Micro iDSD Signature is the latest evolution of the iDSD series. Its predecessor Micro iDSD Black Label is widely regarded as a great device which contributed to building out on the expectaction regarding its evolution.
I did take my sweet time on this one as I’m particularly sensible to the topic. If one thing I learnt of this hobby is that source quality comes ages before drivers quality. I won’t anticipate the conclusions but in a nutshell Micro iDSD Signature is a remarkable player, sporting some great and even some unique features.
Table of contents
|Extremly good DAC performance, top class in mobile devices category||Too basic USB dejitter / regen features|
|High quality upsampling applied by GTO firmware sensibly improves reconstruction quality on sub-hires material||Extremely tight transients and supercontrolled bass reconstruction – especially on GTO firmware – may be not everyone’s preference|
|Very good AMP quality in mobile devices category||Too tiny battery status led|
|Unique easy AMP reconfig capability optimally match extremely diverse driver needs||Lack of third party (and even native brand) accessories to fully exploit S-Balanced ports|
|Nice 3D+ crossfeed option||Not inexpensive|
|XBass+ option may be welcome by some to add some bass body back|
The device is easily transportable, but nowhere near to pocketable. Not really eligible if you are looking for a walker. Good as a sitter though.
Micro iDSD Signature exclusively works from its internal battery, it never gets power from the USB VBUS bit.
This is good from the sound performances standpoint as while playing it relies solely on local battery generated power which is apriori less electrically dirty compared to the power coming from an uncontrolled host, or from a budget power supply unit.
On the other hand this means Micro iDSD Signature needs to be charged, and this happens from a separate USB port, a USB-C one.
The battery charging circuit is compatible with quick chargers and protected vs excessive voltage. Based on a direct interview I had with iFi tech people the maximum exploited charging amperage is 1.5A – whatever above that will not harm the unit, but will be wasted.
Next to the USB-C port dedicated to battery charging there is a tiny color-phased led: that is the sole visual indication informing us about the battery charge level.
Battery capacity is above decent. iFi declares circa 12h on Eco mode, 9h on Normal mode depending on load and volume of course. My experience matches such values, give or take.
Lastly, micro iDSD Signature has no “sleep” feature: if you leave it on while not playing batteries will go on discharging.
Micro iDSD Signature has no analog input. So unlike its predecessor Micro iDSD Black Label it can not be used as a standalone amplifier. Which is a pity, as the amp section is not bad at all as you’ll read later.
Two digital input are available: USB and S/PDIF.
The S/PDIF port accepts either 3.5mm coax or Tosink optical connections (a Toslink mini-plug adapter is supplied), but exclusively supports PCM only up to 192KHz sample rates.
The USB port is iFi’s “usual” recessed-USB-A-male connector (same as on Nano iDSD BL, Hip Dac, etc). Depending on firmware, up to PCM 768KHz / DSD512, in addition to MQA, are supported through this channel.
As mentioned above, the USB-A port is for data only and no power charging happens from this end. Which is good, as VBUS is usually a major source of electrical noise and therefore distortion.
The bad news though is that Micro iDSD Signature includes only limited, anyhow insufficient, “USB filtering” features. And it shows.
Ifi’s description talks about an “intelligent memory buffer” relying on a high precision internal clock. The presence of a (legacy) iPurifier circuitry inside is also in the specs. That said, I tested Micro iDSD Signature both natively plugged onto my PC and plugged through my Nano iUSB3.0 conditioner – and the output quality difference is significant. USB dejitter inside Micro iDSD Signature is sadly not something to write home about and this is bad when looking at its DAC module quality (more on this later) which does deserve a much better effort on this front.
My assessment has been conducted using Nano iUSB3.0 upstream. For your curiosity, here you can find some info on Nano iUSB3.0 and the general digital stream conditioning topic.
Micro iDSD Signature offers both 6.3mm and 4.4mm phone analog output, and 2xRCA line output.
The Line output works on fixed parameters: > 2V voltage, < 240 Ohm impedance, > 117dB (A) are the key declared values.
Phone outputs come with some very interesting modulation features such as the option to select 3 different amplification power levels (labelled Eco, Normal and Turbo mode), and engaging a built-in iEMatch circuit. Much more on these down below, in the Amp module section.
