JSmith
Master Contributor
No, it isn't... they break the sampling theorem’s assumptions, leave in ultrasonic images, have measurable high frequency roll off, and rely on an analog filter that can’t actually do the job.
JSmith
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Fiio K13 R2R - Owners Thread
- Thread starter Guddu
- Start date
JSmith
OP
Guddu
Major Contributor
- Joined
- Feb 10, 2020
- Messages
- 1,861
- Likes
- 1,928
- Thread Starter
- #162
I started this thread to introduce this new DAC as well as seeing how most of us approach to such a product.
R2R DACs have been subjectivist’s favorite and seems like you can’t escape out unless you try one, and I am getting one knowing that objectively it won’t match any similarly priced DAC (chip).
Anyone who has it and explored PEQ properly, can you tell if it works without any quirks?
R2R DACs have been subjectivist’s favorite and seems like you can’t escape out unless you try one, and I am getting one knowing that objectively it won’t match any similarly priced DAC (chip).
Anyone who has it and explored PEQ properly, can you tell if it works without any quirks?
solderdude
Grand Contributor
They seem to be ... just like analog tape, vinyl and tubes does.
What these have in common is inferior signal quality and the fact that they are older techniques which have an 'aura' of days gone by where everything sounded 'better' (analog).
Some of their signal degrading properties are not as sound detrimental as some make it out to be and some aspects are even preferred by some people.
Of course everyone is free to use whatever brings them joy.
R2R (or other techniques) are audibly just as fine as DS (or other techniques).
Filterless Non Over Sampling (emulated or not) which basically is 'sample and hold' in reality and has little to do with R2R conversion (or other techniques) and this is what most audiophiles mean with 'R2R'. They actually mean 'sample and hold'.
Funnily enough the 'older R2R chip CDP' all use filtered oversampling as the reconstruction filter is essential but some
Fortunately the 'kind-of treble roll-off' and the 'look at that beautiful square-wave reproduction (uncompromised 'speed' and impulse response) and lack of dreaded ringing is what they see and hear. For most (usually older) audiophiles that 'roll-off' is above their hearing limit anyway so they may not even notice it.
Usually this filterless NOS works out really well and sounds just fine despite of the compromised signal quality and (inaudible) image frequencies.
It just isn't 'what is on the recording'.
Filterless NOS R2R (preferably with tubes) is a hype and manufacturers see potential profit as it has a higher profit margin because audiophiles don't mind paying more.
FiiO (and similar brands) also jumped on that wagon and had the great idea to bring it at a lower price...
Last edited:
To it's credit it's actually better than Mscaler for a fraction of the price ? Mscaler seems to have obvious artifacts that a well implemented software can easily sidestep by just using floating point math .
But as you say why ? It's done to a good enough level in your DAC anyway ( normally ) .
So it's a solution to a problem you should not have in the first place ? Buy DAC without proper reconstruction of the signal, apply the missing filter externally .
.. Caveat you can design any kind of filter you want so i've heard ? And a whole cult of mythology around that in other forums .
So you can end up with a clearly audible result with your filter tweaking , with this software . And different is better in audiophile lingo
especially with no point of reference .Did you do a blindtest or research for this statement?
JSmith
Master Contributor
You mean ol' Rob Watts who claims to hear things at -300dB? No, he has not published any such testing or peer reviewed research... tap, tap, tap.
JSmith
OP
Guddu
Major Contributor
- Joined
- Feb 10, 2020
- Messages
- 1,861
- Likes
- 1,928
- Thread Starter
- #167
Agreed and I do get it.... who can deny this when you have numbers showing its inferior but my small quest is to see if that inferior SQ really something you can straight away deny or not.
THE Question and claim is that "R2R output is pleasant or more favorable to listen to. I know many would jump in quickly with "BLIND TEST" analogy but I don't want to blindly ignore this and want to try myself. Its really really weird that everyone (subjective, youtubers) sounds madly in love with it calling it like best buy of the year or something.....
I am still trying to process these and understand properly.
They are influencers, they are paid to influence.
Some are not even smart enough to get paid
That’s not always how corruption works , you have add revenue and sales reps that becomes your friend, and free stuff and discounts shows up all the time .
But I’m sure direct payments also occur .
And there is also a natural selection, they provide content that their viewers want . And as most information in audiophile forums and magazines are wrong and the whole audiophile culture took a wrong turn decades ago , you get the videos you deserve.
What’s wrong with most audio magazines is also what’s wrong with many YouTube channels.
oleg87
Addicted to Fun and Learning
Well, you'll likely hear something, but this is a question of psychology, not technology. The actually audible differences between this thing and any other half-comptent dac/amp are going to be very marginal and swamped by illusory differences, as is generally the case with audio electronics working inside their limits. Pretty much all of this is a product of whatever psychological reaction this equipment and the discourse surrounding it produces in the reviewer's mind.
