[Gurra1980]

Quote:

Originally Posted by HighLife /img/forum/go_quote.gif I would think thats cause of the ohm of the 650-600? (i believe the 580-600-650 all 300ohm?) I find the same with my 580s. I need more volume to match the 701s at normal listening volumes. However, the 701s sound much cleaner at lower volumes then the 580s do. It almost seems to me that my 580s lose alot of detail at lower volumes. Just my opinion....take it with a grain of salt.

Agreed, the HD650 need some volume to be at there best, Beyer dt990 was good at lower volumes because of the bass and treble exaggeration.

 

[D2K700]

Gosh, this discussion has become very complicated! To get back to the original question - I'm listening to the cowon d2 at the moment on a pair of senn hd-555. Not the best headphones, but i just had to turn down the volume a moment ago to 29. the volume at 44 becomes very uncomfortably loud. And as you might know, 50 is max. so the D2 seems to have no problem driving these headphones. Am i missing something? these are, after all, 120 ohms.
anyway, my k701s should arrive tomorrow and i'm feeling confident that the D2 should do magic with them if it makes these very plain headphones sound good!

 

[JaZZ]

Quote:

Originally Posted by D2K700 /img/forum/go_quote.gif ...my k701s should arrive tomorrow and i'm feeling confident that the D2 should do magic with them if it makes these very plain headphones sound good!

But don't expect too much from a fresh pair before the 300-hour mark. Anyway, finally someone willing to try this pairing!

.

 

[HardHeadCase]

I think its because the headphones they give you are 16 ohm. Not 62 ohm. Doesn't that translate into 5 or 10 milliwatts? That is what you guys are saying right, 5 or 10 milliwatts or whatever it translates to is enough to drive the 701s.

 

[JaZZ]

Quote:

Originally Posted by HardHeadCase /img/forum/go_quote.gif I think its because the headphones they give you are 16 ohm. Not 62 ohm. Doesn't that translate into 5 or 10 milliwatts? That is what you guys are saying right, 5 or 10 milliwatts or whatever it translates to is enough to drive the 701s.

I just had a look at Cowon's specs: They define the output power with respect to a 16-ohm load -- that's all. No recommendation.
.

 

[majkel]

Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif Your list of advantages from an external amp has some merits (the bigger buffer capacitors being the most glaring advantage IMO). But you still seem to forget that you deal with an additional amplification stage which may actually be dispensable in many cases, since the built-in headphone amp is meant to drive headphones. And the theoretical advantages of the external-amp concept may in fact fall short in many cases since the amp is fed with the (supposedly) poor signal from the (supposedly) poor internal amp and moreover is the source of additional signal degradation..

No, the problem is that you most probably don't understand what is the advantage of unloading the headphone out of a DAP and moving the load to the output of the amp. Each output stage has got its I/V characteristic which is never totally linear, however, if you reduce the current swing, the voltage changes in a more linear manner. The higher the current (lower impedance of headphones), the less linear are the voltage changes. Headphone amp is generally a voltage amp, so you listen to voltage changes mostly. DAP headphone out is capable, say, of 50mA max., while the amp's output might deliver 250mA output (when BUF634's are on board). So, to make it easier, at 50mA you are using 100% of the DAP's output I/V characteristic, and 20% of the amp's capabilities. While 20% of maximum current delivers just slight distortions, 100% gives huge distortion level. Moving to 10% against 50% is still significant difference, etc.

 

[JaZZ]

Quote:

Originally Posted by majkel /img/forum/go_quote.gif No, the problem is that you most probably don't understand what is the advantage of unloading the headphone out of a DAP and moving the load to the output of the amp. Each output stage has got its I/V characteristic which is never totally linear, however, if you reduce the current swing, the voltage changes in a more linear manner. The higher the current (lower impedance of headphones), the less linear are the voltage changes. Headphone amp is generally a voltage amp, so you listen to voltage changes mostly. DAP headphone out is capable, say, of 50mA max., while the amp's output might deliver 250mA output (when BUF634's are on board). So, to make it easier, at 50mA you are using 100% of the DAP's output I/V characteristic, and 20% of the amp's capabilities. While 20% of maximum current delivers just slight distortions, 100% gives huge distortion level. Moving to 10% against 50% is still significant difference, etc.