Very appealing is also the adoption of a full-analog volume control, technologically offering better quality compared to a (cheaper) digital volume modulation option.
Finally, Micro iDSD Signature offers 2 switchable sound shaping options called XBass+ and 3D+, which I again I will cover in better detail later below.
Power mode selection, iEMatch circuit, Xbass+ and 3D+ only apply to headphone outputs. All of these are totally uneffective on Line output.
Differently to what happens on most similar devices, Micro iDSD Signature provide exactly the same power levels either on its 6.3mm or on its 4.4mm phone out ports.
This is evidently not unrelated to the fact that both ports are actually linked to the same “S-Balanced” internal circuit.
S-Balanced is the name of some iFi’s technology, short for “Single-ended compatible Balanced”. iFi also adopts it inside Pro iCAN, xCAN, xDSD and Nano iDSD Black Label. Refer to their own whitepaper for a nice technical description.
Also, if you are not familiar with what TRS / TRRS means, this may help.
Simply put, a cabling scheme is put in place behind both phone ports on Micro iDSD Signature:
- When plugging TRS plugs – the port delivers “normal” single-ended output. All single ended drivers on the market will seemlessly work in there. In addition to that, thanks to how internal cabling is designed, they will also get 50% reduced crosstalk compared to what they would get from an ordinary single-edend port – for free.
- When plugging TRRS plugs – the port delivers full “balanced-ended” output to balanced-cabled drivers, resulting in quite apparently cleaner and more dynamic sound.
I’m a strong supporter of the S-Balanced concept. So much so that I think iFi should dedicate more attention to it and better “close the loop” in terms of offering their users all the tools needed to fully exploit their technology. Let me explain.
For how hard I tried, I never found a 6.3 TRRS M connector available for purchase. Nor I would know where to buy a 4.4mm TRRS M to 3.5mm TRS F adapter, for that matter. iFi themselves do not offer any of such adapters on their options catalogue – and this is really odd to say the least.
Long story short, while and right because I understand the value of the S-Balanced option included on iFi Nano iDSD Black Label much better – where no 4.4mm nor 2.5mm native-balanced port is available – in the Micro iDSD Signature case I think no user possibly can take advantage of the connectivity flexibility opportunity, which will rest as an unexploited value.
So in daily practice the user will expoit Micro iDSD Signature’s balanced-ended and single-ended phone outputs “the old way”, “as if” they were a single-ended-only port (6.3) and a balanced-ended-only port (4.4).
Unlike what happens on most competition the user will get equal power from either port, and very similar cleannes too (the S-Balanced circuit behind Micro iDSD Signature’s 6.3 port will deliver uncommonly low crosstalk to single-ended loads as per design – you did read the whitepaper didnt you?).
The DAC module
Like most if not all other iFi DAC devices, Micro iDSD Signature can run a range of firmware variants, each offering specific features or optimisations. I find iFi’s approach of leaving the user free to choose amongst such different options a very welcome added value.
Firmware packages and the apps required to flash them are freely available on iFi’s web site, here. The flashing process is really easy and straightforward, at least on Windows platform (did not try on Mac).
The 3 significant versions to choose from for Micro iDSD Signature are:
|Supports||Does not support|
|7.0||MQA, DSD up to 256 on Windows, 128 on Mac, PCM up to 384KHz||DSD 512, PCM 768 KHz|
|7.0c||iFi’s proprietary GTO filter, MQA, DSD up to 256 on Windows, 128 on Mac, PCM up to 384KHz||DSD 512, PCM 768 KHz, GTO on S/PDIF input|
|7.2||DSD up to 512 on Windows, PCM up to 768KHz||MQA|
I’m not much into DSD (I’ll explain why in a later article maybe) and I don’t de facto currently own nor plan to own music files sampled above 192KHz, so the two options which get my attention are 7.0 and 7.0c.
Their fundamental difference is one only but a significant one at that: with 7.0c iFi’s own GTO (Gibbs Transient Optimised) filter replaces Burr Brown’s native reconstruction filters.
I strongly recommend you read iFi’s whitepaper about why and how this may be technically desireable, or not.
