Pros: Excellent sound: detailed, transparent, and precise with very little grain, headphone amp even more capable than prior version
Cons: No support for 88.2kHz, no async USB
Yulong D100 MKII
A few years back I did a fairly extensive evaluation of the Yulong D100. It was my first experience with the brand and I was exceedingly impressed with its performance – so much so that I’ve tried almost all the Yulong products since then. Each has been impressive for its cost: The U100 is a nice compact USB DAC for a low price. The A100 amp complements the D100 nicely and offers extreme clarity and focus; I can’t believe it has not become more popular. And the new Sabre D18/A18 flagship combo is quickly becoming one of my favorite setups – I just can’t stop listening to it. Yet the D100 remains Yulong’s “bread and butter” product. It’s cheap enough to not be too far out of reach for people new to the hobby. And it sounds good enough to be used in a very high quality system. I like the fact that it grows with you – start with using it as an all in one DAC/amp. Later add a nice external headphone amp. Want to go balanced? The D100 has XLR outs. Of all the many items I’ve reviewed over the past few years, the D100 may be the one I’ve recommended to the most people.
When something is that good, how does the designer possibly improve it without raising the price by a significant margin? And does the D100 even need to be improved? From where I’m sitting it still looks like one of the top dogs in the sub-$500 category. Realistically Yulong didn’t really need to do anything with it. But they decided to go about it in a way that would offer a modest improvement without raising the price – incremental tweaking to the design based on several years of feedback from numerous customers. If you know where to focus, you need only to change a few things rather than rework the whole design.
Worth noting is the fact that Yulong now has an official North American distributor – Calgary based Grant Fidelity has an extensive track record for customer service, and offers full warranty and after sales support. They even give a 30-day in home trial period. That takes away the uncertainty of ordering from an overseas distributor.
First up, if you haven’t read my original review of the D100, I suggest you take time to do so now. The general description of the MKII is the same and I don’t intend to cover all the same material - I’d rather focus on the specific changes to the hardware and the resulting difference in sound.
I admit that when I first became aware of the MKII, I was under the impression that it had only very minor changes. Specifically, I knew that the original had some random compatibility issues with the USB circuit: it just didn’t get along with certain computers. So I knew that had been redone. I also heard that the headphone amp section received minor changes to achieve a slightly warmer sound, due to some complaints that the original lacked enough body. So I really wasn’t expecting much of a difference at all. But I had to find out for myself.
As it turns out, there are some fairly significant changes that I was not aware of. Through the always helpful Rachel at Grant Fidelity I was able to conduct an interview with Yulong himself, the man behind the D100. Once again I’m grateful for her assistance since my Chinese is a bit lacking.
Q) Can you describe the changes that make the headphone amp sound a little warmer than the original model? I don’t see any changes to the circuit itself.
A) The original circuit design of D100 is very mature in our own opinion. There is no real need for modification of the design. The MKII version changes are mostly on sound tweaking through adjustment of circuit parameters. We have been collecting customer feedback on sound since we started selling D100 years ago and D100MKII is the result of delivering a sound that matched the feedback we collected.
Q) Is it still the case that the ASRC function resamples all data to a sample rate which you find to be optimal for the components in this design?
A) D100 re-sampling rate was at 110kHz and this remains the same on D100MKII. We have chosen to use Japanese NDK high precision low noise crystal clock oscillator in the D100MKII so the final measurement is superior to the D100.
Q) Did you notice any difference in jitter performance from the first version, when using a USB input? I know some people think the Tenor TE7022 receiver is inferior to the TI TAS1020B and I'd like to dispel that argument.
A) Jitter performance is mostly affected by the ASRC circuit, not by the USB chip. As mentioned above, we used NDK clock oscillator in the MKII so the jitter measurement is superior to the original D100. As to which USB chip to use, TE7022 is for USB2.0 standard while TAS1020B is for USB1.0 standard. From real world applications, it has been confirmed by industry designers that T7022 is more reliable with better sound performance. Overall, D100MKII sound warmer, richer and offers better tonality balance / less fatiguing than the D100, plus more reliability through USB input with TE7022. We believe this will offer customers a better listening experience overall than the previous D100.
Q) Has there been any change in the function of the sound mode 1 and sound mode 2 options? Or does it still only deal with very high frequencies above 15kHz?
A) The D100MKII offers sound differentiation from 10kHz and up - this is better than the previous 15kHz benchmark. The sound difference between Mode 1 and Mode 2 is more distinct than before.
So the D100MKII remains very similar to the original with the exception of the changes mentioned above. Hardware is still impressive - CS8416 DIR accepting toslink, coaxial, and AES/EBU inputs up to 24-bit/192kHz. AD1896 asynchronous sample rate converter. AD1955 DAC chip. Output stage with ADA4075 opamps for I/V conversion and OP275 for LPF and buffering. MCU controller. Linear power supply with large Plitron brand toroidal transformer. Headphone amplifier using OPA2134 opamp driving a discrete transistor diamond buffer. Though it was released several years ago the D100 is by no means out of date in terms of hardware. The changes are such that you won’t really spot anything different just by opening the case.
