ODAC Released

CAREFREE ENJOYMENT: I’ve been listening almost exclusively to the ODAC for several weeks now and I’ve been smiling a lot—not so much at the ODAC but at the music. When you know your audio gear is genuinely transparent it opens a worry-free window into the music. And, while I suffer from expectation bias just like everyone else, I’ve run a second blind listening test and can report the O2+ODAC held its own against the $1600 Benchmark DAC1. They both are audibly transparent.

I’M NOT SELLING ANYTHING: Some have claimed because others may profit from the ODAC I’m profiting from it. But that’s not the case. My involvement was simply to help design and measure it--just like the DIY open source O2 amplifier. YoYoDyne and various other vendors are responsible manufacturing and selling the ODAC. They’re the ones taking the financial risks, so they’re also the only ones to profit. This is an entirely not-for-profit blog and will remain so. You can’t buy anything from me.

LIKE O2 LIKE ODAC: The O2 amplifier was created as a simple, low cost, minimalist design delivering 100% transparent performance into nearly any headphone. The ODAC takes the same minimalist approach to transparency. But it’s not quite as simple as it looks in the photo above. It has around 60 components and many of those were carefully chosen through a lot of detailed measurements and trial and error.

NO SNAKE OIL REQUIRED: Many audiophiles want to believe more elaborate or exotic DACs offer higher fidelity. The ODAC demonstrates you do NOT need any of these for 100% transparent performance:

• Asynchronous USB
• UAC2 (USB Audio Class 2) Support
• Asynchronous Sample Rate Conversion (ASRC),
• Minimum Phase Filtering (no pre-ringing)
• Non-oversampling NOS DAC chips
• Dual DAC chips
• Balanced Outputs
• Vacuum Tube Stages
• Elaborate and/or High Current Power Supplies

THE PROOF: I can confidently say none of the above are required for DAC audio nirvana—i.e. having your DAC disappear from the signal chain. Some of the above, like NOS designs and tubes, degrade fidelity. I’m confident because detailed measurements and blind listening tests verify the ODAC’s real world performance. I’ve even tried several different PCs and the ODAC’s performance is relatively consistent between them. So while some of the above might improve a few measurements, if they don’t improve the sound, it’s like taking 4 pills to get rid of your headache when 2 do the job nicely. Once the headache is no longer perceptible, more pain reliever doesn’t help anything. I encourage others to compare the ODAC blind against other DACs, at any price, that measure reasonably well.

ASSEMBLED BOARD: As explained in the ODA/ODAC and ODAC Update articles the ODAC is not DIY friendly. This is true for most 24 bit USB DACs. Neither the USB interface nor the DAC chips are available in small quantities. Both require signed OEM agreements. minimum purchase quantities, and the USB chip also requires custom programming. The ODAC has many extremely tiny 0603 surface mount components and uses a fine pitch 48 pin IC making it very tedious to build by hand. Given all that, the ODAC requires commercial automated assembled in fairly large batches. DIYers can add the ODAC board to the O2, build it into a stand-alone DAC, or add it to the upcoming Objective Desktop Amp (ODA).

NO S/PDIF: To hopefully avoid most of the “how come you didn’t add S/PDIF” or “when will you add S/PDIF” questions, please see the previous ODA/ODAC and ODAC Update articles. Sorry, but it’s not possible.

PARTNERS REVEALED: Because of the above requirements, the ODAC development has been a joint effort with YoYoDyne Consulting. George has supplied other open source USB DAC boards at reasonable prices and seems to be well respected on diyAudio. He’s taking considerable financial risk having a large batch of ODAC boards assembled and hence is coordinating all the manufacturing and distribution. I conducted all the measurements, refined the design, and helped optimize the PCB layout. Just to be clear, I’m not getting any money from the ODAC. Here are the ODAC resources so far:

• YoYoDyne Consulting ODAC Announcement (Canada)
• JDS Labs ODAC Announcement (USA)
• Epiphany Acoustics ODAC Announcement (UK)
• Head-N-HiFi ODAC Announcement (Switzerland)
• Audio Poutine ODAC Source (Canada Facebook Page)

AVAILABILITY: YoYoDyne is estimating boards will be available before the end of May assuming nothing goes wrong. It might take a few days longer to get boards to the UK and Europe. For other information please see the links above.

SCHEMATIC: YoYoDyne has received approval from Tenor to release the schematic once the production boards are verified and released to the various vendors above. YoYoDyne is also interested in a possible future version that’s more DIY friendly.

GUILT-FREE VOLUME ADJUSTMENT: The majority of USB DACs only support 16 bits over USB. That means when you turn down the volume in software you’re getting less than 16 bits of resolution. At background music levels you might only be listening to 11 or 12 bit audio. But the ODAC has a 24 bit USB interface and enough dynamic range to allow guilt-free use of software volume controls.

OTHER DETAILS: There’s a lot more in the previous ODAC articles, but in summary, the ODAC is designed to fit inside the standard O2 enclosure in place of the batteries. A few internal wires need to be soldered to connect the ODAC’s output to the O2’s input jack. Other line level sources can still be used with the O2. It will also fit inside the upcoming ODA. And it can be used standalone with either the on board 3.5mm output jack or panel mounted RCA output jacks. The USB connector is a standard USB-Mini-B as used on the FiiO products, the Sansa Clip, cameras, etc.

FOUR REVISIONS: The ODAC has been through four lengthy revisions—two of the earlier boards are shown to the right. Despite the fact we started with essentially the reference design from the datasheets, the devil was in the details. The  first version played music and sounded OK. Many companies and DIYers that “design by ear” would have stopped there. But that first version didn’t come close to delivering what the DAC chip is capable of. Each revision cycle took at least several weeks, cost hundreds of dollars, and involved countless hours of work. But, in the end, it resulted in much better performance compared to where we started..

REAL WORLD PERFORMANCE: Most DACs and PC sound interfaces priced under $200 fail to come close to the published specs for the chips they use. The FiiO E10’s Wolfson DAC chip is rated at 117 dB of dynamic range but the E10 only delivers a modest 98 dB. The power supply, PCB layout, grounding scheme, I2S waveform fidelity, clock quality, and more, often degrade the performance to well below the manufacturer’s carefully optimized reference design used for the datasheet specs. This is especially true for USB sourced signals and DACs running from USB power. But the ODAC, despite being USB powered, managed to come very close to the “Holy Grail” datasheet specs. See the Tech Section for more.

THE REAL NUMBERS: Here are the real numbers (versus the April Fool’s Day numbers I published two weeks ago) and they all meet the requirements for audible transparency. The letter following many of the results is the same letter grade (A-F) I’ve used in previous reviews with A being the best, and F being a “Fail”:

Measurement	ODAC	FiiO E10	DAC1 Pre
Freq. Response 10 hz – 19 Khz 24/44	+/- 0.1 dB A	+/- 0.1 dB A	+/- 0.1 dB
THD+N 100 hz 0 dBFS	0.0029% A	0.005% A	0.0009%
THD+N 20 hz –1 dBFS	0.003% A	0.004% A	0.0009%
THD+N 10 Khz –1 dBFS	0.003% A	0.004% A	0.0007%
IMD CCIF 19/20 Khz –3 dBFS	0.0011% A	0.013% B	0.0005%
IMD SMPTE –1 dBFS	0.0004% A	0.004% A	0.0004%
Noise A-Weighted dBu 24/44	--102.8 dBu A	-98.3 dBu C	-105.4 dBu
Dynamic Range –60 dBFS A-Wtd	--111.1 dBr A	-97.6 dBr C	-110.9 dBr
Linearity Error -90 dBFS 24/44	0.0 dB A	0.0 dB A	0.2 dB
Crosstalk 0 dBFS Line Out 100K	--93.5 dB A	N/A	-106 dB
USB Jitter 11025 hz J-test 24/44	Excellent	Very Good	Excellent
Maximum Output Line Out 100K	2.0 Vrms	1.65 Vrms	2.5 Vrms (1)

NOTE 1: Maximum output of DAC1 is configured with internal jumpers

BOTTOM LINE: The ODAC has been released to production and will hopefully be available by the end of May. I’ll soon be publishing more detailed measurements, results of listening tests, etc. I’m confident the ODAC is audibly transparent. And, especially when it’s installed inside the O2 or future ODA, it offers a level of performance that’s difficult to find without spending substantially more. It also offers detailed measurements and blind listening tests to back up its performance which is something very few other DACs offer at any price. And, paraphrasing from credit card commercials, worry free enjoyment of music can be priceless.

 

---

TECH SECTION

---

SPECIFICATIONS: Here are some ODAC basics:

• Audio Formats: 16/44, 16/48, 16/96, 24/44, 24/48, 24/96
• Interface: USB Audio Class 1
• Native Driver OS Support: Windows XP & Later, OS X x86, Linux
• Operating Systems Tested: XP, Vista, Win7 (32 & 64), OS X Snow Leopard, Ubuntu 9.1 32 bit
• Line Output: Approximately 2 Vrms into 5K ohms or higher
• Dynamic Range: > 110 dB A-Weighted
• Distortion: < 0.005%
• Dimensions: 49 x 58mm (see drawing to right)

BLIND EVIDENCE: So far I’ve run two relatively informal blind tests with the ODAC. The latest one used special software on the PC to play the same track on both my Benchmark DAC1 Pre and simultaneously on the ODAC plugged into the same PC (both connected via USB and running at 24/44). The ODAC was connected to an O2 headphone amp, and a switchbox allowed the headphones to be rapidly switched between the DAC1 and the O2+ODAC. The two sources were carefully level matched (using their respective volume controls) using a test signal and wideband DMM. I tried both my Sennheiser HD650 and Denon AH-D2000 headphones with a variety of well recorded favorite tracks. One other listener and I could not reliably tell which was playing.

FUTURE BLIND TESTS: I’ll hopefully be running a more comprehensive and rigorous blind test in the future. But, ultimately, it’s best to have listeners who expect to hear a difference, and someone else who understands the technical issues to supervise the test, perform level matching, sync the sources, etc. It would be ideal, for example, to have Mike and/or Lieven from Headfonia be a listener and George from YoYoDyne oversee the test. But sometimes geography and other factors restrict what’s realistic. If anyone is willing to help coordinate such a blind test, please contact me privately with the link in the right hand column.

USB INTERFACE: There are only a few USB interface chips capable of 24 bit operation. The best option we could find is the Tenor TE7022. It’s used in the Violectric USB 24/96 and several other commercial 24 bit USB DACs. Notably, it does not require proprietary drivers to work with any popular operating system including Windows because, like the Benchmark DAC1, it’s a USB Audio Class 1 interface. It also has respectably low jitter. The XMOS solution requires an expensive license and proprietary windows drivers and offers no audible benefit. The TAS1020B is being discontinued, requires extensive firmware, and also offers no audible benefit. A custom microcontroller creates even more hurdles with no audible benefit.

