GO Bar is iFi Audio’s first and long awaited take on a battery-less mobile dac-amp (a “dongle”).
As I will try to outline, from multiple standpoints GO Bar fully partakes into iFi’s overall general product philosophy both in terms of components and general application. It promises outstanding results, and surely comes with a price tag (€ 329 in EU) positioning it at the top market level. It can as always be purchased directly from iFi, on their Amazon space, or from one of their distributors around the world.
Table of contents
- At-a-glance Card
- Features and description
- Other features
- Sound and power
|Very good power delivery on mid and high impedance drivers||High host power demand, not USB2 compliant|
|Above average power delivery on low impedance drivers||Host Volume Sync buggy on Windows, not supported on Linux.|
|Good balanced output sound quality||Unimpressive single-ended output sound quality|
|Four selectable reconstruction filter options||“Yellow” LED light too easily confused with “White”|
|Built-in IEMatch||Uncommon IEMatch settings, limited benefits|
|XBass and XSpace features|
|Selectable High Gain|
Features and description
GO Bar’s housing shape recalls that of their Nano / Micro iDSD line – just many times smaller. Within the “dongle” category, GO Bar falls more or less midway in terms of dimensions and weight: 65x22x13mm and most of all 28.5g are not certainly huge yet not entirely disregardable either when I have GO Bar connected to my transport. It’s more than twice the size and weight than an E1DA 9038S, yet much smaller and lighter compared to a Groove.
On one small ends we find two phone outs, on the opposite end a USB-C connector. On one of the long sides there are two + and – buttons, another multifunction button, and a microswitch dedicated to IEMatch.
Finally, on the bottom face there’s a raw of 9 little holes corresponding to as many LEDs, with engravings already hinting about those being dedicated to signaling which PCM sample rate is being played, or if DSD is being received, whether MQA is being detected and finally wheter XSpace and/or XBass+ are activated.
GO Bar is supplied in a small size carton box with adequate internal protections. The bundle is quite generous insofar as it includes a USBC-USBC short cable, a Lightning-USBC cable, a USBC-USBA adapter and a black leather pouch with enough room for the GO Bar itself plus at least one of such cables.
iFi traditionally offers quite some macroscopical information about the contents of their devices, but scarce or no fine details about them.
USB communication is taken care by a XMOS 16bit chip, which I would call an obviously good – for quality – and at the same time brave choice – when I think to the power that element alone requires to operate.
Unlike many other iFi DAC devices, “traditionally” equipped with Burr-Brown DAC chips, GO Bar is built around a 32bit Cirrus Logic DAC chip. The rest of the components are coming from the usual manufacturers normally enlisted by iFi: Murata and TDK capacitors, and Texas Instruments power supplies. I won’t go too much down through this as, in lack of better specifications coming from the manufacturers, I find this information to be useful essentially at the marketing level only.
GO Bar offers a single input, being the USB-C port which is therefore supposed to carry in both digital data and power.
The range of accepted digital resolutions is nothing short of extended:
- PCM up to 32bit / 384MHz (ASIO drivers required on Windows to exceed 24bit / 192KHz)
- DSD direct up to 256 (DSD-direct requires ASIO drivers on Windows)
GO Bar is also an MQA full decoder. That means that it can both unfold non-authenticated MQA tracks, and (automatically) pre-authenticate with the MQA provider to guaranteer even higher quality and especially fidelity vs the digital content distributed by the publisher.
On the housing’s bottom face there’s a generous stripe of LEDs flanked by engraved markings.
One of the topmost 6 LEDs of the stripe will light up in white to indicate which digital resolution is being received by GO Bar: 44/48, 88/96, 172/192, 352/384, DSD64/128 or DSD256.
The 7th LED will light up of different colours depending on the situation:
- when an MQA stream is being received : green for MQA, blue for MQA Studio and magenta for “MQB” (Original Sample Rate).
- when a PCM stream is being received: cyan for BP filter, white for GTO, red for STD, yellow for MIN and off in case of DSD stream. (Read more down below about filters).