Nothing against your arguments, but the call still stands that an additional amplification stage introduces additional signal degradation due to its inherent colorations -- which may absolutely more than make up for any theoretical advantages of a «better» amp. Moreover the technically superior amp may in fact be the sonically worse. And as the example with the excellent Corda Opera as external amp shows, the direct connection will often sound better despite reasonings such as yours above.

BTW, the «load» is by no means removed from the headphone out if you connect an external amp to it: you just replace a complex low impedance with a less complex high impedance. It's not clear that the DAP's output will fare any better under this condition. And higher power reserves don't automatically mean lower distortion, just as little as higher power means higher distortion in every case. As long as the amp operates in a range clearly below its upper limit, distortion can absolutely play an inferior role anyway.

Note that I never pretended that an external amp can't sound better, but it is by no means written in stone that it has to. Again: additional amplification means additional signal degradation. You seem to highly underrate the coloration every amp suffers from more or less -- even high-end amps. Which means with every additional stage you sacrifice transparency. The fact that some amps may drive a sound transducer better doesn't change much, although the loss may be compensated in one or the other criterion. Add to this that some sorts of coloration sound pleasing to the ears and may be perceived as higher sound quality.
.

 

[nightfire]

Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif And remember: D2 and i7 don't have line outs. Amplifying the signal from the headphone out can't make it any more accurate, in fact it can only make it more inaccurate. The only thing an amp could cure is the obligatory bass drop-off at low impedances shown by the majority of DAPs.

I didn't read the rest of the thread, so sorry if someone .. addressed this before.

This is absolutely not true. Distortion of an amplified signal is usually proportional to the load imposed on it.

The output of a DAP might be very low distortion (across the FR spectrum) when connected to an efficient, high impedance headphone, and yet horribly distorted, at the same volume, driving (to use the extreme case) a standard 8ohm speaker.

Chaining amplifier stages DOES contribute to overall noise and distortion when no significant load is applied, but this figure can be - and often is - lower than what you would achieve with a single stage.

Having said all that, I'm sure the K701 sounds OK out of a powerful, well designed DAP such as the D2, and the sky is the limit from there.

Quote:

BTW, the «load» is by no means removed from the headphone out if you connect an external amp to it: you just replace a complex low impedance with a less complex high impedance. It's not clear that the DAP's output will fare any better under this condition.

Ok first of all, every amp relieves the load of the speaker/antenna from the source. That's the whole point.

Secondly, the base/gate of a transistor or mosfet is not significantly inductive. There are no coils used on the input of an amplifier. If you want to get fancy and start talking about negative feedback, and base->emitter/base->collector loads, go ahead, but it's not the same thing and the numbers are far smaller.

Also, the input to a mosfet (assuming a mosfet amp) is not coupled to the output in any way. No current flows from the gate, and so the output load cannot influence the input signal (again, ignoring negative feedback).

 

[majkel]

nightfire, thanks for your input, I didn't want to go so detailed but hope it helped.
As simple as possible: Headphones are roughly 13~600 ohms, amp input is > 5000 ohms which makes DAPs output doesn't distort as much at 64 ohm level, and the amp does not distort as much at 64 ohms of load as the DAP because of better, bigger transistors which due to their size have the linear region of the characteristic reaching much lower impedances then the DAP. This is mostly due to much higher saturation current of the amp's output which produces the most nasty IMD and TIM distortions.
So, the assumption is that loaded DAP output distorts stronger than unloaded DAP output + amp. Now it's just a matter if somebody believes in physical facts, or not.

 

[JaZZ]

O.k -- I believe you. That doesn't mean your distortion scenario is of importance in every case, though. As long as the amp (the headphone out) is driven clearly below its upper limit, distortion may in fact be low enough, so there's no advantage from an additional amplification. Hence the double-amplification scenario takes effect, where no distortion reduction compensates for the loss of transparency.

Once I switched two high-class amps (HA-2 MkII and EMP) in series to hear how and how much the «monitor» amp colors the input signal (I remember the EMP being the reference and the HA-2 being the monitor): Double amplification made the sound much denser and less transparent and obviously heavily colored.
.

