Suggest Hifi rose RS151 add upsample to DSD2048 features

Function
#43

The difference in the capability of the human ear to hear frequency response 20-20K and timing differences is enormous. This is often conflated to the detriment of arguments about equipment capabilities.
Frequency Response vs. Timing Differences (ITD)

  • Frequency Domain (20 Hz - 20 kHz): Sensitivity peaks around 3.5-4 kHz due to ear canal resonance, meaning sounds here seem louder. Human hearing is not flat; sensitivity drops off significantly at high and low frequencies.
    Timing Domain (0-700+ ):** Interaural time differences (ITD) are based on the time it takes sound to travel between ears. A max ITD is roughly 600-700ms (the time to travel around the head), though echoes are detected over longer delays (30-50 ms).

Interaction:** While ITD detection is generally thought of as independent of frequency, studies show that ITD thresholds are lowest (most sensitive) around 800-1000 Hz, rather than low frequencies.

Neural Tuning:** Midbrain neurons are tuned to specific frequencies and corresponding time delays, matching the acoustic data of the environment.

PubMed Central (PMC) (.gov) +5
‘measurements’ when it comes to audio, are related to the Frequency Response (pitch) and not timing. A visual equivalent might be Audio is Color Spectrum and timing is “Frames Per Second”.

Maybe all the in-fighting over the topic is this misunderstanding. On the one side you have the equivalent of frequency response people focusing on the ‘color reproduction’ saying “You can’t even see Infrared light!” or “If you adjust the color, then the two pictures are exactly the same”. But then team “timing” is talking about resolution and motion fidelity, not necessarily color reproduction.

For example taken from Reddit. How do we determine the location of sounds? The difference in timing between when audio reaches the left and right ears. It can be as low as 10 microseconds according to this article:

It takes some really sensitive audio equipment to resolve 10 microseconds in timing differences.

StandardModel

1 Like
#44

You claim both that “correct” upsampling can only be done with some secret proprietary algorithms on super high powered hardware and in the same breath ask that Rose do it on a cell phone. Sure, audiophiles aren’t big on internal consistency and logical thinking, but this is quite ridiculous.

Yes, a lot of audiophile snobs are terrified of the idea that a mass-market product can be (and is :slight_smile: ) as good as the expensive ju-ju that they bought because they have been brough up to be good little consumers and believe everything marketing tells them.

Ain’t happening though. If you had any idea what you are talking about you’d be asking Rose to add NAA support, which is at least possible (EverSolo did do that).

They look boring; when I upgrade my Roses it will be Lyngdorf.

Nope. Even if we wdre to believe that there’s some unknown thing about hearing that is affected by something that does not show on measurements (and there isn’t much indication that this is true, but let’s assume for a moment that it were) DBT ABX is the only way to distinguish between something that does mke any difference and something that is purely snake oil.

Which is why neither cable manufacturers, nor HQPlayer developer, not “high-end” DAC designers ever agree to perform any.

Except that I am not, this is your personal straw man.

Good. Alas, this isn’t something that would be affected by what you are advocating.

Which, incidentally, are not available on any Rose device, so your initial premise is as nonsensical as everything that you have written afterwards.

Yawn. Good for it. Absent any proof that it results in anything audible it’s as useful as your cable lifters.

Either they are or they are not indistinguishable. Actual test shows that they are, whether you like it or not.

Studios use Merging etc. for many reasons, but not for the ones you think.

I’m sorry that you you still do not want to understand that (would make you question your life choices; can’t have that) but it does prove absolutely conclusively that even the hobbyist gear is far more transparent, even without any external clock DSD upsampling quantum dot enhanced ju-ju than anything that you could possibly hear. Period.

False equivalency because the former does result in real changes in the sound (if done correctly for the better) while the latter has as much audible effect as a fuse.

The only person who keeps bringing up that chart here is you. Which is to be expected – when you do not have any real argument you have to fight straw men.

Alas, no. You prefer to believe marketing, which makes you a very good consumer, and ignore that you do not hear any nonexistent difference. But you won’t acknowledge because that would mean acknowledging that you’ve been had.

Of course, I would have it either way.

So, how many cable lifters do you have?

#45

In addition to the temporal difference you mentioned, our brain uses two other fundamental tricks to “map” space.

