First, came the E1DA COSMOS ADC that brought to the masses who were willing to deal with the well, usability issues, very close to the power of the 'big-dogs' like Audio Precision to the home audio lab.
Then, came the APU (Audio Processing Unit), that added an analog 1khz/10khz notch filter to get THD and THD+N measurements even closer to those big dogs.
Finally, now the Euphonic Review lab has the EIDA COSMOS SCALER, which is a buffer and scaler that will further refine our in lab measurements.
I have read from a reputable source that we are talking about accuracy and quality of measurement somewhere between the Audio Precision SYS27XX and APx555, with the addition of the this new auto-scaler with a high enough impedance to accurately measure pretty much any DAC on the planet, is probably closer to the APx555 in performance!!! Not in features, mind you. The Euphonic Review lab will be limited in features, but what features we DO have we consider very, very useful and more than anyone making purchasing decisions really needs.
The GREATER point is, that as good as our measurements have been so far, with the addition of this new Scaler, we will be on an accuracy level that competes with anyone. You name it. Online or magazine. I know that is a big statement, but the reality is just finally here that the hardware has trickled down to the 'little guys' and there is no more monopoly on state-of-the-art audio measurements.
But measurements have NEVER been the primary source of pride here at Euphonic Review. Our source of pride has always been our HONEST reviews. And what we believe to be excellent 'audiophile' ears to go with it. That is an explosive combination, and combined now with impeccable measurement equipment, we think that explosive combination should be going BOOM at any time. We may be the new kids on the block, but we think we have a helluva lot to offer.
Thanks, so many thanks to those who have read from the beginning, suffered through our growing pains (which will surely continue as running a webpage ain't easy), and endured and hopefully enjoyed.
Here's to the rest of 2023 AND to an explosive 2024!
An FIR filter is like a simple sieve at the end of the slide. As the marbles roll down, the sieve lets only the blue marbles through.
An IIR filter is like a smarter, but more complicated, sieve that remembers some of the previous marbles. It might let a blue marble through, then think, "Hmm, the last few marbles were blue, so I'll be a little more lenient with the next one."
Both types of filters are used to process audio signals, either to enhance certain features or remove unwanted noise, much like how our marble sieves are used to sort the marbles.
DO IIR Filters have Delay Lines like FIR filters??
In digital signal processing, the term "delay line" is often associated with FIR (Finite Impulse Response) filters, where the filter uses a series of past and current samples to compute each output. Essentially, the delay line in an FIR filter holds onto past samples of the input signal, which are then used along with the current sample to calculate the output.
IIR (Infinite Impulse Response) filters, on the other hand, use not only past input samples but also past output samples to compute the current output. While they don't have a "delay line" in the same sense as FIR filters, they do have a form of memory that stores past values. This "memory" isn't typically referred to as a delay line, but it serves a similar purpose: it holds past information that the filter uses to calculate its output.
So, in simple terms:
Both kinds of filters can introduce some amount of delay to the signal, but the way they use past information is different. FIR filters rely solely on past and current input values, while IIR filters use both past input and past output values.
What is a Cascade Comb type Integrator vs. a Moving Average Filter explained very simply?
Let's imagine you're trying to figure out how fast a group of people are running. You could use different methods like a Cascade Comb Type Integrator or a Moving Average Filter. Both aim to give you a general idea, but they work a little differently.
Moving Average Filter:
Think of this like taking a quick glance at your stopwatch every few seconds and averaging those speeds. This method will give you a pretty good idea of how fast people are running right now.
Cascade Comb Type Integrator:
Imagine instead you're using a more complex method where you don't just look at the most recent times but also consider times from way back, maybe even from when the race started. This would be like calculating your grade point average (GPA) throughout all your years in school, not just the recent tests.
Both have their pros and cons, and the right one to use depends on what exactly you want to know.
How are Comb Filters used in Audio Processing?
In the context of audio signal processing, a Cascade Comb Type Integrator (often referred to simply as a "comb filter") is a filter used for various applications like reverb, echo, and other time-domain effects. It's different from a moving average filter and has its own characteristics.
How it works in audio:
A comb filter adds a delayed version of the signal to itself, creating peaks and troughs at regular frequency intervals (hence the name "comb"). When cascaded—or chained together—multiple comb filters can provide a more complex and tailored frequency response.
What does it do to the audio?
Imagine you're in a hallway and you clap your hands. The sound of the clap will bounce off the walls and come back to you at different times, creating an echo or reverb effect. Each bounce is like a single comb filter. When you cascade them, it's as if you're in a more complex environment where the sound bounces off multiple surfaces at various distances, creating a more intricate echo or reverb effect.
So, in audio, a Cascade Comb Type Integrator (or cascaded comb filter) is often used for creating complex time-based effects like reverb or echo. It's different from a moving average filter, which is often used to smooth out a signal but doesn't create the time-based effects that a comb filter does.
How are these Cascade Comb Filters used in DSD Processing?
Digital Stream Digital (DSD) is a high-resolution audio format that uses Pulse-Density Modulation (PDM) instead of the more commonly used Pulse-Code Modulation (PCM). In DSD, audio is represented by a series of 1s and 0s, where the density of 1s indicates the amplitude of the signal. It's a bit like morse code for audio; a bunch of 1s close together represent a high point in the audio wave, and a bunch of 0s represent a low point.
A Finite Impulse Response (FIR) comb filter can be used in DSD audio processing to achieve certain effects or improvements in sound quality.
FIR Comb Filter in Simple Terms:
Imagine you are jumping on a trampoline. Each time you bounce up, you touch a hanging bell, which rings. If you bounce higher, you can touch bells that are hanging further away. The first bell represents the immediate sound, and the bells hanging further away represent echoes or repetitions of that sound.
In a similar fashion, a comb filter creates a series of echoes or repetitions in your audio. An FIR comb filter does this by adding delayed versions of the original signal to itself.
Application in DSD:
So in essence, an FIR comb filter in DSD can help in shaping the sound, either to improve quality or to add specific effects. It's like choosing which bells you want to ring as you jump on the trampoline, each creating a unique bounce pattern and sound.
But isn't is still true a Cascade Comb filter is a type of Moving Average Filter?
