How do I get round a hi-fi amplifier's "poor" headphones output?

[derbigpr]

 
You sure didn't offend me, and I think your post was quite responsible and topical. Never hurts to have people make me think and less experienced users can benefit from the different viewpoints so I say no worries mate, good points.

 
 
All you need to consider is the fact that people who design these amps are not some random guys, they know what they're doing, they're engineers and audiophiles, people who love good audio as much as you and me, on top of that, we're talking about big companies with decades of experience, and a lot of people, money and knowledge that allows them to do far more research and development than say a small company that makes a couple of headphone amplifiers.
 
If they can make the amps sound great with speakers, what makes you think that they couldn't do the same thing when it comes to headphones?  People on head-fi too often seem to act as if headphones are some magical, special, super-complicated devices from the future that take extra special care when it comes to powering them, which is just not the case if you get the basics right, which good sounding integrated amps do. Again, a good sounding speaker amp will sound good with headphones as well, unless they screw up the headphone output somehow, which again, is not the case with any of the amps I've ever tried, for the reasons explained in the first sentence of this post.
 
 
What gets most people confused in my opinion are prices. How could a 500 dollar, huge, 15kg heavy integrated amplifier, with lots of functions and purposes, full of circuits and components, have the same quality of components as a single-purpose small 1000 dollar headphone amp that only has a handful of components inside and is only designed to do one thing? Surely the quality of the components and quality of engineering in integrated amps must be horrible in comparison. And the headphone output must be just an afterthought. Right?
 
You'll find that if you take almost any headphone amplifier on the market (apart from some really cheap ones like O2, Fiio's, etc), especially as you go higher up in the price range, that the price of their electronic components (if you were to buy them separately, especially in large quantities) is a small fraction of what you actually pay for when buying the product, and in almost all cases, the housing of the amp costs close to or even more  than the electronic components inside, and a lot of your money goes into covering the cost of R&D, since again, we're talking about small companies that don't have multimillion dollar budgets they can invest into developing a product line. Yes, usually headphone amps are simpler in design, which in theory means that the sound will be less compromised and the signal will have to travel a simpler and shorter distance from the point where it enters the amplifier to the point where it exists. And yes, dedicated headphone amplifiers usually have top of the line passive components on their PCB boards, while entry level integrated amps don't, but do you really think you'll hear a difference between an amp that uses top of the line Nichicon caps instead of top of the line Panasonic caps? You'll probably measure minimal differences if you use the exact same design with different components, but will you hear a difference? That is a big question. And all the complicated stuff like DAC's or pretty much all chips on headphone amps, as well as connectors, transformers, volume pots, etc. are made by the same manufacturers that make those same components for integrated amplifiers produced by dozens of manufacturers from all over the world.   As I said again in my original post, what all the integrated amps I've tried had in common was that they got the essentials right...like...no audible noise at any volume, no channel imbalance at any volume, flat frequency response and proper extension in both directions, transparent sound with no obvious coloration or changes in the characteristic sound of a certain headphone, black background, plenty of power, lots of headroom when it comes to volume leading to great dynamics and lively sound with a lot of authority and no hint of distortion, etc.  And this cannot be said for all entry level headphone amps.
 
Is a dedicated headphone amp better for headphones? Yes. Compared to equally priced integrated amps, headphone amps usually have a cleaner design, higher end components, and are engineered to do one thing only. You get the freedom of using any headphone, regardless of impedance, and you'll get (in theory) the same performance levels with all of them. Will they sound better than integrated amps? Depends on which integrated amp, which headphone amp and which headphone we're talking about. I'm sure you're aware that "better" in audio is a very broad term. I'd use the word "different" in this case. But certainly not bad when it comes to integrated amps, and integrated amps usually offer far better value for money. I've heard some really bad sounding headphone amps full of problems like hissing, imbalance, pops / clicks, heating, thin sound, lack of power, etc.  I'm yet to hear an integrated amp that didn't get those essentials right.  Even my old Onkyo A-8820 (made in Japan around year 2000) that I got used for like 40€ sounds freaking awesome with T1's and speakers and earned a permanent spot in my room, and honestly makes any dedicated headphone amp under 300 dollars I've ever heard seem like a bad joke in comparison both in terms of build and sound.  And just look how pretty it is, it's like a match made in heaven for T1's.  (yes, this a metal front panel, with metal buttons and knobs, volume knob moves with smoothness unmatched even by my Musical Fidelity M1HPAP, and that source selector has some of the most satisfying "clicks" I've felt in any audio gear, some of us are into that...

