Posted on 1 Comment

NOS vs OS (no oversampling vs oversampling)

Sample rate, non-oversampling and oversampling


We know that in the real world, music is continuous, the image is continuous. TV or film, in fact, is composed of a static picture of the screen, the static picture of the rate of rapid switching, the human eye to see a continuous picture. So, how fast the continuous picture? Is a second 50, as long as this rate, the human eye will not be aware of the intermittent switching between the screen, of course, a second 50 is a minimum requirement, if you can do more faster, it can be Improve the visual effects. For audio, it is the same concept, sound information, as long as the intermittent information is enough, they can be interpreted as a continuous ear music. So, the human ear how much intermittent information can be interpreted as continuous information? Let’s first talk about the human ear can hear the frequency range, 20-20000 Hz. So, if you want to reflect the highest frequency of 20K, then how much intermittent information is needed? Here, I need to introduce a proper noun “Nyquist frequency”, the specific principle we do not need to go into here, we only need to know the results on it. As a result, at least twice the “intermittent information” of 20K is required to express a frequency of 20K. Here, we will be a “intermittent information” as a sampling point. In other words, If we have to be able to reflect the 20K audio, you need at least 40K a second sampling point. So, we come back to see the standard CD, CD standard is 16bit / 44.1Ksps. In this article, we will not explain what is 16bit, which in the follow-up article I will explain. The 44.1Ksps is the sampling rate, we can see, from the theoretical minimum of 40K, only about 10% more margin, a little reluctantly.

In the CD has just appeared, there is no oversampling technology, but as people found that the sampling rate of 44.1K is not enough, how to solve the problem? So over-sampling technology was born. So what is oversampling? Simply put, because 44.1K sampling rate is too reluctantly, if there is a higher sampling rate is bound to have a better effect. However, we recorded in the CD information only 44.1K a second, how to become 88.2K, 176.4K or even higher? This needs to do the interpolation operation, through the mathematics method, calculates the missing sampling point, then insert to original sampling points.

In the oversampling operation, an integer multiple of oversampling is the easiest to do, not prone to problems. Therefore, we should ensure that the oversampling is an integer multiple, if not an integer multiple, such as 44.1K oversampling to 48K, which requires higher computing resources, very prone to problems, it should try to avoid this situation appear.

Today, digital audio in the oversampling of the road has farther and farther, 4 multiplier, 8 multiplier is not surprising, 16 multiplier, 32 multiplier or even higher, this is the mainstream trend. If it is 32 multiplier, then only 1 of 32 sampling points, is from the original record information, and the remaining 31 sampling points are derived from the interpolation calculation. So, can we really trust the 31 sampling points from interpolation? This is what we need to discuss the following topic.

Have to say, the mainstream audio technology, and even designers, have confidence in the 31 calculated points of the sample. But the real fact is that audio-specific applications are not suitable for use with oversampling technique. Now more and more people are aware of this problem and found that they had previously believed in things, in fact, there is a problem. Yes, we are now in the process of negating the current mainstream technology. It is like China has finally realized the importance of environmental protection, began to cut a lot of high-polluting industries, although those highly polluting industries have brought us wealth, but we now pay more attention to the environment, about wealth, we can use other methods.

The advantages of oversampling

Before talking about the problems of oversampling, let’s talk about the benefits of oversampling. Audio technology industry are not all fools, If there no good things, no one will be interested, how it will be developed till today?

First of all, If we analyze the spectrum rate of a signal with sample rate of ‘F’,  Then will have a center in F / 2, the left and right part of spectrum is mirrored at this center point, assuming a 1K audio signal at 48K sampling rate, then, in the spectrum, you can see that there is a signal at 1K, also have another signal at 47K.

Therefore, we can see, half of 48K is 24K, 24K as the center, 1K signal and 47K, is a mirror image relationship. Similarly, if a 5K signal, there will be a 43K of the mirror signal.

The frequency of these mirror images are distributed in the 24K to 48K, although the human ear can not hear more than 20K frequency, it can be ignored. But if the following amplifier is not ideal, two different frequencies may cause intermodulation distortion (IMD) to produce a new frequency falls within the audio range. Therefore, in order to eliminate the hidden troubles, reduce the demand for the following amplifier, these signals outside the audio bandwidth should better to filtered out. Then the problem is that the analog filter, if you want to protect the 20K bandwidth in the amplitude and phase to be flat, but also filter out those very closed 24K-48K signal, which becomes impossible, although a high-order analog filter can be very steep cutting, but it has great impact on the audio bandwidth inside 20KHz, the result is not worth the candle.

The other advantages of digital filters is that they can be perfected in the frequency domain. Digital filters can be used for oversampling and filtering easily. For example oversampling 4X times, 48K * 4 = 192K. 192K half is 96K, then 0-20K audio signal to 96K as the center of the mirror image is 176K-196K. Which has been opened in the middle of the buffer zone is large enough to allow the analog filter to work well.