GO Bar offers 2 distinct phone out ports: a 4.4mm Balanced and a 3.5mm Single Ended one.
iFi’s web site is quite stingy on information about GO Bar’s output power so I asked them directly. Here’s two tables giving a much more complete picture of the situation:
|Load impedance (Ω)||Output Power (mW)||Output Voltage (V)||Output Current (mA)|
Single ended output
|Load impedance (Ω)||Output Power (mW)||Output Voltage (V)||Output Current (mA)|
So it’s quite clear that GO Bar offers very nice output power on high impedance loads, while it is severely limited in terms of current delivery which turns into not particularly enticing figures from 16Ω down.
Don’t take me wrong here: GO Bar is still more powerful on low impedance loads, and delivers more current on medium impedance loads than many other “dongles” out there. However, GO Bar’s marketing tag line (“World’s most powerful headphone amp of its size”) is, simply, not true when taken in the absolute terms used by its very wordage.
Looking at the figures, it comes out that the device has a hard max current cap around 130mA which it reaches on a 32Ω (or thereabout) load, and to stay at safe distance from that it’s tuned such that drawn current is progressively reduced as load impedance goes down.
Running some simple math it’s also quite clear that GO Bar issues its max power on Balanced Ended output vs circa 55Ω impedance, that being a bit less than 1W, and from there on down it rapidly starts winding down both voltage and current flow.
Neither of the two output connections is configurable as a pure Line Out.
Output impedance on both BE and SE ports are declared to be equal, both < 1Ω. A nicely low value, although not a superlow one.
I also find it interesting to note that SNR and THD+N promised values on BE and SE ports are dramatically different. On BE port iFi declares 132dB(A) SNR and <0.002% THD+N. The former value in particular is really outstanding.
On the SE port they instead declare SNR at 108dB(A) and <0.09% THD+N which I would both call “unimpressive” to be generous. And match unenticing subjective audition experience (more below).
Lastly, the 3.5mm phone out on GO Bar offers iFi’s proprietary “S-Balanced” connectivity – which is a sort of superset of the usual 3.5mm Single Ended standards, offering some more cleanness, and full backwards compatibility to all existing 3.5mm terminated drivers.
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 drawing may help.
Simply put, the S-Balanced “special” cabling scheme behind GO Bar’s 3.5mm port works as follows:
- When plugging TRS plugs – the port delivers “normal” single-ended output. All single ended drivers on the market will seemlessly work in here. 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.
If we except the case of fixed-cables earphones carrying TRRS 3.5mm plugs from factory (I don’t personally know of one), exploiting GO Bar’s 3.5 TRRS option would require swapping cables, and as such pretty much anybody at that point would swap onto a 2.5mm or 4.4mm terminated one, and exploit the full-blown Balanced port (the 4.4mm one) on the GO Bar.
So in the case of GO Bar – much the same as in the case of Micro iDSD Signature or GO Blu – I guess that the practical value of the S-Balanced technology applied behind the 3.5mm port is limited to the xtalk improvement – which is nevertheless nothing to bin.
Host power requirements
I didn’t analythically measure GO Bar’s power requirements, yet there are quite a few things that can be said based on iFi’s published data, and subjective experience.
First: iFi declares a (maximum) host power draw of 4W. That’s huge. It corresponds to 800mA which is far beyoned USB 2.0 limits.
That’s a maximum absorbtion figure – GO Bar will work on usual smartphones and tablets, and even on iPhones, but if the host (the smartphone, the tablet, or the PC) is not capable of delivering up to 800mA current / 4W power then GO Bar’s effective powering capabilities will be limited by the host’s cap. Or oppositely stated: to get the full power “promised” by its specs, GO Bar must be hooked onto a USB3-capable host, or a smartphone compliant with USB PowerDelivery standard.
Second: GO Bar does drain the smartphone’s battery quite fast, even when not “pumped to the max” power by the way.
As my FifteenReaderstm know, the above is no big concern to me as I don’t use my general purpose smartphone as a transport, rather a separate dedicated device, paired to a dedicated small powerbank and wired with an appropriate custom cable. Nevertheless my personal use case is quite evidently not so common, so the average user looking into adopting GO Bar as a classic “dongle” to be paired to his main phone while commuting should keep its power absorbtion needs in due count.