Did you read the paper? C’mon do it! Seriously…
As you’ve seen the paper focuses on throughly illustrating GTO’s output features while leaving another important aspect in the background: with 7.0c Micro iDSD Signature will systematically upsample all digital input coming from the USB port up to 32 bit / 384KHz resolution prior to feeding the DAC chips. For what I seem to have understood this is fundamentally required for the GTO filter itself to work as intended.
It’s at this point worth noting (or remembering, for those who follow my articles) that I already experienced iFi’s GTO implementation in conjunction with Micro iDSD Signature’s smaller sibling, the Nano iDSD Black Label.
My full take on that is here, but in short: on Nano iDSD BL the GTO option “sounds worse” than the native ones – for my tastes at least. My suspect is in that case the upsampling effort ended up not adequately turned into higher sound quality delivery due to inherent dac bandwidth limitations.
Micro iDSD Signature on firmware 7.0 offers very nice DAC performances.
Range is superbly extended, sub-bass is fully and correctly rendered, bass is bodied and especially phenomenally controlled, mids are present without exaggerations and trebles are powerful and vivid.
Particularly significant are cleanness, note separation, imaging and layering.
Directly compared to my reference DAC which is Apogee Groove (my take here), Micro iDSD Signature on firmware 7.0 has a deeper lowend extension on one side, a less extended treble span on the opposite side. Tonally it comes accross even more controlled than Groove on the bass (up to delivering a “leaner” flavour there), very similar on the mids and trebles. Draws on 3D space very, very well, although still not precisely at Groove level. On space rendering alone, Micro iDSD Signature is the single DAC that comes closer to Groove that I heard as of yet, and that’s saying quite something.
A very evident feature of Micro iDSD Signature DAC which is worth underlining is tight transients.
The “Bit Perfect” option is very tight. All notes are snappy, razor cut. Switching onto Minimum Phase or Standard transients get a tad more relaxed, less “dry”, yet the general impact stays way into “analythical” territory, especially when compared to a more “musical” alternative e.g. Apogee Groove.
In terms of transients rendering Micro iDSD Signature is actually more on Chord Mojo ballpark – for those who have experience with that. Mojo stays a bit ahead of Micro iDSD Signature on its unique capacity to close the gap between front and back instruments, however Micro iDSD Signature provides quite evidently better results in terms of extension, lack of coloration, and detail, and seriously beats Mojo on space rendering (which never was Mojo’s specialty, there’s that…).
With firmware 7.0c Micro iDSD Signature will upsample all digital traffic incoming from the USB line prior to passing it to its DAC chips.
While this does not produce any “dramatic” difference when the original samples come at an already high resolution (96, 176 or 192 KHz), an evident improvement is audible when 44.1 or 48KHz material is being supplied, and especially so of course if the track mastering is good.
The improvement is of course mainly on spatial reconstruction, size shaping and imaging.
Pulling back the comparison vs Apogee Groove, firmware 7.0c comes out as a major point of advantage for Micro iDSD Signature. Both DACs show great mastery beyond the 40KHz mark (where most of the information on space, reverberations etc come from) but that only is applicable if there actually is some digital data in that region to be decoded. On native 96KHz++ material the two DACs compete head-to-head. On lower sample rate material Micro iDSD Signature offers a built-in, automatic and very well implemented (!) upsampling option, while Groove has to rely on the same happening on the host (if ever.
Another interesting note: Micro iDSD Signature is very much able to exploit the GTO upsampling and convert it into a tangible benefit to the user when busy with sub-hires music at the very least, while the same does not happen on Nano iDSD Black Label. Why? I don’t know. I suspect this is either because the GTO algorithm is better implemented on fw 7.0c (usable on Micro iDSD Signature) vs fw 5.3c (usable on Nano iDSD BL), or because Nano’s lower-tier DAC section is unable to exploit the opportunity. Or both.
Another very important note to make about GTO is that it renders transients even tighter compared to the already tight Bit Perfect option on non-GTO 7.0 package. And, there’s no escape: when 7.0c is installed only one filter is available – no Minimum Phase or Standard alternatives are attainable. You better like it as is…
And simply put, I don’t. I find it excessive. I do appreciate of course clarity and cleanness, and precise rendering of each note – it’s a matter of “levels”. This is of course totally subjective. I have friends considering GTO’s reconstruction “supremely natural”. I’m afraid I can’t anticipate which one you will prefer.
