Jump to content

gmgraves

  • Posts

    8605
  • Joined

  • Last visited

  • Country

    country-ZZ

3 Followers

Retained

  • Member Title
    Senior Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Headphone listening has become pretty much the norm for much of the music listening public. People on the go, and, of course, audiophiles, especially, have demanded that high-quality headphones and the portable electronics to support them be made available. In response to this demand, many high efficiency, high quality headphones have come on the market from such companies as HiFiMan, Audeze, Sennheiser, Sony, etc. Several years ago AudioQuest, a company mostly known for their audio cables, stunned the Hi-Fi world by introducing a tiny USB “dongle” the size of a portable thumb-drive memory stick, which contained both a high quality DAC (using an ESS SabreDAC chip) and a decent amplifier module capable of driving many of the premium dynamic headphones on the market with good volume and amazing sound. Dubbed the “DragonFly” by it’s maker this device has spawned a plethora of imitators at price points both below and above that of the DragonFly brand. Of course, AudioQuest has not rested on their laurels either. In the interim between the introduction of the original Dragonfly and today, the company has released two higher performance models to accompany the original black model (which has also been upgraded), a DragonFly Red and a premium DragonFly Cobalt. Earlier this year, the Chinese company, Questyle, introduced the M15, a DragonFly sized device that similarly contains a DAC and a high-quality headphone amplifier. Also powered by USB, the similarity to the DragonFlys pretty much stops there. Rather than having a USB ‘A’ connector sticking out of one end of the device (hence the resemblance to a memory thumb drive) The M15 comes with two USB “pigtail” cables which plug into the M15. One is terminated with the standard USB ‘A’ connector and the other is terminated with a USB ‘C’ connector. The M15 itself is equipped with a USB ‘C’ female connector as its digital interface. Physical Description The Questyle M15 is a small, rectangular metal box measuring 2.44” (62.3 mm) long, X 1.06” (27.2 mm wide, X 0.47” (12.0 mm) thick. One side of the unit has a clear glass cover which shows the unit’s interior with all components clearly visible on the PC board inside (see legend in illustration, above). On one end of the M15, there is a ‘C’ type USB connector, and on the other end there are two miniature headphone jacks. One is a standard 3.5 mm (1/8th”) TRS jack and the other is a 4.4 mm TRRS jack (more about these, later). On one side there is a small slide switch labeled “gain” with two settings: “low” and “high”. There are no other controls or ports on the device. The M15 is very well made with the case machined from a solid block of aluminum using CNC. Overall, the quality is extremely high and the look and feel is that of a premium device. The M15 comes with a one year warranty from the date of purchase. Operation The M15 has been designed for full compatibility with Android devices running v.5.1 of the Android OS or higher, with Windows 10 v.18.3, MacOS, and iOS. To operate the M15, first select either high or low gain based upon upon one’s phone’s efficiency. Then plug the phones into either the standard 3.5mm headphone jack or the 4.4 mm jack (note: it is not recommended to use both output ports simultaneously). Included in the box with the M15 (beneath the cut-out for the converter itself), are two “pig tail” cables of approximately 4.4” (112mm) length. Each has a ‘C’ type USB connector on one end for connection the M15, but one cable has an ‘A’ type USB male connector on the other end and the second cable has a ‘C’ type connector on both ends. If one is using the M15 with a current Apple iOS device (iPhone, iPad, or iPod) one will have to supply one’s own “Lightning” connector to either USB ‘A’ female or USB ‘C’ male adapter. Apple and other vendors sell a so-called “Camera Adapter” which adapts the Apple proprietary Lightning connector to USB ‘A’ female. There are also Lightening connector male to USB ‘C’ male cables available from vendors such as Amazon at various price points. Once the M15 recognizes the attached headphones, the device will begin operation. The gain switch will cause a light on the PC board (viewable through the window on one side of the M15) to glow green when switched to low gain and red when switched to high gain. Low gain is generally preferred for low impedance phones and high gain is usually used for high impedance phones. Experimentation with your particular headphones will tell you which setting is best for your particular needs. The high gain position should also be used if the M15 is to be connected to an external amplifier to drive speakers. When connecting the M15 to a computer as a source, be sure to select “M15” as the USB output device in your computer’s settings or preferences section. If your device has USB ‘C’ as the output, it may be necessary to turn-on the OTG function in the system settings or preferences to insure proper operation of the M15. Performance of the Questyle M15 The M15 is an MQA (Master Quality Authenticated) enabled DAC. Which will complete the final unfold of MQA encoded content. In the clear window, one will find three colored lights which will illuminate to confirm the various DAC modes. If the DAC is playing digital content with a sample-rate of 48 KHz or less, the light will show GREEN. The light will turn RED when decoding high-res lossless audio files of 88.2 through 384 KHz sampling rate or for DSD files of DSD64 through DSD 256. Finally, the light will glow MAGENTA when the DAC detects and performs the final unfold of an MQA file. While the M15 utilizes TOREX power management to maximize battery life of the unit to which it’s connected, actual playing time is a function of the device powering the unit and will vary accordingly. As nifty and a high of a build quality as the M15 is, the final analysis is, of course it’s audio quality. To assess that, this reviewer uses two pairs of headphones. My main listening was done using a pair of HiFiMan Edition X phones as well as a pair of Stax electrostatic phones which took the output of the M15 and routed it through