The Audio Engineering Society has specified a measurement standard for DAC and digital amplifier measurements, called AES-17, which requires a complete filtering off signals outside the 20K audio range to avoid problems affecting measurements. If the DAC is over-sampling, even if not in accordance with the standard will not have any impact, because the out-band of the signal has been filtered out off very clean. However, if the non-oversampling DAC, if not in accordance with this standard will cause the measurement results will be wrong.

In addition, in the high frequency part, if the oversampling processing is not performed, the amplitude of the sample will be attenuated due to the lack of samples. If at 48K sampling rate, signal at 20K will be 2-3db attenuation. Over sampling process will not have this problem.

The problems of oversampling

As you can see, oversampling is really a very good technology and it solves a lot of problems. It can be said that, based on the digital filter, over-sampling technology is perfect at frequency domain, even the very simple digital filter, the performance in frequency domain is awesome. Well, then, we have to talk about is the problem of oversampling, which is why we want to reject this technology.

First, oversampling will have a ringing artifact. A pulse signal without oversampling should be as shown in Figure 1, but the oversampling technique we are using now leads to distortion in Figure 2, or Figure 3.










Figure 1                      Figure 2                                       Figure 3


In Figure 2, we see that when a signal is transformed, it produces some ringing before and after the transform. There is some new technology which can result in Figure 3, which can effectively remove the pre-ringing, but the post-ringing will become quite large.

From a musical point of view, the pre-ringing is like there is something came out before a note should appeared, this will cause some oddly feeling. Post-ringing is like echo, resulting in some additional reverb, the playback environment will usually have a certain reverb, for example, if we in the room, the sound is from multiple souces. First, the sound source generate Direct sound to you ear, then there sound will reflect on the wall then reach to you ear, and some are reflected several times before it reaches to your ears. It should be noticed, the post-ringing in Fig. 3 is much bigger than in Fig. 2, so when it reaches that level, you will feel the sound is lagged and the transient will be poor. In addition, the post-ringing should disturb the original room reverberation, resulting in the positioning of the instrument in the sound field has also been a problem. The reverberation mode of a room is fixed, but the post-ringing generated by oversampling is subject to change all the time. So the sound field will be confused, the position of instruments are flying. From the listen experience point of view, I personally think that Figure 2 has smaller problem than Figure 3.

If from the waveform you can not directly understand how the ringing artifacts can affect the listening experience. The following picture may be more intuitive to understand, because the ringing effect also exists in the image processing.




The left is the original image, the right is ringing artifact. Maybe some people will think that the right is the high-resolution details, edge sharpening will highlight some of the lines, but at the same time the loss of delicate texture and natural transition. Have you noticed behind this men, there is a men in black, he is watch back. In the original image, we hardly notice that person, but in the right image, it is much easier to noticed. So, as an audiophile, which do you like?

Why the above problems look so bad, but still haven’t been noticed as a big problem for a very long time? Because the audio industry using spectrum analysis to measure various standard specs. The digital filter which we used for over-sampling, has perfect performance in frequency domain. But in the time domain, it is very bad. In an easy word, the frequency domain is equivalent to the content, time domain is equivalent to the order.

For example, the following statement, the content is complete,

  1. Little monkey can not find the father and mother, very anxious.
  2. Little monkey can not find the father, and mother very anxious. (A small change in the order, but has caused ambiguity)
  3. Dad can not find mother, little monkey very anxious. (Greater variation in order, greater ambiguity)
  4. Dad not find can little, very mother monkey anxious. (More chaotic order, although the content is still complete, but has no meaning)


Moreover, the missing information, whether the mathematical calculations can be very good to calculate the missing information and make up for it?

Here, we first talk about people’s vision and hearing, eyes and ears is just a pick up information organ, the information is transmitted to the brain, the brain has a special part responsible to interpretation of these informations. The brain is very powerful of these analysis Functions, Our modern science, even with a super computer, can not compete with a brain. How brain can analyze these information so powerful, is still a myth to our modern science.

As a visual example, we know that we need two eyes to form a three-dimensional vision, that if you try to close one eye, you will experience some impact, but not very big, you can still know how far the objects is away from you. You can still have daily life without much impact. The 2D information capture with one eye only, after the analysis of brain, can easily make up three-dimensional information which the original information do not have.

As an example of hearing, we know that we have only two ears, in a horizontal line. So, we distinguish left and right in the sound field is very easy to understand, because the two ears are in a horizontal line. But why can we tell if the sound is coming from up side or down side? This is also because of the brain has done some very complex calculation, and make up those vertical information which was not able picked by two horizontal ears.

Above these two examples, we can not use modern science to achieve, even if we ignoring the cost. But the human brain can achieve it like a piece of cake.

If there is a very simple sentence


The middle one is lost


The lost information, the computer is very easy to fill. For the human brain, it is also very easy to fill


Again, if there is now a slightly more complex sentence

Little monkey can not find the father and mother, very anxious.