Last edited:
Also subjektive YouTubers are not even wrong thier sigthed methods are so bias filled that it drowns the subject matter.
As it would for you and me , but we would not tell the results to the world as if its where an absolute truth.
They can be rigth they can be wrong it’s undecided due to the lack of controls.
And you pretty much assume they are fooling themselves when reviewing marginal effects beyond and at the limit of human capacity.
OP
Guddu
Major Contributor
- Joined
- Feb 10, 2020
- Messages
- 1,861
- Likes
- 1,928
- Thread Starter
- #173
I definitely agree but I thunk I myself can’t really say if they are biased or wrong. It’s their opinion and understanding which may or maybe match with others.
Normally I avoid all that but pulled plug on K13 R2R and trying to understand all NOS / OS thing, but I also am realizing that there is some difference. For example, I have a few different versions (or releases) of a few same recordings (like Eagles etc) and I know which one I prefer and normally get most of the times if someone else plays it in my setup me not knowing. However everything is sounding same with K13 NOS now and I am investing my time with honesty to understand this.
Anyways, this is my first time with R2R and hoping I end up saying that its just crap and stick to my E50 or Wiim Ultra.
I see some striking parallels to what's happening in the musical instrument industry right now.
There's the brand new, just announced Waldorf Protein synthesizer, a fully digital emulation (of sorts) that proudly sports all the original 8-bit oscillator (without interpolation and antialiasing) glory of their old super famous 90s Microwave synth, by meticulous analysis and reproduction of said original. By the very same company. Source: company CTO.
The discussion, just within the last few days, is utterly furious. Omg it sounds great, BUT does it really have the original sound? Nevermind it's not even intended to be, expressedly not, as said so by the CTO. Does it use the original ASICs? Ofc it does not, Waldorf CTOman Rolf said it would be way too expensive, but nevermind friends, let's speculate and assume freely from other misleading uToob reviewer sources!
Musicians (and/or wannabes) continue to ramble. Omg it got digital (instrument level) filters, the original had analog ones. Nevermind the very second historical series from the early 2000s (Microwave 2, of which I had the XT version of for a few years, trust me it sounds glorious ) was fully digital! What sounds better? Original 8bit glory run through fully analog resonant filters, or fully digital, or digital filters on analog oscillators? Microwave 2/XT/XTK already had deliberate, adjustable aliasing settings? Who ever used those fully intentional "lofi" features? Nobody knows, nobody ever did anything even remotely resembling a half-blind test.
Meanwhile, sober minds always agree: the biggest influence on the sound - and let's not forget we're talking about instruments aka actual sound generators, not reproduction - is the user. Skill trumps everything.
It's ridiculous. The rambling and cluelessness in the music community more or less reflects that in the audiophile one. The craziness and level of delusion differs in detail, but overall it's the same.
Last edited:
Michael Larkin
Member
- Joined
- May 9, 2021
- Messages
- 6
- Likes
- 19
Sorry to reply -- I can't seem to find a way to make a fresh post, so please treat this as one.
I have five or six delta sigma DACs and they all sound quite similar. I also have a couple of R2R DACs, of which the K13 is one, and the Gustard R26, the other. The K13 is in my secondary, nearfield system where I listen on Quad Revela 1 speakers. Prior to getting it, I was listening to my Earmen Tradutto, which was IMHO marginally the best of the delta sigma bunch.
At less than half the price, I decided to give the K13 a punt, seeing as I liked the Gustard in my main system in the lounge so much better than all the delta sigmas. Now I don't know or care about measurements -- I just like to listen to music and whatever sounds best to me is what I'm after.
I also don't know well how to describe sounds or why I like R2R as much as I do. I can only say that I very definitely do. Maybe the K13 is the worst measuring DAC in the universe, but if so, I couldn't care less. Maybe using NOS is for whatever reason wrong. Maybe it offends the measurement purist. But again, I'm unconcerned -- I'm not governed by dogma either way. If I ever come across delta sigma DAC costing around, or even a little more than the K13 that sounds significantly better to me, I'd be prepared to give it a punt.
All I can say is that what I'm hearing through the K13 is not far off as gorgeous-sounding to my ears as the R26 is in my main system (and at less than 1/5 the price). Along with speakers, I suspect that DACs are the most important element of a digitally-sourced system like mine. Just so long as halfway decent amplification (in my secondary system, that's an Audiolab 6000A) accompanies them, all should be well.
I've hinted that IMHO there's dogma on both sides -- "self-deluded audiophile snobs" on the one hand and "indoctrinated measurement junkies" on the other. I say to blazes with both of them. Just sit back and listen, and choose whatever sounds best to one's own ears. Don't let it start to sound worse just because someone in either camp opines it is so. IMHO, one can usefully learn to live without external approval and validation.
