[odigg]

PLEASE NOTE - I already know the standard answer to this question. Yes, by most accounts an amp is required. So please don't come here and simply say "An amp had an audible improvement so yes an amp is required" when the test was not done in any sort of a controlled fashion. Also, please don't turn this thread into an argument about DBT or controlled testing.

More than anything, this is a challenge to DBT supporters.

So here is my overall question. In a controlled test, will you conclude a dedicated headphone amp is really required when you compare a headphone plugged into that amp versus plugged straight into a decent (something that measures well) sound card.

There are a number of DBT supporters on this site. Still, some DBT supporters seem to stop short of testing everything on their chain. For example, a lot of DBT supporters will say all cables and DACs/sources sound the same.

But a lot of people say that dedicated amplifiers are required to drive headphones. There are the technical arguments (skew rate, transient response, dynamic range, controlling the drivers) and then of course the claims that, subjectively, certain headphones need an amp to sound their best.

But have DBT supporters tested if a dedicated amp is required versus plugging your headphones into your sound card? Are you making these statements based on actual controlled tests? After all, in DBT we are looking at audibility, not just technical superiority.

I have done a number of volume matched tests with headphones ranging from the Denon D5000 25 ohms to the AKG 240M 600 ohms. I have an A/B switch that allows me to switch between two inputs (sound card and amp) easily and almost transparently.

With the high impedance headphones I've tried I have heard no difference. A dedicated amp does nothing. With the low impedance headphones there is a very subtle bass rolloff when plugged straight into the sound card. I might not have even detected it if I had to move the jack of my headphones from the sound card to the amp resulting in a break between the music.

Some sound cards have some electrical noise, but in my experience this noise is not audible when music is playing.

Yes, I'm making a claim that totally goes against all conventional statements on Head-Fi.

I realize doing a DBT or even an SBT to test such a thing may be difficult. I encourage doing a blind test. If nothing else, please use an A/B switch and make sure you match volumes using a multimeter.

This is a challenge to DBT supporters. Take what you say about testing cables and sources and apply that same rigor to testing if an amp is required. I'd like to know your results.

**edit**

I should have originally stated that the headphone amp should probably not be a tube amp and should have a flat frequency response in the audible range. I'm not all that familiar with tube amps so the statements I've made may not hold for tube amps.

 

[Uncle Erik]

What, precisely, do you mean by "decent" soundcard? If you would please provide models with specifications, including power output and output impedance, I would appreciate it.

And which headphone amps are you comparing against?

There is a wide variation in power amps. Do you suppose to make a blanket statement against all of them? What about ones with speaker taps?

One problem with soundcards is that they run off a PCI slot. Even if you have a Turbo 8,000W supply in your box, it only delivers 3.3V or 5V to the slot. There, you run into much the same problem portable amps face - not enough power. Well, for power hungry headphones. Again, there is wide variation in headphone power requirements, so it is difficult to make broad statements as to "all."

But back to the power supply. 3.3V or 5V with limited Amps ain't much to work with, especially when compared to 120V out of the wall with enough Amps to charge an electric car. Yes, you really do need that kind of power to drive an amp into class A and have enough power to handle a difficult headphone.

Speaking of which, do you know which class your soundcard's amp operates in? Is it class A? AB? Or does the manufacturer not tell you?

Further, the soundcard has to be powered from a switching power supply, the one in the computer. Go try to find any other piece of audio gear that runs off a switching supply. They aren't used, and for good reason. Amplification relies on excellent power. Linear supplies are necessary for that. You don't need to DBT, you can put a linear supply and a switching supply on an oscilloscope and literally see the difference. Switching supplies are used in computers because they work well enough and are really cheap and lightweight. For precision, you need a linear supply. Preferably a well filtered, regulated and rectified one. If you don't have precise power, you don't have precise amplification. Again, go look at *every* piece of audio gear and check to see if it has a linear or switching supply. Even the $199 boxes at Best Buy use linear supplies, let alone the audiophile field.

If you want clean power, enough power and class A operation, you just are not going to get that from some mass market 3.3V PCI slot. You just aren't.