Volume and gain control
GO Bar offers the user both the option to control volume by pressing the + and – buttons on its housing, and software syncing with the host’s OS.
When changing volume via +/- buttons the LEDs on GO Bar’s bottom faceplate briefly light up to give a visual representation of the volume level.
Host Volume Syncing is supposed to make such that when moving the system volume slider on the host (the machine onto which GO Bar is connected, be it on Windows, MacOS, Android…) then GO Bar’s internal volume changes, and viceversa changing Go Bar’s volume by pushing the +/- buttons will change the host’s volume.
Host Volume Sync is off by default. To turn it on one needs to keep the multifunction button pressed >5 seconds. The switch on is confirmed by an animation played on the upper 6 LEDs on the bottom faceplate. Same procedure to turn it off : keep button pressed >5 seconds, and a (different) animation gets played on the LEDs to confirm.
For my direct experience, Host Volume Sync works as intended on my Android devices, and on my proprietary-Linux small transport, the Tempotec V1. It works “erratically” on my Windows 10 laptop. It does not work at all on any of my different-distro Linux boxes.
After inquiring with iFi’s tech support I got no fix for the Windows problems, and I got confirmed that Linux is indeed not supported “by design”. This is bad, as this de facto prevents GO Bar to be used in pretty much any Linux Client-Server configuration e.g. in a Roon, or LMS, or similar infrastructure, unless by slamming its volume to 100% via its HW buttons, and then actionate on host’s Digital Volume which is of course far from ideal in terms of output quality.
GO Bar also has a +6dB High Gain mode. Unlike what happens e.g. on the GO Blu where gain selection follows an automatic system, on GO Bar it’s the user who has to manually set the device on High or Low Gain mode.
To toggle Gain H / L one needs to push both + and – volume buttons at the same time for >2 seconds. Similarly to the Host Volume Sync case, the “usual” uppermost 6 LEDs on the device’s bottom faceplate will play an animation to indicate the action has been carried out.
MQA Full Decoding
I won’t spend a word on what MQA itself is, of course. Google around if you wish and you’ll be overflooded with info.
What matters here is: GO Bar is a “MQA Full Decoder”. This means that GO Bar not only can (like any “MQA Renderer”) fully unfold MQA tracks on its own hardware, but that it can also:
- Authenticate the provenance of MQA tracks.
- Authenticate the Original Bit Rate of the MQA tracks.
I’ll try to vulgarise the rationales about such extra features.
Singers/players/bands/publishers record their tracks, and eventually release their albums. Prior to the digital music distribution era, there could be very little doubt about whether the music we were listening to was the “original” version of that album as its creator/publisher intended or not; if we had a legit copy of that LP or of that CD, that was it.
In the digital music distribution system, instead, the end user has no “solid” way to make absolutely sure that he’s receiving an unaltered version of those tracks. For what he knows, he might be getting a subsequently remastered, equalised, anyhow manipulated version of that album.
The MQA offers a way to “certify” this. An “MQA Studio” track is a file which containes some sort of “certification codes” that guarantee that track is indeed “the original” as released by the authors. A sort of digital signature, if you wish. Anyone might process, EQ, remaster, etc, that track, and re-encode it under MQA but the new file wouldn’t carry the original author signature anymore.
“MQA Original Sample Rate” (a.k.a. “MQB”) tracks are MQA Studio Tracks for which a further certification is given that not even the mere sample rate has been altered (in particular: oversampled) compared to the “original version” as released by the authors.
Any MQA-capable device can play back all MQA encoded tracks, but only MQA Full Decoders are able to identify such additional “digital signatures” and tell the user “hey, this is an original track” or not.
Ifi GO Bar, Gryphon, HipDac-2 are all Full Decoder devices. Ifi HipDac, Micro iDSD Signature, Nano iDSD Black Label are Renderers.