You're correct that a cascaded comb filter can be viewed as a specialized form of a moving average filter, particularly when it's implemented with Finite Impulse Response (FIR) characteristics. Both are linear filters, and both can be used to smooth out or otherwise modify a signal. However, the way they're commonly used and their resulting effects can be quite different.
Moving Average Filter:
In a moving average filter, you average the past 'N' samples to get the current output. It's like looking back over the last few seconds to see how fast you've been running on average. This kind of filter smooths out sudden jumps in the signal and is often used for noise reduction or signal smoothing.
Cascaded Comb Filter:
A cascaded comb filter, on the other hand, takes the current sample and adds to it a delayed version of itself, possibly scaling it by a factor. When you cascade multiple of these comb filters together, the effect becomes more complex, but the basic operation is similar: current sample plus delayed sample.
How They Relate:
Imagine you're checking the temperature every day. A moving average filter would be like saying, "The average temperature over the past week is 'X' degrees." This gives you a smoothed-out view of what the temperature has been like.
A (single) comb filter would be like saying, "Today's 'feels-like' temperature is a combination of today's actual temperature and the temperature from a week ago." A cascaded comb filter would then involve multiple such 'feels-like' calculations based on temperatures from various past days, leading to a more complex profile.
Now, if you set up a cascaded comb filter in a certain way, it can act similarly to a moving average filter. But they are usually employed for different purposes: moving average filters for smoothing, and comb filters for adding echo, reverb, or other time-domain effects.
So, while you can view a cascaded comb filter as a specialized type of moving average filter, in practice they are generally used for different tasks.
Andrew's comments on the above
The Cascaded Comb filter, a special case of moving average filter, is the type of filter used to filter DSD in native DSD DACs. Multiple streams of the original 1-bit DSD stream are 'cascaded' upon each other, delayed by a sample.
The Signalyst DSC Native DSD DAC has 32 of the 1-bit DSD streams cascaded on one another, offset by a single sample each. This is a 33 level unary/thermometer coded FILTER/DAC. Burr-Brown Native DSD chips like those in iFi products do something very similar, the difference I wish to highlight here is the amount of levels... there are only 8 cascaded/delayed streams, meaning a 9 level unary/thermometer coded FILTER/DAC. Finally, I will mention T+A, the outstanding high end audio hardware producer out of Germany. I can only repeat what I was told by one of their representatives, but their Cascade Comb filter has only 4 offset DSD streams, for a 5 level unary/thermometer coded FILTER/DAC.
Why do the call them 'comb' filters? Because when viewed in FFT , they look like this... a bit like a comb. (this is the S.M.S.L D300 native DSD filter output at DSD64, fc at 13khz. )
I named the S.M.S.L D300 DAC my current King of Budget Esoterica. It sounds way, way better than just another one in a line of nice budget kit, as long as certain conditions are met. Rather than re-hash all of that, I encourage you to click here and check my review if you have not already done so.
The issue isn't with my ears. I can hear just fine even as I approach 50 years. Not that it helps with reviewing audio equipment, it can't hurt that I am a classically educated college level musician with perfect pitch. It's nothing to me to notice overtones, also known as harmonics, both what should be natural to the music, and what is 'added' as harmonic distortion by the 'bad, bad' equipment. I never have really understood the aversion to harmonic distortion. It's, well, HARMONIC. The first harmonic is the fundamental, the second harmonic is the Octave, which means its the exact same note just twice the frequency rate. The 'mean old nasty' third harmonic? Its a perfect fifth above the second harmonic, and therefore is also consonant with the fundamental.. they call it perfect for a reason. Fourth harmonic? Its a perfect fourth. Which guess what? Is the exact same note as the fundamental and the second harmonic! The are all climbing in an octave sequence. Yes, I get that equipment that adds or accentuates these things MAY not represent the musical performance accurately. But I also 'get' that I LOVE the way music sounds on tube amplifiers. I have NO technical or scientific argument or treatise to give you on what makes for the best musical reproduction. All I know is I know what I like when I hear it. It stands out from amongst all kinds of 'copy cats' in a saturated market.
You know what else I like? I am about to welcome you into the cognitive dissonance in which I daily live. I like to know what my hardware is doing on the electronic level. Measurements mean things to me. Surprisingly, they are important. I mean, we have to have SOME BASIC standards, right? I think we passed those benchmarks a long time ago actually, and the crazy low noise and distortion stuff out there is way past our ear's resolution. Yet no one is suddenly claiming digital audio is cured of all its 'digititis'. No one, except maybe a few of that same old cult are proclaiming perfect audio. Same guys who have been yelling 'bits are bits' since the CD hit the scene. (Reminds me of an episode from back in the early 90's when a friend of mine insisted that CD was the thing of the past.. DAT was the real deal and the future... I will let you laugh and clear before we move on. )
So what is my problem?? I am obviously talking about the S.M.S.L D300 DAC. Its PCM measurements are impeccable and it doesn't sound in any way clinical or digital; it frankly sounds amazing.
The PROBLEM came when I started seeing the DSD measurements. The DSD filters are very low, and very slow and gentle. All good so far for the DSD crowd, like me, who wants gentle filters to maximize transient performance and minimize ringing. DSD should be different than PCM. Many DACs just make DSD as much like PCM as possible without actually saying that to the buying public in order to quietly deal with its major issue... all that ultrasonic noise. Ummm... THAT is where we run into the problem with the S.M.S.L D300. Something is off with the analog stage.
The numbers not only are worse than their PCM equivalents, as we go up in oversampling speed, they get much worse. Certain websites would be panning this thing with trophies that have missing parts.
My best guess is the issue is in the secondary analog filter that should come after the initial FIR conversion filter. For instance, iFi has their DSD FIR filter converter followed by a RC analog filter at 80khz to keep the DSD time domain superiority intact while dealing with the remaining ultrasonic noise in a way that adds no new ringing at all, and maintains the signal to noise ratio and harmonic distortion at levels that match its PCM performance. Same with pretty much every other 'native' DSD DAC I have sitting around.
But not this S.MS.L. And damned if it defies the measurements and still sounds amazing anyway, especially at the 'sweet spot' I identified.