a HiFiMan Jade II amplifier to power ES phones. Don’t ever let anybody tell you that all Digital-to-Analog Converters (DACs) sound alike. I keep running into this particular bit of mythology, mostly from highly opinionated audiophiles on one audio forum or another. Over the past few years, this reviewer has had, pass through his system, such disparate DACs as the Schiit Yggdrasil, the Denafrips Pontus II DAC, the Denafrips Terminator II (with the Gaia D-to-D converter/system clock) as well as the Schiit Modi Multi-Bit and the Topping D90 as well as all three AudioQuest DragonFly models. They all sound as different from one another as any group of amplifiers or even speakers (but to a lesser degree. Speakers, being transducers vary in tonal quality much more than any electronic components). While all of these DACs do a credible job of converting a digital audio stream to an analog audio signal, they vary greatly in what portion of the audio spectrum they emphasize and what portion they tend to gloss over. While it is true that, generally speaking, the more the DAC costs, the fewer compromises one will find with the sound, even some relatively modest DACs do an admirable job of gleaning most of what’s encoded in those ones and zeros that represent our favorite musical performances. The Questyle M15 is actually two devices in one. The first device is a DAC and the second is a headphone amplifier. The DAC section of the M15 uses the ESS ES9281AC PRO “SabreDAC” chipset. This decoder was first introduced last year (2021) and is used in a number of other products. There are two versions of the chipset, the AQ and the AC suffix versions. The difference seems to be that the AC version contains circuitry to decode the final fold of MQA encoded files while the AQ version lacks this feature. Many consider the ESS Delta-Sigma DAC chips to be the best IC-based DACs, but this reviewer believes that the AKM AK4499EX/AK4191chipset, also a Delta-Sigma design, offers superior audio performance. We are waiting, as I write this for a review sample of a DAC from a Florida-based company called Geshelli Labs that uses this new AKM chipset. As you might remember, AKM had a devastating fire in 2020 which caused them to cease production of their highly regarded DAC chips. I’m sure that we are all delighted that they are back with a new, improved DAC chip set and wish them well. I am anxious to hear the new Geshelli J2 DAC to see if the new AKM DAC chip set is as good, and hopefully an improvement over the DAC chips interrupted by the fire. The amp section of the M15 uses current mode amplification (as opposed to voltage mode). Questyle has been using current mode amplification in both its desktop and portable offerings almost since the beginning. Current mode amplification has the advantage of its performance being largely independent of load impedance. This allows for much more efficient drive for a wider range of headphone types from planar magnet designs such as HiFiMan and Audeze to apex driven transducer such as those used by Sennheiser and Focal as well as some that are notoriously hard to drive such as the Abyss AB1266 and some IEMs (In Ear Monitors), but not the HiFiMan Susvaras! This can result in much lower distortion over the entire audio spectrum. The 4.4 mm output jack on the M15 is for balanced output. Balanced drive gives approximately twice the drive power of the standard single-ended output. This helps for many hard to drive ‘phones and I recommend using that output if possible. Unfortunately balanced cables for many popular headphones sporting removable cables are thin on the ground. The reason for this is the 4.4mm TRRS plug required. I was able to find only one that would fit my HiFiMan Edition X phones but several that fit more recent models. Sound Quality of the Questyle M15 Of all the “dongle-style” portable headphone electronics that I have auditioned over the years the M15 is by far the best. I have owned all of the AudioQuest DragonFlys as well as models from IFI such as the Hip-DAC and the XDSD as well as the FIIO Jade Audio KA3s. While all of these competitors have their advantages, when compared to the Questyle M15, the ones that I have had in-house all exhibit a two dimensional quality that, when coupled with the usual shortcomings of headphone listening, present a rather artificial soundstage, that while delivering, in most cases, decent audio performance, do not really satisfy with content that relies on good imaging to totally immerse the listener in a realistic experience. While mostly, this characteristic emerges with true stereo recordings as well as with binaural recordings, it is often the difference between a vocalist or soloist appearing inside the listener’s head rather than in front of the listener as in a real-life concert environment. The M15 excels as this type of presentation. Images are well defined, and produce a panoramic sound stage reminiscent of listening on a pair of really good speakers. Now, I don’t mean to infer, here that listening to headphones could ever equal the soundstage and image specificity of a good pair of speakers, no headphone setup could do that, but the M15 presentation on good phones can be startling if one has never experienced the phenomenon before. Also, on top of this totally immersive spatial experience the M15 is also very clean sounding. Distortion is much lower than say, the AudioQuest DragonFly Cobalt (frankly, I’ve never spoken to anyone who can tell the difference in performance between the DragonFly Red and the Cobalt. I certainly can’t and believe me I have tried to hear the difference). The difference in the level of distortion between the M15 and any of the DragonFlys, or the IFI XDSD (which is an amp only device) is far from subtle. One can hear the improvement in overall cleanliness even on older recordings which in and of themselves aren’t all that low in distortion. The Heifetz recording of Rozsa’s Violin Concerto, for instance, was recorded by RCA Victor in 1956 with the Dallas Symphony. This Red Seal recording, while an excellent performance, lacks the clarity and cleanliness of the later Telarc with Robert McDuffie on the violin and Yoel Levi conducting the Atlanta Symphony. The Telarc is definitely cleaner but, of course, while McDuffie is a competent soloist, there’s