THE DAC CHIP: As I’ve explained elsewhere you can get transparent performance from most of the better DAC chips on the market from a half dozen chip vendors. What’s most important is choosing one that’s best suited to the particular application. In this case, that means running from a single 5 volt USB power supply, having a buffered voltage output (to avoid needing a single-supply op amp), and operating properly without using a microprocessor. Just those three requirements narrow the choices considerably. We chose the ESS Sabre ES9023 which is used in a lot of commercial designs (it’s an improved version of the popular ES9022). Unlike TI, Burr Brown, Analog Devices, etc, ESS specializes in audio chips and they did a nice job with the ES9023’s feature set and specs.

DIRTY LIES: Many popular “boutique” DACs (especially those being sold on eBay) mislead their fans by quoting only the chip specs for their entire DAC. Basically that amounts to cheating and lying. It further implies either the company is incapable of making the proper measurements or the real measurements were bad enough they didn’t want to share them. The implication is a DACs performance is solely determined by the chip used. But the opposite is usually more true. The implementation matters far more than the chip. the FiiO E10 is just one of many examples. The HA-Info I’ll soon be comparing to the ODAC is another.

CHIP ENVY: Any serious audio engineer designing DACs and making proper measurements knows the implementation matters more than the chip used. The first version of the ODAC, which closely followed the ESS datasheet, only managed about 98 dB of dynamic range. The distortion was also much higher than listed on the datasheet and the jitter was somewhat disappointing. Unless you exactly duplicate the chip manufacturer’s reference design, right down to the PC board layout (which I’ve yet to see any manufacturer do), you really don’t know what you’ll get. To measure all the important parameters you need more than just RMAA. You need a real audio analyzer with performance substantially better than the DAC being measured.

DESIGN BY EAR: Detailed and credible published measurements are missing with most “boutique” DACs being sold including those from Schiit Audio, Audio-GD, AMB, Twisted Pear, Burson, and NuForce. Where’s the credible evidence they got it right? A lot of these companies try to claim specs don’t matter, and they instead design by ear, but that method is seriously flawed (see: What We Hear). It would be like designing a car engine without a dynamometer and having no idea how much horsepower and torque it produced, how fuel efficient it was, etc. Given all the proven problems with sighted listening, and how our ears and brains work, those who claim to design by ear are very likely getting it wrong. Put another way, they’re often designing products with far lower fidelity than they’re otherwise capable of.

LESSONS LEARNED: If this project has taught me anything, it’s that getting much better than 16 bit (96 dB) performance can be challenging. The first version of the ODAC, despite following the reference design, only had about 98 dB DNR. That’s about the same as the FiiO E10. The photo to the right shows a few dozen assorted surface mount parts that were laboriously swapped out one at a time and measurements repeated dozens of times using the dScope. Some improvements were far from intuitive. Audiophile preferred polyphenylene capacitors performed worse than less expensive types. Additional filtering on the digital power supply dramatically increased jitter. Chasing down the last few dB of dynamic range the chip is capable of proved to be especially challenging. When it was said and done, the DNR went from 98 dB to over 111 dB. That’s a huge difference and something the design-by-ear crowd would have never achieved.

THE AUDIOPHILE WAY: All too many small or “boutique” audiophile manufactures and DIYers seem to just slap trendy chips on a board, listen to their creation expecting it to sound good (so that’s what they hear), and call it good. Many don’t even follow the reference design. Instead they include a bunch of “audiophile upgrades” expecting better performance—and they hear what they expect to hear even when it’s not true. But the ODAC demonstrated those upgrades often make things worse. So instead of getting even 98 dB DNR like the first ODAC revision, those following audiophile myths and designing-by-ear probably would have ended up with something even worse. Unless you’re making the right measurements, you really have no idea what you’re getting.

THE POWER SUPPLY: For reasons explained in the earlier ODAC articles, the ODAC is USB powered. This allows it to work standalone, as an internal add-on to the O2, and in the upcoming ODA. There are many obvious advantages to USB power but it often degrades performance due to noise. To get around this, the ODAC uses split digital and analog power supplies each with their own filtering and regulator. The analog supply has additional filtering and the critical reference voltages, and negative supply for the DAC chip, are further optimized. I literally tested more than 100 variations of components, including different brands of capacitors, to get the most out of the ES9023. This level of refinement would be impossible without a serious audio analyzer.

PUMP YOU UP: The ESS chip has the huge advantage of a built-in low noise charge pump. It generates its own regulated negative power supply allowing a Redbook standard 2 Vrms output from a single 5 volt USB power supply and a direct coupled output. This is an important distinction compared to a lot of USB powered DACs. Without the charge pump, or some other negative power supply, USB DACs can’t produce the Redbook standard 2 Vrms which reduces their effective dynamic range, lowers their ENOB, and creates level matching problems. It also requires an undesirable output coupling capacitor and usually results in loud transients on power up. The AMB Gamma, and most USB powered DACs I’ve tested, don’t meet the Redbook standard. The ODAC does. And it produces only a soft click on power up.

ADAPTIVE USB INTERFACE & LOCAL CLOCK: I’ve talked about this before, but just to be clear, the ODAC is NOT clocked by the USB port. So the quality of the audio clock, and any resulting jitter, is largely independent of the PC’s USB timing. It has its own low phase noise 12 Mhz crystal controlled oscillator that’s used to generate the MCLK and SCLK audio clocks.

A NOTE FOR 24/88 FANS: Some have asked about 24/88 high resolution audio support (popular for SACD rips). While the ODAC doesn’t support 24/88, it does support the audibly identical 24/44. It’s trivial to re-sample 24/88 audio to 24/44 with no artifacts as it’s a simple divide-by-two operation (and one the operating system will perform for you automatically). I know many audiophiles probably think they’re losing something, but nobody has proven they are. Meyer & Moran demonstrated in a very in-depth study that even 16/44 audio sounded identical to SACD. Another good read is 24/192 Music Downloads. And if you refuse to believe all that, try resampling some 24/88 audio to 24/44 and compare them yourself with Foobar and the ABX add-on. It’s been done at HydrogenAudio and elsewhere always with the same result: Unless you mess up the resampling somehow, or change the levels, you can’t tell them apart.

TRANSPARENCY GUIDELINES: The What We Hear article offers information and references outlining guidelines as to what’s required for a piece of audio gear to genuinely disappear from the signal path and not alter the sound in any audible way. Here are what I believe to be relatively conservative criteria for audible transparency and the ODAC passes all of them:

• Frequency Response 20hz – 19 Khz within +/- 0.1 dB (Most DACs, due to the Nyquist limit of 22 Khz, start to roll off past 19 Khz when operating at 44 Khz sampling rate—the ODAC is down about 0.4 dB at 20 Khz). The widely accepted, but less conservative standard is +/- 0.5 dB (1 dB total variation) from 20 hz to 20 Khz.
• All Harmonic, IMD, Alias, Modulation, & Crosstalk Components Below –90 dBFS and total sum below –80 dBFS (0.01%)
• All Noise Components below –110 dB and total sum below –100 dBFS
• All Jitter Components below –110 dB and total sum below -100 dBFS

GREEN GUIDE LINES: A few months ago I introduced green guide lines on several of my measurement graphs to help show the worst case ideal performance. Some of these are slightly more lenient than the above criteria or take into account more detailed thresholds (i.e. that power line hum can be slightly higher in level than midrange noise). For now I’m keeping the green guide lines consistent with earlier reviews. But please note the ODAC meets even the tougher criteria above.

PARTIAL MEASUREMENTS: I’ve made LOTS of ODAC measurements including some things I’ve never measured before—such as true latency. For this article I’ve only shown some of the more common measurements. In a future article I’ll cover additional measurements, 16 bit operation, 24/96, as well as several comparisons to the DAC1, FiiO E10, and an HA-Info eBay headphone DAC with a well respected DAC chip. So, in the interest of getting this article done sooner rather than later, and keeping it to a manageable size, only a sampling of 24/44 measurements are shown below.

DYNAMIC RANGE: A DAC’s noise floor impacts Dynamic Range (DNR), audible noise, THD+N, and can even exceed jitter-induced distortion. If you have to pick a single number to evaluate real world DAC performance –60 dBFS dynamic range (DNR) is one of the most revealing. The guys in the white lab coats have determined DNR greater than 100 dB results in transparency under realistic conditions. And, if you want to adjust the volume in software, it’s best to have at least 110 dB DNR to keep the noise floor inaudible even if the downstream gain is left cranked way up. Anything beyond 110 dB is past the point of diminishing returns—it looks nice on paper but doesn’t help the sound quality. The ODAC is a very substantial 14 dB better than the FiiO E10.  Here’s both channels of the ODAC referenced to the 2.03 Vrms at 0 dBFS. Note the channels are very symmetrical indicating a careful PCB layout:

ODAC VS ESS: As explained earlier, the DNR quoted on datasheets is often something of a Holy Grail. The chip specs are typically from a very high quality AES/EBU or I2S laboratory quality signal (as output by high-end audio analyzers like a Prism dScope or Audio Precision). And they’re typically running from expensive ultra low noise bench power supplies costing thousands of dollars. It’s safe to assume the datasheet numbers were not made with USB data while running on USB power. ESS rates the ES9023 DNR at –112 dB A Weighted. The ODAC delivers –111.1 dB A-Weighted under the same conditions. In other words, even using USB data and power, the ODAC comes within a fraction of a dB of achieving the datasheet spec! I’m fairly proud of this aspect of the ODAC. It wasn’t easy.

