Many of us (who are looking at using more than just our smart-phone for audio) will be looking for solutions for particular circumstances. Personally, I have a desktop set-up mainly centered around an iFi stack (iUSD, iTube, iDSD) – which powers my headphones (HD800S, HD600 + others) as well as my active speakers (LSR 305’s). Portably I switch between a few dedicated DAPs, of my iPhone + Bluetooth DAC/amps. At work I take my portable solutions.
But what about those who are looking for a budget solution (desk-top or lap-top), or cheaper and more space-efficient work solution? How about something for the frequent traveller? FiiO recently released their new K3 – replacement for the ageing E10K. For the last 3 weeks I’ve had the opportunity to put it to the test. How did the pocket sized DAC/amp stack up?
FiiO was first founded in 2007. Their first offerings were some extremely low cost portable amplifiers – which were sometimes critiqued by some seasoned audio enthusiasts as being low budget “toys”. But FiiO has spent a lot of time with their on-line communities, and continued to listen to their potential buyers, adopt their ideas, and grow their product range. They debuted their first DAP (the X3) in 2013, and despite some early hiccups with developing the UI, have worked with their customer base to continually develop the firmware for a better user experience. The X3 was followed by the X5, X1, X7 and most of these DAPs are now into their 2nd or even 3rd generations.
They’ve also developed new cables, desktop and portable amplifiers, DACs, ear-buds and earphones. FiiO’s products have followed a very simple formula since 2007 – affordable, stylish, well built, functional, measuring well, and most importantly sounding good.
The K3 DAC/amp that I’m reviewing today was provided to me freely as a review sample. FiiO have asked me to keep it for my personal use, or for follow up comparisons, and I thank them for this. The retail price at time of review is ~120 USD.
If you haven’t read any of my reviews, I suggest starting here, as it will give you an insight into my known preferences and bias.
For the purposes of this review – I used the FiiO K3:
- As a desk-top DAC at home with my lap-top and also my PC
- As a desk-top DAC at work with my lap-top
- With a variety of headphones and earphones including the HD630VB, Brainwavz HM100, Alessandro MS Pro, SE846, 64 Audio U10, VE Zen and many others.
In the time I have spent with the K3, I have noticed no real change to the overall sonic presentation (burn-in). This is a purely subjective review – my gear, my ears, and my experience. Please take it all with a grain of salt – especially if it does not match your own experience.
The K3 arrived in a 90 x 150 x 30mm retail outer box (over internal hinged box). The outer box is predominantly white with a picture of the K3 on the front. Opening the box reveals an inner box with foam insert which holds the K5, and adjacent is a smaller box with the other accessories. The total accessory package with my unit included:
- The K3 DAC/AMP
- 1 x USB mini to USB-C data and power cable
- 2 sets of foam feet (x 2 pieces per set)
- Warranty card
- Quick start operation card
FiiO has a selection of graphs on their website, and they show a nicely flat response from 20 Hz to 20 kHz (under 32 ohm load / 10K for line-out), and less than 0.5dB down at both ends of the frequency spectrum.
I’ve never had much luck measuring DACs with my equipment – its just not good enough to measure what I’d need to for the purposes of an objective review (especially THD / THD+N). I do use RMAA (v6) for a simple frequency response + gain + bass boost measurement just to check FiiO’s measurements. My graph (SE only) shows the same data as FiiO with a bass boost of +6dB and a gain between 6.5-7 dB
The K3 outer body consists of a one piece aluminium casing (CNC machined, anodised and sand-blasted), book-ended front and back by single aluminium face plates. The outer body is nicely rounded.
On the front face (from left to right) is a 2.5mm balanced socked, 3.5mm single-ended socket, bass boost and gain control switches, and digital volume knob. The knob is knurled for easy grip, and has an LED indicator (also lights around the circumference of the knob) which changes with sample rate. Red is indicated when the K3 is connected but not engaged as active device, blue for 48 kHz or less, yellow (sort of a pale greenish yellow) for above 48 kHz, and a deeper green for DSD.