#46

Your fixation on Rose’s current CPU limitations misses the point of a firmware request. The request is about enabling the device to pass-through or handle advanced modulations (like NAA support you mentioned, or raw DSD output) so that the DAC—which is capable—can perform at its peak. It’s about opening the pipeline. To suggest that because a device can’t do ‘everything’ now means it should never evolve is the definition of stagnant engineering. If manufacturers followed your logic, we’d still be stuck with the first generation of limited-buffer streaming chips.

You keep retreating to the ‘cable lifter’ and ‘fuse’ straw man because you cannot address the actual Mathematics of Delta-Sigma Modulation . There is a verifiable, measurable difference in the Impulse Response between a standard 8x tap filter on an ESS chip and a million-tap polyphase filter used in DSD2048 upsampling. One results in significant pre-ringing (time-domain error), the other approaches a near-perfect sinc function. This isn’t ‘marketing’; it’s Signal Processing 101 . If you can’t distinguish between a passive plastic stand and a GFLOPS-heavy mathematical algorithm, you are the one ignoring 'logical thinking.

Your claim that AD/DA cycles are ‘indistinguishable’ is precisely why the gap between hobbyists and professionals exists. If transparency were a solved problem at the $20 chip level, the AES (Audio Engineering Society) wouldn’t still be publishing papers on jitter-induced sidebands and reconstruction artifacts. Studios use Merging and Lavry because conversion errors are cumulative . Just because a ‘cheap’ system is ‘transparent enough’ for your ears doesn’t mean it satisfies the requirements of high-fidelity reconstruction where every microsecond of Inter-aural Time Difference (ITD) matters for soundstage depth.
Demanding a Double-Blind Test (DBT) is the last refuge of someone who wants to dismiss any nuance they haven’t personally experienced. While DBT is a valid tool, it is not the only tool in engineering. We measure phase shifts, ringing, and settling time because they are real physical phenomena. I trust the engineering that seeks to minimize these artifacts. You prefer to wait for ‘proof’ that fits your personal threshold of audibility—that’s fine, but don’t confuse your personal ceiling with the technical ceiling of the industry.

It’s clear you’re happy with ‘good enough’ transparency, and that’s a valid way to enjoy music. But for those of us pushing for DSD2048 and advanced reconstruction, it’s about the pursuit of the 1% —the realism that happens when you move digital artifacts so far out of the band that the music finally ‘breathes.’ Enjoy your Lyngdorf when you get it; I’ll keep pushing Rose to innovate rather than settle for the status quo.

Cheers Mate :slight_smile:

#47

ML, you have come to the wrong place, though with you being an owner of a RS151 I understand why you did. Rose does not innovate, it markets and remediates. Everyone on this forum who has purchased a Rose device and has offered suggestions for improvement or pointed out bugs in the system is a beta tester. They listen to us and our concerns, they prioritize them and remediate the low-hanging fruit. Hi-Fi Rose is a work in progress, an investment made by the parent company to expand the market for its screen technology to an audiophile market where purchase price has less meaning than almost any other industry.

My jaded perspective is that if Rose can implement your DSD2048 at minimal cost and is confident marketing can extract a premium profit margin for doing so, it could find its way into a future product. But in the here and now I see them struggling to make a reliable product that just lives up to its marketing.

1 Like
#48

Your request is, literally, “Hifi rose RS151 add upsample to DSD2048 features” but it’s not like you’ve left any doubt that you have absolutely no idea what you are talking about.

You really should stick to how much your quantum dots improved the sound, and you swear you’ve heard the veils lifted…

And as soon as you, or Juicy, or Rob “Mosquito fart” Watts present some evidence that it matters, we can discuss the particulars. But it does not.

Logical thinking isn’t really something you should be bringing up.

I don’t know what your “profession” is but it is definitely not the one where any use of a brain is required. Real professionals use things that work. That’s why you do not see SFPs, I2S, and other ju-ju in studio gear, but do see it as selling points on something like Rose.

Which it happens to be.

And that would be applicable to home playback how exactly? Sure, audiophiles do lots of idiotic things (our ambulance chaser with Alzheimer’s here runs output of his RD-160 through a MiniDSP and thinks that he’s listening to some magical properties of Rose’s DAC…) but even they do not run dozens of AD/DA conversions. Usually. And of course even those aren’t audible.

This is just a load of bull-crap that one would expect from Uncle Bull McShitter in one of his fireside chats… because (after dismissing that it’s just fancy sounding words thrown together with no underlying meaning) we’ve already proven that even those 40 conversion cycles aren’t enough to cause any timing errors that humans can gear. On a single conversion your DSD2048 DAC will not perform any better than a $10 dongle. In reality it will perform worse because it was made by charlatans.