CCIF IMD: This 19+20 Khz twin tone is a difficult test for many DACs running at 44 Khz. Old style (NOS) non-oversampling DACs especially struggle due to aliasing problems. In addition the output buffer (or I-V stage) in many DACs contributes high frequency distortion because the RC filter can be a challenging reactive load at these frequencies. If you look back through my reviews, you’ll find lots of products struggle on this test. Even the E10 turned in a marginal result. The ODAC, however, due to careful optimization of the output filter, and the superior digital filtering of the ESS DAC, does very well here with everything in the audio band well below 100 dB (both channels shown). Note also the 19 and 20 Khz tones are visibly equal in level which is not the case for many DACs:

 

SMPTE IMD: This twin tone test is more revealing of low frequency problems including power supply interaction. Again, the ODAC does very well with everything well below 100 dB (both channels shown):

 

100hz THD+N @ 0 dBFS: This test checks for clipping of the DAC at 0 dBFS and also shows the maximum output and channel balance error. You can see the ODAC produces 2.03 Vrms which is within an in significant 0.03 volts of the Redbook standard for digital audio. And even at 0 dBFS, the distortion is still 3 times less than what’s required for 100% transparency. The channels are perfectly balanced to within 0.001 dB. This spectrum is shown all the way out to 96 Khz and you can see the ESS DAC is well behaved even above 20 Khz with all noise still below about –110 dBFS. This is excellent performance:

 

THD+N VS FREQUENCY: Here’s the distortion performance at –1 dBFS from 20 hz to 20 Khz into a more challenging 10K load with a measurement bandwidth of 22 Khz. At 1 Khz the distortion is only 0.0027% and it remains around 0.003% over most of the audio band with only a slight rise up to 0.0048% at 9 Khz before the harmonics fall above the audible range. This is excellent performance and both channels are very closely matched (yellow vs blue):

 

NEW JITTER FINDINGS: I did quite a bit more research on jitter during the ODAC’s development. I’m also using a new dScope method that shows the same spectrum as before but now the symmetrical jitter components are marked (with a white “X”) and summed to obtain a total numerical value (previously the dScope was just showing the total residual noise floor). Having the single number (-103.3 dB below) made it easier to optimize the ODAC for the lowest jitter. The objective evidence conservatively indicates  if you keep all related components below -110 dB, and the total below -100 dB, the jitter will be entirely inaudible. Jitter creates dissonant distortion products in the audible band. It’s reasonable to assume if the audible effects of jitter are kept at or below the inaudible noise floor, they too will be inaudible. So the same levels of –100 dB and –110 dB that apply to noise also apply to jitter contributions. This is also consistent with various professional reviewers and their anecdotal opinions on jitter performance as well as my blind testing against the Benchmark DAC1 which has even lower jitter.

ODAC JITTER: The ODAC passes the conservative criteria with several dB to spare even on the worst-case J-Test signal. And the spread at the base of the signal (very low frequency jitter) is extremely minimal being entirely below –130 dB. It’s also worth noting the jitter here looks subjectively worse because the noise floor is much lower than most of my jitter measurements which are done with a 16 bit test signal. A 16 bit noise floors masks most of the “spikes” seen below. The ODAC also has negligible inter-channel phase error and essentially perfect pitch accuracy:

 

CONCLUSION: Hopefully the above provides some good evidence the ODAC delivers transparent performance. The next ODAC article will compare the 16 and 24 bit performance and I’ll be comparing it to several other DACs. I’ll be publishing many more measurements such as modulation noise, channel separation, square wave/impulse response, latency, frequency response, absolute noise, and more.

ODAC Update

ODAC PROGRESS: As my articles have slowed to a trickle lately, hopefully these 5000+ words will be welcome news. After lots of professional and other distractions, I’m once again devoting serious time to the Objective DAC (aka ODAC). There’s been some good progress and I’m feeling better then ever about the ODAC’s performance. I’ve been comparing it with my Benchmark DAC1 Pre and it’s been eye opening in some areas. Keep reading for more details and further background on the ODAC design.

QUESTIONS (FAQ): Before you post questions or suggestions in the comments please have a look at the FAQ section below to see if your issue has already been addressed. I’ve tried to cover the popular questions and issues.

NOT FOR PROFIT: Please note I’m not selling the ODAC, just helping design it. Just as with my other projects, like the O2 portable amp, any profit from sales will be to whatever companies decide to have them made and sell them.

ODAC + ODA: The ODA, for those not following the history, is the Objective Desktop Amp. It’s essentially a future desktop version of the popular O2 portable amp with a few added features and upgrades. ODAC development is taking priority for now because the ODA will be partially designed around it. The ODAC can also be used with the O2 and as a standalone device.

FILLING A VOID: There are simply not many reasonably priced DACs supporting 24 bit operation over USB under Windows without needing proprietary drivers. While there are some low cost options on the market, at least some of them, such as the FiiO E10 and NuForce uDAC-2, offer barely more than 16 bit performance. There are even fewer options if you want a high quality 24 bit DAC with a low impedance output that can drive most popular headphones well (this leaves virtually all pro audio DACs out in the cold). The least expensive option I know of that’s been properly measured and performs well is the Centrance DacPort with the 1 ohm output upgrade. But it’s close to $500. Most DIY DACs, such as those from Twisted Pear and AMB, only support 16 bit operation over USB or they require proprietary drivers under Windows.

ODAC + O2 RETROFIT: The ODAC fits neatly inside the O2 portable amp using the standard B2-080 enclosure by simply removing the batteries as shown to the right. One side sits in an enclosure slot while a mounting hole lines up with the central mounting hole on the O2 PCB anchoring the opposite side of the board. The battery terminals act as a “back stop” to absorb the force of plugging in a USB cable and the rear panel does the same when removing the plug. It worked out surprisingly well and I can’t take credit for this feat—my commercial counterpart came up with the idea and we mutually optimized the PCB layout to work with the O2. The retrofit requires a new back panel with an opening for the USB jack. At least one person has volunteered to design and hopefully sell a laser cut plastic panel for those wanting the O2/ODAC combo. Stay tuned for more but it’s a pretty simple retrofit that only requires soldering 3 wires at each end.

O2 INPUT SWITCHING: Those who only want a USB headphone DAC (like the FiiO E10 or NuForce uDAC-2) can simply wire from the output header of the ODAC to the P1 input header on the O2 board. But for those who still want to use other sources (like an iPod, etc.) with their O2, you can have your cake and eat it too. Once you cut the traces shown to the right you can wire the ODAC to terminals 3 and 4 of the input jack. These lead to internal switches in the jack and will connect the ODAC only when nothing is plugged into the O2.

ODAC STANDS ALONE: If you just want a DAC with no preamp or headphone amp, the ODAC can be used by itself as a USB powered 24/96 DAC. Because it supports 24 bit operation over USB there’s little or no penalty using only a software volume control or you can control the volume downstream of the ODAC. The board measures about 49mm x 58mm (1.9 x 2.2 inches) and will fit in many small enclosures such as the Box Enclosures B1-080. There are also several inexpensive eBay DIY enclosures that would work. Ignoring the output jack, it’s only about 4mm in total thickness as it’s entirely a surface mount design This allows it to easily “piggy back” on top of other boards, etc. In the O2 tradition, it has three mounting holes (geeks know 3 points determine a plane and 4 just mess things up ;). The components are also held back from two opposing edges for slot mounting. There’s a provision on the board for a 3.5mm stereo output jack that’s on the same edge as the USB connector requiring only one machined (or laser cut) panel. DIYers can also panel mount 3.5mm or RCA output jacks or simply build the ODAC into other gear. The ODAC can be connected to any input that accepts standard Redbook 2 Vrms line level audio and has at least a 10K input impedance.

O2 DESIGN PRINCIPALS APPLIED – The O2 was implemented using well proven design techniques and countless rounds of measurements and refinements. The result is performance well beyond what most would expect from the O2’s modest and inexpensive components. The O2’s popularity has shown this design methodology is valid. Put simply, the O2 has proven far greater than the sum of its parts. I applied the same approach to the ODAC design. While it uses relatively modest components, it delivers performance well beyond what most would expect given the cost and components used. The ODAC closely follows the chip manufacturer’s reference design information and has been carefully optimized through several iterations. Using a professional mixed domain audio analyzer, like the Prism dScope, enables a level of performance refinement that would otherwise be all but impossible. This gives the ODAC a huge advantage over most DIY DACs and products from small manufacturers lacking proper test equipment.

THE AUDIO ANALYZER ADVANTAGE: To make meaningful measurements of a DAC you need an ADC with a noise floor significantly (at least several dB) lower than the DAC’s noise floor. Otherwise you’re measuring the ADC as much (or even more so) than the DAC. Such an ADC is increasingly hard to find once you get up to Benchmark DAC1 levels of performance. The Prism Sound dScope, however, has a noise floor of just over 1 microvolt or around 126 dB below the Redbook standard of 2 volts. Even my $1800 Benchmark ADC1 can only manage a best case 119 dB and less expensive audio interfaces are typically significantly worse. It’s also extremely useful to have an audio analyzer capable of real time mixed (analog and digital) domain generation and analysis. A product like the dScope, or a high-end Audio Precision analyzer, is about the only way you can properly measure and develop a high performance 24 bit DAC. Otherwise the true performance is masked by the ADC you’re using and the limitations of using PC software that cannot do anything in true real time because of the PC’s operating system. Input isolation is also a significant issue. Using RMAA typically requires sharing the DAC’s USB ground with the ADC’s ground which, in itself, can create all kinds of erroneous results and ground loops. Properly testing a high performance DAC requires more than just another PC audio interface. You need specialized hardware designed for the purpose.

THE OTHER WAY: Lots of DIY, and even some commercial DACs, seem to be little more than a few trendy popular chips slapped onto a nice looking PC board. If their designers conducted the proper measurements along the way, where are their published results? You generally have to look at expensive products like those from Centrance, Benchmark Media, and Anedio to find meaningful DAC measurements. Some designers and companies simply quote highly misleading specs from the DAC IC datasheet but, as I’ve shown in my reviews, the results often fall far short. A good example is the FiiO E10 that only delivers roughly 16 bit performance even running at 24 bits. Some companies, like NuForce, Audio-GD, and Twisted Pear, argue they “design by ear” but the unavoidable human bias associated with such sighted listening is well documented.

IT’S ABOUT WAY MORE THAN THE RIGHT PARTS: Using high-end parts is meaningless if the implementation is (often unknowingly) flawed. It’s like putting a Ferrari engine in a Yugo or using ultra fast RAM in a PC with a slow processor and chipset. It’s pointless. Just as cars are defined by far more than just their engine, and PC performance depends on several subsystems all working well together, the same is true of audio gear. The best products are from manufacturers that conduct proper testing and have the resources and desire to sweat all the little details rather than just using the latest FOTM parts and making it look nice.

FOLLOW THE PROFITS: Sadly a lot of “specialty” audio designers and companies seem to depend mainly on subjective hype and sighted listening bias, rather than proper design and objective performance, to sell their products. One can argue some are mainly trying to cash in on the latest FOTM craze rather than investing the time and money to design genuinely solid gear. It’s one of the goals of this blog to help not only expose half baked designs for what they are, but also demonstrate better alternatives don’t have to be be expensive or made with dual phase aligned unobtanium.

SMALL CHANGES CAN EQUAL BIG IMPROVEMENTS: I’ve sometimes been amazed how even small changes have made fairly large differences in the ODAC’s performance. Several of the ODAC’s optimizations run counter to typical DIY audiophile beliefs. Here are a few examples of typical design myths:

• Larger Value Capacitors Are Better – “Upgrading” certain power supply capacitors to larger values made the ODAC perform significantly worse compared to using the values specified in the chip manufacturer’s reference design. Bigger value caps often have higher ESR, more inductance, and much higher impedance at very high frequencies. They can also create other problems.
  