What would normally be sacrilege for a 'purist' like me, I caved and am oversampling everything to DSD256 and have chosen to set the filter at 104khz. And it sounds GLORIOUS. And it sounds about as good, as anything I have, including the RME ADI-2 PRO FS R BLACK EDITION in DSD Bypass mode, which not only sounds great, it measures the part. The only native DSD kit I have that really kindof makes quick work of the S.M.S.L is my Signalyst DSC2 discrete DSD converter build, which ironically really doesn't measure that well either since it uses good ol' fashioned transformer outputs like the good ol' tube days.
So with my rant over and my cognitive dissonance as active as ever, I will share with you what the DSD measurements from the S.M.S.L look like on paper. These tests are done either with Multitone's internal modulator that converts all its test tones to DSD, or they are DSD tones tones made by HQPlayer played back via external sources. ADC in use as always is the E1DA Cosmos, so we are in good technical hands here I assure you. Allright.. on with the show.
The rest I will present to you below, so you can see the changes the mostly unseen ultrasonic noise makes inside the 20hz to 20khz band.
I reiterate as bad as some of this might look, it still sounds, well, pretty darn good. There is NEVER anything audible in the noise floor. Never any noted artifacts. Never any noteworthy distortions.
What I believe I CAN surmise that is happening here.... truly excellent time domain performance. Likely very little digital ringing from filters, harmonic distortion that is innocuous or even pleasing to the ear. Masking effects that keep any other nasties or idle tones from being a problem.
And now? I am going to continue to enjoy listening to this merely $400 DAC that has engrossed me with some of the best DSD playback I have heard for the last two hours I have been writing this Blog entry, while I hope you enjoy going through the remaining graphs in the gallery.
Thank you so much for taking time to read my small corner of the interweb here at EUPHONICREVIEW.COM
I thought you might like a look at some data charts pulled from my archives... We all love (or many of us just quite frankly hate) more charts and data! I am starting to develop that love/hate relationship with it... leaning toward the 'hate' part, because every second I am taking measurements or preparing them for publishing is a second that I was at one time and could be at this time actually enjoying the hobby of LISTENING! Maybe someday I can afford what those fancy shmancy sites have.. those people who do things for you for pay. Alas, for now, it is me doing the measuring AND inputing the data.
I came of age with computers as a teenager in the early to mid 1990's. I have been there for it 'all'. Well almost anyway. From the days of 'Quantumlink' dialup on the Commodore 64, to the 'You've Got Mail!' sound every time I logged onto to America Online! from my mid 1990's Macintosh Performa, all the way to the invention of the iPad, Smartphone, etc, and I still feel like a complete idiot around Gen Z who whiz around their smart technology like their smartphone was grown with them in the womb.
That is a long way of saying, yeah, I am more computing challenged than I would like to admit, being a 'pioneer' and all that haha.
Anyway.. all that rubbish said, here is an early preview of the measurements of the iFi iDSD PRO 4.4mm Pentconn output version. Full review finally coming early next week.
Over the past few months I have reviewed several DACs, all of which are DSD capable. Not all of them, however, are what I would consider a 'native' DSD DAC.
'Native' DSD DACs have some differences in their various implementations, but common to ALL of them is digital to analog conversion via an analog filter at the end of the signal chain. I think this rather simple fact is where a lot of people get hung up; however, it is this fact that makes 'native' DSD conversion what it is; an on/off square wave bitstream that reveals the musical signal when band limited by an analog filter.
Unfortunately, this is where what should be an easy conversion to analog becomes somewhat difficult. The DSD ultrasonic noise monster rears its ugly head. The best analog designs will follow the conversion filter with a secondary filter to better manage the noise, because the initial analog filter is virtually always of the 'moving average' type, and is designed to be at its best in the time domain, which is exactly what DSD is - a time-splicing format. Making a filter with lots of taps, or using unequally weighted taps/switches are but a couple of the means used to improve the frequency domain performance, however a balance must be found between time and frequency domain, lest the DSD superior time domain performance be lost. Compromise too much for the sake of frequency domain, and the lines between DSD and PCM really begin to blur.
Finally, one must deal with how any remaining ultrasonic noise affects the 'in-band' audio. This is no trivial matter, because even after the best filtering choices, there can be and there is enough ultrasonic noise left to cause artifacts in-band, exactly where we don't want them! These ultrasonics can and will cause intermodulation distortion, idle tones, and linearity problems.
It is no wonder that the Schiit Audio company in its earliest days made a DSD 'only' DAC because of these various issues. The analog output stage must be optimized for DSD, and here we are with DACs today that use a single analog stage that must somehow effectively passthrough both PCM and DSD formats! Schiit's valid concerns aside, there are indeed many DACs today with a single analog output stage that are very effective passing true 'native' DSD and 'standard' PCM. However, the very issues mentioned by Schiit are exactly why some DSD compatible DAC's do NOT have a 'DSD direct' mode. These 'non-native' DSD converters can optimize the DSD input, add in digital filtering, volume controls, and can re-modulate with their highly linear, high resolution multi-bit Delta Sigma outputs where all is then ready-made for their optimized 'one size fits all' analog output stage. This solves many problems and adds back conveniences that do not exist with 'native' DSD. And indeed does sound quite good if not outstanding in the end. But it isn't 'native' DSD, although many will argue that point. In reality what these DAC types do is actually easier than 'native' DSD strictly defined. It is therefore ironic that the format that can be converted more simply than any other, in the end can become so very difficult.
So why bother with the crusade for the simplest 'native' conversion?
Because: when done well, in my opinion no other other method of DSD conversion sounds better.
Furthermore, if you have been bit by the bug of software conversion to DSD via Roon or HQPlayer, a 'true, fully native' DSD DAC is a MUST. One is especially lucky if he or she has available to them a build of the open hardware Signalyst DAC made of completely discrete parts, that will compete anyday with such commercial discrete converters as those made by DCS.
The DACs I have had in the Euphonic Review lab which convert 'natively' are all iFi DACs with the Burr-Brown DSD1793 chipset, the S.M.S.L D300 DAC with the latest ROHM BD34301EKV chipset, the RME ADI-2 PRO FS R Black Edition AD/DA with the AKM4493 'pre-fire' chipset, and the Signalyst DSD only converter with proprietary chipless converter, which will be looked at at a later date in its own dedicated thread.