One word is lost

Little monkey can ___ find the father and mother, very anxious

If the audio digital filter try to fill it, will be a strange result like this

Little monkey can can find the father and mother, very anxious

But for the human brain, the correct fill this information is not a bit difficult, the human brain analyze the rest of those messages to understand the meaning of the whole sentence should have, so that the correct fill that missing information.

Therefore, the conclusion comes out, for simple and monotonous information, the existing calculation method can be used to fill it correctly. However, the more complex information, the more the existing calculation method will fill in the wrong information. The human brain, is much more powerful in this case.

For the missing information, the traditional audio industry believe “there is always better than nothing”, but this is actually a wrong direction. For people, ” rather go without than have something shoddy “, a wrong information is far more harmful than no information. Without this information, our brain can make up. But if there is a wrong information will be misleading, and also lost the opportunity to make up by human brain. This is like a pot of soup lack a piece of meat, that may not be a big deal, but if you make up a dead rat into the soup. Rats are also meat, for measurement specs, a pot of soup that maked up enough of the meat, is a better result.

Measurement and listening

Here, we can see that, for audio, we should understand that the audio industry’s existing measurement spec. can not correspond to people’s listening experience. We are not able to use those number to quantify our listening pleasure. These spec tend to be analyzed from the completeness of the content. However, this is not the case for human feelings. Complete information, but the order of the chaos will have disastrous consequences. Likewise, more information is not meant to be better than less information, and a well-designed non-oversampling DAC, we will find richer and more natural details.


2 is greater than 1, even primary school students know it, but more advanced wisdom is to know how to do correct trade-off. Less is More

Posted on 18 Comments

Singxer SU-1 KTE (Kitsune Tuned Edition) – Tuned PSU (3.2uV output noise!)

We have spent a lot of hours with the Singxer SU-1 and recommend this product to many of our customers. It pairs well with most any dac! This SU-1 also has proven to bring the most out whatever dac you choose to pair with it.

Recently we have been modding the SU-1 to improve the power supply.


We have a KTE edition we have been working on and a few prototypes approx 3uV and 125db PSRR. Also in the process having had upgrading the caps on the main board and the psu board. Pure silver wire is used for hookup wire, an upgraded fuse, pure copper Kitsune Badging, and these KTE editions will come with an **optional ultra premium HDMi i2s cable (Supra HD5 0.5M HDMi i2s cable). So far we are very pleased with the results. Below are some photos of the KTE SU-1 and we are gauging the interest for these as a product we would offer. Please contact us for any specifics or if you’d like to place a pre-order.  Complete installation modding service on a new SU1 with all components and labor included. HDMi cable is optional extra. Upgrades are available if you own the SU1 and purchased from KitsuneHiFi originally. The cost is between 330-380usd plus shipping cost to and from (price varies if you choose High Retention USB connector or none.  Please contact us if you’re interested.

Solid copper KitsuneHiFi fox badging is available for KTE products (we sell these to customers who would like them for their official KTE dac)

email us with your serial # to verify and the price of the badges vary. Below are a few samples of our KTE badging

Here are some internal photos of our KTE edition SU1

Pure silver OCC solid core wire was used for the PSU to mainboard connection (new film caps and power caps
 Take note the new 5 caps used on the mainboard
KitExcellent components have been chosen for the SU1 KTE. New photos will be updated soon to show current components in use. They are different than photos and improved design since these photos were taken.




Posted on

Audio-GD NOS11 – Custom with many extras!

I’ve owned this Audio-GD Master 11 for a little over a year and it’s served me very well as my end game DAC/Amp for headphone use.  It’s more than capable to drive HE6 headphones and even can power speakers from the dual 3pin output on the front panel! No joke! it’s truly a beast! I’ve done a lot of mods to this m11 and improving its sound signature a bit.

Up for sale is my beloved Audio-GD Master11 – originally purchased direct from Kingwa when it was released. It was sent back to him to update to the NOS11 at factory. and can be changed back if desired, including spare firmware too.

*Updates were Norne Audio Silvergarde S wiring to replace ALL the copper OCC wire in the amp/dac.

*All connectors replaced with EiZZ tellurium copper/gold plated. (XLR and RCA’s.*)

*Kingwa also hardwired or soldered the upgraded gain stages (+6db gain for a total of 22+DB)

*This has the Amanero Combo 384 USB module

*HDMi i2S module options installed. Please note all these upgrades.

*3M AB5050S EMI/RF film on the pcm1704uk’s, and over the front panel output wires/connectors to shield from Eddy Current from the R-Core transformers.

*Updated with the newest firmware in Early December (nice increase of soundstage and detail retrieval. I would say 5-8% increase/improvement from previous firmware/DSP)

*Black Screw update (replaced all silver screws on front/sides etc)

This is a one of a kind dac/amp and a very high end setup. There is tremendous value in this gear. The wire alone is several hundred dollars, and the connectors are several hundred dollars. I shipped it back to China for 260usd to have the owner of Audio-GD do these update/upgrades. So add labor too. I hope this goes to another fellow Audio-GD enthusiast. This is an incredible amp/dac and one of the best i’ve heard. Truly an end game setup. It will drive any headphone and do so with ease!