 

[glct222]

no source got enough power which can dirve K701s without amp

 

[JaZZ]

Another experiment (one of my key audio experiences):

You'd think that the line out of a DAC is not an ideal amp for a headphone; it's designed to «drive» line-in impedances of 50 kOhm and the like, not complex loads such as 62 or 300 ohm. I tried it nonetheless -- the Bel Canto DAC2 with its just 12 ohm output impedance was an ideal candidate. I used a 500-ohm potentiometer for volume attenuation, switched between DAC2 line out and HD 650. The signal was by far loud enough even for very low passages in the recordings, and the quality of sound was excellent: extremely transparent and pure, with smooth and sparkling treble and full, deep and impactful bass. The only negative point was a slight coolishness, a lack of organicalness. Switching a headphone amp into the signal path (HA-2 or EMP) «cured» exactly the latter issue, but apart from this positive aspect all in all sound quality was noticeably lower: less transparent, less clean, less bass extension and control, less high-frequency extension and less smoothness throughout the sonic spectrum.

BTW, I also tested the impact of the low impedance on the DAC2's output signal by using a headphone amp as monitor, while at the same time plugging in and unplugging the headphone: there was no audible change at all. Furthermore I tested the impact of the potentiometer, using low-level passages, which enabled to remove it from the signal path: Again, no or minimal impact, as far as I could tell (naturally this configration makes A/B testing impossible).

You'd think that a pure line out is a pretty bad headphone amp: It must produce high harmonic distortion on such low impedances which it isn't designed for. But that's not the case, at least not audibly so. I later repeated the experiment with the K 701 -- with the same result. I even did it on my UDP-1's 200-ohm line out. The K 701 was too low a load: not enough current to produce enough volume save for a few loud recordings, but the HD 650 sounded pretty good, even better than on the DAC2 (because of the UDP-1's better sounding DAC), although the volume level was more limited, and an additional parallel resistance served to make the bass a tad leaner, to make the direct/indirect comparison more fair.

I have only encountered three headphone amps which have managed to passably reach the transparency and transient response of the «direct connection»: DynaMight, HeadCode DM, Corda Opera. But when it comes to coloration, each of them has a serious amount of it: The DynaMight has a cold midrange, the HeadCode sounds too warm and euphonic, with a sweet treble not integrated into the rest of the spectrum, and the Opera has a slight tendency to edginess. These are the highest-resolution amps I've heard, and still they don't exactly reach the transparency, the ease and the accuracy of the direct connection. When you think about it: It's even a bit unfair to compare two balanced-drive amps and one balanced-ground amp to a single-ended line out...
.

 

[Gurra1980]

Majkel and Nightfire you seems to know what you are talking about, I did some testing with my iPod classic HP-out through my x-can v3 and the sound got a lot better, could not hear any distortion at all, but when I tested with my zen the sound got worse, one word to describe it with is "smeard" Tested with px-100 32 Ohm and D2000 28 Ohm could it be that the zen is made to match lower ohm headphones and not would do very well with higher ones like 300Ohm and that iPod is the other way around?

 

[JaZZ]

Quote:

Originally Posted by Gurra1980 /img/forum/go_quote.gif Majkel and Nightfire you seems to know what you are talking about, I did some testing with my iPod classic HP-out through my x-can v3 and the sound got a lot better, could not hear any distortion at all, but when I tested with my zen the sound got worse, one word to describe it with is "smeard" Tested with px-100 32 Ohm and D2000 28 Ohm could it be that the zen is made to match lower ohm headphones and not would do very well with higher ones like 300Ohm and that iPod is the other way around?

The iPod's headphone out shows a severe bass drop-off with 28-32 ohm headphones, so the amplification stage serves for curing this issue. With the Zen, there should be a similar issue, but maybe it's less pronounced, however, this configuration seems to make the signal degradation by double amplification come to effect.
.

 

[majkel]

JaZZ, it looks like the DAC has got a decent, current efficient output. Some good high current op-amps don't distort significantly even at the 100 ohm of load. I put some AD8022's in the output stage of my CDP, in the datasheet you can see the distortion curves for low impedance loads. It's actually a very good chip for portable amps - not very powerful but not as weak as most regular chips recommended for the Cmoy, either.
Regarding your experiments with external amps and the DAP - well, you just heard what your amp does to the sound.

I don't recommend to anybody hooking up the DAP to the amp they don't like or hardly find improving overall sonic impression. This is actually what I do when I want to hear what my another DIY amp deviation makes to the sound - I set the volume on the DAP as for the headphones, then set the amp's gain to around 1 and compare the results while switching the cables rapidly. Of course, there is some interconnect influence as well, which some people will find as another disadvantage (or advantage, for those satisifed with its sound signature).