no comparison between him and Heifetz. Anyway, The difference in distortion on the Heifetz recording between the Questyle M15 and any other portable DAC/Amplifier is immediately noticeable and on a cleaner recording such as the Telarc, it’s profound. While the M15 is capable of powerful bass (headphone dependent, of course) and the HiFiMan Edition X phones excel in bottom end performance, in my opinion, where the sound of the M15 stands-out is in the midrange. It is warm and liquid and vocals, both male and female are uncanny in this regard. Whether it’s Pavoratti belting “Nessun Dorma” or Kiri Te Kanawa singing “O Mia Babbino Caro” the vocals are detailed and natural in a way that seems to strip away any of the artificiality imposed by the recording process. Switching to lighter fare, 1966’s Reprise live recording of ‘Sinatra at the Sands’, Frankie’s “It was very good year” has never sounded so poignant, so emotionally satisfying. On the Album ‘The Best of Brazil’ Gal Costa’s vocal rendition of “Aquarela do Brasil” (my favorite song, BTW. It means “Water Colors of Brazil”) is simply sensational sounding. Her voice has a sweetness that comes through the M15 unlike any other headphone experience I’ve heard with a portable player. Conclusion The only fault that I can find with the Questyle M15 is the absence of a volume control, either physically or in software. The DragonFlys all have a software volume control that allows for very fine control of the unit’s volume. The M15, on the other hand relies on the source component’s (phone, pad or computer) volume control to set the SPL of the playback. I don’t know of any device which doesn’t digitally control its sound output, but I’m sure some exist. This is, in my humble opinion, a serious omission on Questyle’s part. Perhaps they couldn’t find a way to include a volume control without compromising audio performance, in which case, I kind of understand, but still… I heartily recommend the Questyle M15 for all of one’s portable headphone listening. When introduced early in 2022, the M15 sold for US$249. Since then the price has increased to US$269, but in this writer’s opinion, this is still a reasonable price for such a premium product. Whether your private listening preferences run to IEMs or high-end, over the ear headphones, the Questyle M15 will drive them with ease (Susvara’s excepted, of course!). Product Information: Questyle M15 $269
  2. I guess I can understand that. Luckily, I don’t suffer from tinnitus. regards George
  3. That is interesting. I had the use of a Chord Hugo 2 for a while and thought it to be wonderful, but I had no need for the headphone interface or portability of the battery power that it afforded. When the Qutest came out claiming to be the same digital circuitry as the Hugo 2, for about a thousand dollars less, I jumped on it. But, no matter what I did, I could not get the Qutest to sound anywhere near as good as the Hugo 2. So I sent it back after reviewing it! The original Topping D90 (not the current one) sounded (to my ears) so much better than the Pontus II, that I sold the latter. When I received the T+/Gaia, I was literally flabbergasted and ended up buying it. I will guarantee that the Topping D90 is head and shoulders above the Qutest; at least it is in MY system. So you are right, SQ is largely subjective. But, I have come to the inescapable conclusion that synergy between the various components in a system makes the difference between a particular component being gold to one listener and dross to another. Also, I believe that one’s taste in music, as well as one’s expectations has a lot to do with one’s likes and dislikes, sound wise. For instance, I hear a lot of live acoustic music (classical and jazz almost exclusively) and rock/electric guitars, essentially, never. I want my reproduction to sound like live, acoustic music, playing in a real space. Somebody who listens to studio-made rock music will likely prefer a totally different type of sound. My experience is that most rockers I’ve known go for a sound that enhances the excitement of a rock performance; ie, bright, aggressive highs and big bass. Play classical music through such a system, and frankly, to my ears, it sounds awful. Are the rocker’s wrong? Absolutely not. They are seeking one type of musical experience, while I’m seeking another. We are both right in our respective pursuits, but that pursuit will yield wildly different opinions about what constitutes “high fidelity”. regards, George
  4. I’m not so sure that I would go so far as to say that I was “advocating” the Raspberry Pi solution, but I did try it with an SPDIF coax/Toslink output daughter board. It worked fine for one listening session, and then It died. I have never been able to get it to work again since. That is when I acquired the Auralic Mini. Believe me, I would have preferred to not have to make THAT expenditure! The Raspberry solution looked like an adequate server (but it REALLY needs the daughter board output module, as the Raspberry’s USB output is less than audiophile quality). It certainly was inexpensive enough, and the iPad/iPhone application allowed me to stream from both Tidal and my ripped library on my Desktop computer. Too bad it proved to be so “fragile”. To answer your question, the server doesn’t seem to matter much. I noticed no SQ differences between the Raspberry Pi server, the Auralic Mni, and a US$6000 Lumin U1. All of the differences seemed to be down to features, software quality, and “bling”. There is quite a difference between Pontus and the T+ as far as sound is concerned. The Pontus II is good the T+/Gaia sound is more akin to A dCS “Vivaldi” than it is to a Pontus. regards George
  5. I agree. I once had a DAC from Sonic Frontiers in Canada which could be used to decode HDCD OR 24-bit, 96 KHz, but not both at once. Before I permanently altered this DAC to play 24/96, I listened to all of my Reference Recordings in HDCD. I thought they sounded great, but I also realized that HDCD would never become an industry standard. Based on that I chose to go with 24/96. After getting a better, more modern DAC, I decided to go back to the HDCD configuration, but alas, in the ensuing years, I lost the HDCD module that would allow me to swap out the 24/96 module for the HDCD one. Sigh!