)
 

 

[pp312]

What he said.

 

[pp312]

 
The only receivers I've heard were Marantz and Denon, all that I've heard from Onkyo were integrated amps.  As I said, in general integrated amps sounded better than receivers, obviously, but none of them sounded bad.

 
You know, I'm not even sure that's true. The theory is that a) the tuner will take power from the power supply, but if you look inside you find the tuner is powered by its own mini power supply with separate transformer; and b) the tuner components will somehow contaminate the amplifying side, but again, if you look inside a good receiver you'll find more than enough shielding to prevent that happening. I've had as good results from receivers as integrateds.  

 

[Sonic Defender]

   
 
All you need to consider is the fact that people who design these amps are not some random guys, they know what they're doing, they're engineers and audiophiles, people who love good audio as much as you and me, on top of that, we're talking about big companies with decades of experience, and a lot of people, money and knowledge that allows them to do far more research and development than say a small company that makes a couple of headphone amplifiers.
 
If they can make the amps sound great with speakers, what makes you think that they couldn't do the same thing when it comes to headphones?  People on head-fi too often seem to act as if headphones are some magical, special, super-complicated devices from the future that take extra special care when it comes to powering them, which is just not the case if you get the basics right, which good sounding integrated amps do. Again, a good sounding speaker amp will sound good with headphones as well, unless they screw up the headphone output somehow, which again, is not the case with any of the amps I've ever tried, for the reasons explained in the first sentence of this post.
 
 
What gets most people confused in my opinion are prices. How could a 500 dollar, huge, 15kg heavy integrated amplifier, with lots of functions and purposes, full of circuits and components, have the same quality of components as a single-purpose small 1000 dollar headphone amp that only has a handful of components inside and is only designed to do one thing? Surely the quality of the components and quality of engineering in integrated amps must be horrible in comparison. And the headphone output must be just an afterthought. Right?
 
You'll find that if you take almost any headphone amplifier on the market (apart from some really cheap ones like O2, Fiio's, etc), especially as you go higher up in the price range, that the price of their electronic components (if you were to buy them separately, especially in large quantities) is a small fraction of what you actually pay for when buying the product, and in almost all cases, the housing of the amp costs close to or even more  than the electronic components inside. Yes, usually headphone amps are simpler in design, which in theory means that the sound will be less compromised and the signal will have to travel a simpler and shorter distance from the point where it enters the amplifier to the point where it exists. And yes, dedicated headphone amplifiers usually have top of the line passive components on their PCB boards, while entry level integrated amps don't, but do you really think you'll hear a difference between an amp that uses top of the line Nichicon caps instead of top of the line Panasonic caps? You'll probably measure minimal differences if you use the exact same design with different components, but will you hear a difference? That is a big question. And all the complicated stuff like DAC's or pretty much all chips on headphone amps, as well as connectors, transformers, volume pots, etc. are made by the same manufacturers that make those same components for integrated amplifiers produced by dozens of manufacturers from all over the world.   As I said again in my original post, what all the integrated amps I've tried had in common was that they got the essentials right...like...no audible noise at any volume, no channel imbalance at any volume, flat frequency response and proper extension in both directions, transparent sound with no obvious coloration or changes in the characteristic sound of a certain headphone, black background, plenty of power, lots of headroom when it comes to volume leading to great dynamics and lively sound with a lot of authority and no hint of distortion, etc.  And this cannot be said for all entry level headphone amps.
 
Is a dedicated headphone amp better for headphones? Yes. Compared to equally priced integrated amps, headphone amps usually they have a cleaner design, higher end components, and are engineered to do one thing only. You get the freedom of using any headphone, regardless of impedance, and you'll get (in theory) the same performance levels with all of them. Will they sound better than integrated amps? Depends on which integrated amp, which headphone amp and which headphone we're talking about. I'm sure you're aware that "better" in audio is a very broad term. I'd use the word "different" in this case. But certainly not bad when it comes to integrated amps. I've heard some really bad sounding headphone amps full of problems like hissing, imbalance, pops / clicks, heating, thin sound, lack of power, etc.  I'm yet to hear an integrated amp that didn't get those essentials right.  Even my old Onkyo A-8820 (made in Japan around year 2000) that I got used for like 40€ sounds freaking awesome with T1's and speakers and earned a permanent spot in my room, and honestly makes any dedicated headphone amp under 300 dollars I've ever heard seem like a bad joke in comparison both in terms of build and sound.  And just look how pretty it is, it's like a match made in heaven for T1's.  (yes, this a metal front panel, with metal buttons and knobs, volume knob moves with smoothness unmatched even by my Musical Fidelity M1HPAP, and that source selector has some of the most satisfying "clicks" I've felt in any audio gear, some of us are into that...

