@rob,yes exactly. this post is just what i have been trying to emphasize in the recent past. one need only two ears to get all the depth perception in the real life scenarios . similarly if properly recorded you need only a stereo mic to portray all the directional cues in the recordings. after all a binaural recording uses only two mics, even if it is a special case. dave with its accurate depth perception portrays those cues in a much more accurate way.
So the implication appears to be that DAVE feeding headphones is more distorted than DAVE + amplifier, because the latter has a wider soundstage. It seems that soundstage width can be used as a measure of transparency/distortion.
This is directly against what Rob has been saying, and against all logic. You simply can not add another piece of equipment to something and take away distortion, you can only add to it.
An analogue amplifier has more distortion the louder the output signal.
An analogue amplifier has more distortion driving a load that demands more power.
So, it is possible that DAVE has more distortion driving headphones than an amplifier that is added to DAVE. In the latter scenario, DAVE's distortion is the minimum it can be. In this scenario the amplifier might have less distortion driving those headphones, than if DAVE were to drive those headphones directly.
I'm not saying the add on amplifier will always be less distorted. Merely that a headphone that is hard to drive (defined as a headphone that requires high volume setting and/or a headphone that requires more power) might cause DAVE to be more distorted than if an amplifier was used instead.
DAVE's distortion changes just like any other amplifier depending on how it's used and what it's used with.
Jawed, I think «THD and noise at 2.5 volts RMS 0.000015%» is hard or impossible to beat by any amp, although it's not clear if and to which degree a complex load would raise these figures. But even if we implicate a factor of 10, it's still virtually unachievable.
Loading Dave at 2.5 V RMS with 33 ohms barely makes any difference at all. THD and noise goes from -127 dB to -124 dB, and the only change is an increase in second harmonic. No other amp on the planet is so load intolerant, and it is due to Dave's unique second order noise shaping analogue amplifier - among other things.
The issue of perceived width is complicated. When depth is degraded, and adding an extra stage is guaranteed to degrade depth - you invariably perceive a wider sound-stage. So a increased width is almost always an indication of poorer transparency. The analogy is like a picture that is out of focus - things look bigger, and an out of focus sound-stage will lack depth but appear wider. But increasing width is not always an indication of poor transparency - there are exceptional circumstances when you can perceive out of the box (that is outside of the speakers - to one's side, or behind you) sounds and this is due to encoding of certain resonances, phasing and complex frequency response. Ever wondered how it is with just two ears you can perceive sounds behind you and above? If you cut off your ear lobes, (do not try this at home!) you lose the ability to perceive rearward sounds and height. Its due to the small reflections and resonances in the ear lobe that gives one the perception of height and rear. Now if the brain can accurately separate the sounds of instruments out, and the timing relationship of transients is accurately maintained, then you can better hear the out of the box effects that are very occasionally encoded in multi track music. So in the instance when out of the box encoding has been put in, then a better DAC will give better width, height and rear perception. If you listen to classical music, this is never encoded, so an increase in width using classical music is always an indication of poorer transparency. So in 99 times out of 100, a report of better width is just due to poorer sound-staging and worse transparency.
Thanks, Rob! Now what's going on – metrologically – when a hard to drive headphone like the very insensitive Abyss is perceived as slightly bass-shy driven by the DAVE, whereas other, stronger amps produce a very strong bass (and the measurements indicate that this is closer to the truth)? Some people consider the DAVE very capable with easy loads such as a Focal Utopia or Sennheiser HD 800 and superior to attached headphone amps, but with relatively insensitive headphones like an Audeze LCD-4 they are undecided which configuration be better, or they even lean towards the amp. I take it that in these cases harmonic distortion is still very low (also since the +4 dB limit is respected) – so what other factor is in play here, electrically speaking?
So the implication appears to be that DAVE feeding headphones is more distorted than DAVE + amplifier, because the latter has a wider soundstage. It seems that soundstage width can be used as a measure of transparency/distortion.
This is directly against what Rob has been saying, and against all logic. You simply can not add another piece of equipment to something and take away distortion, you can only add to it.
An analogue amplifier has more distortion the louder the output signal.