Also comes with aluminum remote control (not shown in photos.) perfect shape.

This will come shipped in original box. Please allow 80USD to ship to USA only. and buyer pays paypal fees 2.9% – Email us if you’re interested in this DAC/AMP

price is 1900usd + 80 shipping/paypal fees. USA only

dsc00151-large dsc00153-large dsc00155-largedsc00159-large dsc00163-large dsc00167-large dsc00169-large dsc00172-large dsc00173-large dsc00174-large dsc00175-large dsc00182-large dsc00132-large

Posted on

Introducing iFi’s newest and smallest Amp/DAC

Nano iDSD ‘Light Edition’

Pocket rocket.nano_idsd_leSouthport, UK – 9th December 2016

Small, dark and handsome The original nano iDSD redefined the US$200 price class for sonics with a worldwide following of portable/home music lovers enjoying all formats from DSD to MP3.
The new nano iDSD ‘LE (Light Edition) in stunning dark titanium is a $129 reboot:

1. Same stunning ‘True Native’ Burr-Brown PCM/DSD chipset

2. Headphone power output of 130mW with 10-hour battery life

3. RCA outputs on analogue volume control

Below US$150, there is nothing as advanced as the nano iDSD LE:


•Burr-Brown ‘True Native’ chipset

• AMR trickle-down ZeroJitter Lite® for low-jitter/phase-noise

• DC-Infinity circuit, DirectDrive® Servo-less circuit

•Battery-life ~10 hours (Compared to original nano DSD, no SPDIF out, no filter switch and DSD128)


•Power Source: Battery/USB Bus power

•Battery Life: ~ 10 Hours playback*

• Formats: 44.1/48/88.2/96/176.4/192/384KHz PCM

2.8/3.1/5.6/6.2/11.2MHz DSD 353/384KHz DXD

• Bit-Perfect DSD & DXD DAC by Burr Brown (1-DAC Chip; 2-Channel; 4-Signals)

• Input: USB 2.0

•Output Audio RCA 3.5mm Headphone

•Dynamic Range: >104dB

• THD &N (35mW) < 0.005%

• THD &N (Line) < 0.005%

• Output Power (16R): > 130mW

• Output Voltage: >1.65V (>100Ω)

•Output Impedance:(Zout): <1Ω

•Power Consumption: < 3W (charging battery and playback together)

• Dimensions: 106(l)x67(w)x28(h)mm

• Weight: 162g(0.43lbs)

# requires Camera Connection Kit or OTG adapter. Cable sold separately.

Specifications are subject to change without notice. The retail price of the nano iDSD LE US$129

Posted on

Torq’s Review – the HoloAudio KitsuneTuned Edition Spring Dac.

Holo Audio Spring DAC (Level 3) Review by Torq

Head-Fi’er @Torq got his hands on the Level 3 version of the Holo Audio Spring DAC. See why he says “It’s the most musical and enjoyable NOS DAC that I’ve heard so far, and doesn’t give up anything on the technical side” by clicking here: Holo Audio Spring DAC Level 3

Head-Fi Spring Dac

Posted on

Cool projects going on at KitsuneHiFi

The spring dac was special for RMAF 2016 , we decided to promote and show off it’s beautiful design. As it’s exterior is quite elegant and sleek. The interior is something we are very proud of. The pcb is painstakingly designed with care and precision. This is part of how it measures so well with linearity. The left and right channel’s are perfectly mirror images of each other with exactly the same trace lengths and component placement. check out what we did to show this off… Customers are now asking us for this option, It “MIGHT” be available in our store, but the cost is yet to be determined. Likely will be 100usd for a full color, gloss laminated, with squeegee installation tool and shipping included. These are printed on a top of line Latex printer that is capable of 1200DPI resolution. print quality is excellent.

Printing them on our 30k latex printer Stickers for kitsunehifi

Full color stickers have also been made and can be sent on request with orders if we have in stock. All KitsuneTuned Edition items are exclusive to owners/buyers from KitsuneHiFi ONLY.  We are HoloAudio USA and only offer to our direct customers exclusive VIP items for their special edition Dac.

Spring Dac with PCB stickerour end results was a beautiful gloss laminated finish and high resolution image of it’s internal beauty. We are proud of this dac and all the materials that went into it. Each component is specially paired and all are factory authorized and tested. These are backed by Jeff Zhu and fully factory warrantied.  This shows off the beautiful solid silver O-type transformer, our new wiring updates with 1.5mm solid silver wire, direct soldered to pcb with mundorf supremem 9.5% silver solder. also pure silver Oyaide faston connectors, Audio Horizon Platinum reference fuse.

Also we will be offering optional copper badging for some of our new product, check this out. we even got new business cards for RMAF.

My wif'e's Business is always professional.
My wif’e’s Business is always professional.