Between parentheses: HipDac and HipDac-2 being virtually identical in terms of sound capabilities, power, etc, with the sole major difference represented by their different MQA capabilities, offered me the interesting opportunity to check the differences on a quite similar if not virtually identical situation and I could tell a quite obvious SQ improvement when listening to a few particular tracks just Rendered (HipDac) or Full Decoded (HipDac-2).
That said, I don’t personally care about MQA, nor about any of the existing digital distribution catalogues for that matter, due to the fundamental lack of good editions of the music I prefer on there.
XBass+ and XSpace
“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.
“XSpace” 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 use it quite often.
My main application for XSpace are those original jazz masters from the 60ies where mixing tended to be executed by hard panning each instrument on a single channel only: crossfeed is almost magical in those cases.
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.
To turn either or both features ON or OFF all is needed is short pressing the Multifunction button on GO Bar in sequence. The two bottom LEDs on the stripe on GO Bar’s bottom faceplate will light up or down accordingly.
Alternative reconstruction filters
Reconstruction filters are an extremely technical topic and some fundamental knowledge needs to be acquired to even start to understand what the heck are they about.
If you are technically inclined a good starting point for your homework about what are Reconstruction Filters is actually this Wikipedia page.
If you alternatively would love a more vulgarised approach, there’s my article about this (or many others on the web).
Assuming you are at least somewhat familiar with these concepts, let me say that GO Bar offers the user 4 options :
- BP (“Bit Perfect”) – this actually corresponds to not applying any digital filtering. No pre nor post ringing is involved (of course).
- STD (“Standard”) – a modestly fast filter with modest pre and post ringing
- MIN (“Minimum Phase”) – a slow minimum phase filter, with minimum pre and post ringing
- GTO (“Gibbs-transient optimised”) – iFi’s proprietary filter, very fast, with no pre ringing and little post ringing. When GTO filter is applied all digital input is automatically up sampled to 352 or 384KHz prior to decoding c/o GO Bar’s internal hw.
To set the preferred filter, hold the Multifunction button down >3 seconds. The MQA LED on GO Bar’s backplate will start flashing of a specific colour. Short pressing the Multifunction botton once will cycle through the 4 possible filters, and the LED will start flashing of a different colour. Long pressing the Multifunction button again >3 sec will “set” the filter and keep it selected until the procedure is done again.
LED color codes are as follows:
- Cyan – BP filter
- Red – STD filter
- Yellow – MIN filter
- White – GTO filter
The first 3 options really sound like 3 of the 5 standard options made available on stock Cirrus CS43131 chips (full spec sheet here)
The “BP” option (Cirrus’ “NOS” filter) – will of course avoid the slightest risk of introducing any ringing, at the cost, however, of leaving artifacts all there where they are. Conceptually not recommended for redbook (44.1/48KHz) tracks, becomes a viable alternative to assess for highres (>96KHz++) tracks. But that’s theory if you ask me: you *do* want to cut the high frequency crap out. Always. I am not using this option myself.
The “MIN” option is a minimum phase, slow filter. More recommended on high(er) res tracks than Redbook ones, it introduces very modest pre and post ringing.
The “STD” option is a step in the direction of a faster filter. More recommended on Redbook tracks, although not really “vertical” as other fast filters are. Ringing is a tad more important then MIN’s one.
Finally, “GTO” is ifi’s proprietary filter called “Gibbs-Transient Optimised. It’s a very fast, minimum phase filter. I strongly recommend you read iFi’s whitepaper about why and how this may be technically desireable, or not.
As mentioned above, selecting the GTO filter also adds a pre-reconstruction systematic up-sampling passage, executed by GO Bar’s internal hardware. All incoming PCM tracks get up sampled to GO Bar’s maximum sample rate of 384KHz (if the original track’s sample rate is 48, 96, or 192Khz) or 352.8 KHz (if the original track’s sample rate is 44.1, 88.2 or 176.4KHz).
