Armaegis
Modern Modder Man of Manitoba
HTML... uphill, both ways!
- Joined
- Oct 18, 2009
- Posts
- 12,548
- Likes
- 1,402
It seems I've been acquiring a lot of Nuforce gear lately. Well what can I say; I find a brand with good service and quality offerings, and I keep climbing up the chain. This review will be covering two high end devices in their “100” line: the DAC-100 and the HAP-100, a dac/headphone amp and a headphone amp/preamp respectively.
Most here will be familiar with the Nuforce HDP and uDac/uDac2, both of which are fairly popular in their respective categories of head-fi-dom (roughly mid-fi and budget-fi if you will). I have owned or extensively auditioned both of these at some point, and was very curious what their bigger brethren could offer. To note, this is the furthest up (in terms of cost) that I have gone for dacs and amps. For this review, I borrowed a friend's Schiit Bifrost and Lyr (a dac and amp respectively) which are fairly well regarded on head-fi, and as a known entity to serve as a point of reference for comparisons. I feel that they are somewhat close in price point to offer reasonable comparison (yes I am aware of the price difference, but the Nuforce products are also more multifunctional, and this is the best I could get short of buying a ton of more gear myself). I have also owned or reviewed various other gear, which can be seen in my profile.
So without further ado, let's get started.
[size=large]DAC-100[/size]
On the surface:
- a DAC capable up to 24/192k and all the fiddly numbers in between
- four selectable inputs: 1 x usb, 2 x coax, 1 x optical toslink (there is no analog input)
- digital volume control, 100 steps, logarithmic
- preamp out (volume controlled), ~2V at max
- Class A headphone output, ~3.4V into high impedance loads (recommended for headphones with impedance of 100 ohm or higher)
- wireless remote
From the product page: http://nuforce.com/index.php?option=com_k2&view=item&id=5:dac-100&Itemid=187
Quote:
Packaging
Compared to their earlier offerings of the Icon2 speaker system which had very nice packaging, I was a little surprised/disappointed at the very plain packaging the DAC-100 came in. A very standard off white box, no pictures, nothing at all to make you think there was a thousand dollar device inside. Just a little sticker stating what was inside. Nonetheless, it's the contents of the box that matter correct? Inside was your typical foam spacers, the manual, remote.... and that's about it. Oh yeah and the DAC-100. I received the HAP-100 at the same time, and it comes in an identical box.
I suppose if one were in the habit of sneaking new toys into the house, the lack of fancy packaging might be a plus. For those who have a revolving door for their ever-changing gear, the packaging is easily reuseable and you could pack it up like brand new.
Build Quality and Aesthetics
This is a solid unit. In sleek black (also available in silver) and relatively small form factor, it sits comfortably and unobtrusively on a desktop. It won't win any awards for design aesthetic, but it doesn't draw any attention to itself either. It's just a little black box.
The chassis and front plate are a matte textured metal, smoother than brushed aluminum, but still a tiny bit rough to the touch. The black casing actually has a very annoying tendency to pick up scratches... at least that's what I thought at first, but then I realized it wasn't scratches in the casing, but actually my fingernails scraping and leaving a mark. A quick brush with my fingertip removed the scratches and it looked like new again. So note to self: trim my fingernails, and maybe drink some more milk.
There are four buttons on the front which let you select which digital input you're using. There's no indication of which is which though, so you'll just have to remember which button corresponds to which input. The main power switch is on the back, but the front power toggle is actually embedded into the volume knob by pressing into it. A bit of an odd choice, but it does keep the front panel nice and minimalistic. The unit is turned off by simple pressing in on the volume knob again, or using the remote. When in use, there is a panel on the front which lights up to indicate the bitrate of the file being played.
A word on the remote: meh, it's a remote. Power button, input selection, volume control, mute. Really not much beyond that. It's pretty tiny though, so be careful not to misplace it. When controlling the volume via remote, note that the knob does not turn since it isn't motorized. This does result in a discrepancy between the volume played and the knob position. Touching the knob immediately jumps the volume to the knob position, so be careful if the manual volume control was set very high.
In terms of temperature, it does heat up a little bit over time, but nothing drastic. My temperature gun reads roughly 35°C when it's fully warmed up along the back and right edge, where the power supply and class A circuitry are located (and there are vents here). The rest of the casing is maybe ten degrees cooler. In comparison, the Schiit Bifrost is also 35°C and the Lyr is about 40°C across most of the casing, with the area around the tubes hovering about 60°C (it's hard to focus the reader on a small area, but in any event don't go touching hot tubes). Stacking the DAC-100 on top of the HAP-100, the DAC-100 temperature reads marginally higher at 40°C.