I’m certainly glad I asked about the “sound mode” update. I suspected there was some action in that department because there seemed to be a much more obvious different between sound mode 1 and sound mode 2. The original model made it very hard to discern and I always thought it was a missed opportunity – the idea was welcome but the implementation was off. Setting the -3dB point at 10kHz rather than 15kHz makes a much more significant impact in day to day listening. It still isn’t night and day between each mode but it can be useful in certain instances. I particularly like the results with the various Utrasone and Beyerdynamic models I tried – models which I find to have some excess energy in the highs that quickly becomes fatiguing. I realize that a similar result could be accomplished via EQ, but that isn’t a possibility unless a computer is involved in your playback chain, which usually isn’t the case in my house.
For the sake of clarity, allow me to summarize all the changes that I’m aware of between the original version and the MKII update.
Upgraded system clock
1) Different USB receiver
2) Upgraded system clock
3) Minor circuit tweaks in the headphone amp section resulting in a warmer sound
4) Lower threshold for “sound mode” options
5) Updated Yulong Audio badge on front panel to match the Sabre models
It isn’t a vast amount of changes but as I’ll explain, it is enough to keep the D100 at the top of the sub-$500 DAC category.
Associated gear used for this review:
Source: JF Digital HDM-03S music server, Squeezebox Touch modified with Enhanced Digital Output, Acer Aspire One laptop
Amps: Violectric V200, Analog Design Labs Svetlana 2, Yulong A100, Matrix Quattro Balanced, Yulong Sabre A18
Headphones: HiFiMAN HE400, Lawton LA7000, Heir 6.A, Audio Technica W1000x, UM Merlin, Lear LCM-2B, V-MODA M80, Earproof Atom
Power handled by a CablePro Revelation conditioner and CablePro Reverie AC cables. Interconnects include Signal Cable Analog Two RCA and Pailiccs Silver Net XLR. A variety of headphone cables were used including CablePro Earcandy (single ended) and Toxic Cables Hybrid (balanced) for HiFiMAN, with Beat Audio Cronus and Supreme Rose for custom IEMs.
After burning in the MKII for a week or so, I got right to it – direct comparisons to the original were the main goal on my agenda. Initially I did some listening on the new unit and from memory didn’t think it was much different than the original. I even said so in one of the forum threads. But when I put them head to head I was surprised to notice that the improvement was more significant than I had anticipated.
Starting with the headphone amp – I did see how the original could be a bit cold at times, especially with certain headphones. It was not my first choice with bass light models like the Audio Technica W1000, or even the more well rounded W1000x. Nor did I love it with my 600 ohm Beyer DT990s. The MKII is noticeably warmer in those instances. It won’t turn a Grado into a Denon. But in cases where a headphone is on the fence between “too cold” and “just right”, the MKII will tip it to the favorable side. When I heard the term “warmer” to describe the update I was concerned that it might lose some clarity in the process – clarity that people have come to enjoy from the original model. Thankfully that isn’t the case. If anything, the new model seems slightly smoother in the upper mids and highs. No less detailed, but smoother in the best possible sense of the word. This means some potentially grating recordings have more of a chance: I’m currently listening to Something to Write Home About by The Get Up Kids - certainly not the last word in audiophile quality recordings. The MKII presents it in a more tolerable and well rounded manner compared to the original. As we scale up to higher quality recordings the focus shifts from making things tolerable to extracting a bit more emotion and realism. Listening to Pancho Sanchez Conga Blue in 24/88.2 resolution, the MKII gives a more convincing sense of impact to the various percussion instruments on display. Maybe it stands out more to me as a drummer, but I do hear a decent improvement. And after listening to some of my favorite classical albums from 2L Records and Reference Recordings, I do hear a subtle improvement in refinement as well – the MKII handles complex pieces more gracefully, with improved air and separation being the main benefits. At the end of the day this amp is still not going to satisfy a hardcore headphone geek by replacing a high-end dedicated unit. But an extra helping of refinement and “maturity” certainly never hurt anyone, and for some folks this will be all the amp they ever need.
Moving on to the DAC section – the thought came to me that perhaps some of the improvement in the amp section was not due to changes to the amp at all, but rather a byproduct of the improved DAC which feeds it. I was hoping that I could discern a concrete difference when using the DAC only, in order to prove this theory correct. After listening back and forth with several nice amps including the Analog Design Labs Svetlana 2 SET amp, the Violectric V200, and Yulong’s own flagship Sabre A18 balanced amp, I was able to determine that the DAC itself is indeed subtly improved. Resolution seems turned up a small notch overall, while the top end becomes smoother. I would never call the original a “grainy” DAC but the MKII is smoother still. I believe this is what contributed to the less fatiguing sound in the amp section. Replacing the UMC oscillator (already a high quality clock) with a higher spec unit from NDK is very likely to blame for this improvement.