In general on higher resolution tracks one tends to prefer slower filters as those don’t risk to cut out on treble air nor spatial cues, their slowness not being a problem thanks to the higher sampling rate. iFi’s GTO filter is a special implementation offering super-fast operation, zero pre ringing and minimal post ringing. This, in addition to the up sampling conducted on the incoming stream, results in very sharp transients and “sculpted” notes. Which – as everything in audio – will then be subject to personal preference!
For my personal experience with iFi devices, and for my tastes, I found the GTO filter viable on all iFi models I tried it onto, including GO Bar, with the sole exception of Micro iDSD Signature, where I found it “excessive”. You may want to read my take about that case here.
For a through description of what IEMatch is, there’s my article which I recommend you to read if you are unfamiliar with the concept or I guarantee you won’t understand what follows.
Much like it happens on many other iFi’s models, GO Bar carries built-in IEMatch circuitry. The implementation does not exactly follow the same specs as the standalone IEMatch devices though.
Firstly, next to the IEMatch switch on GO Bar we don’t find the usual “Ultra” / “High” engravings, but rather “3.5” and “4.4”. The GO Bar manual quite smokily says that “iEMatch reduces the output level, so that even the most sensitive In-Ear-Monitors (IEMs) can be matched to the GO bar”. Which is only a part of what a full-blown IEMatch does. And does not offer precise figures in terms of attenuations nor output impedances to help the user anticipate what he will get by plugging IEMs of specific impedance or sensitivity.
Long story short, I asked iFi’s tech support and they provided me with the following table:
|IEMatch switch position||3.5 output port||4.4 output port|
|Off||Output impedance : <1Ω|
Attenuation @0dB: 0dB
|Output impedance: <1Ω|
Attenuation @0dB: 0dB
|“3.5”||Output impedance: 7.5Ω|
Attenuation @0dB: -6dB
|Output impedance: 7.5Ω|
Attenuation @0dB: -2.5dB
|“4.4”||Output impedance: 3.6Ω|
Attenuation @0dB: -5,7dB
|Output impedance: 3.6Ω|
Attenuation @0dB: -12,5dB
In spite of my repeated requests, iFi didn’t supply me with the other relevant information which is the Input impedance value on all those cases. Or at least they didn’t yet at the time of this article’s publishing. Looking at the figures, and comparing them with those of the standalone IEMatch models, I can only “guess” that input impedances might be in the ballpark of those featured by IEMatch 4.4, so around 40-50Ω. Such “guess” is also corroborated looking at the Single Ended output figures: there, the lower output impedance option does not feature a much higher attenuation as it usually is the case on IEMatch devices, but rather a slightly lower one, with a phase flip involved.
After taking their sweet time, IFI did provide the missing bit of information in the end: the internal IEMatch module features the following internal upstream impedances (the impedance “seen” by the amplifier module when the relevant IEMatch setting is on).
|IEMatch switch position||Impedance applied to Amp module|
Sadly, these figures mean that GO Bar may result not too powerful especially when switched onto its 4.4/4.4 IEMatch output combination, considering the -12.5dB attenuation involved.
Lastly, Going back to GO Bar manual’s recommendation to use the IEMatch switch to attenuate output in order to cope with extrasensitive drivers, looking at the figures it’s quite clear that the rec stands as stated only when the drivers also carry a not very low impedance (ideally, no lower than 30-ish ohm). Low or very low Z drivers (Dunu ZEN, Oriveti OH500…) will show some midbass bump due to reduced dampening, which shall have to be compensated by EQing – or just avoided by plugging a “regular” IEMAtch-2.5 onto GO Bar’s balanced output (via a 2.5-4.4 adapter of course).
Like for most if not really all iFi devices, for GO Bar too iFi makes firmware package availables for the user to download and easily apply.
At the current time there’s only one package available, version 1.7 (in two sub-versions with just a minimal, almost “aesthetic” difference). I do recommend checking that is the version installed on the device when you get it, as the previous one (v1.48) which was installed on my review sample when I got it was quite buggy.
I won’t be surprised if iFi will make more alternative fw packages available going forward, e.g. offering different filtering options as it happens on other iFi models.
GO Bar comes in a small box but with the right bundle accessories, and premium quality ones at that too.
