[daleb]

I've been wondering truly how much of the amp set ups are placebo and how much has been proven to matter (or as close to proof as one can get). I've seen tests on low-end DACs/Amps vs high end, but nothing about just plugging directly into sound cards are motherboards.
 
edit: I can't figure out how to make this not sound rude. Sorry!

 

[Ruben123]

Well amps and dacs SHOULD sound the same but in reality there can be differences... Amp imp can have big consequences for low imp BA iems.
Also, daps can sound different because of standard eq settings that are not flat.

 

[stv014]

There are some loopback recordings of various "low end" devices, including motherboard HD audio codecs, in this older thread. You can probably find more if you search the forums.
 
Using onboard audio as a line output can very well be fine in terms of sound quality, although audible interference from other parts of the computer is not uncommon. The headphone outputs are more likely to be problematic without an external amplifier, as they often have high output impedance and/or are capacitor coupled, the power output could be limited, and the noise floor (and possibly interference) may become audible at low volume settings with sensitive headphones. Note that the majority of these issues, with the exception of low voltage output, actually affect high impedance headphones less.

 

[stv014]

Also, daps can sound different because of standard eq settings that are not flat.

 
Indeed, software issues can also affect audio playback on PCs in a number of ways.

 

[Ruben123]

Yea but I mean also normal DAPs, like the Pono. Maybe DAPs could sound different because of some smart EQ in there.
 
i dont know though, just a thought.

 

[daleb]

I went looking around and there are tons of double blind and ABX studies that prove that amps don't do anything when they get expensive, but these studies always compare $100 amps to $1000 amps and above. I can't find anything that compares plugging headphones directly into the source (e.g. phone/computer) to expensive amps/dacs. Yes, we know that physics proves there is a difference, but psychology can not be solved with just physics equations. The discussions found on these topics are always just conjecture. It is a shame that there is no peer reviewed studies done at universities for this branch of psychoacoustics... or at least none that I have found. 

 

 

[jcx]

amps have to be able to drive the connected load to loud enough without error
 
that's not always the case - headphone loads vary a great deal in the required I and V as well as having huge differences in efficiency
 
its far too simple to say "all amps (should) sound alike" when some simply can't drive some particular headphone well, whether through lack of I, V output drive ability form limited supply V or too small output transistors
 
or in the case of sensitive iem the amp may have too much gain and you hear a annoying amount of hiss, have to turn down the volume control to where they often track poorly...
 
there are other audible errors such as too small a output coupling cap which cuts out some low bass - with low Z headphones, but would be fine with 300 or 600 Ohm headphones
 
so the same amps that audibly fail with some headphone could be quite acceptable paired with headphone with Z and sensitivity that fits well to the amp's capabilities
 
 
soundcards vary in what they do between line level  outputs and "headphone outputs" - when they have the latter at all
 
line level outputs may have 100s of Ohms of series R, headphone outs are expected to have less but it still may be too high to give flat response with say multidriver iem which have lumpy impedance vs frequency curves
 
 
you really need to name the headphone model and the soundcard to get any useful advice - and soundcards and amps seldom have comprehensive specs in their manuals - even for the limited purpose of checking if I, V capabilities are adequate

 

[daleb]

I'm sorry, but I feel what you are saying, and what everyone else has said, goes entirely against the spirit of the question. Everyone is posting this information on how amps affect the sound of a pair of headphones, and I have found this information over and over, posted by many different people, but the question I am asking is about where the scientific evidence for this information is? I want peer reviewed articles, academic publishings. Yet, I can't find them anywhere. Yes, everyone is telling me that amps do have an affect, and they explain in what way, but testimonials aren't what I am looking for.
 

you really need to name the headphone model and the soundcard to get any useful advice - and soundcards and amps seldom have comprehensive specs in their manuals - even for the limited purpose of checking if I, V capabilities are adequate

This also goes against the spirit of the question, as this will not provide any sort of scientific evidence and understanding, it will just provide more testimonial evidence on a specific product. I'm not looking for any specific product, the assumption is that I don't own any (because I don't, the only amp I have is a FiiO E11,and the highest impedance headphone I have is the HD 280s).
 