The rear of the device (again from left to right) has a 3.5mm line-out, switch for USB (1.0 or 2.0 standard), USB-C power/data input, coaxial digital out, and optical digital out.
I can’t really fault the overall build, and my only comment is more on the basis of usability. Because it is such a light-weight and small device, when changing the volume it is quite easy to knock it. Likewise the switches for gain and bass boost are quite difficult to access with a headphone plugged in.
I wonder if an option might be to mount both the volume control and/or switched on top of the device? Its just a thought for future development – interested to see if anyone else sees this as a good design option?
Internally the K3 uses the AK4452 DAC chip which has very good SNR and low distortion metrics. It provides PCM sampling rates up to 32/384 and DSD up to DSD256. The analog signal is then amplified by an LPF/voltage and current drive combination consisting of the TI OPA1612 and dual OPA926.
USB input is provided via the XMOS XUF208 – which provides excellent stability and reliability. It also has high bandwidth capability (including being able to pass DSD via DoP or natively), and using dual crystal oscillators for seamless pass-through of different sampling rates.
Volume control is provided by a digital ADC solution which allows smoother transitional steps, and also eliminates channel imbalances toward the low end of the pot. When combined with the two step gain, it allows fine adjustment of a wide range of headphone / earphone types.
The bass boost is a 2 option switch (on or off) and provides an ~ 6dB boost which is centered at approx 40 Hz (sub-bass), but the effect is noticeable from 800Hz down. The lift in the lower mid-range is pretty benign (only about + 1-2 dB) and increases as it nears the sub-bass region. When engaged it adds some richness and warmth to the lower mid-range, and quite a lot of impact and punch to the sub and mid-bass.
The gain, like the original signal is very linear, and from my measurements adds about 6.7-6.8 dB to the overall volume. For me it is a good boost, adding volume / headroom if required, yet still having a usable range with the volume pot.
The thing I like about the line-out is that it is variable, and thus the pot can be used as a volume control with active monitors. It doesn’t have enough power to use with passive speakers, but with my JBL LSR305’s it is a really handy feature. Possibly the only thing I would have also liked was a fixed line-level out (preferably rear RCAs) which could have been used to go to a powered amp (eg tube amp) – utilising the very good DAC on the K3. IMO this would have been a better use of the real estate than the current coax and optical outputs.
The K3 has two digital outputs – Coax and Optical – both effectively passing digital audio from the USB stage as a direct stream to another DAC/amp. This effectively bypasses the K3 DAC (which is its best overall feature), and turns the K3 into a USB to Optical (or Coax) converter. To others this may be a handy feature. To me its a bit of a waster – when there would have been a much better option (fixed rear RCA). Perhaps an opportunity missed rather than a feature.
USB 1 vs 2 Switch & DAC Support
One of the features with the K3 is being able to switch between the USB 1 and 2 standards. With the 1.0 standard, if you’re using an older device (eg my eeePC running Win7 starter), the K3 is instantly recognised and it allows playback up to 24/96 with just a generic driver. The good news is that even with the 1 standard at 24/96 video syncs almost perfectly with audio and the quality is very good.
Switching to the 2.0 standard unlocks the additional features of the DAC including higher sample rates and DSD playback. For this you’ll need to download and install the 4.47 DAC driver (from their website). If you are running Windows 10, even plugging the K3 without the driver still loads generic support with USB 2.0, but the driver is required for DSD.
Linux (and presumably OSX) need no drivers installed, and support on my Linux box was instant, and I was able to play DSD and highest sample rates (up-sampling using Jriver Media Center). Virtually the only issue I’ve seen so far with the DAC is cases of video and audio not syncing. My advice is to try up sampling to 24/194 or 24/96 if this happens as it does seem to solve any latency issues.
POWER AND HEAT
So far the amp has never run above slightly warm, regardless of the load or power level.