Internals
The unit is somewhat heavier than one might expect, and upon opening up you'll see why. Inside is a decent sized toroidal transformer and power supply board, which is overall actually larger than the dac section. This same power supply is found in the HAP-100 as well. Plenty of caps for regulation and heat sinks off to the side. Note the shielding wall around the transformer. While it is often glossed over in reviews, I cannot overstate the importance of a good power supply. Many folks will espouse "source first". Well, it doesn't come any more "source" than your power source since everything stems from there. All your processing and generated signals are all converted energy from somewhere. The cleaner it starts, the better everything will be.
The dac board layouts appear quite clean, and I was actually almost a little disappointed. I have these conflicting thoughts in my head where I think the more expensive something is, the more complex it should be. On the other hand, simplicity is also very good. So I think this sits at a good middle ground (I did not disassemble further to look underneath; for all I know it could be a jungle on the bottom side). Without trying to decipher all the traces in the circuitry, one can still make a basic assessment of the board. There's the usb receiver child board (*very* similar to the one found in the LPS power supply, review [here]), the main dac section, then the output area with two socketed LM4562 opamps and the class A amp. Simple is good right?
A quick note on the opamps:
From emails with Nuforce, each opamp handles one channel separately, with one half of the opamp feeding the preamp and the other half feeding the headphone amp. Nice; minimizes crosstalk that way. I will add at this point that Nuforce does not officially support users rolling their own opamps. If those sockets are just too tempting, then proceed at your own risk. That's all I can officially say on the matter.
The volume and input controls are all contained within a board connected to the front plate. Upon first using it, I felt the volume knob was a bit loose (off-axis) compared to the Lyr which is rock solid. Then again, the DAC-100 uses digital volume control so it doesn't need a super solid pot so perhaps they saved on some costs there. As mentioned before, the on/off toggle is also built into the pot so it needs to have some give there. This is just a minor quibble, but I suppose on a $1k+ product I wanted it to feel more solid.
Installation:
The DAC-100 requires that you download and install drivers from Nuforce's website. Nothing too fancy beyond that. After installation, I didn't have to do anything. I played music predominantly through JRiver using WASAPI after selecting the DAC-100 as my output device and have had no issues. Using other modes such as Direct Sound will require the use of the provided software, which is quite straight forward.
Measurements
Output DC measurements:
- headphone: 3.1/6.0 mV
- RCA: 1.5/3.1 mV
RCA output voltage, playing test tone 60Hz -1dB, into resistor load:
- 1M ohm resistor load = 1.9 V
headphone output voltages, playing test tone 60Hz -1dB, into resistor loads
- 33 ohm = 2.3 V
- 180 ohm = 3.1 V
- 560 ohm = 3.4 V
- 1M ohm = 3.4 V
Attempting to calculate output impedance resulted in some very odd results. Normally one can calculate it from measure output voltages with differing loads (as given above), but these don't work out. According to their R&D department, there is an output resistor in series to provide short circuit protection and to stabilize low impedance loads (though they recommend headphones with >100ohm impedance to start with). They also said that impedance could not be measured directly since the class A circuitry reacts in an odd way with measurements, resulting in a negative effective impedance.
Given my own attempts to measure the output impedance, I would have to agree that something odd is happening there. I also suspect that there is some active circuit component lying in parallel with the headphone load, as some measurements vary depending if the amp is turned on or off. Anyhow, here are some numbers on my end (done with a cheap multimeter). Do with them what you will.
Measuring resistance (not impedance) with a resistor load in parallel:
- 1M ohm ---> 18.4 ohm
- 22 ohm ---> 10.9 ohm
- 10 ohm ---> 7.4 ohm
- no resistor ---> 15.6 ohm (6.9k ohm when turned off)
Given my caveat about the odd interactions of the circuitry above, do not take these numbers to directly correlate to output impedance. A lot of people seem to think that lower is better, yet some very popular and well regarded amps around here sport high output impedances. The Dacport/dacmini have 10 ohm outputs, as does the FiiO e9 and other TPA6120 chipset based amps. One of my favourites - the Bottlehead Crack tube amp - sports 120 ohm at its output. Do note however that the higher output impedance will have an interaction with headphones that have variances in their impedance curves (as an over-simplification, this generally results in a mild midbass boost)
NOTE: After some early feedback to Nuforce, they have informed me that they are working on a module for lower output impedance which they estimate will be available in Q1 of 2013.