)
 

I guess you can't call it vintage gear, but that is a nice looking piece of kit. Onkyo has always offered some really good gear throughout the year. I certainly agree with much of the substance of your post, those points are obviously quite hard to not accept. I do think the thin edge of the wedge is where we decide what is good enough, and what is good? That is of course subjective and I can't say I have any answers that generalize beyond myself. I guess it depends if one feels a bunch of resistors in a circuit path make any audible difference or not (which is how most headphone jacks are implemented as an afterthought).
 
I remain undecided for various reasons. I would like to say that while I can agree that designers and manufactures in general try to make good gear, that in itself does not guarantee that compromises aren't made in the gear itself to serve the needs of the production versus design goals. I guess we would need to look carefully at the consumer marketplace when the gear was mass produced. If in the 80s for instance, there were few consumers who used headphones, or very few headphone varieties to contend with. it is conceivable that the headphone jack was often just a throw in feature. Then we need to look at the importance the headphone section may or may not have had to the designers/manufacturers. If they were themselves avid users of headphones, perhaps even in the absence of market factors supporting quality headphone sections they might still have opted to put time into the feature.
 
Then we can get right into that circular debate about what is required to produce good sound. Some tell you that the minimal specifications being met with the minimally competent components should yield no sonic compromises when compared with over-built gear under the same operating conditions. So many ways to look at things, so many opinions. Even given that objective measurement is crucial and compelling, are we able to accept that objective measurement is sensitive enough to accurately detect what the brain detects? How do we know that? Perhaps, but perhaps not. In 50 years we may learn more about how the brain hears and realize that there is more than we were able to measure with the devices commonly deployed. Frankly, we understand very little about much of the brain so it is reasonable to suggest there may be some secrets yet to unearth.  

 

[pp312]

 
I remain undecided for various reasons. I would like to say that while I can agree that designers and manufactures in general try to make good gear, that in itself does not guarantee that compromises aren't made in the gear itself to serve the needs of the production versus design goals. I guess we would need to look carefully at the consumer marketplace when the gear was mass produced. If in the 80s for instance, there were few consumers who used headphones, or very few headphone varieties to contend with. it is conceivable that the headphone jack was often just a throw in feature. Then we need to look at the importance the headphone section may or may not have had to the designers/manufacturers. If they were themselves avid users of headphones, perhaps even in the absence of market factors supporting quality headphone sections they might still have opted to put time into the feature.
 

 
This is a very nice post, but there are a couple of things I just can't let pass. The first popular consumer headphone was the Sennheiser HD414 in 1968. By the 80s the headphone market was booming, with just about every major manufacturer having several models and interlopers like Stax grabbing the top end with their electrostatics. There was no 'afterthought' about headphones or headphone jacks. However, all this is moot if you simply accept that right from the beginning the simplest and probably best way to implement a headphone out is to take a line from the main amp through resistors. Do that and you don't need to consider how much thought manufacturers put into it, whether they were avid headphone users, whether it was 'a thrown in feature' or anything else. All that only applies if you believe headphone jacks conceal op amp circuits of individual and eccentric design. By far the majority didn't and don't; they take a line from the main amp. Consequently the sound you get is the sound of the amp as a whole, excluding the rare and probably inaudible impedance mismatch. I've looked inside scores of amps and I can confirm what the manufacturers have told us--that in general they do not use separate headphone amps. Why would they, when it's so much cheaper to use resistors?
 
So for me the path to headphone bliss is simple: audition and choose a 250--300 ohm model, of which there are dozens, select a good sounding integrated or receiver and decent source, and settle back.

 

[derbigpr]

   
This is a very nice post, but there are a couple of things I just can't let pass. The first popular consumer headphone was the Sennheiser HD414 in 1968. By the 80s the headphone market was booming, with just about every major manufacturer having several models and interlopers like Stax grabbing the top end with their electrostatics. There was no 'afterthought' about headphones or headphone jacks. However, all this is moot if you simply accept that right from the beginning the simplest and probably best way to implement a headphone out is to take a line from the main amp through resistors. Do that and you don't need to consider how much thought manufacturers put into it, whether they were avid headphone users, whether it was 'a thrown in feature' or anything else. All that only applies if you believe headphone jacks conceal op amp circuits of individual and eccentric design. By far the majority didn't and don't; they take a line from the main amp. Consequently the sound you get is the sound of the amp as a whole, excluding the rare and probably inaudible impedance mismatch. I've looked inside scores of amps and I can confirm what the manufacturers have told us--that in general they do not use separate headphone amps. Why would they, when it's so much cheaper to use resistors?
 