All of these DACs use a certain variation on the same theme. The 1-bit signal at the input is completely unadulterated until the conversion filter. These conversion filters are 'moving average' filters, with the slight exception that many of them will use unequally weighted taps/switches that allow for change in frequency cutoff while maintaining the same bitstream speed. A great example of this is the DSD1793 used in the iFi converters. Although iFi in recent years chose to software lock-in a single DSD filter, changed only by bitstream speed, the DSD1793 as implemented in their legacy products can change frequency cutoff regardless of bitstream speed. I have no direct knowledge of how the remainder of these chips select their frequency cutoff and rolloff, but simply observing their behavior, I would say most of them are doing something similar. (The Signalyst, as best I can observe, which will be covered in greater depth in its own future review, uses only equally weighted switches and frequency cutoff can only be changed by a doubling/halving of the bitstream speed.)
Regardless of the finest details, all moving average filters are a form of FIR (Finite Impulse Response) filter, and have a delay line, taps, and accumulator. The 1-bit stream is converted into parallel streams, offset by one sample in time. The taps are the output bit-switches, regardless of whether or not they are resistor or capacitor based. This produces an analog current output that when summed together (and converted from current to voltage) is the filtered analog audio signal. Further filtering can (and likely should) be done by a following analog RC filter to remove more of the ultrasonic noise without having to resort to a steeper roll-off by the initial conversion FIR filter. This allows for higher quality analog output stage performance. This can be even more significant with higher DSD rates, which will have by nature a higher filter cutoff and can possibly allow, perhaps counterintuitively, more ultrasonic contamination in-band. With a double filter arrangement, ultrasonic noise is dealt with in a effective and clever way; the secondary analog RC filter will not add any ringing and allows the primary filter to have gentle roll-off characteristics.
I have made a project of measuring every 'native' DSD DAC I have in my current inventory at Euphonic Review. The results are quite revelatory, and actually in some cases reveal the disconnect that can exist in what we hear and how a device measures. Even the worst measuring DSD signals still sounded very, very good. Would it surprise you that the worst measuring signals often are at DSD512 and higher? That the 'sweet spot' by my best deductions poring over the data MAY be at DSD256? At least in my small sampling of DACs. However small the sample, they do represent a wide range of implementations and hardware. For example, we have the new ROHM chipset, which I don't think S.M.S.L did for its DSD performance the most favors in the output stage. Nevertheless, it still sounds very very good, but certainly could be made to sound even better. Also represented is the DSD1793 which carries the same tech pioneered by Burr-Brown in the mid 1990's for the then upcoming SACD format. AKM is here in the form of the AK4493 DAC chip, with a truly excellent analog implementation by RME. Very, very impressive. And finally, one of the most active companies in the DSD software sphere, Signalyst, is represented in a build of its open source fully discrete moving average filter DAC. (Signalyst results to come at a later date.)
Lets begin with the S.M.S.L DAC. It is the most 'tunable' of the DACs here, with 3 filter choices. However across all possible DSD rates, those 3 filters offer 3 different cutoffs per rate, for a total of 6 possible frequency cutoffs. (Remember, they overlap, adding only 1 new, higher cutoff per rate.) This comes in handy, because the greatest DSD performance swings come with the S.M.S.L DAC.
The following charts are sorted by ACTUAL FIR filter cutoff rate. NOT by the cutoff as labeled on the DAC. For instance, if the Data Rate is DSD128, the filter MARKED at 13khz is actually running at 26khz. If the Data Rate is DSD256, the filter MARKED at 52khz is actually running at 208khz. If the Data Rate is DSD512, the filter MARKED at 13khz is actually running at 104khz. Got it? It may be confusing at first, but once you grab hold of the concept, it's no big deal. DSD 64 cutoffs equal 13, 26 and 52 khz. DSD 128 cutoffs equal 26, 52 and 104 khz. DSD256 cutoffs equal 52, 104 and 208 khz. DSD512 cutoffs equal 104, 208 and 416 khz cutoffs.
So in theory, DSD128 played back with Filter 1 (13khz x 2= 26khz) will have the same filter profile as DSD64 played back with Filter 2 (26khz as marked). Indeed, they do have the same filter profile on the spectrum analyzer, but DSD128 has an advantage. The rise in ultrasonic noise starts later, therefore more of the noise is eliminated by the filter, meaning less distortion 'in-band'.
SINAD is the measurement I have chosen to show these changes in fidelity, and indeed doubling the DSD rate while keeping the filter characteristics the same causes a notable increase in SINAD performance, from 96db at DSD64 (26khz filter) to 99db at DSD128 (26khz filter). As far as SINAD in general is concerned, once the 100db SINAD mark is reached, this is an excellent result. As a matter of fact, the DSD1793 used in iFi products is rated to max out right at 100db SINAD. It is true that the latest, most state of the art chips from ESS, AKM, ROHM, etc can well exceed 100db SINAD. However once you get past the 90db SINAD mark and approach the 100db SINAD mark, you are in true high fidelity land. Beyond that we are in 'space cadet' category where performance metrics start becoming more academic and less practical, especially as we are starting to see SINAD breach the 120db line and steadily marching to the 130db benchmark.
Thankfully the S.MS.L D300 offers many choices of filters, especially if you rely on computer software oversampling to maximize your performance. No, this method of software oversampling is NOT native DSD per se, but it IS the way to get the best performance out of this DAC. I do not hold the ROHM chipset at fault here; I believe the fault is with the S.M.S.L analog implementation. Every other DAC tested here has DSD and PCM performance that match very closely. The S.M.S.L D300 goes down the path of truly exceptional PCM performance, combined with 'merely' decent to very good DSD performance, depending on the oversampling rate and cutoff frequency. I believe the observed phenomenon here is due to the lack of a well implemented secondary RC analog filter, or perhaps none at all. I could be incorrect, and am open to correction, however, something in the analog stage is preventing DSD from reaching its fullest playback potential.
However as was mentioned earlier, there is a notable drop in performance in DSD performance as compared to PCM with the S.M.S.L DAC. SINAD performance with PCM files approaches and exceeds 114db. The best DSD performance, 104db SINAD, is at DSD64 with the 13khz filter, which is quite too low a cutoff for high-fidelity audio. Indeed, the filter is of the very gentle moving average variety; nevertheless by 20khz the rolloff has exceeded 8 decibels! This will cause a very audible treble rolloff. For DSD64 files, surely the 26khz filter and its 96db SINAD is the minimum filter of choice. A far cry from the excellent 114db PCM SINAD, but thankfully it is still good enough that any loss of enjoyment is likely all in your head, because I told you so! If you had never read these measurements, I would be quick to bet if you listened to the D300 you would have heard nothing negative at all. Ah, the power of simple suggestion....