• Top & Bottom Ground Planes Should Be “Stitched” Together – Stitching is the practice of applying a board-wide grid of small vias (plated through holes) that connect the top and bottom ground planes. Some argue this lowers the ground impedance and keeps ground paths shorter. But they’re not true ground planes on a 2 layer board. Instead you have a bunch of ground fill areas isolated by signal and power traces. When you have these “ground islands” rather than a true continuous ground plane, the stitching can easily create undesirable ground loops and send ground currents where you don’t want them. I’ve seen several DIY and commercial designs ignore this issue and other proper grounding practices.
  
• Expensive Audiophile Dielectric Capacitors Work Best – I experimented with various types of poly film caps, including audiophile preferred SMT Polyphenylene Sulphide (PPS) types, and found they sometimes made things worse. I also discovered not all ceramic caps perform equally. There’s no simple rule of thumb that always works. You have to make the right measurements and sweat the details. It’s time consuming but proved worth it. Every capacitor associated with the DAC chip and analog circuitry of the ODAC has been carefully optimized using the dScope. Overall, my capacitor tweaking resulted in lowering the noise and distortion by more than 6 db. And, trust me, the final result is not what your average DIYer would intuitively think is best.
  
• Fully independent Analog & Digital Power Supplies Improve Performance – While there are probably a few high-end DACs that measure slightly better with fully independent digital and analog power supplies, the reality is the DAC chip itself and related digital noise is more likely the dominant limiting factor. I’ve shown, with multiple measurements and tests, the power supply is not holding the ODAC back in terms of performance. And while there’s some carefully designed filtering and ground routing between the analog and digital sections, they’re both derived from the same source. A DAC will only perform as well as its weakest link allows. In an optimized design that weak link is usually the DAC chip and/or a certain amount of unavoidable noise from the USB and I2S digital buses. When that’s true, adding more esoteric power supplies won’t help much if at all.
  

PCB LAYOUT IS EVERYTHING: I can’t stress this enough. Just as the O2 board was designed for “function over form” so was the ODAC. Many DIYers and audiophile manufacturers want to show off their PC boards. So they often lay them out at least partly to look nice. That almost always means the performance suffers. In the case of a 24 bit DAC I’ve learned it might suffer quite a bit. Keeping the noisy digital signals out of the analog side of a DAC becomes challenging once you get past about 17 bit performance. And for those trying to judge if they got it right by using RMAA and a sound card, or an 8 bit oscilloscope with only 40 dB of dynamic range, good luck with that (see The Audio Analyzer Advantage above and my RMAA article for why). And good luck using typical sighted listening tests to verify a PCB layout.

BENCHMARK DAC1 REFERENCE TARGET: The Benchmark DAC1 models have been widely reviewed and praised—especially for their excellent measurements, sound quality, and hassle free 24 bit high resolution USB support. Hopefully many agree the DAC1 Pre is a worthy reference target to judge the ODAC against. While it’s a bit like putting a Mazda MX5 Miata up against a Porsche 911 it’s still a worthy goal. When I present the ODAC’s final measurements I’ll show many of the same results for the Benchmark and everyone can judge the end result for themselves. I’ll also be comparing the ODAC to the FiiO E10 as it’s probably the ODAC’s closest current 24 bit USB competitor in terms of cost and might even include a few pro audio interface results as well.

PRELIMINARY BLIND TESTING: In preliminary blind comparisons between the ODAC prototype and my DAC1 Pre, they sound the same. I’ll be doing more involved blind tests, but the initial results are very promising. If this trend continues I’ll be adding an “ODAC Public Blind Challenge” to my existing O2 and op amp blind challenges. If someone thinks they have a DAC that measures well and sounds better, let’s find out!

DESIGNED AND (for now) ASSEMBLED IN THE USA: This may not matter to everyone, but living in the USA myself, it’s something I take at least some pride in. Steve Jobs famously told President Obama “those jobs are not coming back” referring to Apple making nearly everything in China. And while Apple products are generally made to a high standard in China they’re still designed in the USA. The same cannot be said for a lot of reasonably priced “boutique” audiophile gear. NuForce, Audio-GD and FiiO are a few examples with well documented, and sometimes embarrassing, problems. And, based on what I’ve seen, a lot of the gear being sold mainly on eBay and direct out of Asia can be even worse.

USB POWER ADVANTAGES: A self powered DAC that operates entirely from USB power has a number of practical advantages:

• One Size Fits More – Because the ODAC needs to be commercially assembled in volume using automated equipment it’s much more cost effective to have a single version that works in the O2, ODA, standalone, and can be added to other DIY or commercial designs. USB power is virtually required to make this possible. Higher volume manufacturing of a single board brings the price down for everyone.
  
• O2 Compatibility - The O2 has a dual +/- 12 volt power supply delivering around 200 mA peak to each channel of the amplifier. The ODAC requires around 125 mA at a far lower voltage. The O2 cannot power a high quality DAC without adding a switching regulator or DC-DC converter which would add noise and still tax the O2’s power supply. There’s also no physical room for the added hardware. So USB power is the only viable option for the ODAC in an O2.
  
• Stand Alone Operation – For those wanting to use the ODAC by itself, USB power has a clear advantage. It eliminates the need for another power supply and is especially desirable for portable use.
  
• No Enumeration Problems – Some USB interface ICs are designed to be USB powered. They will not enumerate correctly with the host device if they’re already powered when the USB connection is made. This can necessitate using USB power for the interface chip while using a different power source for the DAC chip. This creates various other challenges including the next point.
  
• No Unpowered Inputs - USB power also simplifies problems associated with having unpowered IC’s connected to powered ICs. Digital ICs that are not powered don’t generally like being connected to ICs that are. It can result in a potentially destructive condition where the normally high impedance inputs of a powered IC represent a low impedance when the IC has no power. The powered IC pumps abnormal levels of current into the unpowered inputs. When power is applied with the IC in this invalid state it can “latchup” and draw large amounts of power supply current destroying itself. Some of the solutions to this problem can degrade jitter performance because the I2S bus is subject to this issue on a USB DAC with a split power scheme. Powering the entire DAC from USB power neatly solves these problems.
  

USB POWER CHALLENGES:  USB power also presents some challenges:

• Potentially Greater Noise -  USB power is more of an unknown compared to a dedicated power source. So it requires extra filtering and careful design to avoid degrading the performance of the DAC. This is especially true when you’re aiming for much better than 16 bit performance. The USB powered FiiO E10, Creative X-Fi Go, and NuForce uDAC-2 all promise 24 bit performance but only deliver around 16 bit performance. Likewise, when running from AC power, the latest MacBook Air also fails to deliver better than 16 bit performance from its 24 bit DAC due to extraneous power-related noise.
  
• Audio Output Voltage - USB power can be as low as 4.5 volts, and when you add in losses from power filtering and the DAC/op amp circuitry, you’re lucky to get 4 volts peak-to-peak of output swing without clipping. That works out to 1.4 Vrms which is a significant 3 dB shy of the 2.0 volt Redbook standard for digital devices. That’s 3 dB of potential dynamic range lost and a 3 db drop in level compared to normal home sources (like a CD player, network media player, etc). Indeed most of the USB DACs I’ve tested, and even popular DIY DACs like the AMB gamma, have this shortcoming.
  
• Capacitor Coupled Outputs - A single-ended DAC power supply usually requires an output coupling capacitor to block the 1/2 Vcc voltage at the DAC output. Such a capacitor, to drive a 10K load, needs to be a fairly large value to avoid low frequency roll off and excessive phase shift. It also should be a high quality film, rather than electrolytic, type. But many USB powered DACs, in the interest of saving money and space, use electrolytic or otherwise compromised output capacitors. This is even true of some audiophile DIY designs like the AMB gamma.
  
• Power Related Jitter - Power and ground “pollution” at various frequencies can have a significant impact on jitter performance. Noise from the power supply can, in effect, modulate the digital bit stream creating jitter.
  
• USB Maximum Current Limit - While the USB ports on any PC or laptop made in the last 6+ years can nearly always supply 500 mA of current there are some exceptions. There are a few ultra low power netbooks that have 100 mA USB ports and unpowered USB hubs are also, at least in theory, limited to 100 mA per port. I’m not sure about iPads and Android tablets but I suspect they may be rated for only 100 mA as well.
  

ODAC POWER DESIGN – The above challenges can be largely or entirely overcome with careful design and by using the right components. Here’s how the ODAC addresses them:

• Power Noise Below the DAC’s Noise Floor – It turns out ground and conducted (electromagnetic) noise are typically as significant as noise on the USB power line. Because a connection to the PC’s ground is required regardless, simply using an external power supply doesn’t automatically mean freedom from USB bus noise. With the ODAC the solution involved careful routing of ground currents, different power supply conditioning for the digital and analog sides, careful capacitor selection, and inductive filtering. The result is the latest prototype’s noise floor is mainly determined by the DAC IC itself not the USB power bus. Put another way, an independent power supply wouldn’t make much difference.
  
• Redbook Standard Audio Output Voltage – The ODAC delivers delivers the Redbook standard output of 2 Vrms without clipping by using a bipolar power supply. This alone adds roughly 3 dB of dynamic range to the ODAC’s performance compared to many USB powered DACs.
  
• Direct Coupled Output – The ODAC uses a ground referenced split supply so no virtual ground or output capacitors are required. This assures accurate low frequency amplitude and phase response with no bass roll off and avoids potential capacitor-induced distortion.
  
• Power Related Jitter – The ODAC’s power filtering was optimized not just for the best noise and distortion performance but also for the lowest jitter. Interestingly some of my attempts at filtering did help lower the power supply noise but increased the jitter. The dScope’s J-test proved invaluable in optimizing the PC board layout and power supply design for simultaneously low jitter and low noise.
  
• USB Current Limit – I’m not aware of any 24 bit USB interface and high performance DAC ICs that, combined, come in safely under 100 mA total. The bipolar power supply that allows a 2 volt output does require a bit more power. And the choices in 24 bit interface chips are very limited. The only way to keep the total budget under 100 mA would involve a much lower performance DAC and that’s not an acceptable trade off. The good news is I’ve tried the ODAC on many PCs, including two netbooks, and it works fine. It even worked on an unpowered USB hub despite being slightly over the 100 mA limit.
  
• Worst Case – If someone does encounter a USB port that is either extremely noisy or can’t provide 125 mA of current, the solution is simply to use a powered USB hub. They can be purchased for as little as $20 or so.
  

SO HOW QUIET IS IT? Using the industry standard A-weighted dynamic range test with a –60 dBFS signal, the current ODAC prototype has an impressive 112 dB of dynamic range. How good is that? My $1600 Benchmark DAC1 Pre, on the exact same test referenced to the same 2 volts, is slightly worse at 111 dB. CD quality audio, in comparison, has only 96 dB of dynamic range. I should note if you have an application where you can use the DAC1’s full 7+ volts of output, it can manage 116 dB of dynamic range referenced to it’s maximum output. So you do get something for your extra $1500.