Yet another example of you having no clue. Seriously, was your degree in intersectional grievances or in basket weaving? DBT is the only known way of determining whether something is audible (in this case we’re talking about audible phenomena) or not, regardless of any single person’s preconceived notions. And that shows that unless you see the price tag, no, you do not hear any of the alleged advantages that you are claiming.

But if you have any money left over from your DAC, there’s a Brooklyn Bridge I could sell you. There’s a great story about different filters in it…

So is the skin effect (one of the cable makers’ selling point), Casimir effect, and virtual particles popping out of nothingness in vacuum. So is the price of tea in China. All of them have about as much to do with the sound reproduction quality as what you are asking for.

If you mean pursuit of pushing owners of Chord or dCS into the 1% club, then maybe, although even that is unlikely, sales volume ain’t there, even though margins are quite astronomical.

If you are talking about the quality of your system, then definitely not; it is guaranteed to sound worse than what anyone with some basic knowledge about audio could build with Wiim gear.

At least Lyngdorf does innovate. If you are looking for that, you’re barking up the wrong tree (well, that’s what you do all the time anyway, isn’t it?) Rose makes some nice-looking gear, but innovation isn’t part of their corporate strategy, it’s just an underperforming Novatron clone that’s still in business.

#49

ELEVEN ,

Thank you for sharing your perspective. I appreciate the honest take on the current state of Hi-Fi Rose. As a new owner, I’m still navigating the balance between their impressive hardware and the software experience you described. It’s helpful to have this context as I manage my expectations for future updates like DSD2048.

Cheers mate :slight_smile:

#50

It’s clear you have strong opinions on audio engineering, but personal insults about my profession or intelligence don’t strengthen your technical arguments—they just undermine the conversation.

My request for DSD2048 is about exploring the technical capabilities of the hardware I purchased, which is a standard part of this hobby. While you may believe a $10 dongle suffices, many in the audiophile community enjoy pushing the boundaries of upsampling and filter performance. We can disagree on the audible benefits without resorting to toxicity. Since this has devolved into personal attacks rather than a constructive discussion on audio, I’ll leave you to your views.

Cheers mates :slight_smile:

#51

THe conversation would work much better if you were arguing in good faith to begin with. Having some idea about what you are arguing for would help, too. Instead, apart from not even knowing what it is that you are asking for, you deflect everything to regurgitation of marketing materials totally irrelevant to the point you pretend to argue.

If some “accuracy of spatial reconstruction” or whatever nonexistent thing you claim is improved by DSD2048 upsampling to the point of being audible (“it sounds better”) it would show up in DBT just fine. When you claim, in the same breath as you claim that there is an improvement, but DBT does not matter because something, impukse response, something, technobabble, technobabble, either you are being malicious (do you work for iFooFoo’s marketing department?) or, sorry to say, you have absolutely no understanding of what you are arguing about.

Hardware you purchased (a cell phone with decent DAC and a big screen) has no capabilities for upsampling to DSD2048 that you’ve requested. Might as well ask Rose to add a 1000Wpc power amplifier via a firmware upgrade.

They also enjoy pushing the envelope with directional cables, isolation platforms, CD demagnetizers, and green markers. With identical pseudo-scientific explanations and identical claims of incredible improvement in sound. Of course, only if you have a “resolving system” and golden ears (i.e. are of a Stereophile reviewer’s age, haven’t heard any treble in a few decades, and have a tube amp that clips at 2 Watts).

It’s quite telling that at least one of the existing DSD2048 DACs, iFooFoo’s one, has tubes, completely blowing out of the water any claim that DSD2048 is about any kind of more accurate reproduction.

Just like everything audiophiles do, all of this has nothing to do with enjoying yhe music, and everything with boating rights – look at me! My cables cost more than your system! I upsample to DSD16384! Look at me!

You could have tried arguing the points instead of deflecting and beating up whatever straw men you’ve imagined for yourself, but “waah, you don’t believe me when I say the Earth is flat! I’ll take my toys and go home!” is ages old common audiophile behavior and was to be expected.

#52

Wow ! It is clear this you have moved away from technical debate into personal attacks and “bad faith” accusations to avoid addressing the actual signal processing science. Your logical fallacies (specifically obsession with tubes/cables as a distraction) and double down on the mathematical reality of reconstruction filters.