Last edited:
JSmith
Master Contributor
Blah, blah, blah... DAC's are not meant to have a sound and rarely do unless very poorly designed/broken.
Exactly, don't worry about the DAC so much and enjoy some tunes.
JSmith
oleg87
Addicted to Fun and Learning
"Dogmatic" implies a stubborn adherence to some set of belief irrespective of evidence, an absence of curiosity, and so on - but I've spent far, far more energy testing this stuff than it's worth over the decades. I know for an absolute fact that my hearing is extremely prone to hearing imaginary differences. I'm very open to (technically coherent) cases for how conventional engineering wisdom about audio fidelity might be wrong - they just don't really exist (or demand that a staggering percentage of audio electronics out there be wildly misdesigned or malfunctioning, which I have not found to be the case in testing my admittedly modest selection of equipment). I think it is an extremely safe bet to assume that an audiophile is more likely to hear things that aren't there, than highly sensitive test equipment failing to capture large differences that are, based on plenty of experience both with measuring and listening, not blind dogmatism. No one's stopping anyone from (mis)placing trust in their ears and marketing, and indulging their taste for placebo, but as your heap of converters suggests, this tends to get expensive.
For what it's worth, the Gustard R26 is a DAC that would hardly make the top of Amir's chart, but based on L7AudioLab's review there's no measurable evidence of it having a "sound" (the usual broken-by-design NOS mode aside, though if you're 75, even that might be indistinguishable from anything else you have by ears alone).
Last edited:
Michael Larkin
Member
- Joined
- May 9, 2021
- Messages
- 6
- Likes
- 19
When I bought it, the Earmen Tradutto delta sigma cost around £799 (that's currently just over $1000). But I don't think it's that much better than the cheapest c. £100 delta sigma I have. If DACs aren't meant to sound like anything, or as I take that, are meant to sound "transparent" to the source, then my conclusion might be that R2Rs don't sound transparent in that sense.Blah, blah, blah... DAC's are not meant to have a sound and rarely do unless very poorly designed/broken.
Exactly, don't worry about the DAC so much and enjoy some tunes.
JSmith
It's not that delta sigmas sound bad to me, so much as just a little boring. Based on the 2 I have, maybe R2Rs are "coloured" or "inaccurate" -- perhaps even "broken". But if so, I like that. For whatever reason, an R2R gets me involved, so that now I listen quite a lot more on the secondary system on the desktop in my study rather than always taking a break to listen on the main system in the lounge.
I have little idea, technically speaking, why R2R gets me more involved -- I can only speak about subjective responses to tone, soundstage, musical interplay and so forth. If anyone feels that a delta sigma gets them more involved, that's fine by me. I don't believe there's right or wrong in all this, just people and the pairs of ears they happen to have. But IMHO no one should allow themselves, either way, to be unduly influenced by others.
When I restarted this hobby about five years ago after a long hiatus, that's a mistake I made and it led me on a merry dance. I bought all sorts of stuff and ended up not liking quite a lot of it, often after convincing myself for a while I did like it. Only when I started to rely on my own perceptions and being more discerning about equipment reviews/reviewers could I begin to really enjoy my music.
oleg87
Addicted to Fun and Learning
Thing is, the Gustard is neither colored nor inaccurate. If we want to give points for competent reinventions of the square wheel, it is a respectably high-fidelity R2R converter that clears CD quality, and is far more akin to an allegedly "boring" sigma-delta than this FiiO.
Do not conflate R2R and NOS, they are united only by association in being used to market fashionably needlessly-low-performance hardware. The former is a way to implement a converter, which tends to underperform the state-of-the-art but still yield solid performance as far as human hearing is concerned; the latter is declining to attempt to reconstruct the sampled analog waveform. That the resulting digital stairstep boogeyman signal is believed to be more "analog-sounding" by the audio cognoscenti (who are too busy worrying about nonexistent filter ringing to worry about the heap of ultrasonic garbage they are getting instead) will never cease to amuse me.
Last edited:
+1000 it would not bee to terrible if they did not also removed the filter when in NOS mode
NOS with filter is how the early DAC's in the beginning of digital audio worked, but they did have a reconstruction filter even if partly analog if i remember .Some modern users of NOS DAC do the OS part in HQPlayer hence first create the problem then fix it elsewhere ?
There is a funny myth that the timing of redbook is 1/44.1Khz which is totally wrong thats not how s**t works ( it's more complicated it involves the bit depth too and is much much better than this ) But as far as i understand if you do it the "digital stairstep boogeyman signal" way with no filter it's actually true
you lower the time resolution to exactly that where a proper reconstruction filter would have restored this to better degree .