ENOB: ENOB stands for Effective Number of Bits and is another measure of a DAC’s performance. No 24 (or 32) bit audio DAC can achieve true 24 bit performance, In fact, 20 ENOB is generally considered the “Holy Grail” of real world DAC performance. The ODAC is just under 19 ENOB and the Benchmark, even referenced to its full 7+ volt maximum output, is 19.3 ENOB. The FiiO E10, even in 24 bit mode, is only 16.2 ENOB.

DISTORTION: The ODAC prototype on a standard –1 dBFS 1 Khz signal has about 0.003% THD+N while the Benchmark DAC1 is only slightly lower at 0.0025%. The ODAC, on every distortion test I’ve run, is well below my ideal 0.01% maximum THD+N. That’s true even at 0 dBFS which wrecks havoc with some DACs (like the NuForce uDAC-2). The ODAC’s performance does not noticeably degrade in any way at 0 dBFS so I can officially certify it Lady Gaga compatible (scroll down to Lady Gaga In Audacity here).

JITTER: You can’t put a single number on jitter performance but I’m quite happy with the ODAC’s results so far. It can’t match the expensive ASRC jitter reduction used in the Benchmark DAC1 but it’s significantly better than most DACs, including the FiiO E10, and several of my pro sound audio interfaces. The ultimate proof will be in the blind listening tests.

IT MAY GET EVEN BETTER: It’s a bit like peeling an onion. As you improve one area of the design, the lowered noise and distortion reveals other areas that can benefit from refinement.While there was some hope the version on my bench now would be close to the final production ODAC, we’ve decided to go one more prototype iteration and incorporate several more incremental improvements. Some of those changes may improve the performance still further but it’s hard to know by how much. When it’s all said and done there will be at least 4 generations of ODAC PC boards. Despite the more than acceptable performance of the current prototype we’re not done chasing the Benchmark DAC1 yet.

---

FAQ (please have a look here before posting questions!)

---

WHAT WILL THE ODAC COST?: The ODAC is still on track to hopefully come in under $100 for a completely assembled, programmed and operational board ready to slip into an O2 or the upcoming ODA. If you want a standalone DAC you have to add at least a few bucks for an enclosure, panel and output jack. We won’t know the final pricing until the production design is bid out for assembly.

WHEN WILL THE ODAC BE AVAILABLE? Given we have another prototype cycle ahead of us the best case is likely late April if all goes reasonably smoothly. But that could slip into May or even later if there are unforeseen problems.

IS THERE AN OFFICIAL ODAC FORUM THREAD? Not yet but I’ll post an update as soon as there is.

CAN I BUILD MY OWN ODAC? Unfortunately no. The ODAC will be offered only as an assembled board. Please see the next question. (PS – You probably wouldn’t want to anyway. It has a lot of 0603 components which are really tiny and a fine pitch quad IC package which are no fun to work with).

WHERE IS THE SCHEMATIC & PARTS LIST? As explained in the first ODAC article, there are no DIY-friendly 24 bit USB audio chips that meet the design criteria. So we’re forced to use components licensed for OEM use that are not available through normal distribution. Even the datasheets are marked confidential and access to the interface chip requires signing an agreement. Given the ICs have to be purchased direct from the manufacturer in substantial minimum quantities it’s not reasonable to expect someone to make a substantial initial investment only to undermine them by giving the same design to other commercial interests. It would be a bit like designing a new car for Ford and then saying “oh by the way we’re also giving this to the Volkswagen Group, General Motors, and anyone else who might want it. You better hope there’s enough sales volume to divide between all of you.”  Having the ODAC be an open source design would only increase the financial risk and the end result could easily be a “chicken and egg” situation. Everyone might choose to wait and see who else might choose to produce the DAC. And, regardless, individual DIYers can’t get the required parts anyway. Hopefully everyone can understand it’s only fair the person taking all the initial risk, and spending the money up front, should not have the rug potentially pulled out from underneath them by sharing the same design to anyone with the commercial means to produce it.

WHAT DAC AND INTERFACE CHIPS DOES THE ODAC USE? For the same reasons we’re not releasing the schematic, we’re not yet releasing certain other details yet either. Please see the question above.

IS THE ODAC BASED ON THE XMOS USB INTERFACE? No. The XMOS solution requires a proprietary Windows driver that must be licensed for a fee. We also don’t believe the XMOS chip offers any real world benefits in this application. See the next two questions.

IS THE ODAC USB AUDIO CLASS 1 OR CLASS 2 COMPLIANT? The ODAC is USB Audio Class 1 (UAC1) compliant because it’s the only standard that allows native 24/96 support in Windows, OS X and Linux without any special drivers. USB Audio Class 2 (UAC2), while newer, offers no meaningful advantages for 2 channel audio playback and is not yet natively supported in Windows. Its main benefits are for multi-channel audio and recording at 24/192. Some of the best performing DACs available at any price are UAC1 devices (including the Benchmark DAC1 series).

IS THE ODAC ASYNCHRONOUS There’s a lot of marketing and other hype lately surrounding asynchronous DACs. Much of it is myth. Just like op amps got a bad name from the old 741 released many decades ago, so did USB audio based on early synchronous designs. But most modern USB audio devices use an adaptive interface where a local clock controls the DAC and is only loosely coupled to the PC’s timing. Contrary to popular belief, with an adaptive interface the data is not directly clocked by the USB port. This method has been refined over the years and can work very well. It’s also natively supported by all major operating systems. Most methods of true asynchronous USB audio require proprietary drivers under Windows and proprietary drivers are rarely a good thing. Judging a USB DAC by whether it’s asynch or adaptive is a bit like judging a car by the engine configuration—i.e. an inline six in a BMW vs a V6 in a Nissan GT-R. Other aspects of the design are more likely to limit the performance and there are plenty of examples of outstanding USB DACs using adaptive interfaces.

WHAT DIGITAL FILTERING DOES THE ODAC USE? Another trendy DAC topic is the type of output filter. It’s mostly about chip and DAC marketing types dreaming up added ways to differentiate their products. But, ultimately, what matters most is the measured performance at audible frequencies and listening tests using proper blind techniques. Judged by those criteria, I’m not sure either filter has an overall advantage. I have not seen any credible blind listening tests that support one filter type sounding better than the other. The ODAC, like many great DACs, uses a linear phase filter. Just like with asynch vs adaptive, there are many other variables that determine the overall performance. You have to consider the entire product as a whole. It’s foolish to dismiss a BMW or Porsche because the engine cylinders are not arranged in a “V” because, ultimately, that’s only a small part of a much bigger picture.

WHAT AUDIO FORMATS DOES THE ODAC NATIVELY SUPPORT? The ODAC supports 44, 48, and 96 Khz each at 16 or 24 bits. 24/192 is not natively supported by Windows over USB, nor does it offer any audible advantages for audio playback (in fact most DACs that support 24/192 perform worse at 192 Khz than at 96 Khz).

ARE YOU PLANNING TO MAKE MONEY OFF THE ODAC? No. I will not get any money from ODAC sales. Someone else is taking all the financial risk and coordinating all the manufacturing and sales logistics. If the ODAC proves popular, any profit is rightfully all theirs. Frankly, given the 4+ design iterations, steep minimum order quantities, and all that’s gone into it so far, they might be lucky to break even. But it’s ultimately no different than JDS Labs and Epiphany making a profit selling the O2. For me this remains an entirely not-for-profit blog and endeavor.

WHERE ARE THE MEASUREMENTS? Given the performance may change (hopefully for the better) with the next iteration it would be a lot of wasted work to publish detailed measurements and graphs only to do it all over again once we have the final design done. I made that mistake with the O2.

WILL THE ODA BE OPEN SOURCE? Yes it will. The ODA will be just like the O2, and anyone who wants to sell the ODA + ODAC can likely negotiate buying assembled ODAC boards with some sort of quantity discount. They might even want to weigh in before the initial production quantity is determined.

WHY NO S/PDIF INPUT? I know there have been several requests for S/PDIF, but as was explained previously, it’s not a trivial or even logical addition. To summarize, here’s why:

• O2 Compatibility – Having the ODAC fit inside the standard O2 is a pretty cool thing. O2 compatibility seems to have more interest than S/PDIF support. It would be very difficult to add an option for S/PDIF and still have the ODAC fit in the O2. If it’s even possible with a 2 layer board it would require components on both sides which would substantially raise the price for everyone. And it would likely have inferior performance due to a less than optimal PCB layout. Likewise splitting the ODAC into two different boards would also raise the cost in multiple ways.
  
• S/PDIF Doesn’t Fill a Niche – There are lots of 24 bit capable S/PDIF DACs as it’s much easier and cheaper to implement compared to 24 bit over USB. Even the $25 FiiO D3 does 24 bit over S/PDIF. But there is a shortage of reasonably priced 24 bit USB DACs. See Filling A Void near the beginning of this article.
  
• No Native Support - There are very few suitable 24 bit USB audio compliant (i.e. driverless) interface ICs available. Likewise there are not many relatively high performance DAC chips that work well in a simple 24 bit USB DAC design without a microcontroller. The chips used in the ODAC lack native support for an alternate S/PDIF or second I2S input and the lack of a microcontroller complicates things further.
  
• More Jitter – Adding an S/PDIF input option would mean “breaking” the I2S bus between the USB interface and the DAC to allow the insertion of an S/PDIF interface and/or switch. Ideally, for lowest jitter, you want to keep the I2S routing as short, clean, and direct as possible. With the current PCB size adding S/PDIF could easily compromise the jitter performance.
  
• Possible Future Version – There is more room in the ODA and it’s possible there may be a more expensive version of the ODAC someday as a “plug and play” replacement for the current ODAC. Such a DAC would likely use more expensive components and possibly a 4 layer PCB raising the performance bar even higher and could also more easily include an S/PDIF input. That way the cost is kept as low as possible for the mainstream version and those who want more can pay more. I also expect some better 24 bit USB audio interface ICs to come on the market compared to what’s available now. Some might even more DIY-friendly and suitable for an open source design.
  

FiiO E10 DAC

INTRO: First off, I’m sorry this review has taken so long. I started receiving requests to review the FiiO E10 long before it was even available in the US. Finally it became available for a short time from a few US vendors but then disappeared. After many months it appeared again and I managed to borrow one in December to test. But the holidays, travel, and year-end workload, intervened. Since the first of the year I’ve had many unplanned non-audio distractions. But, finally, here’s the long awaited E10 review! Thanks for everyone’s patience.

FiiO E10: The FiiO E10 is a simple Chinese designed and built USB powered headphone DAC designed to outperform the older, but more expensive and feature laden, FiiO E7. It has a USB port, headphone output, line output, coaxial S/PDIF digital output, gain switch, bass EQ switch and a volume control. The E10’s closest popular competitor might be the NuForce uDAC-2 I reviewed in February of last year. The E10 is impressively small and highly portable. It’s typically priced around $80 in the USA making it much cheaper than the flawed uDAC-2.