Sound
The dac:
In general, I do not mind amps injecting some of their own “flavour” into the sound (I can hear the objectivists cringing... subjectively of course), but I expect dacs to be fully transparent. In this regard, I am very pleased with the DAC-100. It is supremely clean and detailed. Fuzzy wording I know, but bear with me.
The bulk of my dac comparisons came against the Schiit Bifrost. Both were fed identical digital streams from my LPS-2U-192k which converts usb into both coax and optical. These digital streams were then fed simultaneously into the Bifrost and DAC-100, and yes I did switch coax and optical between them and was not able to discern the difference.
In general (again), if I am trying to discern an audible difference between amps I will focus on the bass and its overall impact/power and energy delivery. In dacs, I tend to focus on the treble and how it affects detailing. Do not presume this to mean I listen for the low and high frequencies. It's more an umbrella term encompassing waveforms/harmonics (so you'll get “treble” components in a “bass” sound, etc).
So anyways, in my A/B comparisons of the DAC-100 vs the Bifrost, I felt the DAC-100 was the clear winner. I fed both into the Nuforce HAP-100 (preamp/headphone amp) and was able to switch mid-song between them with the push of a button. It wasn't a huge difference, but it was palpable once I knew what signs to listen for. The Nuforce offered just that much better resolution in the upper ranges. Cymbals had more shine to them, strings had a more faithful tonality, and vocals had just that indescribable breath to them. I dislike using such subjective here, but without a suite of measuring equipment it is the best I can describe. Quite often I found myself actually turning down the volume, not because it was too loud or sibilant, but because it felt like there was too much... information for lack of a better word. It felt like a detail overload sometimes, where I could literally pick out these little bits and pieces of all the different instruments and their layering, and it was just too much to handle all at once.
Against the Bifrost, I also noticed that the Schiit dac had a slight warble when playing pure sine wave tones. I couldn't quite tell if they were frequency or amplitude warbles, but they were very noticeable when flipping back and forth with the DAC-100.
Distortion and encoding artifacts were also more noticeable on the DAC-100. It is not much of an issue with newer encoders, but older files may show through with some unpleasant sound.
If there is a weak point of the DAC-100, it would be in the noise floor. This is not the quietest dac that I've heard, and the Bifrost bests it easily here. The noise is consistent regardless of volume setting, but also exists on the preamp output so it is puzzling where the source of this noise is coming from. With high impedance or low sensitivity headphones, this is virtually a non-issue as the point at which the noise is barely audible will have music loud enough to deafen you. For headphones with lower impedance or higher sensitivity, noise may become an issue during quiet passages. The preamp outputs appear to be quieter than the headphone output, and feeding into an external amplifier seems to ameliorate the noise to a degree. More on this in the amp section.
I will make one final note however, that the noise sounds like a 120 Hz electrical hash, which makes me think it might be a ground issue (or possibly an internal grounding). I can't rule out the possibility that this is a ground loop or other contaminant from my power lines, though I doubt it. Though the Bifrost plugged into the same power strip did not exhibit the same noise, that's no indicator either. It might be due to how their power supplies differ, or any other possibilities. The nature/source of ground noise is very difficult to isolate.
March 2013 update
vs Schiit Gungnir
- the Gungnir is a bit more prominent in the uppermids
- which makes it feel more aggressive, leaning towards abrasive in comparison
- bass impact feels about the same
- I like the Gungnir more than the Bifrost upon initial listen
vs Nuforce HDP
- DAC-100 is better in those deep gutwrenching lows (<50Hz), while the HDP feels a little rolled off
- DAC-100 seems more prominent in the lower treble, I actually slightly prefer the HDP here strictly in that regard
- but the HDP soundstage is slightly flatter and feels a smidge dull
vs iBasso D10
- I honestly had a harder time telling apart the D10 vs DAC-100 than I did the Bifrost vs DAC-100
- it was a change in staging, with the D10 feeling a little more shallow
- not so much the shimmer on the cymbals and highs, but the tone feels more natural with the DAC-100
The amp
Given the notes about output impedance that I mentioned earlier, it is recommended to stick with headphones that have a high impedance. In fact, Nuforce's recommendation is to use headphones of 100 ohm or higher. This also minimizes the noise concerns that I touched upon earlier.
I went through a wide array of high impedance cans, notably the Sennheiser HD600 (300 ohm), AKG K240 Sextett MP (600 ohm), and Sennheiser HD25-13-ii (600 ohm). The Sennheiser has the (nearly trademark) impedance hump around 100 Hz. This effect is barely noticeable as a slight bloom in the midbass. Trying with the lower end Senn HD595 makes this effect much more noticeable, as the nominal impedance there is 50 ohm and spikes to over 200 ohm. Even going by the output voltage measurements posted earlier, that is a significant increase in voltage swing going to the headphone at the spike.