Copper KitsuneHiFi emblem Copper KitsuneHiFi emblem 1.25 Copper KitsuneHiFi emblem

Solid pure copper badging is actually for a few projects… one will be our new BendyHead Headphone stands, and the other will be optional badges for the spring dac. such as below. Pricing will vary from 10usd-20usd per badge.


Kitsune Tuned Edition Copper Badge Kitsune Tuned Edition Copper Badge

Posted on

KitsuneHiFi / HoloAudio USA@ RMAF 2016

Holo Audio USA and KitsuneHiFi was fortunate to make this event and we co exhibited with Headamp in the Canjam Tent! This was a WIN WIN for both of our companies as Headamp truly benefitted from our Kitsune Tuned Edition Spring dac just as much as we did from their World class amps. The Blue Hawaii SE (worlds finest electrostatic headphone amplifier) and the GSX-MK2 which is also considered one of the worlds finest SS headamps.  Our customers were able to hear our dac with very transparent and revealing amplifiers.  Many customers told us we had the best sound in show! that was definitely nice to hear this. Overall the results were very spectacular and we’d be doing this every year now!   Our booth was tight and small yet very busy most of the time.  Our listening station had a pair of LCD4, HE1000V2, MS Carbon, Prototype MS Electrostats, and the Stax 009.  All serious summit fi gear.   Our source was a simple basic Microsoft surface book. From there we used a kimber cable USB cable to our JCAT Isolator (intona industrial pcb), and from that we used the JCAT ISO cable to the Singxer SU-1 (one of our best selling products!)  The singxer SU-1 was the transport that allowed us to use the HDMI i2s input of our spring dac. The HDMI i2s IMHO is the best possible input to use for optimum sound.  Why do we use JCAT isolator? Well show conditions had poor power quality, let alone any laptop has a poor PSU, this was key to eliminate any ground issues with true galvanic isolation of power and data at FULL usb 2.0 480mbps. Honestly I won’t have any audio system without my Jcat Isolator, it’s simply more than subtle.  In show conditions it really does help and we like to thank JCAT, one of our vendors. Check out their new USB Femto card, this thing is sweeeet. 😀 OK i’m doing too much product shout outs, but these companies deserve it. Singxer, for instance, is one company that has USB figured out and has developed a digital to digital convertor with spectacular results. Their SU-1 has femto clocks, Crytek 575, and multistage regulation of power, and the newest XU208 xmos chipset.  We choose their DDC as the ideal pairing with the Spring Dac. It delivers slightly more detail and wider soundstage.




Posted on

iFi Audio Comes to Kitsune HiFi!



About them….

iFi is a brand new line of electronics with trickle-down technology licensed from AMR and aimed primarily at the future, Computer Audio generation. All iFi products boast Class A analogue circuitry with no DSP and the signal stays ‘Bit Perfect’ throughout.

How a product looks and performs matters, but so does its impact on the environment. That’s why nearly every iFi product and its packaging are made from highly recyclable materials like aluminum, paper, recycled plastic and why we refuse to use harmful toxins in our components. They do this to ensure that every product they release meets our environmental standards.


We’re so excited to share our new discovery with you we’re offering free shipping on everything iFi


Products we carry from iFi. Why we chose to carry the product…….


 Pro iCAN

We chose to carry the Pro iCan because this amp has more performance bang for its buck on the market. It includes the option of Solid State output, Tube output, and a of Solid State AND Tube combo output. On paper, the measurements, are better than most amps on the market, in practice it lives up to it standards. We feel this is one of the best headphone amplifiers on the market today and would like to share it with you! Outputs include, balanced outputs, 4-Pin XLR, dual 3-Pin XLR, and 3.5 mm TRRS. The remaining outputs are a 3.5 mm TRS and two single ended 6.35 mm. Lastly, its future ready with an output for an electro-static module .


Accessories –

SPDIF iPurifier

 53 mil-spec component (BOM) that stands packed to the brim (top and bottom views)spdif-launch453x327

iFi has been hard at work on changing the notoriously unrecognizable unbalanced wave into a true Square Wave, to even out signal and reduce jitter and is what we audiophile enthusiasts appreciate. Bit Perfect audio ensures the audio signal remains unchanged handling  44.1 / 48 / 88.2 / 96 / 176.4 / 192kHz; 16 to 24 Bit, DSD/DoP, and even Dolby Digital and DTS.  Galvanic isolation prevents noise contamination in the signal. To further reduce noise jitter and drift a large number of complete audio samples helps absorb variation in signal. With co-axial/optical input AND co-axial/optical output enabling compatibility with any and every product from mass market Apple TV to a transport audiophile DAC.  Add that extra, included in box, iPower ‘quieter than battery’ 1uV noise floor adapter, and low power consumption what else could you ask for. <300 femto second crystal clock with a precision of 0.001Hz fine tunes the signal and limits drift.

Active Noise cancellation

Any noise in our power and signal inputs distorts the perfect sound quality we’ve put so much time, effort and money into achieving.