It’s ironic that you accuse me of ‘technobabble’ while repeatedly using ‘tubes’ and ‘cables’ as a shield to avoid discussing Digital Signal Processing (DSP). Let’s separate the engineering from your personal frustrations:

  1. On Hardware Capabilities: You keep comparing a firmware request for high-rate support to adding a ‘1000W amplifier.’ This is a fundamental misunderstanding. Supporting DSD2048 or NAA (Network Audio Adapter) protocols is about data throughput and handshaking, allowing the Rose to act as a high-performance bridge to an external DAC that handles the heavy lifting. It’s about removing bottlenecks, not defying the laws of physics.
  2. The Reconstruction Reality: You dismiss impulse response as ‘technobabble,’ but it is the literal foundation of digital audio. A standard 8x interpolation filter in a mobile-grade DAC chip creates aliasing artifacts and time-domain ringing that are mathematically eliminated by the high-tap, high-order modulators used in DSD2048 upsampling. Whether you value that level of reconstruction accuracy is a personal choice, but calling the math ‘nonexistent’ doesn’t make it so.
  3. The ‘Tube’ Red Herring: Bringing up iFi or other manufacturers using tubes is a classic red herring. The benefit of DSD2048 happens in the digital domain—optimizing the conversion process before it ever hits an analog stage (tube or solid-state). Improving the precision of the digital-to-analog reconstruction is a win regardless of the flavor of the output stage.
  4. DBT and Engineering: If we only built what was ‘proven’ in a 5-minute ABX test, the industry would never have moved past the CD Redbook standard. Innovation is driven by pushing the measurable limits of jitter, phase noise, and filter pre-ringing. I prefer to support manufacturers who push those envelopes rather than those who hide behind ‘good enough.’

You seem more interested in winning an argument about ‘audiophile snobs’ than discussing the evolution of digital filters. I’m interested in the technical ceiling of my system. Since we clearly have different priorities—one seeking the status quo and the other seeking the state-of-the-art—I’m happy to leave the conversation here. Enjoy your Lyngdorf; I’ll keep advocating for better processing.

Cheers mate :slight_smile:

#53

… and then he gets surprised at being treated like a marketing shill clown…

Your request, right here in the thread title, is to add (completely useless) DSD2048 upsampling. Which is as feasible as adding a power amp. It’s not even obvious that adding a DSD2048 passthrough is doable.

DSD2048 and NAA are quite orthogonal concepts, but it is quite obvious by now that you don’t know what either one of them is anyway.

Blah blah blah. You keep repeating some marketing FUD that Josie whispered in your ear completely ignoring the easily tested (and it has been tested) sad reality that it makes no difference. Even after multiple cycles.

And you do it every time. Every time you hear “audibility,” to which you have nothing to counteract, you bring up some meaningless word salad of “time-domain ringing” or whatever. Then you somehow manage to feel insulted when it is pointed out that this is no different from Analysis+'s claims of skin effect in cables having audible impact.

Uh huh. And of course that mythical benefit of DSD2048, well below the threshold of audibility, will magically shine through the analog stage which, especially in case of a tube iFooFoo adds audible distortion. Yup. That’s it :rofl:

Proper DBT takes much more than 5 minutes (not that you would know, since audiophiles avoid them like a plague), and I am sorry to inform you, but apart from great advances in DSP for useful purposes, like DRC, and whatever dCS and their ilk did 20+ years ago (back when they were doing engineering instead of being a marketing-only company) for DAC design, which is now mass prodiuced by ESS/AKM/ROhm etc. etc., there haven’t been anything meaningful on that front in decades.

Let’s fix that. You prefer to support manufacturers who sell you inaudible BS because it makes you feel all warm, fuzzy, and superior to plebeians who do not have your expensive shiny toy. Never mind that their systems sound better. That’s no different from the less intelligent subset of ASR readers who upgrade to each new Topping device because it has 1dB better SINAD. Even when it sounds exactly the same as the one they already have.

You just keep repeating that technobabble about " limits of jitter, phase noise, and filter pre-ringing" even when past some point (that has been reached) these improvements make no difference.

I am interested in audible improvements to the music I listen to. Which Rose could have easily done for future devices with an addition of a $20 extra DAC (does not even need to be super fancy one) for low frequency output, and support for loading convolution filters, which could be easily handled by all existing devices.

Then you should change your handle to “Snake Oil Lover” because useless “improvements” that do nothing for the actual sound quality, even if they are free, is the definition of snake oil.