I also tried some popular planar magnetics like the Fostex T50rp, Audeze LCD-2, and HiFiMAN HE-6. The amp section was good, adequate, and just ok respectively. Moving up to the Nuforce HAP-100 gave enough juice to the LCD-2 for high listening levels, though not quite headbanging levels if you're the kind to push it that far. Do note that all three of these headphones have impedances lower than what Nuforce recommends. However, they appeared to operate just fine by my ears, and noise was negligible as they are all fairly inefficient compared to dynamics.
In terms of actual sound characteristics, I would describe the amp section as lean and mean. This is not a basshead's amplifier, but not a treble lover's either. The sound is quick and aggressive, much like the dac. I mentioned early on in this review that I typically don't mind when an amp injects some of its own character into the sound. In this case, I don't think it really adds much of anything.
It doesn't quite have the kick in the bass like the Lyr, but the sound is cleaner. Even with the Sennheisers with the midbass bloom, there is the added warmth there, but it maintains the clean technical sound. I might go so far as to say that the Sennheisers are an ideal pairing for the DAC-100. Word around the street is that the DAC-100 was voiced with an HD800 in mind. While I can neither confirm nor deny this, nor do I have an HD800, my experiences with the lower end Senns might indicate a similar experience. Some other users have shared similar views, and I may copy some of their posts into this review as appropriate.
March 2013 update - abridged notes on specific headphone pairings
Sennheiser HD595
- you can really tell there's a headphone impedance spike just from the frequency hump at 100Hz
- overall good, though slight noise
Sennheiser HD600/650
- noise not noticeable
- impedance swing is not as noticeable as the 595
- nice and full bodied sound
- not as good a match as the Bottlehead Crack
Sennheiser HD800
- noise not noticeable
- I really like this pairing. Just a hint more warmth to the HD800, though now the rounding of the bass is stronger which perhaps doesn't work for crazy low electronica
- I'm not actually that big a fan of the HD800, but it takes on a slightly mellower characteristic here and I can get into the groove more easily
Fostex T50rp
- noise not noticeable unless I'm focusing with nothing playing
- at first I was meh, but it really grew on me, especially with my modded versions
- bass has a bit of extra energy, but so does the lower treble
- the lower impedance means it'll current clip at high volumes, but that's still plenty loud
Audeze LCD-2/3
- noise not noticeable
- upper midrange seems tamed down a bit (good thing) so vocals aren't as shouty
- it can get loud, but struggles at the top
HiFiMAN HE-500
- noise not noticeable
- not bad, not bad at all
- similar to the Fostex, there's a bit of extra energy in the lows
- the treble feels a bit edgy, but that's just the HE-500 characteristic
HiFiMan HE-6
- haha, no
Optional jack module w/jumper for bypassing output resistor
- the standard jack has a 15 ohm resistor in series for headphone protection; the optional module has a jumper which when set will bypass this resistor (place the jumper in the position closer to the rear of the board)
- except for frequency response changes due to impedance interactions, the sonic changes noted here are very very minute
- in general: bass becomes a bit leaner, treble cleaner - this is mostly from damping factor
- there is more voltage available for low impedance headphones
- the HD800 loses some of its warmth and becomes more neutral
Differences in damping were noticeable with the HD800 (300 ohm) and T50rp (50 ohm)
- bass thumps a bit more with lower damping; this can be pleasant sometimes
- lower damping also has slightly zingier treble
- noticeable with the T50rp, very hard to tell with the HD800
Conclusions
Overall, this is a good piece of kit. It's not cheap (tipping in at $1.1k), but is fairly versatile and can serve as the centrepiece of a multi-device setup (assuming they all have digital outputs). The dac is top notch and at this point possibly the best that I have heard. The amp works best with high impedance cans of moderate sensitivity. This puts it squarely in the yard of the high end Sennheisers and Beyerdynamics, with whom it plays very well.
In terms of value for the dollar, this is a tricky thing for me. As of this time, the DAC-100 is the most expensive piece of gear that I have auditioned and actually exceeds what my personal budget allows. Even if it's better than sliced bread, I will always have my trepidations. Expectantly, and perhaps thankfully, it is also the best performer that I have had the pleasure of listening to. Compared to the Schiit Bifrost/Lyr stack which is comparable in price and feature set, the DAC-100 performs better as dac, and I would say roughly on par as an amp (with a narrower range of choice headphones, but arguably better sound with those so it evens out).