COMPARED TO FiiO E7: The E10 is missing several E7 features. They include battery operation, a line input allowing use as a headphone amp (for say an iPod or other portable device), multiple EQ settings, microprocessor control, electronic volume adjustment, a digital display and the E9 docking option.

FUTURE FiiO E17: If you miss some of the E7 features, but want the better audio performance of the E10, you might wait for the future E17. But, if it’s anything like the E10’s USA introduction, be prepared to wait a while. FiiO seems to be having problems with manufacturing, distribution, and/or releasing products too early.

E10 VERSIONS: Some speculated FiiO revised the E10 during the several months it disappeared from the US market (hence the big gap in shipments). And FiiO recently announced they’ve formally changed it—perhaps for the second time. The version tested here was purchased in December 2011 but is not the latest one pictured on the FiiO website. The latest version has a gold headphone jack.

MADE IN CHINA PROBLEMS? I’m not sure what’s behind the E10 changes. It might be FiiO switched manufacturing to a different factory to save some money. That’s happening a lot in China lately with Chinese labor rates rising rapidly and build quality fluctuating as cheaper labor is being used for assembly. If you’re curious to know more check out this excellent Wired magazine article on recent problems with Chinese manufacturing (it even features a headphone maker). There are many more articles on this topic. The big players like Apple have locked up the best manufacturing in China and the smaller companies, like FiiO, are often having to really struggle to maintain quality at a decent price.

GAIN SWITCH: The recessed gain switch on the bottom adds about 7 dB of gain in the High Gain position. It also causes the amp to clip with a normal input signal at only around 5 or 6 (50% or 60%) on the volume control. The good news is this switch helps adapt the E10 to various headphones. Use the low gain position with sensitive headphones like IEMs. Only use the high gain position if the E10 won’t get loud enough when set to low. The gain switch has no effect on the line output. Do not use the high gain position for headphones 32 ohms or lower as the E10 will only clip rather than produce more output.

BASS EQ SWITCH: The EQ switch only works on the headphone output not the line output. It causes a broad boost that starts at midrange frequencies and is a constant +5 db below about 100 hz. This requires the E10 to produce over three times as much power below 100 hz and can easily push it into clipping.

INSTALLATION: The E10 installed smoothly without needing any drivers in Windows 7 and XP. Depending on where you’re looking, it’s identified as either “DigiHug USB Audio" or “FiiO USB DAC-E10”. I have no idea what “DigiHug” stands for. A Google search brings up an iPad game and cartoon characters. Perhaps it’s a Chinese thing? The PC’s master volume control, unlike with the E7, remains functional.

USABILITY: The E10 is easy enough to use and, for its size, the ergonomics are reasonable. The gain switch is recessed flush into the bottom of the enclosure so it’s difficult to accidently change the gain. The volume knob is a nice size and there’s not much else to worry about besides the bass EQ switch. Like most FiiO products, they mostly did an impressive job with the industrial design for the price.

HEADPHONE REJECTION: One annoying problem is the headphone jack likes to eject the headphone plug with the slightest tug on the headphone cable. Given that most of us move around while wearing our headphones such tugs can happen relatively often. The jack seems to be poorly designed so the spring loaded contact that’s supposed to lock into the groove on the tip of the 3.5mm plug instead barely hangs on to just the tip. I tried several different headphones and they all reacted similarly. Others have reported similar behavior with their E10s. It’s especially bad for headphones with thicker/stiffer cables. This is the sort of problem someone at FiiO surely was aware of but they shipped zillions of E10s anyway. Not good.

DELAYED REACTIONS: Some have complained the E10 delays producing audio when it’s first fed a digital signal (it defaults to being muted when there’s no digital signal or “digital silence”). I confirmed that’s the case but I didn’t try to measure the delay. This is typically only a minor annoyance except for certain applications like live chat/VOIP where the E10 could clip off essential parts of speech.

SUBJECTIVE SOUND QUALITY: The E10 sounded fine with the headphones I tried. Playing some really wide dynamic range music with my Sennheiser HD650s it seemed to clip on the peaks when I pushed it fairly loud in high gain mode. With my very sensitive Ultimate Ears Super Fi Pro IEMs I could hear some hiss even at low gain but it wasn’t objectionable. I also noticed some channel balance error at low volumes with the Super Fi IEMs. I didn’t much like the bass boost with any of my headphones as it mucked up the lower midrange and vocals. It was also very “boomy”.

LACK OF CONSISTENCY: Older FiiO products (even the $20 E5) have better bass EQ than the E10. It’s disappointing there’s almost no consistency between the bass EQ among FiiO products. Every FiiO product I’ve tested is significantly different and they’re getting worse. This implies either FiiO doesn’t care if they get it right or they really have no idea what’s best. Quality audio products, in my opinion, are designed to a consistent set of goals. Things like headphone bass EQ should be based on what works best with popular headphones and listener preferences. That shouldn’t change from amp to amp or DAC to DAC. It’s always sad to see a company go backwards with their new designs.

MEASUREMENT SUMMARY: With some notable exceptions, the E10 measured about as I expected for an $80 headphone DAC. For those who want to use it to directly drive headphones it’s important to know the limitations of the headphone output. For those who want to use the E10 as only a DAC, and will be using the line output, you can skip the next two paragraphs:

• Headphone Output - The E10’s headphone performance was similar to the FiiO E11 and fairly respectable if you can live with a maximum output of 1.5 to 2.5 Vrms depending on the headphones.  The distortion performance was acceptable but a bit marginal into lower impedance loads. There’s also some channel balance error in the sample I tested that was audible with my IEMs and measured higher than I like to see. The bass EQ option is far from ideal and more “boom” than bass. Using the EQ may also overload the amp and/or many headphones at loud levels.
• Headphone Compatibility - There may be some audible noise with sensitive headphones (especially B.A. IEMs like Shures, Etymotics, Ultimate Ears, etc.). Even the low gain setting forces using only about the first 30% of the volume control’s range with these IEMs. At the other extreme, even set to high gain, the E10 is marginal for having enough output for many popular full size cans like the Sennheiser HD650/600. If you listen at “live” levels to highly dynamic music the E10 can easily clip with such cans. For headphones that require even more output, like the popular 600 ohm Beyers, several AKGs, HiFiMan planars, etc., you’re likely to be disappointed. The good news is the E10 has a low output impedance so it’s well suited for most IEMs and some sensitive low impedance cans like the Denon D2000. When looking at headphone reviews on InnerFidelity, look for a 90 dB SPL voltage of 0.15 Vrms or less (up to 0.25 Vrms might be OK if they’re high impedance and you don’t like loud levels). For headphones rated in dB/V look for a rating of at least 106 dB. For headphones rated in dB/mW things get more complex. See the More Power article.
• DAC Performance – The DAC performance of the E10 is a mixed bag. Overall it’s a more respectable DAC than say the NuForce uDAC-2 or any of the inexpensive DACs I recently tested including the UCA202, Creative X-Fi Go, and Asus U3. But its 24 bit performance at 44 Khz was far short of what it should be. The dynamic range from the line out was 97.6 dB which is only 16.2 effective bits of performance. That’s barely better than a good 16 bit DAC. The noise and dynamic range is still likely “good enough” if you use the volume control on the E10 and leave your PC volume turned all the way up. But if you want to control the volume from your PC, you might want to look for a DAC with closer to 18 effective number of bits (ENOB). That allows enough dynamic range to allow penalty free use of the PC’s volume control—the E10 falls short.
• E10 vs E7 – The most obvious advantage of the the E10 is significantly more output than the E7 and its ability to drive a wider variety of headphones—especially higher impedance models. The E10 is relatively free of the excessive ultrasonic “crud” that’s always present in the output of the E7 (and E5/6). How much the “crud” is audible is difficult to say but it’s a potential concern with the E7. The E10 DAC can also run in 24 bit mode, and while it barely delivers better than 16 performance, it is still quieter and has more dynamic range than the E7. The E10 generally has lower distortion than the E7. Finally the E7 disables the volume control in the PC’s operating system and the E10 does not.

Here’s how the E10 measures up to some other headphone DAC’s I’ve tested. The ratings use a letter grade from A to F where A is excellent and F is Fail (unacceptable). In some cases the letter grade takes into account more than just the raw number:

Measurement	FiiO E10	Asus U3	X-Fi Go	UCA202	FiiO E7
Freq. Resp. 10K	+/- 0.1 dB A	+/-1.5 dB C	+/- 0.4 dB A	+/- 0.1 dB A	+/- 0.1 dB A
Freq. Resp. 33 ohm	+/- 0.1 dB A	+/-1.5 dB C	+/-5.0 dB D	N/A	+/- 0.1 dB A
HP Output Imp ohm	0.5 A	23.6 D	7.8 C	47 F	0.13 A
Max Output 10K	1.65V B	2.15V A	1.0V	1.12V B	N/A
Max Output 32 ohm	1.55 V B	0.9V C	0.75V C	N/A	1.3V B
Max Power 32 ohm	75 mW B	25 mW C	18 mW C	N/A	53 mW B
Max Power 600 ohm	10 mW D	8 mW D	2 mW F	2 mW F	3 mW F
THD+N 0 dBFS 10K	0.005% A	0.01%	0.007% A	0.008% A	0.14% C
THD+N 100hz 10K	0.004% A	0.008% A	0.007% A	0.007% A	0.07% C
THD+N 100hz 32 Ω	0.004% A	0.02% B	0.02% B	N/A	N/A
THD+N 1Khz 10K	0.004% A	0.008% A	0.007% A	0.007% A	0.03% B
THD+N 1K 32 Ω	0.004% A	0.04% B	0.009% A	N/A	0.03% B
THD+N 10Khz 10K	0.004% A	0.008% A	0.009% A	0.009% A	0.06% C
THD+N 10Khz 32 Ω	0.015% B	0.01% B	0.01%B	N/A	N/A
IMD CCIF 10K/32 Ω	0.013% B	0.004% A	0.004% A	0.005% A	0.03% D
IMD SMPTE 10K	0.004% A	0.004% A	0.0005% A	0.002% A	0.008% B
Noise A-Wtd dBu	-98.3 C*	-91.6 B	-88.9 C	-88.8 C	-93.7 B
-60 dBFS DNR dB	-97.6 C	N/A	N/A	N/A	N/A
-90 dBFS Linearity	0.0 db A	1.2 dB B	1.5 dB B	3.8 dB C	0.1 dB A
USB Jitter Jtest	VG B	VG B	VG B	VG B	Fair C

*NOTES: The E10 noise is relative to 24 bit performance. The other DACs were tested at 16 bits.