One dufitul last note: I don’t use MQA, so I did not try / test Micro iDSD Signature’s proficiency on that. Not big loss for you as due to what I just wrote I honestly haven’t got any decent experience / opinion to offer on this topic, other than the trivial comments you can easily find everywhere and don’t certainly need me to paste here.
Summarising: Micro iDSD Signature offers very, very good DAC performances. Imaging and spatial drawing in particular are nothing short of spectacular. GTO firmware offers automatic well-executed upscaling to sub-hires audio tracks. Transients are rendered from quite to very tightly, which more musical-sounding presentation lovers might not like.
The AMP module
As I quickly pointed out up above, Micro iDSD Signature offers some very interesting features when it comes to its internal amp section.
Firstly, its internal S-balanced architecture is equally available on both the 6.3 and 4.4 port, although for the reasons explained above in the real world scenario you can bet the two ports will be used as if the former was single-ended only, and the latter balanced-ended only. Too bad.
Secondly, and obviously related to the previous point, both ports provide the same output power levels – so choosing either is not a matter of power delivery, rather of convenience, and of a little cleaner output (better xtalk value) when a balanced connection is established on either port – so de facto on the 4.4 one for lack of avaialble 6.3 trrs connectivity options.
Finally, it allows for power level reconfiguration at the click of a switch to optimally supply high impedance cans – requiring as high voltage swings as possible – or low impedance and low sensitivity ones, like planars – requiring little voltage and very intense currents – or even IEMs. Three different “powering levels” are available, selectable with a switch on the device side, and a built-in IEMatch module is also available for added measure.
At the Normal power level – the intermediate one of the three total – Micro iDSD Signature outputs something short of 2W max power (a bit less than 1W RMS @ 32 Ohm) and 5,5V max swing. That’s already great power.
My Shure SRH1540 (46 Ohm 99dB, but much more current hungry than spec facevalues tell) are biased very satisfactorily on Normal level: bass is full while also staying very controlled, trebles are nicely sparkly – within the limits of a treble-polished driver like 1540 of course.
Indeed, switching iEMatch on makes the situation even better as it tames a further bit of the unneeded voltage swing, making bass come out futher controlled and cleaner. A real pleasure to hear. IEMatch High setting is already sufficient in this case, which is consistent with SRH1540 impedance being still significantly higher than IEMatch-H’s 2.5 Ohm output impedance.
Switching Micro iDSD Signature to Turbo is not a good choice for SRH1540 – and that’s perfectly in line with logic too. The effect of excessive voltage swing on a low impedance load like that is similar to a taxing “Super High Gain” option: dynamics get closed-in, range is compressed, highmids get glary and expecially midbass goes too bloomy.
Same situation takes place with Koss KPH30i (60 Ohm 101 dB). Best biasing is obtained from Normal power + IEMatch-High. Eco power makes them sound muddy(er). Turbo power is totally excessive, KPH30i presentation gets unnatural.
As those who follow my articles probably already know, I’m not into high impedance cans, nor into extremely power hungry planars so I won’t comment on how the Turbo option does in those cases. Suffice here to say that Turbo setting promises 10V max and a tad more than 4W max power (1.5W RMS @ 64 Ohm), not peanuts at all for a portable device and many budget desktop ones either!
On the opposite end of the spectrum, I must stand and vigorously clap hands at how iFi solved the IEM equation instead: the combined benefits of the Eco power option and the built-in IEMatch module are nothing short of spectacular.
As all technically aware audiophiles know, IEMs despise high voltage and most times high power altogether. Impedances below 32 ohms – sometimes as low as 8 Ohm or less – like those on IEMs require (require, not simply prefer) low voltage swings, which is the exact opposite of what a voltage modulation amp powerful enough to drive higher impedance headphones is designed for.
When up to driving IEMs the very first thing to do on Micro iDSD Signature is to set the power switch to Eco setting. This makes sure the maximum delivered voltage is limited to 2V, and the maximum power is 500mW on an 8 Ohm load, which is way more than 99.9% of IEMs out there requires.