And as we stick to the same 26khz filter with 128x DSD rate (by selecting 13khz DSD filter from the menu), we see that there is a noteworthy performance increase up to 99db SINAD with DSD128 due to the ultrasonic noise rise starting much farther from the in-band frequencies (20hz to 20khz), pushing the noise to a higher level where the filter can more effectively eliminate it. Although DSD64 had the very best SINAD measurement of any at 104db, again the filter rolloff is too low to be considered as high resolution audio. S.M.S.L D300 DSD playback is MUCH improved with the DSD128 and DSD256 formats. I see no need to oversample to the DSD512 rate. The most obvious reason being the poor performance at the two highest filter settings. The next reason? There is no SINAD improvement at all over the DSD256 at 104khz filter (26khz labeled in GUI). It is simply a waste of resources to oversample to DSD512 in this case, since you will not see any performance gains. The filter profiles are the same, meaning you will likely not see any gains at all in the time domain, and the noise/distortion performance in the audible band is identical in both. Here we may be running into a very common issue in DACs and 'ultra high speed' DSD.. the ability of the logic itself to keep up. Faster is NOT always better, and it looks to me that the sweet spot for the S.M.S.L DAC DSD converter is DSD256 at either 52khz or 104khz filter. I personally would go with the 104khz filter, to take advantage of the temporal resolution DSD offers while still having good enough SINAD performance. I can totally understand the thinking that would lead one to choose the 52khz option and get that SINAD up close to 100db. Either way, the DSD sweet spot for the S.M.S.L D300 is DSD256, and choose either filter 1 or filter 2 on the GUI (13khz, 26khz).
As I mentioned before, I prefer 'native' DSD at all sample rates, what one might call 'bitperfect'. However, the S.M.S.L D300 is all over the place DSD performance-wise. I will reiterate what I believe is the source of this 'issue', and that is a lack of a secondary analog RC filter following the DSD FIR conversion filter. Yes, one can achieve 'good enough' performance without a secondary filter, and one can 'tune' in the best performance with multiple DSD filter options available on the D300. However, I think the added expense of a secondary filter (or a better performing one) would have been worth it. Bad call here by the engineers. I am on record calling this little DAC a 'giant killer', and this is still especially true for PCM file playback. Its native DSD playback, in spite of any compromises is still quite good, but as you see the performance of the next couple DACs it will become obvious exactly how good the D300 COULD have been.
Now we move onto iFi, where thankfully we can get the temporal resolution promised by the DSD format.
Unlike the 'dac du jour', Thorsten Loesch and iFi didn't just stick with whatever is currently 'popular', such as the latest ESS or AKM chipset. There is more to overall performance/sound than sub-arctic THD and/or stratospheric SINAD. There is still something to be said about overall architecture, in this particular case the Segment DAC architecture of the Burr-Brown DSD1793 chipset. The DSD1793 uses a 64+ level thermometer/unary code DAC, which allows for the top 6 bits of the 24 bit two's complement PCM to be converted directly via bit-perfect PCM (64 levels), with the remaining 18 bits are converted via a 1 bit Delta Sigma converter. This is a particularly ingenious means of conversion, as it avoids the so called PCM zero-crossing error by keeping the 'lower' 18 bits always at "full volume". Obviously it isn't at full amplitude in the time domain, however in the frequency domain zooming down to the sample level, yes, the 1-bit signal is either full on or full off. The chip will use oversampling/time-splicing for the lower 18 bits of resolution. As has been mentioned, according to Burr-Brown, SINAD is expected to be right at 100db.
Another benefit of this type of conversion is it is ready made for 'native' DSD conversion. The 64 level thermometer/ladder bit switches are all unary coded/equally weighted. The delay line for the DSD conversion is only 8 bits long, however. This allows several different combinations of the 64 switches to create 8 groups of switches that function as the 8 unequally weighted taps. These different combinations can mix and match to change the cutoff frequency and the order (steepness) of the frequency rolloff. All the while the 64 individual switches are equally weighted and allow for dynamic element matching and exceptional linearity.
Note this excellent SINAD consistency continues within the iDSD PRO. At all bitrates and their accompanying cutoff frequency, SINAD measures right at 99db.
In addition to the DSD64, DSD128, and DSD256 rates shown above,
at DSD512 the cutoff frequency is 616khz, and the SINAD stays a consistent 99db.
At DSD1024 the cutoff frequency is an astonishing 1,232khz (1.232 megahertz), and still maintains a SINAD of 99db!!!!
As you can see, the iFi products above have output stages that offer equivalent maximum performance via DSD and PCM. (The DSD1793 chipset, repeated here like a broken record, is limited to around 100db SINAD.) The iFi ZEN and iFi iDSD PRO are not configured to have switchable DSD FIR filters, however, their filter cutoff doubles with each oversampling. For example, DSD64 has a -3db cutoff at 77khz. Anything at DSD64 will always have a cutoff at -77khz. Double the speed to DSD128 however, the cutoff will double to 154khz. And this pattern continues right up to DSD1024 on the iDSD PRO. (Note the DSD1793 does indeed offer switchable cutoffs for each DSD speed rate. In recent years, however, iFi has settled on locking into what appears to be the 77khz FIR filter onboard the 1793 chipset, along with an 80khz Analog RC filter to follow.)
And finally in this particular blog entry we will take a look at the very, very impressive RME ADI-2 PRO FS R BLACK EDITION with the AK4493 chipset (pre-factory fire switched capacitor version.) Just as the iFi with its DSD1793 matches PCM and DSD performance, likewise this much more modern AKM chipset matches its PCM and DSD performance. This MAY be the highest performing true native DSD chipset on the market that is NOT made from expensive discrete parts ala the Signalyst Converter.
Note that the following AKM 4493 based DAC boasts an outstanding 114db measured SINAD with PCM material. The 'native' DSD material comes very close to this level of performance. One of the first things to note, however, is the DSD filter numbers in the GUI are incorrect. They are not 50khz/150khz. This was correct for the previous chipset used by this AD/DA, the AKM4490. The AKM 4493 used in this particular upgraded AD/DA uses DSD64 filters at 39khz/76khz.