FIRST CLASS:

• Small size
• Relatively low price
• Decent headphone output within its limitations
• Low headphone output impedance
• Respectable 16 bit DAC performance
• Two gain modes

ECONOMY:

• 24 bit mode only offers 16.2 bits effective resolution
• Audible noise with highly sensitive headphones
• Channel balance error may be audible
• Poor bass EQ is boomy, corrupts midrange, and may cause overload
• Line output is nearly 2 dB below 2 Vrms Redbook standard
• Headphone plug pops out of the jack easily
• Not enough output for some popular full size headphones
• Clips above 50% volume in high gain mode
• Delay may mute start of audio in some circumstances
• Marginally high line output impedance
• Potential problems if connected to grounded equipment
• Marginally high DC offset—especially in high gain mode

BOTTOM LINE: The E10 has some issues but still is a clear step up from $30 to $40 products like the X-Fi Go and Asus Xonar U3. Overall, if it has enough output for your headphones and you plan to use the volume control on the E10 or a downstream amp, it’s worth considering—especially if you’re on a tight budget. But if you want to control the volume at your PC, you might be better off with a 24 bit DAC that offers more dynamic range and effective bit resolution like the HRT Music Streamer II (which requires a headphone amp like the O2) or a Centrance DACport with the low impedance output option. If you have difficult to drive headphones you may need a dedicated high output headphone amp like the O2 portable amp or upcoming ODA/ODAC.

---

TECH SECTION

---

TECH INTRO: I’ve been getting a lot of feedback lately saying things like “nice blog but your reviews are way over my head”. I want to make it clear if you’re not a total audio geek, engineer, scientist, etc. you probably want to stop reading here. All the important stuff has already been summarized above. The rest is mostly just the behind the scenes details for those who care about such things.

TECH INFO: The E10 supposedly uses the setting same AD8397 op amp as the FiiO E11 and AMB Mini3. It’s a high current, and somewhat “high strung” op amp that works better in the E10/E11 than the Mini3. The E10 also supposedly uses the Wolfson WM8740 DAC chip. Unless otherwise noted, I made most measurements using the low gain setting. Several measurements, where applicable, were made from both the line and headphone outputs. I also spot checked some measurements at 24/48 and 24/96 where most performance was similar to 24/44. The E10 is not usable at 24/88 due to a limitation of the USB interface. For more information on my audio measurements, see Music vs Sine Waves, Testing Methods, and Headphone Amp Measurements (InnerFidelity).

VIRTUAL GROUNDS: It would seem the E10, unlike the Mini3 clone E11, does not use a virtual ground. This makes sense because the virtual ground in the E11 prevents using the USB charging port while using it due to the potential severe ground conflicts. The E10, being entirely USB powered, has to operate connected to the PC’s ground. That said, when I connected the E10 to my grounded Agilent scope it freaked out and had massive DC offset. So while there’s no DC voltage between the USB ground and the input or output audio grounds, the E10 seems sensitive to ground loops. Use extreme caution when using it with grounded equipment.

MAXIMUM OUTPUT LEVEL AT CLIPPING: At the low gain setting the E10 will not clip into 35 ohms or higher impedance loads. Below 35 ohms, however, it mildly clips at near full volume with a 0 dBFS signal. At the high gain setting it always clips well before full volume even with no load. The clipping point drops proportionally with the load impedance. This is a bit disappointing as many users will need to be careful not to set the volume too high or they risk severe clipping distortion. But the flip side is having some excess gain for “quiet” recordings. The maximum output above 35 ohms was 1.55 Vrms at the low gain setting and it varied at the high gain setting as the E10 runs out of current below about 35 ohms. It reaches a voltage limit of 2.7 Vrms into no load at hard clipping. This works out to for <1% THD (low gain/high gain):

• 16 ohms = 1.5 Vrms 140 mW / 1.5 Vrms 140 mW
• 32 ohms = 1.5 Vrms 70 mW / 2.1 Vrms 138 mW (or 75 mW at initial 1.55V clip point)
• 80 ohms = 1.55 Vrms 30 mW / 2.5 Vrms 78 mW
• 300 ohms = 1.55 Vrms 8 mW / 2.52 Vrms 21 mW
• 600 ohms = 1.55 Vrms 4 mw / 2.6 Vrms 11 mW
• Line Out 100K = 1.65 Vrms, 10K = 1.55 Vrms, 600 ohms = 0.82 Vrms.

CLIPPING ODDITIES & CURRENT LIMITS: The E10, when set to High Gain, has the same odd clipping behavior below about 35 ohms as the E11 does. This isn’t surprising as the E10 shares the E11’s output AD8397 op amp and perhaps some power supply details as well. The true clipping point is hard to define but distortion rises dramatically above 1.5 Vrms into 32 ohms. This is very likely caused by the DC-DC converter running out of current and letting the supply rails sag badly. This isn’t ideal behavior as it means the amp is generating substantial distortion over a relatively wide range of output levels into 35 ohms or less before it’s obviously clipping. The O2’s current limiting, by comparison, takes place in the final output stage (versus the E10’s power supply) and the O2 maintains low distortion right up to the true clip point.

LINE LEVEL OUTPUT & GAIN: The term “gain” is a bit hard to define for a device with only a digital input. The Redbook standard for digital audio devices is 0 dBFS of input is supposed to give 2 Vrms of output. With the E10 you only get, at best, 1.65 Vrms. That gives a “gain” of negative 1.7 dB (i.e. the E10 is 1.7 dB below the reference standard). The gain switch has no effect on the line output but adds 7 dB of gain to the headphone output in the High Gain position. The low gain position is still arguably on the high side for balanced armature IEMs and other highly sensitive headphones that only need about 0.3 volts versus the 1.5 volts the E10 delivers. So, with sensitive headphones, you’re stuck using only a small portion of the E10’s volume range which makes the channel balance problems worse and small volume adjustments more difficult.

EXCESS GAIN: For lack of a better description, “excess gain” is my generic term for “extra volume control range” above and beyond the normal clipping point (or desired maximum output) with a 0 dBFS signal. For the E10 in high gain mode it’s about 3 dB into high impedance loads and increases into loads below 35 ohms as the E10 runs out of current and clips sooner. The E10, in high gain mode, starts clipping around “5” or “6” on the volume control (50% to 60% of the range) depending on the load with a typical input signal where the peaks hit 0 dBFS.

0 dBFS DAC PERFORMANCE: The E10 distortion only rose slightly from 0.004% to 0.005% with a 0 dBFS input versus –1 dBFS. This means there’s no significant problem with clipping in the DAC chip or related circuitry (unlike the oddly designed NuForce uDAC-2 which rose to nearly 1% at 0 dBFS).

THD+N vs OUTPUT: The E10 has acceptable but not great distortion performance at 1 Khz. The green line in the graph below represents the upper limit of what’s desirable. In high gain mode the E10 flirts with the limit around 0.75 volts output which is marginal performance. The red line shows the line output which maxes out at 1.65 Vrms with no clipping and reasonably low THD+N. The yellow line shows the odd clipping behavior (likely due to power supply limitations) into 32 ohms mentioned above. The channels were fairly closely matched but not exactly. The aqua and blue lines show the imbalance—likely due to PCB layout problems or compromises.  The two blue-ish lines should be on top of each other. At higher output levels into 16 ohms the currents flowing in the PCB traces are highest and that’s where PCB layout is the most critical. See the Legend for a description of each line:

OUTPUT IMPEDANCE: The output impedance is measured at 100 hz as that’s typically where headphone damping is most critical. The E10 dropped from 400 mV no load to 389 mV with a 16 ohm load. This works out to about a 0.5 ohm output impedance which is acceptably low. The line output had a 600 ohm output impedance which is higher than I prefer to see (under 100 ohms is typical). The E10’s line output may have problems with long cable runs, amp inputs with low impedances, passive preamps/volume controls, higher noise, and the input filters in some devices. It will, however, work OK with the O2 and ODA’s input circuitry.

16 BIT RESOLUTION : The following is a 1 Khz undithered sine wave at an extremely low level of –90 dBFS. A proper bit accurate DAC (no upsampling, internal dither, etc.) should reproduce the sine wave as three distinct stepped levels. The E10 performs as expected:

24 BIT RESOLUTION: When the above test is run at 24 bits, if the DAC really supports more than 16 bits, you should get something close to a sine wave. The E10 performs correctly:

FREQUENCY RESPONSE: The frequency response is plotted below from 20 hz to 20 Khz via the line output. It was the same at 16 and 600 ohms from the headphone output. There’s nothing to complain about here and it indicates the E10’s line and headphone outputs are likely direct coupled (no capacitors):

THD+N vs FREQUENCY -1 dBFS: The yellow/orange plot shows the distortion vs frequency via the line output and the blue plot shows the headphone output into 32 ohms at 1.38 volts (-1 dBFS at full volume and low gain). Below about 1000 hz the two outputs are similar. Above 1 Khz the headphone output exhibits rising distortion getting very close to the green threshold line. The drop above 7 Khz is normal due to the intentional bandwidth restriction of the measurement. The line out performance is fine while the headphone output is marginal but acceptable. It should be noted the headphone output is shown at low gain. It was significantly worse at high gain even at the same output voltage:

SMPTE IMD: This is a twin tone test of both low (60 hz) and high (7 Khz) frequencies together in a specific ratio as defined by the SMPTE standard. The dScope ignores the regular THD from each tone and instead looks for distortion created by the two tones interacting with each other. But the raw number doesn’t tell the whole story. It’s also important how much “spread” there is at the base of the 7 Khz tone. Ideally everything should be comfortably under –80 dB and that’s the case with the E10 into 32 ohms at around 1 volt. The line output was a bit better than shown here but similar:

CCIF IMD: This also a twin tone test but at 19 Khz and 20 Khz. This test is far more revealing of audible forms of high frequency distortion than simple THD measurements are above 6 Khz. The reason is the third harmonic of signals 7 Khz and higher are beyond the range of human hearing and beyond the typical cut off frequency of the distortion analysis. But an amplifier with high frequency non-linearities will create even-order distortion products on this test at 1 Khz, 2 Khz, etc. and odd order distortion at at 18 Khz, 17 Khz, etc. All of these are within the audio range and, if above –80 dB, may be audible under some conditions. The E10 struggled a bit (especially in Ch B) with several distortion markers reaching for the green threshold line and one barely exceeding it. it’s notable there are significant differences between the channels. The yellow channel was quite a bit worse likely due to the PCB layout as previously discussed. This is marginal but acceptable performance for 1 volt into 32 ohms. Again, the line output was a bit better:

NOISE & DYNAMIC RANGE 16 BIT: Reproducing a 1 Khz signal at a very quiet –60 dBFS the noise is measured by the dScope removing the 1 Khz signal and its harmonic distortion and calculating the left over noise out to 22 Khz. This test is referenced to the maximum clean output of the device at 0 dBFS. A perfect 16 bit DAC has 96 dB of un-weighted dynamic range. Most audio noise measurements supplied my manufactures are A-Weighted which adjusts for the sensitivity of human hearing to noise. Using the line output, and referencing its maximum output of 1.65 volts, the dynamic range (DNR) is approaching the best you can get from 16 bits which is 96 dB. This shows the DAC, not the analog circuitry after it, is the limiting factor and about as good of performance as you can expect from a 16 bit DAC. The headphone output had only slightly more noise so the DAC chip itself seems to be the dominant noise source in the E10:

NOISE & DYNAMIC RANGE 24 BIT: Switching to 24 bit mode, with everything else the same as above, things don’t improve nearly as much as I would have expected. If you look closely you can see the central part of the noise floor drops about 10 dB. That’s what I would expect. But, unfortunately, there are a bunch of new noise components apparently related to operating the DAC chip in 24/44 mode and reproducing the –60 dBFS 1 Khz signal. The result is the total A-Weighted dynamic range is 97.6 dB. That’s only about 2 dB better than the 16 bit result above and only 16.2 effective bits of performance. While this is still OK for applications that use the E10’s volume control it might be a problem for those wanting to leave the E10 volume turned up and use only the PC’s volume control. Especially with sensitive headphones you may hear some noise from the E10 under similar conditions. I’m not sure what’s causing the extra noise but my guess would be the digital filtering on the Wolfson DAC isn’t optimized for 24/44 operation:

DAC LINEARITY & 24 BIT ABSOLUTE NOISE: Linearity measures how accurately a DAC can reproduce very low levels. A DAC should be within +/- 1 db at –90 dBFS and many remain linear to even lower levels. The E10 does fine in the linearity department, but one channel has much more high frequency noise than the other in 24/44 mode. The extra noise goes away in 16 bit mode. This seems related to the same problem discussed above but here it’s low enough in level to not be a big problem. The absolute noise is 49 uV or –98 dBu which is quiet enough for most full size cans, but not for total silence with really sensitive IEMs—especially if you control the volume at the PC:

MODULATION NOISE: Ideally the noise floor of a DAC will remain constant regardless of the signal level. In practice, this is rarely true and the noise typically rises with the signal. If the noise is high enough this can be an audible problem as the noise will fluctuate in response to the music—something the ear seems especially sensitive to. The important criteria is how flat the noise curve is with respect to the level and how far below the green threshold it is. The blue-sih lines are the two channels in 24/44 mode while the orange/yellow lines are the two channels at 16/44. At lower levels the E10 is quieter in 24 bit mode but, interestingly, at around –12 dBFS and above, it’s quieter in 16 bit mode. Again, I suspect this is related to the problem in the previous two tests in 24 bit mode. Everything is below the green line, and hence likely inaudible, but this is more odd behavior. The blue line should be at least 6 below the orange line across the entire graph:

CROSSTALK: Into 32 ohms the E10 measured a consistent 55 dB of crosstalk across the audio range. The flat result implies it’s almost entirely limited by the impedance of the ground circuitry and shared common ground in the headphone jack. It’s an acceptable result but just below the green threshold at –50 dB. Using the line output the crosstalk is about –95 dB best case at low frequencies and steadily rises to about –55 dB at 20 Khz. This shows high frequency coupling in the PCB layout and volume pot. This is also an acceptable result:

CHANNEL BALANCE: With the volume set at 40% (a typical setting) there was about 0.25 dB of channel balance error as can be seen below. I didn’t do my usual graphing at many volume settings, but adjusting the volume knob and watching the dScope’s live reading of channel balance, the error was somewhat worse than usual overall. At my usual level of –45 dB below full volume it measured 1.3 dB which is over the 1 dB threshold and generally audible. At even lower levels the error was as high as 5 dB while at higher levels it varied randomly with the volume setting but was generally around 0.25 – 1.0 dB. This is marginal performance and a sign FiiO probably didn’t spend enough money on a decent volume pot:

JITTER & INTERCHANNEL PHASE: Here’s the spectrum from the dScope’s J-Test for jitter. It’s a special signal at 1/4 the sampling rate with the lowest bit toggled on and off. It’s not possible to summarize this test in one number. What matters most are the number and height of symmetrical distortion components mirrored on either side of the 11025 hz signal which indicate high frequency jitter components. And the amount of “spread” at the base of the signal indicates the relative amount of low frequency jitter. Ideally nothing should reach above –110 dB (with a slight allowance for exceptions near the main signal). But, it’s worth noting, there is considerable debate about the audibility of various amounts of jitter (see: Jitter Does it Matter?). The E10 stays under the green guideline but barely. The result is shown for both 16 and 24 bit modes. Besides the noise floor dropping in 24 bit mode not much else changes. The phase error, as would be expected with a decent DAC chip like the Wolfson, is essentially zero. This is an acceptable, but not great, performance:

BASS EQ: I’m not very impressed with the E10’s bass EQ. It reaches well up into the midrange which makes vocals (especially male vocals) sound oddly “tubby” and thick. It has almost full boost at 100 hz which is more “boom” than true bass. And, finally, it doesn’t taper off at really low and subsonic frequencies. This means when playing content with a lot of really deep bass energy (like some dance music) the E10 may clip and/or overload your headphones (bottom out the drivers) when using the EQ.  I prefer EQ curves that peak around 50 hz and rapidly taper off on both sides. The E10’s EQ is far from that ideal and notably worse than the EQ in some earlier FiiO headphone products. Going backwards is not progress:

CLIPPING PERFORMANCE: The E10’s headphone output clipped cleanly at around +/- 4 volts peak in high gain mode into 600 ohms. Dropping the load to 16 ohms causes the E10 to be current limited and it clips at just over 2 volts peak. In both cases the clipping was clean and close to symmetrical even when viewed on a high speed scope. There was only a small amount of ultrasonic “crud” from the DC-DC converter in the E10 (not visible here). The E10 is massively better in this regard than the TI chip amp based E5, E6 and E7:

IMPULSE RESPONSE: The E10 preserves absolute polarity and uses a classic linear phase digital filter with pre and post ringing. The same headphones and capacitive test load that gave the MacBook Air fits is handled by the E10 with minimal fuss. Unless you’re a fan of minimum phase filters (no pre-ringing), there’s nothing remarkable here:

DC OFFSET: The E10 had significantly higher DC offset than I’m used to seeing, In the low gain mode it was 14 mV in both channels. In the high gain mode it was 30 and 31 mV. This is marginal for highly sensitive IEMs and also causes some noise when you plug the headphones in. As mentioned earlier it’s apparently a DC coupled design and that may have been a poor choice without otherwise managing the resulting excessive DC offset. FiiO is clearly not sweating the hidden details like they do with the visible ones (like the case, controls, packaging, etc.). It’s form over function.

TECH COMMENTS: With a couple of notable exceptions, the E10 measured about like I expected it to. The exceptions are:

• 24 Bit Performance - The biggest disappointment, by far, was the 24 bit performance using an industry standard –60 dBFS 24/44 signal. Wolfson quotes 117 dB A-weighted SNR at 24/48 (19.5 ENOB) for the DAC chip and the E10 only delivers about 97 dB (16.2 ENOB). I don’t expect any DAC to meet the chip’s datasheet spec, but the E10 missed by a huge 20 dB margin which rates a solid FAIL from an engineering perspective. While the E10 performs better at 48 Khz that’s only useful for watching DVDs unless you want to resample 99% of the digital music out there. Re-sampling likely creates more problems than operating the E10 at 48 Khz solves. The 24/44 issue showed up in several tests so it’s clearly not some measurement fluke. It indicates a  potential implementation problem—likely in how FiiO is configuring (or failing to properly configure) the Wolfson chip for 24/44 operation. When there’s no microcontroller the options are often very limited with communication between the USB interface and the DAC chip. The result is the DAC chip may fall on its face for certain data formats. That seems to be the case with the E10 and 24/44.
• Poor Choice of DAC Chip: If the above is an unavoidable problem without adding a microcontroller it’s my opinion FiiO chose the wrong DAC chip. Lots of small “boutique” audio manufactures seem to put FOTM chip brands and part numbers ahead of common sense. They use parts popular among audiophiles but they’re often completely the wrong part for that particular design. That may well be the case for the E10 and for the E11 which uses the same OPA690 op amp as the Mini 3. The OPA690 is not even remotely designed or specified for audio use and holds back the performance of both the E11 and the Mini3. The WM8740 may have been a similarly poor choice for the E10. I just don’t get the logic behind such design decisions when it so obviously compromises the performance.
• Odd Clipping/Current Limiting – Like the E11, the E10 struggles with loads below about 35 ohms. I’m guessing the DC-DC converter runs out of beans and the supply rails sag badly. That likely creates the very odd distortion behavior above 1.5 Vrms into 32 ohms or lower. The E10 would be a poor choice for low impedance current hungry headphones like most planars. This creates a situation where the amp starts distorting relatively severely before it’s obviously clipping.
• Limited Voltage Output – The line output fails to meet the Redbook standard of 2 Vrms and the headphone output, in high gain mode, maxes out at only about 2.6 Vrms which is well short of the 5+ Vrms many headphones require including the popular Beyer DT880-600. It’s also a marginal output level for even many of the popular full size Sennheisers and AKGs. But it’s only fair to consider it’s an $80 tiny USB powered headphone DAC. Some compromises are to be expected. At least it outperforms all the cheaper USB headphone DACs I know of.
• Channel Balance Error – The channel balance was a bit disappointing and notably worse than devices I’ve tested that use an Alps RK09 volume pot such as the O2 amp. FiiO probably chose a cheaper pot to keep the price down. The parts and labor costs of the E10 likely total less than $20 in the volumes FiiO buys at. Again, you have to expect compromises at the E10’s price.
• Marginal THD+N & CCIF IMD – The noise floor was higher than I would like to see in 24 bit mode and the distortion driving 32 ohm headphones was a bit excessive under some conditions as well (CCIF IMD and above THD 3 khz). This indicates the AD8397 implementation and/or PC board layout are less than ideal. There was also some significant channel asymmetry between the high frequency distortion and noise performance indicating a compromised or flawed PCB layout. Again, given the small size and low price, this isn’t too surprising.
• Marginal DC Offset – The DC offset is significantly higher than I like to see. It’s many times the typical offset of the O2. FiiO probably should have addressed this either using DC blocking caps or a DC offset compensated design. It’s something few users will be aware of but indicates they cut some serious corners in the design and/or consider marginal specs acceptable.

FINAL WORDS: If you’re OK with its limitations, and for the right applications, the E10 is worth its $80 price and a better choice than some of its competitors. I would choose the E10 over the NuForce uDAC-2, for example, without hesitation. Those with more challenging headphones may want to consider adding a higher quality headphone amp like the O2. And those who prefer to control the volume from their PC may want to choose a 24 bit USB DAC with higher dynamic range like the Centrance DACport with the low impedance output option. When used within its limitations, the E10 is decent headphone DAC for $80. But it’s compromised in several areas likely due to the low price, small size, and being USB powered.