  6. I think you missed my point. When I had the Hermes, I was using it with a Pontus II. As that DAC does not allow for clock synchronization, I wasn’t able to utilize the Hermes to full advantage. Therefore, my evaluation of that component is not valid when compared with the Gaia and the T+. Any conclusion I make about it’s performance should be taken with a grain of salt. That’s all I’m saying.
  7. I think I did that in the review, but I don’t mind re-iterating and expanding on that list… 1- T+/Gaia 2- T+stand alone 3- Topping D-90 (not the SE version) 4- Pontus II (with or without the Hermès) and waaaaay down the list 5- Schiit Yggdrasil (latest version) 6- Schiit Modi Multi-bit. I don’t notice much difference in music streamers, sound-wise. So, my comments are for the DACs alone.
  8. I have nothing against the Hermes, I’m sure it’s a good “budget” D-to-D. When I had one in house, I used it with the Pontus II, which doesn’t allow for clock synchronization, so I wasn’t able to take full advantage of it. Under those circumstances, the improvement to the sound of the Pontus was subtle at best, and non-existent at worst.
  9. Loyal readers (Loyal Readers? Sounds like I think I’m a journalist or something!) may remember that last year, I got my hands on a Denafrips Pontus II DAC. At the time, it was, by far, the most detailed, cleanest and most musical DAC I had ever heard. And as good as it was, I decided to ask Denafrips to send me their flagship model, the $4000+ Terminator Plus (T+) and their premier D-to-D converter the Gaia. Well, I expected the T+ to sound somewhat better than the Pontus II (after all, it costs more than twice as much for the DAC alone), but I must say that I was not prepared for the level of difference. It’s one of those cases of “How much better could it be?” The answer, comes back, of course, that in spite of the fact that while I could not imagine “better”, it turned out to be a LOT better! It seems that technologically, there is much more to the depth and breadth of information encoded in a digital signal to be plumbed than many would expect and I suspect that we have barely scratched the surface of what’s available even in an MP3 digital stream, much less in high definition recordings! Along with the T+, I requested the company’s flagship D-to-D converter, the Gaia, as well,. Together these two units tip the scale at more than US$8000! The reason that I don’t give a more precise price is because Denafrips lists their prices in Singapore dollars, and the conversion to US Dollars or Euros or Pounds Sterling fluctuates with the exchange rates. Now, I know that this is a lot of money, especially since a perfectly usable DAC, such as the Schiit Modi Multibit, (to which I’m listening on my computer desktop system as I write this) can be bought for just a hair under US$250. But keep in mind, that there are DACs from companies like MSB and dCS that cost (considerably) more than US$20,000. So, by comparison, a US$8000 solution is really not all that extravagant – relatively speaking. What matters is not so much the actual price as it is what you are getting for your money. And with the Denafrips Terminator+/Gaia combo, you are buying performance that I would not hesitate to compare to many of these stratospheric solutions from the likes of MSB, dCS and others. For this combo is certainly a world-class digital to analog solution! Physical Descriptions Terminator Plus (T+) The Terminator+ is a large, full-component sized unit at: 430 X 380 X 105 mm (17” X 15” X 4”) and it weighs 19 Kg (41 lbs, 14 oz). The case is available either in silver finish or black, and all front panel controls are accessible via push buttons with red LEDs for function indicators. There is no remote control. Front panel From Left to right the controls are: Input –, input +, and mute. In the center, is a larger “standby” button. Next is the OS/NOS button which switches between the oversampling and the non-oversampling mode, followed by the phase invert selector and finally, the mode switch. Above the control is a line of red LEDs which, again from left to right, indicate which digital input is selected: Coaxial, AES/EBU input 1, AES/EBU input 2, Optical, I2S-1 (HDMI), I2S-2 (RJ45), I2S-3 (RJ45) and USB. Finally on the front panel are LEDs indicating whether oversampling was on or off, and whether phase invert was enabled or not and then the sample rate being processed. Two main sample rates are given for Linear PCM (LPCM) at 44.1 KHz and 48 KHz. Following those two numbers are the multipliers: X1, X2, X4 and X8. Following that is DSD which lights up when the T+ recognizes a Direct Stream Digital signal. The sampling rate LEDs show the sampling rate being decoded by lighting up either the 44.1 or the 48 KHz LED along with a multiplier. For instance, a standard 44.1 KHz data stream, as from a CD player will cause the 44.1 LED to light as well as the X1 LED. An 88.2 KHz sampling rate signal will light the 44.1 KHz LED and the X2 LED, etc. Rear Panel The rear panel of the T+ is, as is customary, for I/O. From left to right, are the audio output jacks. The first are the XLR balanced outputs (pin 2 = hot) and then the RCA unbalanced outputs. In the center, at the bottom of the apron, is the IEC mains connector which will take 115VAC, 60 Hz to 230VAC 50 Hz (no switching required. The unit senses the mains and adjusts itself automatically). At the top of the apron are the digital inputs. These consist of two BNC connectors which are the interface to a D-to-D converter if present (in our case, the Gaia) followed by an RCA jack for coaxial SPDIF input. Next is the Toslink connector