What more can be said? Used in DSD Direct Mode (Native DSD), DSD256 using the 'high' filter, in this case at 304khz, while maintaining a 111db SINAD, is truly remarkable performance. This is an astonishingly good choice of DAC for those who use software oversampling. Oversampling in Roon or HQPlayer to DSD256 with the higher of the two filter settings (304khz) with a SINAD over 110db should offer state of the art performance, and then some.
Also, if you are more of a purist, as I am, and wish to have as little DSP touching the DSD stream as possible, The RME ADI-2 PRO FS R BLACK EDITION is also an excellent choice. You can have native rate conversion with every filter save one between 110db and 112db SINAD. The only exception, is the high 76khz filter on DSD64 material. This is not unexpected, as there is a significant amount of ultrasonic noise that escapes the lowpass filter, yet the actual performance penalty is very, very small, as the SINAD is still a high 107db!!
The RME team have built an exceptional native DSD DAC. On an aside and maybe for a future entry, I will publish the RME results with DSD in Volume Control Mode and remodulation to AKM's multi-bit Delta Sigma format. The performance here, even with DSD, is also quite outstanding. The RME continues to impress. Full review coming sometime this year.
MORE TO COME SOON, INCLUDING THE FULL REVIEW OF THE iFi IDSD PRO (part two of the review; part one reviewing the ifi iDSD NEO is currently posted. )
Also a full review of the Signalyst PURE DSD DAC is coming soon! After much listening and mixing/matching amps and headphones, the Signalyst Converter may indeed produce the very best sound I have ever heard.
The S.M.S.L D300 I reviewed was loaded with firmware 1.0. For a time, a short time, a firmware update to 1.1 was uploaded that enabled DoP (for DSD playback for those who do not have ASIO, such as on Macintosh.)
After a late night discussion with a fellow enthusiast, and many measurements, followed by more research, it looks like the S.M.S.L firmware update has been a real nightmare for some people, so bad I believe it has degraded the performance of this DAC overall. Some have even 'bricked' their D300.
Version 1.0? Yes, a GIANT KILLER DAC and a real gift to those who love native DSD conversion via analog filter. I stand by my review and findings. However, it sounds like newer firmware created a nightmare. That firmware was periodically available from the S.M.S.L website. Yes, it came and went, came back and left again. As of now, the firmware update is NOT available on the company website. It CAN be found if you look to the forums, but my advice? Stay away! If you have a version 1.0? Keep it and treasure it.
If you want to buy one? Make sure it comes with firmware 1.0! New ones did indeed ship with the updated firmware that, again, is no longer even available from the company. Bad, bad situation, and I am watching closely to see if customer service and engineering can work this out.
linked below are the user experiences. This is linked to the final page of thread, but if you want a fuller picture, go back about 10 pages and start there.
In my previous post, linked here, I noted that I was using a generic, very cheap variable Switch Mode Power Supply to power the DSC2 Converter temporarily. The supply is meant to help evaluate headphone amplifiers and their ability to power low impedance headphones, as it has up to a 5amp output at the lowest voltages. It is NOT by any means made for critical listening, and the XPower proves that. The changes in noise floor and digital 'hash' are clear when the XPower takes over duties from the generic switch mode power supply.
Now, the exact benchmark being sought at the time (jitter, IMD, SINAD, etc.) doesn't necessarily change, although the measured results CLEARLY show improvement when using the XPower. I have seen similar results from other clean power devices, especially PSU's and USB galvanic isolators. However....
Many an absolute objectivist, especially the most vocal, will point and say 'haha!', or 'gotcha!' because the jitter/IMD/SINAD etc. distortion measured result is the same. I find this to be extremely hypocritical and only self-serving. When faced with real evidence of a difference in the measurements, it is ignored and re-directed into excuses such as, 'you would never hear the difference anyway', or as mentioned before, 'the measurement you were seeking didn't change.' Ironically, they then sound exactly like the 'opposite' camp, who will point out that past a certain point, things like higher SINAD, or lower THD are meaningless because the differences are inaudible. These particular absolute objectivists seem to want their cake and eat it too, but they just can't have it both ways. One cannot laud "product Y" for its 125db SINAD, proclaiming it king of the hill, while at the same time say the obvious changes visible from "product X" are meaningless because they are 'obviously' inaudible.
The fact remains that something major DID change in the measurements, even if it is not the metric or metrics for which one searched. Perhaps what we see IS audible or contributes in some way that when combined with other metrics is audible. For all we think we know about audio reproduction, there are any number of things we don't fully know or understand. Why? Because ultimately what comes out of any audio product is judged inside an extremely complex neural network and bio-sensory system that we are only beginning to understand. If all we do is measure all day and don't listen, thinking our measurements tell us everything we need to know, this hobby is rubbish. Just as much rubbish as a half-million dollar RCA interconnect. (Yes, I find the truth to be likely found somewhere more in the middle. The fundamentalist absolutist objectivist 'know it all', and the 'used car salesman' in the store who calls amps 'class A' because he reads the Stereophile scriptures, not knowing that Class A actually IS a type of amp topology, are equally off-putting.)
Unfortunately the scientific method has been misused time and time again by closed minds to justify their 'position', or their 'ideology'. It is too bad there are not more researchers just interested in the truth, who will follow the data wherever it may lead, and entertain hypotheses in humility rather than dismiss them in arrogance, perhaps holding back and delaying progress.
Now getting off my soapbox, back to the changes the iFi xPower made in the DSC2 measurements. I will choose to focus on two in this blog entry. Intermodulation Distortion and Jitter.
With regards to the specific measurements being sought, the specific data didn't change. But many other things did. The noise floor, once full of spurious tones and quite frankly, a bit of a mess, cleaned up dramatically. And yes, listening tests performed BEFORE the measurements had already revealed a much more natural sound, with less sibilance and harshness than before.
We will start with IMD, 19khz+20khz. The first measurement is with the Generic PSU, the second is with the iFi XPower.
DIY OPEN SOURCE BUILD WIPES FLOOR WITH MORE EXPENSIVE COMMERCIAL DSD DACS
I received a coveted not easy to find product recently. It is a Signalyst (HQPlayer) true DSD DAC. Click here for description. Or, perhaps Converter is simply the best thing to call it, since, it really has no DAC in it, which is one of the points in DSD's favor; simplicity of conversion. IF one choses to make it simple. That tends NOT to be the case with many DSD DAC's, with their digital volume controls, so called 'DSD-Domain' processing, re-modulation, etc.