So for me the path to headphone bliss is simple: audition and choose a 250--300 ohm model, of which there are dozens, select a good sounding integrated or receiver and decent source, and settle back.

 
 
True, I don't really see the big deal with resistors, it's really simple, just basic physics. If implemented properly they won't do anything to the sound apart from making it less loud, that's pretty much it.  It's like taking a speaker output and instead of plugging headphones straight into it and then having to work with first first few degrees of rotation on the volume knob because anything above that would put too much power into headphones, they put resistors into that path to make the volume adjustment easier.
 
 
As you said, the best thing is to get high impedance headphones (unless you intend to use them for portable use without an amp) and just have the freedom of being able to plug them anything anything with enough power output.
 
The best thing is, high impedance models ARE usually the best headphones you can get at their price range. Not because they're high impedance, but simply because they happen to be the best sounding headphones in their classes.  If the world of headphones only had 250-600 Ohm Beyerdynamics and 300 Ohm Sennheisers to offer, we could still say we're doing alright when it comes to headphones for any type of use.   You want closed back studio monitoring headphones?  Get DT770's, DT150's, DT100's, DT250's or T70's.  You want neutral headphones for mixing? Get DT880's or HD600's.  You want headphones for critical listening? Get HD800's, T1's or T90's. You want them bass heavy? Get DT770's, DT150's, DT990 Premiums or HD650's.  You want them bright? Get DT990 Pro's, T70, T90's or HD800's.  You want them warm and enjoyable?  Get HD650's.   You want the ultimate dynamic audiophile headphones? Get HD800's, T1's or T90's.  Pretty simple.

 

[nicholars]

The people building them may be audiophiles etc. but at the end of the day they are a business trying to make money so they have to stick to a price point.

 

[Sonic Defender]

I'm not willing to stipulate that resistors in a signal path have no effect. That may be, but until I have time to research it further I will simply say that I am not convinced in either direction. I was a teenager in the 80s, and headphones were very rare in my experience. Now I am willing to say that might simply have been in my circles, but I don't remember much advertising for headphones Tons of speaker ads. I think that if they were in full swing (and I have no reason to doubt this), they were still a small percentage of the consumer spend on audio and further I suspect there were vey few choices around. I still believe that the headphone sections were not particularly important to companies, but as you have intonated, being cheap to slap some resistors behind a jack, it wasn't a big issue to provide the headphone section.
 
Ultimately I think there will be a variety of results depending on the gear being used, some good, some bad. I have not done any testing myself recently although I should as I have an old Pioneer SX 780 receiver upstairs that I drum to and it has had all the caps redone recently so it should sound good. I do drum with the headphones on coming from the Pioneer, but that is hardly a fair trial so I will have to do better than that. It would be really interesting to have some blind testing done to help add some extra information to the discussion. Anyway, I will say that if resistors in the signal path such as we are discussing have any effect, it is likely to be small.

 

[Mr Rick]

  I'm not willing to stipulate that resistors in a signal path have no effect. That may be, but until I have time to research it further I will simply say that I am not convinced in either direction. I was a teenager in the 80s, and headphones were very rare in my experience. Now I am willing to say that might simply have been in my circles, but I don't remember much advertising for headphones Tons of speaker ads. I think that if they were in full swing (and I have no reason to doubt this), they were still a small percentage of the consumer spend on audio and further I suspect there were vey few choices around. I still believe that the headphone sections were not particularly important to companies, but as you have intonated, being cheap to slap some resistors behind a jack, it wasn't a big issue to provide the headphone section.
 
Ultimately I think there will be a variety of results depending on the gear being used, some good, some bad. I have not done any testing myself recently although I should as I have an old Pioneer SX 780 receiver upstairs that I drum to and it has had all the caps redone recently so it should sound good. I do drum with the headphones on coming from the Pioneer, but that is hardly a fair trial so I will have to do better than that. It would be really interesting to have some blind testing done to help add some extra information to the discussion. Anyway, I will say that if resistors in the signal path such as we are discussing have any effect, it is likely to be small.