I just recently obtained a used HDP (which I've auditioned in the past), and I feel the HDP is an excellent value for the dollar. The DAC-100 is a step up in all respects. If your budget is comfortable playing in the $1k range, you will not be disappointed.
Most here will be familiar with the Nuforce HDP and uDac/uDac2, both of which are fairly popular in their respective categories of head-fi-dom (roughly mid-fi and budget-fi if you will). I have owned or extensively auditioned both of these at some point, and was very curious what their bigger brethren could offer. To note, this is the furthest up (in terms of cost) that I have gone for dacs and amps. For this review, I borrowed a friend's Schiit Bifrost and Lyr (a dac and amp respectively) which are fairly well regarded on head-fi, and as a known entity to serve as a point of reference for comparisons. I feel that they are somewhat close in price point to offer reasonable comparison (yes I am aware of the price difference, but the Nuforce products are also more multifunctional, and this is the best I could get short of buying a ton of more gear myself). I have also owned or reviewed various other gear, which can be seen in my profile.
So without further ado, let's get started.
[size=large]DAC-100[/size]
On the surface:
- a DAC capable up to 24/192k and all the fiddly numbers in between
- four selectable inputs: 1 x usb, 2 x coax, 1 x optical toslink (there is no analog input)
- digital volume control, 100 steps, logarithmic
- preamp out (volume controlled), ~2V at max
- Class A headphone output, ~3.4V into high impedance loads (recommended for headphones with impedance of 100 ohm or higher)
- wireless remote
From the product page: http://nuforce.com/index.php?option=com_k2&view=item&id=5:dac-100&Itemid=187
Quote:
Packaging
Compared to their earlier offerings of the Icon2 speaker system which had very nice packaging, I was a little surprised/disappointed at the very plain packaging the DAC-100 came in. A very standard off white box, no pictures, nothing at all to make you think there was a thousand dollar device inside. Just a little sticker stating what was inside. Nonetheless, it's the contents of the box that matter correct? Inside was your typical foam spacers, the manual, remote.... and that's about it. Oh yeah and the DAC-100. I received the HAP-100 at the same time, and it comes in an identical box.
I suppose if one were in the habit of sneaking new toys into the house, the lack of fancy packaging might be a plus. For those who have a revolving door for their ever-changing gear, the packaging is easily reuseable and you could pack it up like brand new.
Build Quality and Aesthetics
This is a solid unit. In sleek black (also available in silver) and relatively small form factor, it sits comfortably and unobtrusively on a desktop. It won't win any awards for design aesthetic, but it doesn't draw any attention to itself either. It's just a little black box.
The chassis and front plate are a matte textured metal, smoother than brushed aluminum, but still a tiny bit rough to the touch. The black casing actually has a very annoying tendency to pick up scratches... at least that's what I thought at first, but then I realized it wasn't scratches in the casing, but actually my fingernails scraping and leaving a mark. A quick brush with my fingertip removed the scratches and it looked like new again. So note to self: trim my fingernails, and maybe drink some more milk.
There are four buttons on the front which let you select which digital input you're using. There's no indication of which is which though, so you'll just have to remember which button corresponds to which input. The main power switch is on the back, but the front power toggle is actually embedded into the volume knob by pressing into it. A bit of an odd choice, but it does keep the front panel nice and minimalistic. The unit is turned off by simple pressing in on the volume knob again, or using the remote. When in use, there is a panel on the front which lights up to indicate the bitrate of the file being played.
A word on the remote: meh, it's a remote. Power button, input selection, volume control, mute. Really not much beyond that. It's pretty tiny though, so be careful not to misplace it. When controlling the volume via remote, note that the knob does not turn since it isn't motorized. This does result in a discrepancy between the volume played and the knob position. Touching the knob immediately jumps the volume to the knob position, so be careful if the manual volume control was set very high.
In terms of temperature, it does heat up a little bit over time, but nothing drastic. My temperature gun reads roughly 35°C when it's fully warmed up along the back and right edge, where the power supply and class A circuitry are located (and there are vents here). The rest of the casing is maybe ten degrees cooler. In comparison, the Schiit Bifrost is also 35°C and the Lyr is about 40°C across most of the casing, with the area around the tubes hovering about 60°C (it's hard to focus the reader on a small area, but in any event don't go touching hot tubes). Stacking the DAC-100 on top of the HAP-100, the DAC-100 temperature reads marginally higher at 40°C.