I recently made a blog post on how pure DSD works. (Click here to read the post). The purest DSD conversion is still something of a holy grail, involving only an analog filter, such as a RC filter, and never having more than the 1 bit bitstream. The more realistic, and better performing pure DSD conversion still uses just an analog filter, but it's a FIR filter, which is more complex than the RC and other analog alternatives. But a FIR filter, which is usually implemented via digital means, can be done 100 percent in the analog realm, either on silicon or via discrete analog components. The beauty of the Signalyst DSC DAC is it is made of all discrete analog components once you are past the USB input stage.
The DSC DAC is the hardware version of what we call 'open source' software, so its design is open to some tinkering and modifications. My version is known as the DSC2, and is a fully differential balanced output that uses output transformers for final analog stage. Output transformers are a bit controversial in modern solid state equipment, because they have higher harmonic distortion than many find unacceptable, nor are they perfectly linear. The worst distortion is found at the low frequencies, and it lessens as frequency increases. The output transformer will impart something of a tube like sound. However, this does not have to be the analog output stage for this DAC. Going back to the 'open source' nature, it could be anything of the designer's choosing.
Anyone read the current controversy regarding the uber expensive PSAudio Directstream DSD DAC MK2? And its 'high' noise levels and 'high' distortion levels? Well, most of its measurement issues center around the use of output transformers.
I am making NO statement whatsoever on the efficacy of the PSAudio implementation or its actual sound. I have never heard it. What I DO know, is my DSC2 Converter, which costs a tenth as much (that is, if you can find one or the parts to build it. It is unfortunately NOT a commercial product.) significantly outperforms the Directstream in every measurement of the DSC2 I have made, while only needing a simple software such as Roon to listen to PCM files, and needing NO conversion to listen to native DSD files. (HQPlayer, though, is recommended as it and the DSC2 converter are designed to work in tandem.) It is bitperfect with DSD, something the Directstream cannot claim.
The 'worst' measurements will be the THD+N, which inverted is called SINAD. Anything over 90db SINAD is certainly decent enough performance, and 100db SINAD is probably more than good enough for most ears. All of this, however, is the consequence of the use of output transformers at the output stage in the DSC2. I can gladly report, though, that the transformers used are quite linear, for transformers. It stays above 90db SINAD from approx. 50hz to over 15khz. Not only that, the NOISE floor is well below -120db at virtually all frequencies. The low level amplitude linearity is ASTONISHINGLY good for a 1 bit DAC. And jitter? Not even a factor. Less than 20 picoseconds. The worst thing the the lower SINAD will do is impart a sweet, tube like sound, which is exactly what it does, but NEVER at the expense of treble extension, detail or transient response. It shifts a little to the warm side of neutral, and as is the case with many a piece of kit, will need careful equipment matching. Yet overall its sound is both state of the art digital and beguilingly 'analog like' at the same time. Its a beautiful sound that you will just want to listen to all night long, and not feel guilty about the internet mob mocking the noise floor, low linearity, or jitter while you fall asleep, because there is nothing there for the absolutist zealots to mock.
Yet, I am still hesitant to post too many measurements, by experience I have become acutely aware of how important clean power is. I am currently running the DSC2 on a variable output Switch Mode Power Supply that is of unknown noise quality. I will be more comfortable when the XPower 9 volt arrives. Apparently it was out of stock on Amazon and is delayed to a Saturday delivery. More refined measurements and graphs to come.
UPDATE 7/2/2023- I now have a 9v iFi Xpower SMPS powering the DSC2 DAC. It did exactly as expected, lowered the 'hash' and anomalous 'stuff' in the noise floor. See the new graphs below.
The resolution of the DSC2 is exceptional. Both the noise floor, and the linearity. (Both of these are required for true high resolution. Often we focus on the excellent in-band noise floor of DSD, but ignore the low level resolution that is exposed in a linearity test)
No worries here. As I said, the linearity of the DSC2 is truly first class.
All these measurements are at DSD256.
At -100db, linearity is virtually perfect. deviations are LESS than 0.1db. Much less.
At -110db, we are STILL at <0.1db deviation. This beats out some DACs and come close to others made by that other company with the letters DSC in it, lol.
At -115db, we finally hit some loss of resolution, with a deviation of an 'entire' lol -0.5db. Wow.
-119db we finally have a deviation of -2.5db.
So with my measurements we have practically perfect 18 bits of resolution, still very, very linear to 19 bits of resolution. Linearity doesn't quite make it to a 'perfect' 20 bits, but for a 1 bit converter this is exceptional. Heck, for any converter it is impressive.
Couple that with the actual noise floor that is lower than -120db, I cannot help but be impressed by the measured performance here.
SINAD (THD+N) in my latest, better calibrated measurements stays in the mid 90db range, approaching 100db at times.
Even deep into the low frequencies, it only drops once you get under 50hz, and more like 30hz. Our ears are not as sensitive to the distortion that low anyway. Once again, I am thinking of a $8000 DSD transformer DAC that does not come close to this level of performance.
And on jitter... my final measurements at DSD256, with a 48khz base rate, (48k x 256 or 12,288,000hz), I got a low result of a mere 18 picosecond jitter.
I will remeasure everything once I get the XPower 9V power supply. I am not sure the numbers will change much; I do expect the FFT noise floor to be cleaner however, in my graphs.
As with the Topping E70 ESS chipset vs E70V AKM chipset comparo, the iFi comparo will have to come out to you in multiple parts. There is just too much info for 1 review post. Be on lookout for the official iFi NEO DAC/Headphone Amplifier this week. I have gathered lots of data and am in the process of sorting what data makes it and what gets left behind... At this point all I will say is, the iFi series of kit is truly amazing stuff. Lots of overlap in their different categories as well, meaning that deciding exactly what to buy isn't just a cut and dry affair. Hopefully the data Euphonic Review has gathered can help anyone still on the fence sort it out.