FiiO has published the output levels (SE 220 into 16 ohm, 120 ohm into 32 ohm / Bal. 320 into 16 ohm, 200 ohm into 32 ohm), but they didn’t include data on delivery to higher impedance headphones. FiiO recommends up to 150 ohm headphones for the K3 and this makes sense given the relatively weaker expected output into 300 ohms (I would guess at single figures for the SE, and approx 20 mW or less for the balanced – calculated from similar devices).
With easier to drive full sized headphones (HD630VB, HM100, and MS Pro), there were no issues, and that was on low gain – achieving my normal 56-75 dB average listening level (pot at 2-3 o’clock, or about 1/2 way). This was measured with a calibrated SPL meter using a-weighting.
I tried the 300ohm HD600s and while volume was easy enough to achieve (SE) with the pot at 3pm (so a little over 1/2 way), it would be fair to say that music was not quite as dynamic out of the K3 as it was out of my iFi rig. It still sounded pretty good though, and if I was on a really tight budget, the K3 could suffice in a pinch. I also tried the HD800S balanced, and again while volume wasn’t an issue – they sounded decidedly flat, and this was far more noticeable than with the HD600s.
With earphones and IEMs, the K3 was pretty impressive. It easily handled HiFiMan’s (60 ohm 103 dB/mW) RE2000’s, and the bass output especially was very well controlled. Switching to the 320 ohm Zen original and Zen V2 both appeared well driven, although I admit that with the bass boost engaged, they were a much better sounding ear-bud (perhaps slight evidence of the K3 needing just a little more oomph?). Mine aren’t balanced terminated, so it was hard to say if the extra power balanced would have helped.
Even the notoriously harder to drive MEE P1 (50 ohm, 96 dB/mW) were easily handled at 2-3pm on the pot on low gain, and again the overall dynamics were everything I expected froma well driven pair of IEMs.
My final test was with the highly sensitive SE846 (9 ohm and 114 dB/mW) which apparently hiss monsters with a lot of sources (I can’t hear this because of my permanent tinnitus). My daughter – with her highly sensitive hearing helped me with this one, and even with the volume at max (low gain, single-ended) the K3 was extremely quiet – nice job FiiO.
I’m going to preface this section with a little critique I received a while ago, and my answer to it – so that you can understand why I don’t comment on some things, and why I do comment on others. I was told my review on another amp was poor because I didn’t include sections on bass, mid-range, treble, sound-stage, imaging etc – yet referred to an amp as warm, full, or lean.
Now I can understand the reference to warm / full / lean – as they are very subjective terms, and whilst I’d like to avoid their use, they are invaluable to convey true meaning. Comparing my NFB-12 to the Aune X1S for example – the Audio-gd does sound richer and warmer. It’s the nature of the DAC and amp which is used.
But I choose not to comment on bass, mids, treble, and most definitely not sound-stage – simply because when we are talking about a DAC/amp – IMO they shouldn’t be discussed. A DACs job is to decode the signal in as linear fashion as possible, and the amp’s job is to amplify the signal with as low distortion as possible. If the device is doing its job properly, there is no effect on bass, mids, or treble – except if hardware boost is concerned. And IME an amp does not affect sound-stage (unless there is DSP or cross-feed in play) – that is solely the realm of the transducers and the actual recording.
So we have that out of the way how does the K3 perform sonically?
To test tonality, I usually compare (using the same source) with my E17K. The FiiO E17K is one of the most linear devices I own, and is essentially ruler flat from 20Hz to 20kHz. Both devices were volume matched and had my PC as source. With Linux, I can seamlessly move from one DAC to another with just a few seconds delay, so it was easy to switch back and forth (the test was sighted though). For headphones – I used the HM100 – mainly because they are easy to drive, quite neutral.
In subjective comparison, the K3 sounds exactly the same (to me) as the E17K, quite flat, and very clean. Essentially wire with gain – which to me is everything I want a DAC/amp to be.