An analogue amplifier has more distortion driving a load that demands more power.
So, it is possible that DAVE has more distortion driving headphones than an amplifier that is added to DAVE. In the latter scenario, DAVE's distortion is the minimum it can be. In this scenario the amplifier might have less distortion driving those headphones, than if DAVE were to drive those headphones directly.
I'm not saying the add on amplifier will always be less distorted. Merely that a headphone that is hard to drive (defined as a headphone that requires high volume setting and/or a headphone that requires more power) might cause DAVE to be more distorted than if an amplifier was used instead.
DAVE's distortion changes just like any other amplifier depending on how it's used and what it's used with.
Jawed, I think «THD and noise at 2.5 volts RMS 0.000015%» is hard or impossible to beat by any amp, although it's not clear if and to which degree a complex load would raise these figures. But even if we implicate a factor of 10, it's still virtually unachievable.
Loading Dave at 2.5 V RMS with 33 ohms barely makes any difference at all. THD and noise goes from -127 dB to -124 dB, and the only change is an increase in second harmonic. No other amp on the planet is so load intolerant, and it is due to Dave's unique second order noise shaping analogue amplifier - among other things.
The issue of perceived width is complicated. When depth is degraded, and adding an extra stage is guaranteed to degrade depth - you invariably perceive a wider sound-stage. So a increased width is almost always an indication of poorer transparency. The analogy is like a picture that is out of focus - things look bigger, and an out of focus sound-stage will lack depth but appear wider. But increasing width is not always an indication of poor transparency - there are exceptional circumstances when you can perceive out of the box (that is outside of the speakers - to one's side, or behind you) sounds and this is due to encoding of certain resonances, phasing and complex frequency response. Ever wondered how it is with just two ears you can perceive sounds behind you and above? If you cut off your ear lobes, (do not try this at home!) you lose the ability to perceive rearward sounds and height. Its due to the small reflections and resonances in the ear lobe that gives one the perception of height and rear. Now if the brain can accurately separate the sounds of instruments out, and the timing relationship of transients is accurately maintained, then you can better hear the out of the box effects that are very occasionally encoded in multi track music. So in the instance when out of the box encoding has been put in, then a better DAC will give better width, height and rear perception. If you listen to classical music, this is never encoded, so an increase in width using classical music is always an indication of poorer transparency. So in 99 times out of 100, a report of better width is just due to poorer sound-staging and worse transparency.
Thanks, Rob! Now what is happening – metrologically – if a hard to drive headphone like the very insensitive Abyss is considered slightly bass-shy driven by the DAVE, whereas other, stronger amps produce a very strong bass, and the measurements indicate that this is closer to the truth? Some people consider the DAVE to be very capable with easy loads such as a Focal Utopia or Sennheiser HD 800 and superior to attached headphone amps, but with relatively insensitive headphones like an Audeze LCD-4 they are undecided which configuration be better, or they even lean towards the amp. I take it that in these cases harmonic distortion is still very low (also since the +4 dB limit is respected) – so what other factor is in play here, electrically speaking?
The abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
The abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
Simply FANTASTIC, I did not have found any problem to drive the Abyss with Dave alone...
No problems with bass distortion at all... The lowest volume recording I have is a Classical Brahms's Requiem Mass (Pentatone classic) that I listen with +4 or less ...
Thanks!
The abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
( Just for your info, the Abyss got the greatest dynamic Bass / low end on the market today in the Ultra High End range, so it is not a numb headphone, so a slight adjustment in the low end for the less sensitive headphones should have bin great as an option in a future upgrade ( i know it will not be totally transparent then, but in some recordings it is needed)
This feature should be great for the ones who are not using a PC/MAC + EQ as a source )
Then if you ever come across the Abyss i recommend a serious listening with DAVE, the synergy is immense ( besides the low end energy) =)
Yes, link works for me too. An interesting shoot-out.
I assume you listened with stock cables. I have both the Utopia and a microRendu/LPS-1. I think the former needs a better cable (I will soon be able to use both Prion and Black Dragon cable with it) and I am getting a SOtM sMS-200, which I expect to be better than the mR, which I feel is a little too hard sounding at times. I've only listened briefly to the HE1000 v2 at my dealer while he took care of something for me and I think my thoughts at the time were close to your assessment. I would need to listen to them again in my system to be sure.
The abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
This exactly mirrors my listening observation with the HugoTT versus using a separate amp with it. With the external amp (Cavalli Liquid Carbon), the bass seems more but is sorta one note thump. Direct out of the TT, the bass is more detailed and nuanced.
So The DAVE I believe outputs even more power directly than the HugoTT. I cannot see it not having enough power for virtually any headphones.
The Abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
That's good to hear! Something that still is in the back of my mind, although I'm fundamentally an «amps are overrated» advocate: What's exactly the benefit from extra-strong or -fast power supplies in audiophile-grade amps – again electrically speaking –, especially considering that the ultra caps in the Hugo TT do have a positive effect in the form of increased «authority» and refinement compared to the Hugo? Isn't there an argumentation gap in favor of additional amping? And what about balanced headphone drive (favored by some, although it's hard to detect a technical advantage)?
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
Yes, I agree, but in the case of the Abyss we're talking of a planar magnetic headphone with no fundamental resonance reflected in the impedance response. (Do the impedance variations during membrane movement also apply to planar magnetic headphones?)
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
Indeed, I fully agree with you – as a convinced EQ user and and militant EQ proponent. Because all headphones benefit from equalizing (there's none with a just remotely flat frequency response). A truly linear frequency response (I mean: as linear as it gets) makes the sound smoother, more transparent and less prone to call for euphonic harmonic distortion – it's not just about bright or dark. And it's an illusion to think that there's such a thing as an amp that's just a wire with gain.
The primary reason many Dacs have a balanced internal topology is try to overcome switching noise that has been induced into the Dac chips substrate. Balanced circuitry though causes other distortions that should be avoided. Chord Dacs have no substrate switching noise because the switching elements are seperated from the FPGA and more importantly from the analogue circuitry. So because we don't use standard Dac chips that can suffer from these problems. Therefore we do not have no need to used a balanced internal topology so we don't! The balanced analogue outputs are derived externally in analogue form only from single ended ultra low distortion output. I hope that is not too confusing.
The primary reason many Dacs have a balanced internal topology is try to overcome switching noise that has been induced into the Dac chips substrate. Balanced circuitry though causes other distortions that should be avoided. Chord Dacs have no substrate switching noise because the switching elements are seperated from the FPGA and more importantly from the analogue circuitry. So because we don't use standard Dac chips that can suffer from these problems. Therefore we do not have no need to used a balanced internal topology so we don't! The balanced analogue outputs are derived externally in analogue form only from single ended ultra low distortion output. I hope that is not too confusing.
The abyss is an easy load to drive with an impedance of 46 ohms, so there is no way Dave will run out of current - so you can run Dave at +4dB and the output from Dave will be linear. So assuming you are not running above +4dB, there is plenty of drive in voltage and current available.
The interactions of headphone and outputs are not simple, and for sure a lack of distortion will make it sound brighter, as low order (particularly 2nd) harmonics fatten up the sound. An additional factor is output impedance in terms of damping factor. A high damping factor better controls the drive unit, making the bass sound fast, tight and somewhat leaner. Additionally, low output impedance itself reduces distortion. The load presented by a headphone varies with cone displacement, so the impedance itself varies with amplitude. Now with an amp and cables that have zero impedance, this variation in load has no effect - no increase in distortion. But with a finite cable and amp impedance, you will increase distortion, and because it depends upon bass, this will fatten up the sound too, making the bass much bigger.
Now Dave has only 55 milli ohms of output impedance (0.055 ohms), so it has a substantial damping factor (much lower than any other amp I have seen), which will ensure the bass is tight and accurate; also it will almost eliminate distortion due to load non-linearity (assuming you have a low impedance cable too).
My view is if a headphone is too lean, then use good EQ - or even better, use headphones that innately have more bass to suit your taste. The problem with using distortion to fatten up the sound, is that everything then sounds the same - big and fat. I want my music to be fast, tight and lean when the musician wanted that sound - and big and fat when that was the intent. You can only get that when the system is truly transparent.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