The following accessories include active noise cancellation based on military radar technology actively generating an identical signal in the opposite phase which neutralizes the sound providing a clean power supply signal. When measured, this drops noise by 100 times or 40 dB compared to the Common Noise Filter.

DC iPurifier


A laptop uses a switch mode power supply and this device filters the power, improving it to the level of a linear power supply, this is why we use it on our personal laptop and other mobile listening stations. This provides ultra clean, low noise power, with active noise cancellation,  noticeably reducing the noise floor. Also this features a reversed polarity protection/notification light stopping power input preventing potential damage to your device. You can also improve your the performance of your existing power supply by adding this inline with a DC output cable.

iPower (5V, 9V, 12V and even 15V versions)

ifi ipower ipower61 walloutlet

An ultra low noise floor DC adapter, 1uV compared to the average audiophile linear power supply which measures 20uV. We like to use these for powering devices such as USB modules, DACs, AMPs, DIY projects and other devices such as RPi (Raspberry Pi), Arduino, Microrendu, Logitech Squeezebox, Uptone Regen, Paul Pang USB card, etc.

iPurifier 2

iFi Audio iPurifier 2

Cleaning up the digital USB audio chain by reclocking, rebalancing and regenerating the USB transmission. Clean up your downstream DAC by restoring Signal level, reduce Jitter reduce Common Mode noise, rebalance signal, impedance Matching. We use one of these in our Audio-gd NFB-28 it increases the resolution of the top end detail and removes some of the harshness that delta sigma chips are known for. Future ready for all present and future PCM/DSD/DXD formats – without restriction.

Posted on

KitsuneHifi aka HoloAudio USA brings you a new product that will excite your senses!

[metaslider id=513]

This is the “Level3” KitsuneTuned Spring Dac Above

[metaslider id=568]

This is the “Level2” Rise Ji Edition Spring Dac Above


Kitsune HiFi carefully chooses products that we want to offer to our customers and be unique and/or stand out from all these audio supply stores. We have recently been talked with a very talented engineer that is the brains behind Holo Audio. Jeff Zhu is the man behind the curtain and has impressed us with his knowledge and his newest line of products he offers. The Spring DAC is his newest product and is capable of DSD native on an R2R DAC!  This spring is capable of DSD512 native and actually running on his bespoke discrete ladder DAC! This unique DAC is not your run of the mill DAC, it’s full of unique and well thought out designs and implementations. Patented R2R technology. This is the first discrete DAC that has linear compensation and this allows for ultimate music reproduction accuracy. Dual R2R network for PCM, and Dual R2R network for DSD! We will be offering the first Spring DAC’s in the USA and the Level 3/Kitsune Tuned edition is Exclusive to us.   We won’t stock more than a few at a time, and the wait time is usually a few weeks.

You may be asking… well what is so special about this DAC? The first Discrete R2R dac with Linear compensation…what is this?

There’s an additional R2R ladder in the Spring, it compensates the main R2R ladder. It works like trimming, but trimming is to change the resistor value. This additional R2R ladder is digitally controlled and will accurately compensate the resistor tolerance. For example, the msb of 16 bits should have the value of 32768, but due to tolerance, it represents 32700 in real world results. Then that additional R2R ladder will compensate 68 into it. Thus it now becomes 32700+68=32768.  What this means is that it’s likely the most precise Discrete dac on the market with near flawless linearity, lowest THD and highest SNR of any NOS Dac. The sound is simply something you must experience to fully realize how special this dac is. It’s a patented technology exclusive to HoloAudio.


Product Description:
This new DAC.”Spring” is the first design of a new era, a milestone for HOLO Audio’s own Jeff Zhu. It’s a full discrete R2R type of audio decoder and does not have off the shelf-DAC chip! This is a bespoke custom-designed core DAC module and truly a breakthrough with technology for any DAC chip today. The Spring is here to achieve new heights, new dynamics and simply a full spectrum of audio to please the aural senses.

Here at Kitsune, we have three versions of the Spring DAC. Other places ONLY offer two models. Below are the three models:

HOLO Spring base model – LEVEL 1
(The base model is quite good all by itself and priced so competitively it’s impossible to beat the performance/price ratio.)

HOLO Spring “Rise Ji Edition” – LEVEL 2
(with 4 Jensen 4700uF / 63V added for ultimate power filtration and improved sound quality –  For sound effects are more relaxed, transparent and details are more apparent. This is the most popular version that sell’s most often.)