DSD can be like TNT in the audiophile world. Even as accepted as it is these days, it still evokes controversy in a lot of circles. Less so now than in the 'Great DSD Wars' of the previous decade. But there indeed was a time when having a strong opinion one way or the other about DSD was a path to anathema from someone or some group. No one was pleased. In those days and in the days since, I have studied and studied the pros, cons, ins and outs of Direct Stream Digital. (1 bit PDM, or is that 1bit PCM? Depends on who you talk to I suppose). I have learned enough to be very dangerous. Which means I have crossed the point where I know how truly complex this subject is, and I know I am nowhere near an expert. At the same time, there are a few things that I do know fairly well, with decent confidence. So let us start there as we continue to stir up a little controversy in the name of FUN here at Euphonic Review.
How do you like your DSD?
Personally, I will take mine rare, as straight from the cow as possible, presuming the germs are all killed. I prefer "NATIVE DSD", but what does that even mean?? It seems every product maker has their definition, usually all in the name of selling products over accuracy. If you poll 1000 audiophiles, you will probably get 10,000 different answers. Not even the actual experts can agree, either.
What is MY definition?
Native DSD (not counting the production side, that is another can of worms) is when a DAC maintains the native, bit-perfect 1-bit signal, and the only conversion occurs at the end of the chain via a filter that leaves behind the music and cuts out the noise.
This can be divided into two camps, but only one of them seems to be realistically achievable, although it hasn't stopped many from trying. The 'purest' form I speak of, which is still out there like the Holy Grail waiting to be found, looks something like this (poorly put together) graph.
This extremely minimalistic design has only one bitswitch, which is usually the DSD output of a USB audio processor. Amanero Asynchronous USB input card is a common choice when attempting to build one of these 'dac-less' DACs. This 1 bit 'logic' output, can be treated the same as the unfiltered analog DSD square wave. It is of low signal level, but can be filtered by a RC analog filter to remove the carrier frequency and some of the noise-shaping. A good result will be more achievable with a very high DSD sample rate due to presumably less ultrasonic noise, but as of 2023, I know of no one who has been able to make this work well enough to produce a commercial example. I believe Lampizator came up with a version than used a very particular style of antiquated FM radio to do something similar, but it was too noisy to be viable.
The next 'camp' of Native DSD conversion is ubiquitous in modern Digital to Analog Conversion. I am sure earlier versions of this exist, but the Burr-Brown DSD1700 chip is what I think of as ushering in the SACD age.
This is what you will find in Native DSD products by iFi. It is the technique used by ROHM chips in DACs such as the S.MS.L D300. AKM and Cirrus/Wolfson chips use a similar technique in their Native DSD 'Bypass' Modes. It is also how the Signalyst DSC1 DAC works.
NO, it is not what ESS uses. ESS in this case reminds me of that old Sesame Street song, 'One of these things is not like the other'. More on ESS later. Back to our first two Native DSD 'camps'.
How are they different from one another? In principle, not very. In practice, however, yes, there is a difference, and whether or not this is 1-bit or multi-bit conversion is probably a fruitless argument, because in this configuration, the 'multi-bit' properties exist post coefficients, and are fully in the analog domain of this 'hybrid digital/analog' Fixed Impulse Response (FIR) filter. The output of the filter isn't a digital word, rather, it is an electrical current, whether that be from the switches/coefficients or at/after the accumulator. Again, we are not in the 'standard digital' realm.
FIR filter. There you have it. That is the difference. Rather than a single bit switch and an RC filter, the 1 bit DSD goes through what is a digital FIR filter, however it is implemented in the analog domain, either partially or fully. This FIR filter IS the Digital to Analog conversion. Pretty cool huh? Filter and converter all in one step. Here we still have the simplicity of no DSP, a 1-bit signal all the way until the very end when it reaches the Fixed Impulse Response 'Moving Average' filter. The 1-bit signal moves though a delay line, and the taps/coefficients ARE the bitswitches themselves, their outputs being summed up as the analog signal output that goes to your ears. (Note many of these type DACs also have a secondary RC filter to help deal with the ultrasonic noise).
Note, the graph is crude, but the basic concept of how it works is here, so don't go too hard on me...
So what is next?? What about ESS? Well, ESS is super quiet about DSD. They protect their proprietary info. It is VERY closely guarded, kind-of like a Fort Knox. And that is not a terrible exaggeration. All I can say, it is NOT QUITE like what you read above in the 'analog' FIR filter example. But we can make some educated guesses based on measured behavior, and the experiences of those who work with the chip in their kit.
What about AKM in non Native DSD mode? Cirrus? Well, they all use some variations on the same theme. 1 bit DSD is internally converted to a multi-bit signal. Sometimes it may be decimated to PCM, sometimes it may maintain the same sample rate. It is VERY common to convert 1-bit DSD into a multi-bit 'DSD derivative' via a digital filter.
However, unlike the FIR filter described above, this Filter is in 100 percent digital domain. These fully digital DSD filters are excellent and precise at filtering out the unwanted ultrasonic noise, and due to their multi-bit nature, and being in fully digital domain, other DSP is easily applied in this state. I.E. Sample Rate Conversion, Volume Control, further filtering such as IIR etc. One thing that MUST occur, though, is a re-modulation to some form of Delta-Sigma before the output. ESS would call this their Hyperstream DAC. DCS has a 5 bit (32 levels?) Delta Sigma modulator before its "Ring DAC", which is simply their take on a thermometer/unary code converter using scramble code/dynamic element matching. AKM has various types of Delta Sigma modulators in their various chips. Same for Cirrus/Wolfson and others.
I personally am not as big a fan of this because of all the DSP. Yes, it is true that PCM often goes through lots of DSP, too. But I am personally consistent in not liking that either. I like my PCM at high resolution with Non-Oversampling. Also, simplicity of conversion is one of the 'promises' of DSD. When you bring lots of DSP into the equation, it isn't so simple anymore. Transient response can suffer. The original 1-bit input and whatever is at the output are very, very different. This CAN be a good thing, though, as advanced Delta-Sigma modulators like the ESS Hypersteam can and often do convert DSD with a lower noise floor, lower distortion, and better linearity.
As I always say, in the end, let the ear be your guide!!
For a visual example of these types of DACs, I have gone old school, and will take a stab at showing how Sony converts 1 bit DSD into their 8 bit PCM 'DSD-WIDE'.
I hope you found this trip through DSD land edifying. Thank you as always for reading Euphonic Review!
Until next time,