for optical SPDIF, and then there are two balanced XLRs for AES/EBU 1 and 2. Next in line are the I2S inputs. I2S-A is an HDMI connector followed by two RJ45 connectors representing I2S-B and I2S-C inputs. Finally there is the standard type-B connector for the USB input. Operation The T+, like other Denafrips DACs, is pretty straightforward in operation. Connect your sources and select the one to which you wish to listen with the buttons on the front of the unit (regretfully, there is no remote control. This is something Denafrips needs to address in future models) and then choose whether you want oversampling turned on or not (OS/NOS). That’s about it. While there is, at present, no industry standard for I2S pin-out, it is, according to many, the best digital interface on the market today. To facilitate this, Denafrips has designed the T+ (and others in their line) to adapt to most connection protocols in use today. While it is beyond the scope of this review, there is YouTube video available on the Denafrips website that shows how to use the buttons on the front panel to find the correct set-up for your I2S device. For instance, I use an HDMI connected, third party device to allow me to play SACD discs through the T+ using any Blu-Ray player that supports SACD playback. Once the correct interface is set on the Denafrips, all one need do is select I2S via HDMI as one’s input and insert an SACD into the drawer of a suitable player. Of course, there is no guarantee of the compatibility of any third party I2S device, but in my case, the SACD “adapter” was wired as the default configuration for the T+ I2S interface, so no configuration was necessary for me. The T+ has two OXCO (Oven Controlled Crystal Oscillators) clocks operating at 45.1584 and 49.152 MHz. These two crystal clocks are encased in temperature controlled metal housings on the mother board and were designed to exhibit ultra-low phase noise and very high frequency stability (thus the temperature control) especially for digital audio applications. The oscillators are powered by a constant current power supply to ensure linearity and stability of the OXCO’s frequency and amplitude. Jitter, a major concern for any Digital to Analog Converter, is handled in the T+ by re-clocking the incoming digital signal using a digital buffer device called a FIFO (First In, First Out). This device takes a digital “word” of any length supported, and momentarily stores it in the memory of the device. This means that the spacing of each incoming bit can be quite chaotic, and this would normally be disadvantageous to the performance of the analog conversion, affecting the resultant analog sound quality negatively. The FIFO circuitry won’t let the word out of the buffer until it is clocked out by the ultra accurate OCXO oscillator. Since the distance between each clock cycle is exactly the same from cycle to cycle with the OCXO clock, and each single bit is output for each clock cycle, the asymmetrical nature of the incoming signal is changed to a super symmetrical data stream in which jitter is reduced to just a few parts per million. In other words, to the extent of the accuracy of the clock; jitter is, for all intent and purposes, almost completely eliminated to the point that as a concern for the listener, it can be, essentially, ignored. The T+ is equipped with a proprietary, state-of-the-art USB interface which supports up to a 24-bit/1536 KHz LPCM data stream as well as native decoding of DSD signals up to DSD1024. Available for Windows users is the Thesycon USB Driver (Mac and Linux platforms don’t need no stinkin’ drivers!). The USB circuitry is designed to only trigger when the USB input is selected. This feature is implemented to keep cross-interference between inputs at a minimum and to reduce possible digital noise pollution caused by the USB input which could degrade the audio signal-to-noise ratio. In order to improve the performance of the SPDIF inputs (Coaxial, Optical (Toslink), and AES/EBU inputs) which operate up to 24-bit/192 KHz uses a proprietary on-board FPGA (Field Programmable Gate Array) to decode these signals rather than the customary Digital Audio Receiver chip. This shortens the SPDIF signal path, ostensibly allowing for less noise incursion and reducing coloration. In addition, the T+ is equipped with a 26-bit R2R DAC to decode LPCM and 32-step FIR analogue filters in the hardware decoder to handle a DSD data stream. According to Denafrips, very few designs currently marketed where a R2R DAC can decode, using hardware, both LPCM and DSD data streams. It looks as if the decoding of DSD signals by the T+ is accomplished by converting the Direct Stream Digital signal to Linear PCM. I’ve read that while this approach works fairly well, the noise floor of the converted data stream is somewhat higher than it is for LPCM sources. One thing that’s kind of unusual in the T+ is the output impedance of the balanced XLR outputs. It’s 1250Ω! That means that you want to avoid long runs of more than a meter or two to your amplifier. If you need long runs, it’s best to employ the unbalanced RCA outputs which are 625Ω, or employ a solid-state buffer stage between the T+ and the amp inputs. While 1250Ω seems high to me, balanced audio impedances can be anything as long as both the positive and negative terminals in a balanced circuit are the same. The Gaia D-to-D Converter This review evaluates both the Denafrips Terminator + DAC and the Denafrips Gaia Digital to Digital converter together as a system. While the T+ is, of itself, a breakthrough in DAC performance at its price point, this reviewer feels that