HOLO Spring “Kitsune Tuned Edition” – LEVEL 3 and exclusive to Kitsune.
We offer a transformer upgrade to the highest grade available. The O-core 100VA red label is replaced with a 100VA O-Core green label. Price is significantly more due to the OCC Silver Primary and OCC silver secondary wire used in this transformer. The sound improvement is not subtle, it’s a night and day improvement in our opinion and many of our customers agree. Kitsune has tuned this Spring DAC. We are very proud to offer this Model that is installed from the factory who manufactures the transformers for HOLO Audio. This transformer is our finest available model and makes the Spring’s sound quite elegant and improved detail.
Kitsune HiFi Pure Silver Transformer
Kitsune’s Pure Silver O-Type Transformer

DSD is natively supported for the first time on this R2R Discrete DAC
HOLO Audio is the world’s first to support DSD natively on R2R DAC, so far the only one. This is not the DSD converted to PCM before digital-analog converter, but directly by the discrete components of the DSD digital to analog converter. Supported currently on MAC and Linux, and soon to be Windows/PC. Evaluation driver for native DSD512/256 here

Spring’s input and output interface:

  • Digital input interface includes
  • USB (ground isolation)
  • RCA coaxial
  • BNC coaxial
  • AES
  • optical fiber
  • I2S (HDMI)
  • All digital input interface supports DSD (DOP mode).
  • Analog Output interface: single-ended,  balanced.
Spring DAC R2R structure and design
Modern and popular delta-sigma type DAC differs from R2R within one clock analog value can recover a sampling point, and the delta-sigma is used to represent an analog signal after passing through oversampling and high-speed digital switching 0 and 1 . In comparison, the conversion structure of R2R is most direct and pure, but delta-sigma is essentially a digital chip, high-speed digital signals 0 and 1 switch to the low-pass filter to process the analog signal and this process is prone to various problems, produce digital sound (digititus) and also in the super-sampling process will inevitably cause some ringing and distortion. But DAC R2R structure requires high-precision resistor network which the cost can be very expensive. And the digital delta-sigma DAC chip in comparison is very low cost. Patented R2R technology. This is the first discrete dac that has Linear compensation and this allows for ultimate music reproduction accuracy. Dual R2R network for PCM, and Dual R2R network for DSD!
With diverse and flexible sampling mode conversion mode
1: NOS mode , no digital oversampling, the raw data is directly converted to analog.  Because digital oversampling will produce time-domain distortions such as ringing, so NOS avoid these problems. Generally NOS mode other performance indicators have a significant impact, but the Spring is designed to allow top performance while in NOS mode.
2: OS mode , the PCM super-sampled PCM higher frequency, the DSD DSD super-sampling to higher frequency, then digital to analog converter.
3: OS PCM mode , enter either PCM or DSD input, all the way to the PCM oversampling analog to digital conversion.
4: OS DSD mode , enter either PCM or DSD input over all sampled DSD approach to digital to analog converter.
What are the Spring’s sound characteristics?:
One of the strong points of the spring is the spatial sounds stage presentation. It’s very wide and precise. Never does it sound congested list many DACs. The high end is very detailed and never harsh sounding. It has the R2R goodness that many have come to know and love. R2R has almost a cult like following and this DAC will surely become a sought after DAC.  Everything comes effortlessly for this DAC. Bass is layered and textures, quite full and present without ever masking over the sweet mid tones that are produced. One special thing about the Spring is it’s amazing voltage regulation circuitry that has no more than .2uV noise! This along with the O-core transformer helps regulated the purest power which translates to audio clarity and accurate reproduction. I would say this Spring DAC is better than just about any DAC I’ve ever heard which is saying a lot! Must hear it to appreciate and come to love it’s sound. The Green Label transformer is a step up that brings a new level of clarity not even thought possible. The Red Label already sounds incredible, but if you must have the best and nothing compromised then I would suggest adding the Green Label.

A little note from Jeff:

Many have asked about how does a discrete resistor network accurately reproduce sound. This email is a good read for those who want a technical explanation on how Jeff was capable of achieving flawless linearity and ultra good specs on THD. This was a reply to a customer who had a technical question about resistors and their limitations.

“About the resistor tolerance, his word is partly right, but not accurate. The MSB of 16bits value only represent 32768, not 65536, all 16bits is 65535, but the MSB is only 32768. There are techniques to reduce this requirement. For example, segment+R2R, using segment into the first few bits, then r2r for the rest of them. Rockna is segment+R2R, TotalDAC is all R2R. And you can see, Rockna’s THD spec. is much better than Total. Total uses foil resistor, but they used the simplest architecture. That results an unsatisfied test performance. To my opinion, it’s some kind of wasting foil resistors. Soekris also uses segment+R2R. MSB’s old model is all R2R, but I don’t know about their new models.

Anyway, There are ways to improve the performance, the segment+R2R is one way. Trimming is another way. There is an additional R2R ladder in Spring, it compensate the main R2R ladder. It works like trimming, but trimming is to change the resistor value. This additional R2R ladder are digital controlled and is to compensate the resistor tolerance. For example, the msb of 16 bits should have the value of 32768, but due to tolerance, it represent 32700 in real world. Then that additional R2R ladder will compensate 68 into it. Then it became 32700+68=32768.

Actually there are other ways to improve performance, I just told you two stories. There are more stories behind the design. It is really hard. I must count every via holes in layout, every PCB wire has it’s impedance and must put attention on it. A via hole can be 50 m Ohm, it is 1/20000 of 1K ohm, and you see, it covers 32768 which is the msb of 16bits. Also the switches has self-impedance which is about several ohms to 30 ohms. If you find this issue. Maybe you will go crazy, it seems to be an impossible mission. But a good designer will overcome all these problems. It’s our value at HOLO Audio.

