posted. Thanks amirm. I wonder if the Benchmark amp has a means of suppressing back EMF? If so, how? I mean it makes sense: you have a dynamic driver that at low frequencies has a greater excursion, which would induce a decent voltage on the coil feeding back to the amplifier, or am I mistaken?

 

[Glmoneydawg]

It does make me a little sad

If the RME is all over the place, no doubt my iFi gear is just as bad if not worse. I HAVE BEEN DECEIVED!

Edit: maybe not, after reading the rest of that thread @amirm posted. Thanks amirm. I wonder if the Benchmark amp has a means of suppressing back EMF? If so, how? I mean it makes sense: you have a dynamic driver that at low frequencies has a greater excursion, which would induce a decent voltage on the coil feeding back to the amplifier, or am I mistaken?

If its voicecoil generated voltage you need to look at amps damping factor i think?Although i'm not so sure a headphone driver is capable of generating significant voltage...easy experiment if you're comfortable pumping your drivers back n forth though.

 

[71 dB]

Well said. Check out this graph of testing the headphone amplifier of the RME ADI-2 DAC (one of the best DAC/headphone amps I have tested) with real headphones versus dummy (resistive) load:

There is as much as 40 dB of degradation from headphone load versus dummy load! And of course nobody specs amps with real headphones (you would be lucky if the spec had both channels driven versus one, or was at reasonable distortion).

Luckily ear is not as sensitive to distortion at low frequencies as it is at higher frequencies. Headphone is an electro-mechanical resonator damped partly mechanically and partly electronically. What is the output impedance of RME ADI-2 DAC at lowest frequency (below 100 Hz)? Headphones are also "microphones." If you measure them in noisy environment you can have noisy results. But yes, driving headphones gives you more THD+N than resistors. Does it matter? Probably not. Grado SR-60e is probably good enough and the others overkill in THD+N performance.

 

[71 dB]

Could they be designing headphones to make them deliberately persnickety? I don't know myself. I wouldn't buy portable headphones that require amping. Just one more thing to lug around. I'd rather just use cans that you can plug into everything.

Saying that you are yourself deliberately persnickety.

As you know, electroacoustic transformers are always compromises. Ignorance has partly caused the situation. Output impedance is too complicated of a concept for most consumers to understand so manufacturers don't make a fuzz about it and consumers don't know what to demand.

 

[castleofargh]

deciding to measure the characteristics of low and sub frequencies on a Grado, that's vicious. ^_^

I believe some may read too much into those measures. headphones have high distortions(high compared to DACs and clean amps). this measurement while revealing that not all amps react the same way to various headphones, which is very interesting if other standards measurements fail to show it, let's not forget that we're not measuring the amp's behavior alone, but the amp+headphone. and that was never going to achieve DAC level of distortions.

 

[pwjazz]

let's not forget that we're not measuring the amp's behavior alone, but the amp+headphone

I find that really exciting, as we often talk subjectively about amp+headphone but do relatively little measuring of it. It seems that we have a lot to learn in that area, which seems fun!

 

[bigshot]

Output impedance is too complicated of a concept for most consumers to understand so manufacturers don't make a fuzz about it and consumers don't know what to demand.

It seems to me that what the consumers demand is something you plug in and it works. I would think that impedance is predictable enough to at least come up with a standard so amps and headphones play well together. I understand though that a big part of the audiophile mindset is to overcomplicate things and to plug as many black boxes together as possible between the source and transducers. If you can achieve great sound simply, I don't know why you would want to do it complicated. But that's just me. I realize most audiophiles don't think that way.

 

[amirm]

I'd be interested in seeing the THD components at one of the low frequencies, say like a 50 Hz sine wave with the Sony MDR-V6s or the Grado SR60es you tested. Do you have that perchance?

I have not tested the headphone. But someone else has done so online: https://www.rtings.com/headphones/1-2/graph#306/2085

 

[colonelkernel8]

I have not tested the headphone. But someone else has done so online: https://www.rtings.com/headphones/1-2/graph#306/2085

What I mean is to plot the components of the THD of just one frequency using the same method you’re using now, just instead of a sweep, a single frequency.

I’ve been following that thread on your forum. I’m interested in identifying how the Benchmark suppresses back EMF and if it’s even necessary. I’ll do some paper perusing. They seem to just avoid the issue altogether in Benchmark’s “paper”. I also have Analog Devices’s “Linear Circuit Design Handbook” edited by the great Hank Zumbahlen, who I met in person a couple times.

 

[ev13wt]

Very soon, 90+% of headphones will be bluetooth or similar. Thats plug and play.

 

[amirm]

What I mean is to plot the components of the THD of just one frequency using the same method you’re using now, just instead of a sweep, a single frequency.

Ah, sure. When I get a chance I will plot it.

 

[colonelkernel8]

Ah, sure. When I get a chance I will plot it.

Thanks much. This discussion may merit its own thread. I’ll consider doing so when I’m in front of my computer and not on a phone.