the full potential of this DAC cannot be reached without pairing it to the Gaia D-to-D. What Is A D-to-D Converter, Anyway A D-to-D converter (DDC) is a piece of gear which takes a raw digital stream (in this case, audio) and removes noise from the transmission media, as well as from the AC line noise, then re-clocks the signal, reducing jitter, and generally hands-off the stream to the DAC as a much cleaner, purer digital signal. Design Highlights The Denafrips Gaia DDC accomplishes three main tasks: First, it isolates the incoming data from it’s downstream processes, using 50-Mbps photo-couplers or opto-isolators as they are known. This optical (as opposed to electronic) isolation of the incoming digital signals can lower the noise floor of the digital stream considerably, resulting in fewer errors and requiring less error correction during conversion. This can increase resolution by reducing interpolation (guessing) in the D to A phase and by improving the signal to noise ratio of the resultant audio output. Secondly the Gaia buffers the signal and thirdly, it re-clocks the signal using a local OCXO. This may seem redundant as the T+ also has an on-board OCXO and it also re-clocks and buffers the incoming data stream, but the Gaia actually works together with the T+’s OCXO for synchronization. This means that not only do the two devices use the same clock frequencies, but the two clock’s square-waveforms start and stop at exactly the same time. This synchronization assures extra digital precision further “cleansing” the incoming digital stream with what I can only describe as remarkable results. Gaia Front Panel The Gaia has no power switch as it is “on” all the time. From, left to right, the front panel presents the user with three three rows of lights. Top to bottom, they read 44.1K, 64, and 48K from left to right, the first row is marked 44.1K, 88.2k, 176.4K, 352.8K, 705.6 K, which obviously refer to multiples of the 44.1 KHz LPCM (Linear Pulse Code Modulation) sampling rate. Then there are lights signifying the following digital inputs: DSD, OPT, COAX. The second row is marked 64, 128, 256 and 512 for the Direct Stream Digital (DSD) formats supported. The bottom row is marked 48K, 96K, 192K, 384k, and 768K. These are, obviously, multiples of the base 48 KHz sampling rate. Following there are lights which indicate clock synchronization with the DAC and the remaining inputs AES/EBU and USB. Finally, on the extreme left are a star or cross arrangement of five push buttons which are used to select the four inputs while the bottom-most button is the Setup button. Gaia Rear Panel From left to right are the Gaia digital outputs: AES/EBU1, AES/EBU2 (both XLR), I2S-A (HDMI), I2S-C (Ethernet RJ-45) Optical (Toslink) and Coax (RCA). Next are the Digital inputs: 49.152 MHz, 45.1584 MHz (BNC female – both from the same clock speed inputs on the DAC), Optical (Toslink) AES/EBU (XLR), and USB (USB “B” female connector) and finally, the EIA mains connector (115 V/60Hz or 220 V/50 Hz). Setting Up the Gaia To setup the Gaia for interface with a suitable DAC (in this case the T+) the user presses the Setup button once to enter the configuration mode. Then one presses the OPT button twice. After that, pressing the OPT button will toggle the clock light on or off to indicate whether the internal clock is being paired with the clock on the T+. I.E. Clock On = Enable Clock In, Clock Off = Disable Clock In. To confirm the setting, press the Setup button again to save the selected mode and exit the Configuration mode. Connecting the Terminator Plus and the Gaia Clocks Together To connect the clock OCXO’s of both machines together it is necessary to procure two 75Ω BNC male to BNC male cables of sufficient length to reach between the Gaia and the T+ (in this case). The two units should connect the same clock speed to the same clock speed; I.E, the T+’s clock inputs to the Gaia’s Clock inputs (see below). Example showing the T+ and the Gaia’s clocks tied together A Glimpse of Audio Nirvana After living with the Pontus II for about a year, and then living with the original Topping D-90 DAC (with the original Asahi Kasei AK449 DAC chip, not the later D-90SE with the ESS SabreDAC) for several months, I decided to contact Denafrips about procuring their top-of-the-line DAC/D-to-D solution, the Terminator Plus/Gaia. As good as the Pontus II was, I found that the less than US$800 Topping D-90 bested it in a number of ways. First was, of course, price. The D-90 provided significantly superior SQ for less than half the price of the Pontus II. Secondly, the Topping supported full hardware decoding for MQA and Denafrips doesn’t support the protocol at all, and thirdly, the Topping has a full remote control, something no Denafrips solution offers, and which I consider almost de riguer. The Topping blew me away. I couldn’t see how anything could be better than the Pontus II, but after hearing the D-90 in my system, I started to wonder if perhaps the T+/Gaia DAC combo might raise the bar on D-to-A conversion significantly further. I contacted Alvin at Denafrips in Singapore, and soon a T+ and a Gaia were on their way to me. I expected the T+ setup to be better than the Topping and I already suspected that the hierarchy would be, in ascending order of SQ, the Pontus II, the Topping D-90, and the T+ setup. What I did not suspect was that the T+ setup would be the difference between viewing the road in front of one’s car through a muddy windshield, and viewing the road through one just cleaned with Windex inside and out! The only thing better, would be no windshield at all. We aren’t there yet (AFAIK). To say that I was astonished, would be an understatement! After listening to the T+ alone for several days, I could hardly believe how much better it sounded than anything else to which I had ever listened! The Gaia can’t improve on this, I thought. So I hooked them together with a pair of 1 meter 75Ω BNC cables, and routed the digital output of the Gaia through the HDM I2S cable to the HDMI input of the T+. After connecting all my digital inputs to the Gaia: MacBook Pro to the USB port, Auralic Mini (for Tidal and locally stored files) to the coaxial SPDIF input, and my Logitech Squeezebox Touch (for Internet radio ONLY) to the Toslink input, I let the whole rig “cook” for about a week before sitting down to do any serious listening. Oh My Gosh! The T+ by itself had been a revelation, and I really wasn’t expecting much from the Gaia, I must admit. I had tried the Gaia’s little brother, the Hermes with the Pontus II and had noticed very little if any improvement, so my expectations weren’t too high. Of course, the difference was that the Pontus II does not provide the clock interface between the DAC and the D-to-D like the T+/Gaia does; although the Hermes is capable of this with the right DAC (the only other DAC that Denafrips makes that can use the Hermes’ or the Gaia’s clock synchronization feature is the Terminator (without the plus). The Venus II, Pontus II and Aries II lines of DACs don’t support clock synchronization). Well my first listen to the combo made me spill my adult beverage (and believe me, I’d as soon see a church fall down than spill my drink)! The T+ by itself was an eye-opener, but with the Gaia connected between my digital sources, and the T+, the sound quality is elevated to another plateau! I wouldn’t have expected that much of an improvement in SQ from a D-to-D converter. For many years I have had a Sonic Frontiers D2D converter box which re-clocked the incoming signal and up-converted 16/44.1 and 16/48 to 24/96, but the difference with, and without, the D2D box was subtle, at best. This improvement over the T+ alone is anything but subtle! All at once the soundstage opened up. It was already wider and deeper than any DAC I had ever tried, but it got even wider and deeper! To make sure that I wasn’t hallucinating (audiophiles are prone to confirmation bias. I.E., “I just got a new widget. It was expensive, and it’s highly touted. It must be better than the widget it’s replacing”, and so it is – in our imagination, at least!), I played one of my own symphony CDs. These were made from 15 ips (38 cm/sec), half-track analog master tapes that I recorded of the San Jose (CA) Symphony Orchestra under the world-class baton of the late Maestro Georg Cleve. I chose their performance of Ravel’s Daphnis et Chloe, the complete ballet. Here the orchestra is joined by the large Schola Cantorum choir of San Jose State University. The chorus was arrayed behind the orchestra on stage and stretched clear across the stage from one side to the other and in two rows. So from front to back, the ensemble consisted of viols and cellos on the front row, woodwinds on the second, brass behind the woodwinds on risers, and then the two rows of chorus on risers behind the brass. The recording was true stereo, as are all my recordings, made with a pair of Sony C-37P large-capsule condenser mikes in the cardioid pattern and mounted 7-inches and 90 degrees apart on a stereo T-bar. They were hung about 15 feet over, and slightly behind the conductor’s head. If anything could show the difference in soundstage between the Terminator Plus alone and the combination of the T+ and the Gaia, this recording, which images magnificently anyway, would surely be it. Immediately, I was struck by how the entire ensemble opened-up in all three dimensions. The orchestra spread out to extend way beyond the edges of the speakers, the ensemble also telescoped out, front to back, to take on the true depth of the stage in the venue where the performance was recorded! And suddenly, one could easily tell that the brass and chorus were on risers behind the woodwinds. The triangle shimmered over the left-side of the orchestra just as it does if you were sitting in the fifth row center in the audience. I’ve never heard this rendered so convincingly before from the CD, but I had heard it on playback from the master tape (which belonged to the orchestra, not to me). I had figured that digital simply couldn’t render that kind of detail. Indeed, I had only “glimpsed” this kind of image height from past DAC setups including the Topping D90 as well as the Denafrips Pontus II! There is a new CD (December, 2020) of some never before released alternate takes of the famous Dave Brubeck album “Time Out”. This Album, on the Brubeck Editions label is entitled “Time Outtakes”. And also on the album are some never before released cuts that weren’t included in the original LP. Now I’ve had (and still have) several copies of the original release from 1959. I once had it on LP, I have it on CD as well as on SACD (from Sony) and on Tidal as a streaming album. When I played the outtakes I had a real hard time believing that the material on these two albums came from the same session in the same venue, recorded using the same equipment. The new release sounds so different; so modern, so palpably alive, that it really points out what modern recording practices and equipment can do, even with master tape material laid-down over 60 years ago! Having seen pictures of the recording sessions for this famous album, I can tell you that they used Ampex 350 tape decks running Scotch 206 mastering tape. The electronics