All the resistor tolerance, switch impedance, line impedance, via hole impedance, finally reflect as linearity. There is a chart that shows Spring’s linearity, it’s excellent. But I suggest you to look at THD performance. Bad linearity must result a bad THD performance. But good linearity doesn’t mean there will be good THD performance. THD is dynamic performance, more critical than linearity which is a static performance.

If you look at the THD performance, you will find Spring to maybe the best among those competitors. I’m applying a patent which contribute a lot of that performance. But I can’t tell you how. It’s confidential now.

Best Regards

Jeff Zhu”

And another note on Linear Compensation. as Some have asked what happens when the Resistors values change over time? Jeff responds:

No matter it is a digital switch or analog switch, they are built by transistors, can be either bjt, mosfet or jfet. They have self-impedance call Ron(impedance when at on state). If we expect a lower Ron, the a bigger transistor is needed. But bigger transistor has bigger self-capacitance that will cause switching speed to be lower. So there is no perfect components in the real world.

About long term drift, the modern resistor technology has made it much better than older process. But the most important thing is, we don’t care absolute drift, we only care relative drift. That means, if all resistors are drifted in the same direction and same speed. Then it will be no impact for this application. So, we make sure to use the same batch of components in a single board. That will minimize the problem causing by long term drift.

As I mentioned in the earlier email. Spring is not using trimming, but using compensation. Not matter it is trimming or compensation, their goal is same, to make the resistance much more accurate so the linearity and distortion can be improved a lot.

Best Regards
Jeff Zhu

Also, another customer asking about the oversampling, Jeff replied and has some good information to those who are curious.

The oversampling is done by AK4137, ‘OS PCM’ converts the input stream to PCM no matter it is PCM or DSD, then send to DAC. ‘OS DSD’ converts the input stream to DSD no matter is is PCM or DSD, then send to DAC. ‘OS’ over-samples the input stream. If it is DSD input, it over-sample to DSD256. If it is PCM input, it over-samples to PCM384K or 352.8K ‘NOS’ does no oversampling. send the input stream directly to DAC module.About DSD ‘native’, it is the way the USB transfer the stream. You know DOP, which is DSD Over PCM. DOP put the DSD into a PCM stream. At the receiver end, it just unpack the PCM stream and get the DSD back. DOP does NOT change any bit of DSD stream, just sending it with the package. So it waste some data rate. At the same data rate, DOP is 1/2 of DSD native, the other 1/2 is the package. So DOP and DSD native are all bit perfect. The DSD stream sending to DAC is all the same.So, now you should know, DSD native is just about the way it transfers the stream. Not describing the way it does digital to analog conversion. At the DAC stage, the DSD stream can convert to PCM then do the final digital to analog conversion. Actually, most DAC chips do this inside the chip. One point needs to be remembered, DSD streams can’t do volume attenuation. If this chip can do digital volume control under DSD mode, then is must converted to PCM already.Let’s go back to Spring. If Spring works in ‘NOS’ mode. There is no
oversampling, no conversion, just the original data doing the final digital
to analog conversion. I will not explain how it does PCM conversion, you
should already know quite well about it. There is a lot of resistor ladder
DAC, they are born to do this. I just explain how Spring can do ‘discrete
DSD conversion’. Actually it is like DCS and CHORD. If you looking to DCS
and CHORD’s PCB, you will find ‘discrete DSD conversion’ is also using
switches and precision resistors, like resistor ladder DACs. So, no matter
it is ‘discrete DSD conversion’ or ‘discrete PCM conversion’, they all use
the same discrete devices. Just work in different architecture. That makes
possible to combine these two modes in one device.

Best Regards
Jeff Zhu

Measurements –

Please take a closer look by clicking on the image to open in new window. Take note of the Linearity of this dac. And also take note of the THD level’s being ultra low along with the Signal to Noise Ratio being better than average in general, but remember this is a NOS Dac so it’s actually extremely good and definitely one of the best NOS dac on the market today! 

[metaslider id=514]

Here are some more photos that also explain some details 

[metaslider id=515]


This photo shows squarewave NOS vs. Oversampling.

THD @ 1K @ NOS
THD @ 1K @ DSD
THD + N @ 5K @ NOS (0.00068%)
Red Label O-Core 100VA
Holo Spring Photos
Base Model Power Caps
Red Label Transfomers
“Rise Ji Edition” w/ red label Transfomer
Jensen Caps 4700uF 63V
“Rise Ji Edition+” w/ green label silver transfomer
Beauty of discrete R2R
Spring Logo
take notice of custom HOLO IC’s.
A small sealed transformer is ONLY used for standby power. The spring dac was designed to be POWER EFFICIENT
no off the shelf USB modules, this one is HOLO audio brand.