- USBC-USBC 10cm cable
- USBC-Lightning 10cm cable
- USBC-USBA passthrough adapter
Cable quality is apparently top notch.
Same can be said of the black leather travel case, offering enough space for the GO Bar device itself and one or actually both of its USB cables.
Sound and power
GO Bar sounds well, and I should actually remark “very well” indeed, from its balanced output port.
As for voicing GO Bar definitely marks a diversion from that warm and midbass-accented iFi’s “house sound” typical of many other models e.g. Hip Dac, Nano iDSD BL, etc.
GO Bar is much closer to neutrality (although still somewhat into warm-ish territory). Its sound is well bilaterally extended, with very good note body accross the board, good clarity and good detail, with very good but not over-accented bass presence and a good treble rendering.
About trebles it should definitely be noted that GO Bar delivers unoffensive high notes, and a nice, unfatiguing and nicely musical experience on one hand, while staying south of some competitor’s last mile in terms of treble energy and detail retrieval on the other. Pick your poison I guess, and as for all compromises appreciation for iFi’s choices on this will strongly depend on users’ preferences.
From the power delivery standapoint GO Bar is definitely a musclar device, although some notes are in order on this respect.
Regarding voltage swing into very high impedance drivers (600Ω) GO Bar easily promises (and delivers) the highest figure on the “dongles” market today, a whopping 7,2V. That’s significantly higher even compared to Apogee Groove’s 5V on 600Ω. Ifi does not declare (and I couldn’t measure) the swing on 300Ω (Groove’s stays just a bit below 5V there).
Truth be told, as most if not all high impedance cans are equipped with dynamic drivers, I’m not sure to understand what the purpose of a 7V+ swing really is (“stunning” spec sheet figure apart, I mean…).
GO Bar delivers circa 1W onto a 55Ω load (always talking about the Balanced Ended output), which is definitely a huge lot for a dongle, and why it drives the likes of Shure SRH1540 wonderfully well, and SRH1840 near perfectly, too.
It delivers circa half Watt into a 32Ω driver, which is really a lot in a sense, indeed overkill for most DD, BA or other technology IEMS out there, yet (!…) not enough for higher demanding planars, which require even more current and/or they require it at lower load impedance values.
Going further down with load impedance GO Bar’s power drops rapidly (as noted above the device has a sort of hard cap on output current at approx 115mA), thus delivering “only” circa 200mW on 14Ω and circa 140mW on 10Ω. Again, such figures are higher than those on most of the direct competition, yet not quite at dongle market’s top (E1DA’s 9038SG3 delivers something similar to 600mW onto 10ohm…).
Consistently to this, GO Bar drives the likes of Final A3000, Tanchjim Darling, and even Final E5000 waaaay better then most other dongles, but does not have enough power for RHA CL2, nor of course any demanding planar overear.
As I repeatedly mentioned, all the above refers to GO Bar’s Balanced Ended output. The Single Ended output is not at the same level, neither in terms of output power nor – most of all – in terms of sound quality. Even on easy to drive loads GO Bar Single Ended Out is perceivably duller, much scarcer in microdynamics and more closed-in on space reconstruction.
Simply put, if you ask me GO Bar’s Single Ended output is to be disregarded, in favour of its Balanced Ended sibling.
Cayin RU6 ($250)
GO Bar is by far better than RU6 on pretty much every single count, although this is much more due to RU6 being an overall disappointing device to be honest – which makes the comparison meaningful only due to RU6’s ungrounded hype than anything else really.
Won’t spend more time on this for now, stay tuned if you wish for my piece about RU6, due Soontm.
E1DA 9038SG3 (€126)
Another thing is power. GO Bar is more powerful on high and medium impedance loads, 9038SG3 wins big on loads from 20Ω down. In more practical words, GO Bar’s edge on mid/high impedance drivers proves useless (9038SG3’s power is enough for most drivers, and for those where it is not, GO Bar’s higher power is not enough either), while 9038SG3’s higher power on low and superlow impedance drivers allows translates in E1DA’s dongle being much more agile in driving certain “difficult” IEMs then GO Bar is.