As with the original, minor differences between the single ended outputs and balanced outputs can be heard. Nailing down exactly what those differences are is tricky and I don’t think either one is necessarily “better”, but merely “different”. I did very much like the combo of D100 MKII and A18 amp. I know the A18 is part of the Sabre series and an ideal match for the D18 DAC, but the D100 also does a fine job. In fact I think some people may even prefer the MKII to the D18 depending on their preference for sound. The D18 is more spacious and live sounding while the MKII has a bit more of a “monitor-like” presentation, and will show more of the flaws in your bad recordings. I prefer the D18/A18 combo but for someone who already owns the MKII the A18 is a completely viable upgrade. Interestingly, I still prefer the original D100 when pairing with the A100 amp. I know the A100 was voiced with the original D100 in mind, and for some reason that combo just has synergy together. But for the other half dozen amps in my house right now the MKII is the better match.
Probably the biggest question on my mind was about the new USB implementation – like many around here I was under the impression that the Tenor TE7022 was lower on the scale of quality as compared to the TAS1020B from Texas Instruments. And perhaps in some situations it is. I do think it is easier to work with, a fact that contributes to its ubiquity among no-name eBay type DACs and other questionable designs. Yet from experience I have seen it perform transparently in highly resolving DACs like the Violectric V800. And again in the case of the MKII, the USB input sounds no different to me than the other options. My conclusion is that if paired with an excellent ASRC circuit the Tenor chip offers no disadvantage whatsoever. Factor in the increased reliability (you rarely hear of compatibility issues with the TE7022) and it seems like a good way to go. I do feel that around here we have a tendency to ascribe too much importance to each individual chip used in a design. I’m guilty of this myself. We need to keep in mind that we are not “listening” to any particular chip itself. Rather, what we hear is a complex network of components – the USB receiver sends an I2S signal to the DIR which then passes it along to the SRC, the out to the DAC, I/V conversion, low pass filter, then out through the headphone stage or rear outputs. The whole thing works as a system. We can speculate all day long about how much each individual part contributes to the end result, but if the sound is pleasing then there is no point in arguing. In the case of the MKII, I do think the result is as described by Yulong – smoother and less fatiguing, warmer, improved balance and tonality. That applies to all inputs including USB.
The downside is the lack of support for the 88.2kHz sample rate. This never used to bother me but I’ve steadily built up a decent selection of albums in the format so it does become a minor problem now. And of course I would have loved to see an asynchronous implementation added – but remember that we are talking about a refreshed design that does not cost any more than the original. What do you suppose the prices of components have done in the past few years since the original D100 was released: go up, or go down? Yet Yulong has managed to add some improvements while keeping the price identical. That’s impressive enough for me to forget about async USB at the moment. I do know that Yulong is at work on a dedicated USB to SPDIF converter. After trying various chipsets like XMOS, Via VT1731, and C-Media 6631, he went with the Tenor TE8802. So at some point it can be expected that the async USB design would work its way into higher end Yulong DACs. It just seems like a lot to ask from a sub-$500 all in one unit to include it.
The MKII update to the Yulong D100 qualifies as evolutionary rather than revolutionary. It addresses the few complaints I’ve heard about the original model, while maintaining the good aspects that many users have come to love. It even refines and expands on some of those. All while keeping the same price. Impressive, and enough to keep my strongest recommendation for a DAC in the $500 range.
To put things into perspective for those who are unfamiliar with the D100 – this DAC is absolutely competitive with such highly regarded models as the Benchmark DAC 1, Lavry DA10, and Grace Design m902. It’s even superior in several ways, especially with the MKII update. And why shouldn’t it be? Many of the same hardware components are being used, in a well thought out design that is no less complex than the others. The only thing lacking here is the prestige of those well-known brand names. Worth noting is that the D100 sells for less than half the price of those models, which helps take away the sting of owning a “lesser brand”. I’ve owned all of those models and am confident about what I’m hearing. Just in case I doubted myself, there have been many other folks who posted similar impressions – a small sample can be found here http://www.head-fi.org/products/yulong-d100/reviews.
The only thing left to address is about current owners of the original D100. Should they upgrade to the MKII or stick with what they have? That’s kind of like asking if someone should upgrade from an iPad 2 to the recently released “new” iPad. The upgrades are significant but not overwhelming, meaning the answer varies with each situation. If a user has compatibility issues with their USB connection, and a desire to use that input more often, the MKII will take care of that. Or if the D100 is used as an all in one unit, and the user finds the headphone out to be occasionally too thin, the MKII will address that as well. But for people who are pleased with the sound and functionality of their original D100, an upgrade to the MKII is probably not necessary. Yes, it does sound noticeably improved, but the difference is not massive. If an upgrade is desired it would make more sense to climb the Yulong ladder to the flagship Sabre D18 unit instead.