You can roll opamps all you want, but that's like putting a special Indy car wing on your 1986 Hyundai. Yeah, that might be the best wing in the world, but you're just polishing a turd and it's not really going to help anything.

I could go on. There is a difference - and not a subtle one. There's a sense of magical thinking at Head-Fi where if someone thinks something sounds good, then it automatically trumps everything that costs more. That is not the case. I'll come back into the thread after the real fighting breaks out.

 

[odigg]

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif What, precisely, do you mean by "decent" soundcard? If you would please provide models with specifications, including power output and output impedance, I would appreciate it.

To be honest it really doesn't matter. Use the output of your laptop, your motherboard's onboard sound, HT Omega Striker, our whatever you have at home. If it's a relatively new product (made in the last year or two) that's better as the newer sound cards, even onboard sound on many motherboards and laptops, measure quite well and support 24bit/192khz.

If people ran RMAA tests (loaded and unloaded) with the equipment they are using all the better as that'll just give us more info to play with.

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif I'll come back into the thread after the real fighting breaks out.

As I said, I'm asking people to do a controlled listening test. I understand that there are technical arguments for the superiority of a dedicated amp, but many people cannot hear technical/measurement improvements beyond a certain point in a DBT. If people want to have a technical argument that's fine with me. I'm won't be a part of it.

 

[moogoob]

I must say that I am a huge skeptic when it comes to things like cables and whatnot and I could easily tell the difference (lower sibilance and greater dynamics after versus before) between switching from a Headroom Total Bithead (portable amp) alone with my Grado 325i Goldies and running the Bithead into a Headroom Micro (non-portable desktop amp). Things like the cases of sibilance could not be explained by placebo alone, so I had to conclude that my Micro was a good purchase. Needed? Not really, but it certainly helped even out problems that I was chalking up to my recordings.

 

[cswann1]

Quote:

Originally Posted by odigg /img/forum/go_quote.gif So here is my overall question. In a controlled test, will you conclude a dedicated headphone amp is really required when you compare a headphone plugged into that amp versus plugged straight into a decent (something that measures well) sound card.

No. And no testing is needed to answer that question. An amp is designed to drive a load to higher levels (than would be allowed without one) while keeping distortion to a minimum.

But to listen to music all you need is a source that gives an analog signal and a speaker or headphone. No amp is required in the case of headphones. You'll hear the music just fine.

It sounds like you are sidestepping the question of whether an amp is required to give an overall better listening experience vs and un-amped setup. If that's the case, Uncle_Eric is right on the money when he points out that you can't make a blanket statement to answer that question. You need to specify what gear you want to compare.

 

[odigg]

Quote:

Originally Posted by cswann1 /img/forum/go_quote.gif It sounds like you are sidestepping the question of whether an amp is required to give an overall better listening experience vs and un-amped setup. If that's the case, Uncle_Eric is right on the money when he points out that you can't make a blanket statement to answer that question. You need to specify what gear you want to compare.

I asked exactly the first question you've stated in your first sentence above. I'm asking if people can hear a difference in a controlled listening test. If people hear a difference they usually say it is an overall better listening experience with a dedicated amp.

Run this test with any gear you want. I'll add these rather broad specifications for what gear you should use.

Sound Card - Preferably use something made in the last two years. Use your motherboard's onboard sound, your laptop's built in sound, the sound card you purchased for whatever price. Turn off all DSP/EQ effects.

Amp - Something with a high input impedance, low output impedance, and a flat frequency response. Many solid state amps (opamp based or not) meet these basic requirements.

Do the test at a volume that is loud enough where you can hear all the details you want to hear in your music, but not so loud you damage your ears.

Finally, don't run the sound card or amp into clipping or distortion. I've found all sound cards I've tested can run the headphones I've tested to loud volumes without such issues.

Also, it would be nice if you could do a loaded and unloaded RMAA test so we can see the frequency response of the equipment you are testing.

 

[dvw]

All sound cards technically have a built in amplifier. So the real question is are the dedicated amplifier better than the integrated amplifier?

It really depends on the load. With speaker systems, the answer is easy. Better amplifier can drive more difficult load. With headphones, the answer is not so easy. First of all headphone is a relatively easy load to drive.