Probably due to its performances on higher impedances, or to lesser efficiency, or both, GO Bar, unlike 9038SG3 or 9038D, is a power w**re (it absorbs up to 4W while working, which is 800mA – so it is not USB2 compliant and by far so). Oppositely, 9038SG3 is modest in terms of power needs vs its output power capabilities, and fully USB2 compliant.
GO Bar misses the harmonic compensation and masterclock customisation infrastructure available on 9038SG3, and that’s not small stuff, and offers only 4 different FIR filters to choose from instead of 7. On the flip side GO Bar literally covers the user with features one nicer and/or sexyer than the other, all of which are totally missing on 9038SG3: XBass and XSpace analog-domain effects, selectable low/high gain, integrated IEMatch, high quality integrated power filtering, and (for Tidal’s aficionados) MQA full decoding.
Sound quality wise 9038SG3 out of the box is definitely cleaner and comes across as more analythical and more energetic compared to GO Bar, which sounds more musical and more relaxing. Actionating upon its multiple tweaks 9038SG3 can be made “sweeter/smoother” though.
In the end GO Bar does give more than 9038SG3 especially in terms of overall features package in return for that much higher purchase price and much higher host power need. On the flip side 9038SG3 can power some IEM drivers which GO Bar can’t trigger well enough.
Apogee Groove ($220)
As extensively reported on my piece about it, Apogee Groove is an oddball. A badass of an oddball if you wish, but still an uncommon device, with the pros and cons one may after all expect from oddity.
Groove’s output stage is based on proprietary technology and does not support crossover filters or similar circuitry, and all too often it also powers Balanced Architecture drivers (even single-driver models) very quirkily. To cut it short, Groove is mainly if not solely intended for Dynamic Drivers, which is of course an apriori fact to seriously consider when looking instead for a “universal application” DAC/AMP dongle.
Groove swings 5V into 600Ω impedance cans which is a lot. It is indeed way short of GO Bar’s huge 7,2V although it’s worth noting that per se there’s little need for those extra 2V when driving high impedance dynamic divers.
On the opposite end Groove is less powerful than GO Bar onto 32Ω loads, but its current cap is a bit higher than GO Bar’s so it ends up delivering more power vs very low impedances like 14Ω or even 10Ω. As a consequence, GO Bar is (power wise) more agile than Groove when paired to the likes of Shure SRH1840, but the coin flips when considering Final E5000.
Groove is quite demanding in terms of host power (340mA, circa 1.5W) but with that it still stays well within USB2 compliance limits, unlike GO Bar which requires almost 3 times as much at full power levels. Beyond these differences, at the end of the day for both sticking an external powerbank onto one’s own preferred transport, and using a suitable single-leg-powered Y-USB cable is the right way to go.
Power profiles aside, Groove and GO Bar are quite different in terms of sound presentation.
Groove is way superior in terms of micro-dynamics and even more so in terms of spatial drawing: I hardly can name a single mobile DAC device better than Groove on this. GO Bar is less colored and may deliver some more subtlety in terms of sheer detail retrieval. Groove is no doubt “more musical”, GO Bar is “more neutral” (just in comparison to Groove – it’s not a “dry neutral” device taken per se).
Lastly, GO Bar’s additional features (selectable reconstruction filters, high gain option, MQA decoding, Xbass, Xspace) are totally alien to Groove.
In terms of their product line, GO Bar covers an evident lack in iFi Audio’s range which never offered a battery-less device before. Now they do and quite expectably their first attempt is definitely a hit.
GO Bar is a very good device. It’s in facts very powerful. While nitpickingly maybe not the single most powerful dongle around (yet still one of the top… three?) I can hardly name a direct competitor offering half of the extra features GO Bar makes readily available under the users’ fingeritps.
GO Bar is superbly design, solid, and – last in the list, but of course first for importance – sounds very well.
At the end of the day I guess its single relevant downside is the price – which is not low at all. A few other downsides are also there, but none of those seriously shadows its positives.
This article previously appeared on audioreviews.org