All of my IEMs are in facts perfectly driven by Micro iDSD Signature, on Eco mode and – on a case by case basis – with IEMatch set to Ultra or High. Same for my final Sonorous-II (16 Ohm 105 dB) which are closedback overears electrically behaving very much like IEMs: the best setup for them is Eco, with IEMatch turned off, although Normal + IEMatch Ultra is also a strong contender.
I am Soon™ going to publish a comprehensive article about IEMatch but I guess it’s worth to synthetically recall what IEMatch is here.
In its standalone incarnation iEMatch is a device to be plugged in between an amp’s headphone port and a IEM or Headphone cable, and vulgarly said it does 3 things:
- It “tricks” the amp into sensing a predetermined load impedance of 16Ω, regardless of the IEM/Headphone’s real (average) one.
- On the opposite end it also “tricks” the IEM/Headphone into sensing a predetermined amp output impedance, regardless of the amp’s real one. The user can flip a switch and choose between 2.5Ω or 1Ω.
- It attenuates – think about it as if it “sinked” – the amp’s output by a predetermined amount: -12dB when output impedance is set to 2.5Ω, and -24dB at 1Ω
Micro iDSD Signature already comes with a number of features making some of its IEMatch built-in circuitry redundant, read useless: I’m talking about very low output impedance, support for loads as low as 8 Ohm and a volume pot in the analog-domain. That said, the IEMatch module stays beneficial to 2 main purposes: further reduce output voltage swing, and eliminate sibilance on hyper-sensitive IEMs.
The former of the two benefits is especially intriguing: on a case by case basis some drivers do sound better under Eco power other sound better under Normal power with IEMatch Ultra switched on on top. And for some others… it’s hard to tell – e.g. final Sonorous-II, Tanchjim Darling, final E5000.
Mind you: from a purely technological standpoint the topic is arguable upon to say the least. Post-attenuation does come with benefits and drawbacks like, I guess, everything else in life. The information I want to convey here is not how exactly I recommend to set Micro iDSD Signature to work best with this or that IEM, rather that the device offers the flexibility to try different ways to that target.
A truly multipurpose amp – finally!
As I mentioned before, to my knowdge at least there’s no single amp implementation out there yet which is capable of feeding different impedance/sentivity drivers with equally optimal results. And to my understanding, there’s a solid technological reason for this : high impedance cans “sound better” when submitted to high voltage swings, low impedance HPs and IEMs distort in the same condition. Simple as that.
Conceptually we have 3 possibilities then:
- Use different amps for the various cases.
- Use a high voltage amp, and aposteriori cut its output voltage down. That’s where attenuators e.g. IEMatch may help.
- Use a sort of “configurable” amp…
Micro iDSD Signature’s amp can be reconfigured to optimally support very high impedance cans (Turbo mode) or low sensitivity cans (Normal mode) or low impedance + high sensitivity IEMs (Eco mode) offering each category its own best welcome powering profile.
When facing low impedance loads, in addition to selecting Eco power mode there may still be need to engage the built-in IEMatch module to furtherly reduce output voltage, which will benefit current delivery to particularly sensitive drivers, and/or cancel some hiss out of the most sensitive of those.
Of course a curious question as this point might be “which of the two features should I preferably use before the other: Eco power mode, apriori limitating voltage swing, or IEMatch, cutting it down aposteriori” ? I had my own opinion on that already but I asked iFi’s designers’ take on this.
With regards to Turbo/Normal/Eco modes vis-a-vis headphone matching, our AMR audio background means that we are of the opinion that while the impedance of a headphone is a factor to consider in matching, the over-arching one is actually power > sensitivity.
This is why we developed the headphone calculator which takes the power output of a headphone amplifier and compares it to the headphones to be used. The resulting volume level will give the customer the best insight into the ‘matching’. Just like a 1,000W Mark Levinson would be a poor match for high-sensitivity horns rated at 110dB sensitivity. Or a 10W 300B SET would not drive 87dB Magicos.
iFi’s headphone calculator is indeed an informative but most of all educative tool. Playing with it we can find out how wrong are common assumptions about this or that driver (IEMs or Headphones alike) requiring “high power”… 😉
I can’t begin to stress how brilliant I find iFi’s choice to equip Micro iDSD Signature with what it takes to allow the user to substantially change its amping behaviour to cope with the dramatically diversified nature of those drivers out there. A really, really welcome idea, and a unique one in the mid-tier segment this device partakes into at the very least, but to my knowledge in the one above too.