To be brutally honest and probably make a ton of people really mad? I don't believe amps do anything that a human ear could discern. Machines and computers can, of course, but not humans. I believe I could plug some HD 800s into my motherboard and have no difference between that and a $2000 amp. But you know what? I have no proof of that. I have zero evidence that what I say is true. In fact, it is very likely I am wrong. And thats why I'm here. Me saying that amps are useless is just as useless to everyone as saying amps are the most important piece of audio equipment to have, unless the person saying it has empirical evidence to back it up. But so far, it looks like that evidence doesn't exist. So, the other question is, why? How does an industry where equipment that goes for thousands of dollars not have any scientific studies that they are actually useful? That makes no sense to me. Or maybe I'm just bad a Google...
 
edit: bolded for emphasis

 

[arnyk]

  I went looking around and there are tons of double blind and ABX studies that prove that amps don't do anything when they get expensive, but these studies always compare $100 amps to $1000 amps and above. I can't find anything that compares plugging headphones directly into the source (e.g. phone/computer) to expensive amps/dacs. Yes, we know that physics proves there is a difference, but psychology can not be solved with just physics equations. The discussions found on these topics are always just conjecture. It is a shame that there is no peer reviewed studies done at universities for this branch of psychoacoustics... or at least none that I have found. 

 

 
It is very unlikely that there were or will be comparisons of audio gear such as headphone amps by universities and reported in peer reviewed papers. These institutions are about pure science, and audio gear is applied science. So, the academic papers may tell us  whether an audio system has 0.000001% THD can be detected by listeners (and of course it can't) you are unlikely to find data about comparisons between cheap and expensive DACs in peer reviewed papers, ever. Never were any, never will be.
 
Of course psychology won't be addressed by physics studies, but keeping psychology out of listening evaluations of the physical capabilities the human ear/brain is one of those things that science does.
 
The most thoroughly conjecture-ridden discussions are those among audiophiles about their expensive audio jewelry. Their listening tests are such utter jokes that they aren't even appropriate to call tests because there are no absolute standards for sound quality. There is a list of generally accepted (by science) list of confusing influences that they never control, and this leads to both false positives and negatives.
 
No, audiophiles aren't going to ever be saved from themselves by science, if it were to be done they'd have to do it themselves!

 

[arnyk]

  There are some loopback recordings of various "low end" devices, including motherboard HD audio codecs, in this older thread. You can probably find more if you search the forums.
 
Using onboard audio as a line output can very well be fine in terms of sound quality, although audible interference from other parts of the computer is not uncommon. The headphone outputs are more likely to be problematic without an external amplifier, as they often have high output impedance and/or are capacitor coupled, the power output could be limited, and the noise floor (and possibly interference) may become audible at low volume settings with sensitive headphones. Note that the majority of these issues, with the exception of low voltage output, actually affect high impedance headphones less.

 
I've personally observed just about every flaw mentioned above in onboard and even very expensive sseparate audio interfaces with headphone jacks. Not all audio gear has problems driving headphones, and reasonably priced gear that drives headphones well is readily available, but it is not true that there is an absence of audible differences in this performance area.

 

[prot]

http://www.head-fi.org/t/754695/asrock-z97-e-itx-ac-outperforms-my-bifrost-and-crack-hd-600/0_50

http://www.tomshardware.com/reviews/high-end-pc-audio,3733.html

 

[sonitus mirus]

http://www.tomshardware.com/reviews/high-end-pc-audio,3733.html

 
The takeaway from the linked review:
 

Of course, we're ready for the audiophile community to rise up in arms about the statement you'll read next, but it's true that neither an intermediate enthusiast nor a serious one with ~$70,000 in gear at home were able to reliably tell apart any of the four devices once we properly set up a blind test with accurate volume-matching. We actually enjoyed them all as great audio experiences.

Using world-class headphones, a $2 Realtek integrated audio codec could not be reliably distinguished from the $2000 Benchmark DAC2 HGC in a four-device round-up.

 

[prot]

The takeaway from the linked review:

That's a very tough takeaway for many ... better not to mention it outside the science forum

 

[arnyk]

That's a very tough takeaway for many ... better not to mention it outside the science forum

 
It is all about science.
 
(1) +/- 0.1 - 0.2 dB frequency response variations across the audio band cannot be distinguished.
 
(2) < 0.1 % THD cannot be distinguished.
 
(3) SNR 90 dB down or better is perceived the same as being noiseless.
 
(4) Jitter  90 dB down or better cannot be distinguished.
 
What else is there except the state of mind (prejudices) of the listener?

 

[prot]

It is all about science.

(1) +/- 0.1 - 0.2 dB frequency response variations across the audio band cannot be distinguished.

(2) < 0.1 % THD cannot be distinguished.

(3) SNR 90 dB down or better is perceived the same as being noiseless.

(4) Jitter  90 dB down or better cannot be distinguished.

What else is there except the state of mind (prejudices) of the listener?

I'd be curious to know where did you get those threshold numbers ... IIRC there was such a thread around here but cant find it.

I'd say no3 is a bit low .. and no4 kinda controversial