Portable devices with limitation on power supply and heat dissipation exhibits the most difference. Soundcard with a wide variance in quality are more difficult to determine. In theory you can build an "audiophile grade" sound card with good amplifier build in. Actually, that's what the DAC/amps are, outboard sound card.

 

[jellojoe]

Which headphones are we suppose to use with this test? Big difference between driving an IEM vs. vintage AKG with an onboard sound card.

 

[ILikeMusic]

Quote:

Originally Posted by dvw /img/forum/go_quote.gif All sound cards technically have a built in amplifier. So the real question is are the dedicated amplifier better than the integrated amplifier?

Yes, of course. The answer depends on a host of specific criteria, there is no universal answer.

 

[Punnisher]

It is worth the extra money for a dedicated amp, whether portable or desktop.

The amount of current needed to drive low impedance headphones is simply more than an mp3 player (or soundcard most likely) can deliver. Bass rolloff is easily fixed by an amp.

 

[krmathis]

Not required, as long as you use a moving coil headphone which can be driven by the headphone out on your soundcard/DAP/...
But a dedicated amplifier certainly can improve the sound quality, compared to the said built in one. Definitely worth it imo!

 

[Jingo Lingo]

I'd be eager to see a DBT on this.

 

[Hirsch]

First, you need to use a headphone amp to drive headphones. Some sound cards, receivers, players, and other gear has headphone jacks on it. That allows you to use the headphone amp built into that piece of equipment. The real question is whether the amp on that gear is better than an aftermarket headphone amp that plugs into a line (or USB) output on your gear. If your soundcard has paid good attention to the amplifier section, and you're testing against a cheap POS, then no, you don't need an external amp.

However, there are some pretty stringent limitations in terms of space for components and power capacity that can limit what a sound card's headphone output can do. Yooour soundcard is also located in an environment where a fair amount of electrical noise might be present. Those limits do not apply to external amps that have their own cases, shielding, and power supplies. The inverse square law may also help out with electrical noise. A good external amp will be better than anything that can fit on a soundcard. It will not necessarily be cheap, and in some cases can reveal limitations of a soundcard as a source that you never knew were there.

So, there is no right answer to the question that you posed it. Both amplifiers, either the one on the soundcard or the external amp, are doing exactly the same thing. So, the question is not whether or not you need an amp. You do. The real question is "which amp is better?" and that answer depends entirely on the components in question. An external amplifier has fewer inherent limitations, so the odds are in its favor.

 

[odigg]

Quote:

Originally Posted by dvw /img/forum/go_quote.gif All sound cards technically have a built in amplifier. So the real question is are the dedicated amplifier better than the integrated amplifier?

Yes. Exactly.

Quote:

Originally Posted by Jingo Lingo /img/forum/go_quote.gif I'd be eager to see a DBT on this.

If nothing else, I'd like people to do a proper volume matched test. Everybody says "A dedicated amp improves the sound quality" or "The improvement was so obvious I don't see how it could be placebo."

But just making a claim based on technical or subjective arguments is exactly why some people support DBT in the first place.

As stated, I've done a number of volume matched tests. Doing a DBT has not been possible because of practical reasons. There is some bass roll off with low (25 ohm) headphones, but I found this roll off to be small and easy to miss. With the highest impedance headphone I've tried (600 ohms) I could not tell the difference.

If there are a bunch of people in the Raleigh/Durham, NC are who are interested in doing a DBT or SBT, we could try to put something together so see the results. I'm as curious as most people.

 

[ILikeMusic]

Quote:

Originally Posted by Jingo Lingo /img/forum/go_quote.gif I'd be eager to see a DBT on this.

Such a test is designed to compare two identical items and it's not really applicable to a 'with an amp and without' sort of scenario.

Also, in a consumer equipment environment an amp will almost always sound different in some way as there usually is no attempt to carefully match the interface between amp and source, and whatever mismatch exists can affect the sound, often even more than that provided by the additional gain alone. In order to really test something like this it would have to be done in a laboratory environment... incidental tests with consumer equipment would be useless.

.