While the K3 is quite neutral, the last thing I would call it is harsh or clinical, and during my tests with the HM100 and RE2000 this week, I’ve been surprised how many times I’ve drifted away from critical listening to be sidetracked into simply browsing my music library. Part of this (I think) is the level of overall transparency with the HM100 and K3 combo, and absolute fun with the RE2000 and K3. Because of the very clean background (which in my opinion aids a feeling of depth or space if its in the recording), I’ve been captivated and gone on little journeys of discovery. Its why the review has also taken a little longer. Anyway – back to transparency – all the micro details are present, and transient notes are very clean and clear – this is particularly noticeable with the HM100. Micro details are not smoothed or hazy. And imaging is very clean and consistent.
Always a difficult topic with DAC/amps, and especially when you have a device which is desktop oriented, but your comparable devices are mostly portables. But I’ll try with what I have, and the devices I’ve chosen are:
- FiiO’s Q1ii ($100)
- Fiio E17K ($120)
- IMS’s hybrid valve DAC/amp X1 ($270)
I used the devices with the HM100, and all testing was performed single ended. With each test – the same files were used, volume was set and matched with test tones and a calibrated SPL meter. Listening was at my average normal level of 65-75 dB. Please remember that these comparisons are completely subjective and are my biased view of the differences.
FiiO K3 (USD 120) vs FiiO Q1ii (USD 100)
Build / Design / Overall Package
Both are immaculately built devices, with clean lines, feel good in the hand, and appear to be made from quality materials. Both are light-weight (82 vs 101g), and very portable (hand sized). The Q1ii does come with more accessories – but thats because its primary role is more aligned as a portable DAC/amp than a desktop device. Both are USB powered when used as a desktop device, but the Q1ii can also be used portably with its own power (built in battery). Both have the same DAC and OP amp combination.
Both have gain and bass boost (implemented very similarly too). Both are capable of DSD output and same PCM sample rates. Both have balanced output. Both have a variable line-out controlled by the volume pot. The major difference between the two apart from form factor is the built in battery of the Q1ii.
The other major difference between the two devices is in power output. Whilst both are targeted toward easier to drive headphones/earphones, the K3 does have considerably more SE power, and slightly more balanced output.
Sound / My preference
As I stated earlier, both devices essentially use much the same architecture, and they pretty much sound the same. If anything I sometimes hear very slight warmth in the Q1ii, but I’d be the first to admit that it could be subjective sighted bias. Both are much more alike (identical?) than different. One thing is sure – if you’ve heard and liked the tonality of the Q1ii – you’ll feel right at home with the K3. As far as preference goes – mine depends on the use. For a desktop setting, I’d take the K3. For portable I’d take the Q1ii. For portable lap-top only – it would be the K3. Both are excellent devices.
FiiO K3 (USD 120) vs FiiO E17K (USD 120)
Build / Design / Overall Package
Both are again immaculately built devices, with clean lines, feel good in the hand, and appear to be made from quality materials. Both are light-weight (82 vs 110g), and very portable (hand sized). The E17K does come with more accessories – again because its primary role is more aligned as a portable DAC/amp than a desktop device. Both are USB powered when used as a desktop device, but the E17K can also be used portably with its own power (built in battery). They use different chip-sets.
Both have gain (E17K is 3 stage) and EQ, but where the K3 has bass boost only, the E17K gives full bass and treble tone controls. The E17K has the option for digital input as well as USB (via coax), but is limited in out to 24/192 PCM and does not have DSD decoding support. The K3 has a variable line-out where the E17K is fixed. And of course the E17K has the portability being battery based.
The other major difference between the two devices is again in power output. Again whilst both are targeted toward easier to drive headphones/earphones, the K3 does have considerably more power (almost double), and significantly more via balanced output.