Sound shaping addons
Similar to what is also offered on other iFi devices, Micro iDSD Signature’s amp section features two optional circuits providing bass enhancement and imaging improvement at the flick of a switch.
Both features are according to iFi’s documentation entirely implemented in the analog domain. No DSPs are involved which promises the minimal impact on sound quality of course.
“XBass+” behaves like what an EQ expert would call a low shelf positive filter. By ear I would say it pushes lows up by 2dB-ish from 100Hz down. Very personally speaking, I don’t like these types of options in principle – irregardless of their implementation quality that is – my fundamental position being: if I want more bass than the one delivered by the driver I’m using right now… I swap on a different driver. In the special Micro iDSD Signature case, XBass+ may be actually welcome to “compensate” the device super-lean bass presentation, especially as delivered by the GTO filter (fw 7.0c).
“3D+” is a “crossfeed filter”, i.e. a function that puts “some” of the right channel output into the left one and viceversa, simulating on headphones what happens when listening to loudspeakers. Within its limits (it’s not parametric, configurable etc – just a mere on/off) and situationality (effects are totally evident on some tracks, minimal on others) the trick is really nice, and I used it quite often. My main application case are those original jazz masters from the 60ies where mixing tended to be executed by hard panning each instrument on a single channel only: 3D+ sounds almost magical in those cases.
Conclusions and evaluations
Well, I’m certainly not known for generosity when it comes to assign good scores. This time I must say that from the technical standpoint Micro iDSD Signature deserves a very high mark.
DAC reconstruction quality on high-res lossless material is at the absolute top I found below 1K$ and possibly above. The sole other devices who can play an even match with Micro iDSD Signature on this part are Apogee Groove and Questyle QP1R, all the rest being a full class below at the very least.
Not only: adopting firmware 7.0c Micro iDSD Signature automatically upsamples all USB input thereby granting a significant share of the same quality to sub-highres tracks. Upsampling is not black art, it is very possible to implement that on the host, and feed competing DACs with the same improved input, and obtain improved output from those too – but it must be done, and done right, which is an extra burden and cost, while Micro iDSD Signature does it “out of the box”, and does it right too, and this is a major value on my scorebook.
Micro iDSD Signature’s DAC is so good that a no brainer rec is to exploit the Line Out option and interconnect into a serious quality desktop amp any time that’s possible (nothing short of a Jotunheim-2 at the very least will make Micro iDSD Signature’s DAC decent justice), but mobile DAC+AMP standalone operation will not disfigure at all when compared to no matter which top alternative device in its same price class.
Always talking about standalone mobile features, Micro iDSD Signature’s ability to reconfigure its amping section on the fly to optimally cope with IEMs or low impedance cans or high impedance hps or “nasty” planars is totally brilliant, and deserves – that alone – a standing ovation for the idea, and the implementation quality too.
As always where there’s light there’s some shadow too. The DAC section is so good that the lack of adequate built-in USB dejitter is very evident, and frankly I find it almost disappointing. On an even more subjective level, all filters in general and the GTO filter in particular deliver extremely tight transients – some may find them “more natural”, I find the opposite.
Last but not least, the price – and the value. At € 699,00 EU list price Micro iDSD Signature is not an inexpensive device. And please add another € 50 at the very least (iSilencer) or better another € 150 (iPurifier3) to dutifully add some very deserved USB cleansing.
Is it worth it?
Well, evaluating it in terms of a truly mobile (if not pocketable) device, not relying on host batteries, offering top class DAC competence, and truly capable to optimally bias anything, from the “easiest” IEM up to the “nastiest” planar overear well… Micro iDSD Signature is an easy win.
Such consideration does not make its price tag cheaper of course, but at least for my experience finding another battery driven device with similar output quality at a significantly lower price – at least to my knowledge – is today a hard task.
Totally different is of course the perspective if we plan a mainly static, “desktop” application. In such case Micro iDSD Signature stays a very significant device, but the price of its unexploited portions would make its cost/result score poorer.
This Micro iDSD Signature device has been provided as a temporary loaner unit by iFi for the sole purpose of my assessment.
This article was previously published on audioreviews.org