I want Rose (not that they had shown any interest in that) to get to even Wiim’s level of functionality. And, yes, sound quality in a real setup.

And you are looking for meaningless improvements that only increase few vendor’s bank accounts and do nothing to improve the sound quality.

Lyndorf already does better processing.

Mind finally telling us who are you shilling for? Is it HQPlayer? Or iFooFoo? Maybe Gustard?

#54

this conversation has devolved from a technical discussion into a collection of personal attacks and logical fallacies. To address your misconceptions one last time:

  1. Protocols vs. Amplification: Comparing a firmware request for DSD2048 pass-through or NAA support (which is a data-handling protocol) to ‘adding a 1000W power amp’ is a massive technical category error. One is about software handshaking and data throughput—which modern network streamers are designed for—the other is a physical impossibility. Suggesting that a high-end streamer shouldn’t evolve to support higher-rate data pipelines is simply an argument for stagnation.
  2. Filter Topology is not ‘FUD’: You dismiss Impulse Response and Time-Domain accuracy as ‘technobabble,’ yet these are the literal metrics by which DAC performance is judged in the industry (AES/IEEE standards). The fact that you equate measurable digital filter ringing with ‘cable skin effect’ shows a fundamental refusal to distinguish between active signal processing and passive accessories.
  3. The ‘Audibility’ Shield: Relying on the ‘threshold of audibility’ as a universal stop-gap is a common way to dismiss the incremental gains that define the high-end audio industry. If we followed your ‘mass-produced ASIC is enough’ logic, the entire pursuit of high-fidelity beyond a basic dongle would be a ‘Veblen’ exercise. Some of us choose to pursue the technical ceiling where spatial cues and transparency are optimized by offloading the heavy lifting to superior external modulators.
  4. Innovation vs. Stagnation: You’re happy with the status quo and ‘audible-only’ improvements like room correction. Those are valid, but they aren’t mutually exclusive with reconstruction accuracy. A true high-fidelity system addresses both.

I’m not ‘shilling’ for anyone—I am a consumer advocating for feature parity and technical excellence in a product I own. If you’re content with your current setup and ‘boring’ functionality, that’s your prerogative. I prefer to push for the limits of what the hardware can facilitate.

I am always open to a rational, evidence-based technical debate, but if your goal is simply to seek a sense of accomplishment through insults and labeling others, then that is truly a pity. We clearly have irreconcilable views on what constitutes ‘progress’ in audio. I’ve stated the engineering case; you’ve stated your skepticism. I’ll leave it at that. Enjoy your music

Cheers Mate :slight_smile:

#55

Then maybe you should have had some clue what you were asking for. Which was “DSD2048 upsampling” but that would require actually knowing what it is.

Especially when there is no benefit in it, sure :clown_face:

And you keep ignoring the reality of it being, past a certain point, completely inaudible and therefore irrelevant. Which tells a lot about you arguing in bad faith. And having no idea what you argue about in the first place.

Skin effect is perfectly measurable, too. Only inaudible, just like what you are asking for.

“Incremental gains” that bring no audible benefit are nothing but snake oil. That you refuse to ascknowledge it only exposes you as a shill for snake oil manufacturers. Whether they make cables or upsampling software isn’t particularly important.

When your “true high-fidelity” system sounds indistinguishable from a $300 mass-market product it is not an improvement, only a desperate money grab.

Shouldn’t you be asking for something that would actually benefit consumers? Like software that works, support for streaming services people want to use, DRC that would bring real SQ improvements? But somehow you only keep asking for something that would do nothing to improve the experience. Most likely because you’ve paid too much money for a “DSD2048” DAC and found out that it does not work well with any existing recordings. But instead of just getting a decent DAC (hint, it would not come from BS Audio or iFooFoo; actually your 151 would do a perfectly good job of it, within limits of lacking sub integration and DRC, something that one expects now from even a $300 device…) you want Rose to waste time adding support for something that would have no benefit whatsoever.

You would be a little bit more believable if you were to actually to care the least bit for the evidence. Which, contrary to claims from vendors (who have obvious interest in spreading FUD) shows that DSD2048 does not provide any audible benefit.

When you go to McD to get your quarter-pounder, do you insist that they account for relativistic effects when measuring out that 1/4 of a pound of pink slime? That math also would be far more “correct”…

Except that you did not. You’ve only stated the marketing case for spending more money on something that does not provide any audible benefit. If you are shilling for it without even getting paid that’s straight out sad.