were all tube which were very noisy by comparison to today’s electronics, and Scotch 206 was likewise very noisy, even at 15 ips. However, the cuts on “Time Outtakes” sound like they were recorded yesterday using the latest in solid-state digital equipment! I really would recommend that you acquire this album. It is truly amazing sounding. If you have either Tidal or Qobuz, you can easily stream it, and if not, I encourage you to buy the CD or the vinyl release or listen on Amazon music Tidal or Qobuz. You can also hear this remarkable album by going to the “Brubeck Edition” Website (www.brubeckeditions.com) and follow the “Time Outtakes” thread to the page where you can listen to the album. The opening few bars of “Take Five” on this alternate take features drummer Joe Morello playing the cymbal with his hard drumstick. With the T+/Gaia, one can hear the percussive tap of the stick on the cymbal and then the cymbal note as separate and distinct sounds! On most DACs this sound is smeared to a greater or lesser extent, and frankly, I’m surprised that the primitive recording equipment of that era could even capture such microdynamics. Paul Desmond’s alto sax has a creamy, slightly “wet” texture that allows one to hear the breathy quality of a saxophone. Again, the separation of the player’s breath and the notes emanating from the sax’s bell is a real eye opener. I’m listening to this cut on my desktop system as I write this, and as good as the Schiit Modi Multi-Bit DAC is (for the price), none of these characteristics are even remotely revealed. For Bass, I turned to the Saint-Saëns Symphony #3 in C minor, the “Organ Symphony” with the Kansas City Symphony Orchestra, Conducted by Michael Stern on Reference Recordings (RR-136). The beginning of the fourth and final movement (Maestoso –Allegro) opens with this stupendous full C-major chord on the organ. If you wish to test your woofer/subwoofer, shake your house to the veritable foundations, this is the cut for you! Likewise, it is also the cut to test the bass authority of your DAC. With the T+ alone it is thunderous, but add the Gaia and it is not only a deep, tight organ note but it is also an extremely articulate one. The level of detail; the chuff of the pipes as air is pumped into them is distinct from the actual note that follows. When the note ends, it trails away in a very realistic and “analog” fashion. Something seldom heard on a digital recording. Overall the T+ is a world class DAC. Alone, it produces a liveness and a palpable realism that just a few short years ago would have seemed impossible at any price. I don’t really know how to describe the refinement of that amazing performance made possible by the addition of the Gaia D-to-D. But all I can say is that whatever magic the T+ is bringing to the table with it’s incredible performance, the Gaia takes that magic and turns it into sorcery! The difference between the T+ without and then with the Gaia simply cannot be adequately described. I don’t know what the Gaia is actually doing at a musical level, but whatever it is, it’s beyond anything I, certainly, have heard before. I know that US$8,000 is a lot of money, but nothing else in my experience comes even close, and I’m talking about DACs that cost many times what the Denafrips Terminator Plus and Gaia combo will set one back. If there is better (or even as good) I’ve yet to hear it. OTOH, like I said at the beginning of this review, I believe that there is much more to be gleaned from our digital sources than current technology is able to extract from them. I’m sure that while the T+/Gaia stands at, or at least near, the pinnacle of what’s possible today, I’m sure that next year or the year after, they will be bettered, and perhaps at a more reasonable price. After all, the Topping D90 (again, not the Topping D90 SE) comes “close” to matching this pair, and does it for one-tenth the price. But at the moment, if you want a glimpse of what audio Nirvana is really like, The T+/Gaia is the price of admission. Product Information: Denafrips Terminator Plus Digital to Analog Converter $6,451 Denafrips Terminator Plus User Manual Denafrips Gaia Digital to Digital Converter $1,705 Denafrips Gaia User Manual
  10. Thanks for the clarification. There is damn little information available from the usual sources about these products, and the information I gave about the sources for components and the differences between the EX-M1 and EX-M1+ were from sources I assumed were accurate. Well, you know what they say about assumptions! Thanks again.
  11. First of all, the EX-M1 and the EX-M1+ are not the same amplifier. The + is an altogether better built unit and the speaker terminals are very high quality and made by Furutech as are the gold RCA connectors. I also understand that the circuit of the + is likewise upgraded with better parts and a somewhat modified topology. Frankly, I don’t see how they can sell the EX-M1+ for the price that they sell it for. I’ve seen Integrated amps costing several times the Kinki Studio’s $2900 selling price that aren’t built half as well. Of course, as usual, YMMV!
  12. I once had a Hafler amp. The DH 200, perhaps? It's been a long time. I don't remember it being particularly "soft" but on the other hand, lots of amps use MOSFETs as their output devices and I don't remember any them being characterized as "soft" so I'm not sure that it would be correct to associate the technology with "softness". But I could be wrong about that.
  13. I too would be interested in their new top-of-the-line preamp and the monoblocks. Let me see what I can do about that 😉!
×
×
  • Create New...