Sound / My preference
I tested these two side-by-side earlier and I couldn’t tell a real difference. They both sound neutral to me, and isn’t that the way a good solid-state DAC and amp are supposed to be? What you’re getting with both devices is a clear and clean window to your music, and for me with the right headphones, both are perfect examples of top notch and affordable audio devices. As far as preference goes, I’m petty biased here. Unless I was specifically looking for a desktop device to be used with speakers, I’d be taking the E17K. I use it a lot for testing, and its tone controls are brilliant. But I’d imagine my preferences will be different to most people’s. Again – two examples of brilliant devices – each with their own strengths.
FiiO K3 (USD 120) vs IMS X1 Hybrid Valve Amp (USD 270)
Build / Design / Overall Package
Both are devices with very good use of materials, but I’d be first to admit that the overall finish on the K3 is slightly better than the X1. This time the X1 (with DAC) weighs in at considerably heavier (82 vs 150g), but both are still very portable (hand sized). Accessory package is similar for both, although the X1 does have extra cabling due to its design for portability. Both are USB powered when used as a desktop device, but the X1 can also be used portably with its own power (built in battery). They use different chip-sets, with the X1 utilising an SA9227 master and Burr Brown PCM5102 slave DAC. The X1 also passes all audio via twin Raytheon 6418 valves (tubes). Based on pure aesthetics, the K3 looks cleaner and more modern, while the X1 looks more retro and I’m a sucker for the warm glow of tubes.
Both have gain, but the K3 also has significantly better overall volume control (more range). The K3 also has a bass boost, where the X1 has none. The X1 has line-in (K3 is limited to USB in), but the X1 has no line-out, so can’t be used in a desk-top role with powered speakers. The X1 overall is quite limited in feature-set, but where it shines is in tonality (with the tube output stage). One issue with this though is that The X1 can distort if the valves are over-driven (this can be fixed easily by attenuating the input signal). As far as supported formats go, the K3 does have a wider overall range (up to 32/384 PCM & DSD) where as the X1 is little more limited (24/192 and no DSD).
Despite the tube output, the X1 does not have a lot of power output, and the K3 is more than twice as powerful SE and even more so balanced.
Sound / My preference
First up – both devices sound really good, but there is no doubt that the X1 is the warmer of the two, and I can pick this one blind. The 2nd order harmonics on the X1 give a warm and rich tonality, but not at the expense of sounding overly dark or lacking resolution. Its almost impossible to pick a preference here either as the K3 shines for desktop use and the X1 for portable. For its intended use, the K3 does a much better job, and for less than half the money – for desktop and transportable use, the choice is obvious.
This will come down to intended personal use. If you’re adding the K3 for the specific task of having an easily transportable DAC/amp (especially for a laptop, or for a space limited desktop situation), it really is a lovely little device. I do think FiiO could have squeezed more value out by dropping the digital outs and having either fixed RCA line-outs, or possibly an alternative digital in – but for the overall price and features offered, the K3 really does offer some pretty good value.
FIIO K3 SUMMARY
The K3 has grown on me over the last few weeks of use, and especially in its intended use – as a simple desk-top DAC/amp. The build quality is very good (as most of FiiO’s products are), and the feature set for such a cheap device is pretty good – both bass boost and gain are well implemented.
Add in the balanced output (with additional output power), and the ability to cover virtually every sample rate up to double DSD, and you have a truly versatile device.
Power is pretty good for a device with such a small footprint, but don’t expect miracles – you won’t be properly driving 300 ohm headphones or anything which needs a strong current input. Where the K3 does shine though is in its extremely quiet operation, and very low noise floor, making it ideal for those with more sensitive IEMs.
For desktop use, the ability to have a variable line-out with powered amplifiers is brilliant, and its ultra portability makes it an ideal companion for work lap-tops (especially with driver-less 1.0 USB, or even 2.0 with Win 10 if you don’t care about DSD).
Once again FiiO proves that a device does not have to be expensive to deliver good features and sound. Thanks once again to Lily and the team for allowing me to review this little device.