Internals
The unit is somewhat heavier than one might expect, and upon opening up you'll see why. Inside is a decent sized toroidal transformer and power supply board, which is overall actually larger than the dac section. This same power supply is found in the HAP-100 as well. Plenty of caps for regulation and heat sinks off to the side. Note the shielding wall around the transformer. While it is often glossed over in reviews, I cannot overstate the importance of a good power supply. Many folks will espouse "source first". Well, it doesn't come any more "source" than your power source since everything stems from there. All your processing and generated signals are all converted energy from somewhere. The cleaner it starts, the better everything will be.
The dac board layouts appear quite clean, and I was actually almost a little disappointed. I have these conflicting thoughts in my head where I think the more expensive something is, the more complex it should be. On the other hand, simplicity is also very good. So I think this sits at a good middle ground (I did not disassemble further to look underneath; for all I know it could be a jungle on the bottom side). Without trying to decipher all the traces in the circuitry, one can still make a basic assessment of the board. There's the usb receiver child board (*very* similar to the one found in the LPS power supply, review [here]), the main dac section, then the output area with two socketed LM4562 opamps and the class A amp. Simple is good right?
A quick note on the opamps:
From emails with Nuforce, each opamp handles one channel separately, with one half of the opamp feeding the preamp and the other half feeding the headphone amp. Nice; minimizes crosstalk that way. I will add at this point that Nuforce does not officially support users rolling their own opamps. If those sockets are just too tempting, then proceed at your own risk. That's all I can officially say on the matter.
The volume and input controls are all contained within a board connected to the front plate. Upon first using it, I felt the volume knob was a bit loose (off-axis) compared to the Lyr which is rock solid. Then again, the DAC-100 uses digital volume control so it doesn't need a super solid pot so perhaps they saved on some costs there. As mentioned before, the on/off toggle is also built into the pot so it needs to have some give there. This is just a minor quibble, but I suppose on a $1k+ product I wanted it to feel more solid.
Installation:
The DAC-100 requires that you download and install drivers from Nuforce's website. Nothing too fancy beyond that. After installation, I didn't have to do anything. I played music predominantly through JRiver using WASAPI after selecting the DAC-100 as my output device and have had no issues. Using other modes such as Direct Sound will require the use of the provided software, which is quite straight forward.
Measurements
Output DC measurements:
- headphone: 3.1/6.0 mV
- RCA: 1.5/3.1 mV
RCA output voltage, playing test tone 60Hz -1dB, into resistor load:
- 1M ohm resistor load = 1.9 V
headphone output voltages, playing test tone 60Hz -1dB, into resistor loads
- 33 ohm = 2.3 V
- 180 ohm = 3.1 V
- 560 ohm = 3.4 V
- 1M ohm = 3.4 V
Attempting to calculate output impedance resulted in some very odd results. Normally one can calculate it from measure output voltages with differing loads (as given above), but these don't work out. According to their R&D department, there is an output resistor in series to provide short circuit protection and to stabilize low impedance loads (though they recommend headphones with >100ohm impedance to start with). They also said that impedance could not be measured directly since the class A circuitry reacts in an odd way with measurements, resulting in a negative effective impedance.
Given my own attempts to measure the output impedance, I would have to agree that something odd is happening there. I also suspect that there is some active circuit component lying in parallel with the headphone load, as some measurements vary depending if the amp is turned on or off. Anyhow, here are some numbers on my end (done with a cheap multimeter). Do with them what you will.
Measuring resistance (not impedance) with a resistor load in parallel:
- 1M ohm ---> 18.4 ohm
- 22 ohm ---> 10.9 ohm
- 10 ohm ---> 7.4 ohm
- no resistor ---> 15.6 ohm (6.9k ohm when turned off)
Given my caveat about the odd interactions of the circuitry above, do not take these numbers to directly correlate to output impedance. A lot of people seem to think that lower is better, yet some very popular and well regarded amps around here sport high output impedances. The Dacport/dacmini have 10 ohm outputs, as does the FiiO e9 and other TPA6120 chipset based amps. One of my favourites - the Bottlehead Crack tube amp - sports 120 ohm at its output. Do note however that the higher output impedance will have an interaction with headphones that have variances in their impedance curves (as an over-simplification, this generally results in a mild midbass boost)
NOTE: After some early feedback to Nuforce, they have informed me that they are working on a module for lower output impedance which they estimate will be available in Q1 of 2013.
Sound
The dac:
In general, I do not mind amps injecting some of their own “flavour” into the sound (I can hear the objectivists cringing... subjectively of course), but I expect dacs to be fully transparent. In this regard, I am very pleased with the DAC-100. It is supremely clean and detailed. Fuzzy wording I know, but bear with me.
The bulk of my dac comparisons came against the Schiit Bifrost. Both were fed identical digital streams from my LPS-2U-192k which converts usb into both coax and optical. These digital streams were then fed simultaneously into the Bifrost and DAC-100, and yes I did switch coax and optical between them and was not able to discern the difference.
In general (again), if I am trying to discern an audible difference between amps I will focus on the bass and its overall impact/power and energy delivery. In dacs, I tend to focus on the treble and how it affects detailing. Do not presume this to mean I listen for the low and high frequencies. It's more an umbrella term encompassing waveforms/harmonics (so you'll get “treble” components in a “bass” sound, etc).
So anyways, in my A/B comparisons of the DAC-100 vs the Bifrost, I felt the DAC-100 was the clear winner. I fed both into the Nuforce HAP-100 (preamp/headphone amp) and was able to switch mid-song between them with the push of a button. It wasn't a huge difference, but it was palpable once I knew what signs to listen for. The Nuforce offered just that much better resolution in the upper ranges. Cymbals had more shine to them, strings had a more faithful tonality, and vocals had just that indescribable breath to them. I dislike using such subjective here, but without a suite of measuring equipment it is the best I can describe. Quite often I found myself actually turning down the volume, not because it was too loud or sibilant, but because it felt like there was too much... information for lack of a better word. It felt like a detail overload sometimes, where I could literally pick out these little bits and pieces of all the different instruments and their layering, and it was just too much to handle all at once.
Against the Bifrost, I also noticed that the Schiit dac had a slight warble when playing pure sine wave tones. I couldn't quite tell if they were frequency or amplitude warbles, but they were very noticeable when flipping back and forth with the DAC-100.
Distortion and encoding artifacts were also more noticeable on the DAC-100. It is not much of an issue with newer encoders, but older files may show through with some unpleasant sound.
If there is a weak point of the DAC-100, it would be in the noise floor. This is not the quietest dac that I've heard, and the Bifrost bests it easily here. The noise is consistent regardless of volume setting, but also exists on the preamp output so it is puzzling where the source of this noise is coming from. With high impedance or low sensitivity headphones, this is virtually a non-issue as the point at which the noise is barely audible will have music loud enough to deafen you. For headphones with lower impedance or higher sensitivity, noise may become an issue during quiet passages. The preamp outputs appear to be quieter than the headphone output, and feeding into an external amplifier seems to ameliorate the noise to a degree. More on this in the amp section.
I will make one final note however, that the noise sounds like a 120 Hz electrical hash, which makes me think it might be a ground issue (or possibly an internal grounding). I can't rule out the possibility that this is a ground loop or other contaminant from my power lines, though I doubt it. Though the Bifrost plugged into the same power strip did not exhibit the same noise, that's no indicator either. It might be due to how their power supplies differ, or any other possibilities. The nature/source of ground noise is very difficult to isolate.
March 2013 update
vs Schiit Gungnir
- the Gungnir is a bit more prominent in the uppermids
- which makes it feel more aggressive, leaning towards abrasive in comparison
- bass impact feels about the same
- I like the Gungnir more than the Bifrost upon initial listen
vs Nuforce HDP
- DAC-100 is better in those deep gutwrenching lows (<50Hz), while the HDP feels a little rolled off
- DAC-100 seems more prominent in the lower treble, I actually slightly prefer the HDP here strictly in that regard
- but the HDP soundstage is slightly flatter and feels a smidge dull
vs iBasso D10
- I honestly had a harder time telling apart the D10 vs DAC-100 than I did the Bifrost vs DAC-100
- it was a change in staging, with the D10 feeling a little more shallow
- not so much the shimmer on the cymbals and highs, but the tone feels more natural with the DAC-100
The amp
Given the notes about output impedance that I mentioned earlier, it is recommended to stick with headphones that have a high impedance. In fact, Nuforce's recommendation is to use headphones of 100 ohm or higher. This also minimizes the noise concerns that I touched upon earlier.
I went through a wide array of high impedance cans, notably the Sennheiser HD600 (300 ohm), AKG K240 Sextett MP (600 ohm), and Sennheiser HD25-13-ii (600 ohm). The Sennheiser has the (nearly trademark) impedance hump around 100 Hz. This effect is barely noticeable as a slight bloom in the midbass. Trying with the lower end Senn HD595 makes this effect much more noticeable, as the nominal impedance there is 50 ohm and spikes to over 200 ohm. Even going by the output voltage measurements posted earlier, that is a significant increase in voltage swing going to the headphone at the spike.
I also tried some popular planar magnetics like the Fostex T50rp, Audeze LCD-2, and HiFiMAN HE-6. The amp section was good, adequate, and just ok respectively. Moving up to the Nuforce HAP-100 gave enough juice to the LCD-2 for high listening levels, though not quite headbanging levels if you're the kind to push it that far. Do note that all three of these headphones have impedances lower than what Nuforce recommends. However, they appeared to operate just fine by my ears, and noise was negligible as they are all fairly inefficient compared to dynamics.
In terms of actual sound characteristics, I would describe the amp section as lean and mean. This is not a basshead's amplifier, but not a treble lover's either. The sound is quick and aggressive, much like the dac. I mentioned early on in this review that I typically don't mind when an amp injects some of its own character into the sound. In this case, I don't think it really adds much of anything.
It doesn't quite have the kick in the bass like the Lyr, but the sound is cleaner. Even with the Sennheisers with the midbass bloom, there is the added warmth there, but it maintains the clean technical sound. I might go so far as to say that the Sennheisers are an ideal pairing for the DAC-100. Word around the street is that the DAC-100 was voiced with an HD800 in mind. While I can neither confirm nor deny this, nor do I have an HD800, my experiences with the lower end Senns might indicate a similar experience. Some other users have shared similar views, and I may copy some of their posts into this review as appropriate.
March 2013 update - abridged notes on specific headphone pairings
Sennheiser HD595
- you can really tell there's a headphone impedance spike just from the frequency hump at 100Hz
- overall good, though slight noise
Sennheiser HD600/650
- noise not noticeable
- impedance swing is not as noticeable as the 595
- nice and full bodied sound
- not as good a match as the Bottlehead Crack
Sennheiser HD800
- noise not noticeable
- I really like this pairing. Just a hint more warmth to the HD800, though now the rounding of the bass is stronger which perhaps doesn't work for crazy low electronica
- I'm not actually that big a fan of the HD800, but it takes on a slightly mellower characteristic here and I can get into the groove more easily
Fostex T50rp
- noise not noticeable unless I'm focusing with nothing playing
- at first I was meh, but it really grew on me, especially with my modded versions
- bass has a bit of extra energy, but so does the lower treble
- the lower impedance means it'll current clip at high volumes, but that's still plenty loud
Audeze LCD-2/3
- noise not noticeable
- upper midrange seems tamed down a bit (good thing) so vocals aren't as shouty
- it can get loud, but struggles at the top
HiFiMAN HE-500
- noise not noticeable
- not bad, not bad at all
- similar to the Fostex, there's a bit of extra energy in the lows
- the treble feels a bit edgy, but that's just the HE-500 characteristic
HiFiMan HE-6
- haha, no
Optional jack module w/jumper for bypassing output resistor
- the standard jack has a 15 ohm resistor in series for headphone protection; the optional module has a jumper which when set will bypass this resistor (place the jumper in the position closer to the rear of the board)
- except for frequency response changes due to impedance interactions, the sonic changes noted here are very very minute
- in general: bass becomes a bit leaner, treble cleaner - this is mostly from damping factor
- there is more voltage available for low impedance headphones
- the HD800 loses some of its warmth and becomes more neutral
Differences in damping were noticeable with the HD800 (300 ohm) and T50rp (50 ohm)
- bass thumps a bit more with lower damping; this can be pleasant sometimes
- lower damping also has slightly zingier treble
- noticeable with the T50rp, very hard to tell with the HD800
Conclusions
Overall, this is a good piece of kit. It's not cheap (tipping in at $1.1k), but is fairly versatile and can serve as the centrepiece of a multi-device setup (assuming they all have digital outputs). The dac is top notch and at this point possibly the best that I have heard. The amp works best with high impedance cans of moderate sensitivity. This puts it squarely in the yard of the high end Sennheisers and Beyerdynamics, with whom it plays very well.
In terms of value for the dollar, this is a tricky thing for me. As of this time, the DAC-100 is the most expensive piece of gear that I have auditioned and actually exceeds what my personal budget allows. Even if it's better than sliced bread, I will always have my trepidations. Expectantly, and perhaps thankfully, it is also the best performer that I have had the pleasure of listening to. Compared to the Schiit Bifrost/Lyr stack which is comparable in price and feature set, the DAC-100 performs better as dac, and I would say roughly on par as an amp (with a narrower range of choice headphones, but arguably better sound with those so it evens out).
I just recently obtained a used HDP (which I've auditioned in the past), and I feel the HDP is an excellent value for the dollar. The DAC-100 is a step up in all respects. If your budget is comfortable playing in the $1k range, you will not be disappointed.