You are a tough nut to crack my friend. LOL

 

[derbigpr]

  I'm not willing to stipulate that resistors in a signal path have no effect. That may be, but until I have time to research it further I will simply say that I am not convinced in either direction.

 
 
What if I told you that dedicated headphone amplifiers have resistors in the signal path as well? Or that the volume pots themselves are actually resistors?

 

[Sonic Defender]

   
 
What if I told you that dedicated headphone amplifiers have resistors in the signal path as well? Or that the volume pots themselves are actually resistors?

I would say that they as well might have some effect on the sound. Again, is it an audible effect? That is what I wonder. Plus my real point is whether the amount of resistance has a sum effect. I can have two drinks and feel no effect, give me six and that is a different story. Might there be such an effect with resistance? I just wish I had the time to really start studying the topic, but I'm virtually a single father, I work full-time and I go to university part-time so the little free time I have I find I like to kick back here and just read and chat lighter stuff. It is clear I need to spend some time on this topic if I am going to be able to add anything to the conversation beyond pointed questions.

 

[pp312]

  The people building them may be audiophiles etc. but at the end of the day they are a business trying to make money so they have to stick to a price point.

 
This was actually discussed in an earlier thread on this subject and it was pointed out by one poster, correctly I think, that large speaker amp manufacturers have much better scale of economy than small dedicated HP amp manufacturers. In short, Marantz can make a 100W integrated just as cheaply as your average dedicated amp manufacturer can make a 5W HP amp, with just as good components.
 

  I'm not willing to stipulate that resistors in a signal path have no effect. That may be, but until I have time to research it further I will simply say that I am not convinced in either direction. I was a teenager in the 80s, and headphones were very rare in my experience. Now I am willing to say that might simply have been in my circles, but I don't remember much advertising for headphones Tons of speaker ads. I think that if they were in full swing (and I have no reason to doubt this), they were still a small percentage of the consumer spend on audio and further I suspect there were vey few choices around. I still believe that the headphone sections were not particularly important to companies, but as you have intonated, being cheap to slap some resistors behind a jack, it wasn't a big issue to provide the headphone section.
 

 
I was not just a consumer in the 70s & 80s but an occasional contributor to a Hi-Fi mag, and I can assure you that in your average Hi-Fi store there was considerable interest in headphones. In the early 70s I had a Pioneer SE140 (still sought after), later replaced with a Pioneer piezoelectric SE700. Then I went over to Stax and got a Sigma (virtually an electrostatic driver in a small speaker box) and later a Lambda (not Pro). Virtually everyone had at least a Sennheiser HD40, a great little phone. I can't imagine what you and your friends were doing to miss all the headphone fun. A little too much vino perhaps?

    

 

[Sonic Defender]

   
This was actually discussed in an earlier thread on this subject and it was pointed out by one poster, correctly I think, that large speaker amp manufacturers have much better scale of economy than small dedicated HP amp manufacturers. In short, Marantz can make a 100W integrated just as cheaply as your average dedicated amp manufacturer can make a 5W HP amp, with just as good components.
 
 
I was not just a consumer in the 70s & 80s but an occasional contributor to a Hi-Fi mag, and I can assure you that in your average Hi-Fi store there was considerable interest in headphones. In the early 70s I had a Pioneer SE140 (still sought after), later replaced with a Pioneer piezoelectric SE700. Then I went over to Stax and got a Sigma (virtually an electrostatic driver in a small speaker box) and later a Lambda (not Pro). Virtually everyone had at least a Sennheiser HD40, a great little phone. I can't imagine what you and your friends were doing to miss all the headphone fun. A little too much vino perhaps?

    

Sounds like you have had a great deal of audio fun my friend. For the record, we had a set of Sony headphones, not sure what model, but they wouldn't have been very expensive. I spent hours upon hours with those headphones. Makes me wonder whatever happened to them?

 

[nicholars]

   
This was actually discussed in an earlier thread on this subject and it was pointed out by one poster, correctly I think, that large speaker amp manufacturers have much better scale of economy than small dedicated HP amp manufacturers. In short, Marantz can make a 100W integrated just as cheaply as your average dedicated amp manufacturer can make a 5W HP amp, with just as good components.

 
Ok then you are definitely right and infact integrated amps are actually better for headphones than headphone amps

 

 

[Mr Rick]

   
Ok then you are definitely right and infact integrated amps are actually better for headphones than headphone amps

 

I don't think you can make such a blanket statement with any certainty. Too many variables involved.