#56

MusicLover,

Would you rank DACs like this:
1* dCS
2* Chord FPGA
3* Merging
4* Lavry
5* AKM / ESS DACs (like used in the RD160)

StandardModel

#57

Ranking them is difficult because it’s a trade-off:
dCS gives you the most natural detail,
Chord gives you the best 3D soundstage/timing,
Merging gives you the studio truth, and
Lavry gives you the most analog warmth.
AKM / ESS DACs depands on the brand’s design , some of them incredibly analytical , but it feels a bit clinical for my taste , some of them offers a very warm and musical sound, though it lacks the resolution of its competitors.

The ‘best’ one is simply the one that makes your specific speakers sound the most like live music.

Cheers Mate :slight_smile:

1 Like
#58

MusicLover,

Thanks.
Horses for courses.

I have the impression that in general FPGA based DACs are more likely to be temporally correct - in addition to accurately reproducing audio notes in frequency response 20-20kHz (which pretty much all DACs do today) when compared to chip based DACS-AKM and ESS.

The Dacs I listed are in a fairly high price range -at least for me. Are there any less expensive DACs that are FPGA based with high Tap numbers to be spatially accurate? My last DAC before my RD 160 was a LAiV Harmony DAC which I believe is FPGA conversion based. Did I downgrade by going to the RD 160 and AKM?

My understanding is that taps are a measure of iterations in integral calculus image where n equals the number of iterations or “taps”. Is this correct?

StandardModel

#59

@StandardModel you’re mixing things a bit.

Taps are not “iterations of integral calculus”. It’s simply the length of the FIR filter. More taps = longer time window = better reconstruction of the signal in time domain, especially transients.

FPGA vs chip DAC is not automatically “better”. It depends entirely on implementation. Good AKM/ESS can sound just as coherent if done right.

No, you didn’t “downgrade” by going to RD160. You just moved to a different approach. If anything, your current setup is more balanced and less “theoretical”.

And don’t chase tap numbers blindly. Soundstage and timing are not only about taps, but about the whole chain and analog stage.

Also, you said it yourself before: the audible difference is not that significant, while the money involved is.

1 Like
#60

Thanks for the clarification Vic,
As I interpret your explanation, FIR taps are just the sampled version of the impulse response. So my calling taps “iterations of integral calculus” was conceptually sloppy.

Fir filters are:

  • coefficients of the impulse response
  • used in convolution
  • representing a finite time window

More taps do equal a longer time window.

I think MusicLover was saying that chip based Dacs are more cost and real estate constrained to use a shorter time window with fewer taps than a Dac created through the use of a more expensive FPGA.

StandardModel

1 Like
#61

The transition from the LAiV Harmony to the HiFi Rose RD160 isn’t necessarily a “step backward,” but rather a radical shift in construction philosophy and sonic character. While the LAiV Harmony focuses on the organic musicality of a discrete architecture, the RD160 leverages state-of-the-art integrated technologies to achieve precision and transparency.

1 Like
#62

StandardModel,

It’s a great question. You’re right that FPGA-based DACs (like Chord or LAiV) often focus on custom digital filters and high tap counts to improve temporal accuracy, which helps with imaging and ‘space.’

However, the RD160 with its AKM implementation is a powerhouse in terms of dynamic range and low-level detail. Going from the Harmony DAC to the RD160 isn’t a downgrade—it’s a shift from a lush, holographic presentation to a more precise and transparent one.

If you still crave that FPGA spatial accuracy at a lower price point, Chord’s Qutest or Mojo 2 are the most famous alternatives. But don’t let the tech specs dictate your enjoyment; if the RD160 feels more ‘alive’ to you, then the AKM implementation is doing its job perfectly.

Another crucial factor you should consider is Clocking (Jitter Management). While FPGA-based DACs excel at processing data through high tap counts, the temporal accuracy you’re looking for depends heavily on the precision of the clock signal.

Even with a great FPGA, if the clock jitter is high, the spatial cues and ‘air’ around instruments will be lost. The RD160 uses high-precision Femtosecond clocks that are specifically designed to minimize this jitter.

So, did you downgrade? Not necessarily. A flagship-level AKM implementation with a top-tier Clocking system can often provide better spatial definition and blacker backgrounds than an entry-level FPGA DAC with a standard clock. The RD160 focuses on technical perfection—it’s a very ‘stable’ and ‘timed’ sound, even if it uses a chip-based approach.

Cheers Mate :slight_smile: