[Review] Astell and Kern AK320 - Ramblings of a Young Old Fool.
Apr 20, 2016 at 12:01 PM Thread Starter Post #1 of 23
Jul 22, 2011
There are very few things about my life that I would consider certainties. Skepticism is my natural reaction to information that is not founded in some sort of established fact. Amplifying this behavior is the last several years of post grad work that has bent my mind into doubting anything that isn't peer-reviewed and based in solid science. As a result, I'm quite reluctant to believe many things.
I'm also very quick to question my own actions and emotions. I often find myself in a retrospective state, wondering why I exploded at the jaywalker who nearly bloodied my windshield with his forehead, or to a lesser extent the Subway employee who forgot my oatmeal raisin cookies. My most fundamental internal struggle comes from the desire to reconcile the contrast between my faith that has been instilled in me since childhood and the struggle for significance when I know I my existence is simply a whisper in the evolutionary timeline that has extended billions of years before me, and will continue long past everything I know and love fades from memory.
There are only two things I know for sure. First, that I love my mother with all my heart and, secondly, music is the greatest gift I've received.
I've always had a love for music, as I believe most people do, but music has been for a fail-safe language for things I can't fully communicate verbally, from apologies, to playing John Mayer's  Slow Dancing in a Burning Room to a significant other to end the relationship. My music library is comprised of songs that have been the soundtrack to my life, and it is as eclectic as I am: the Jamaican who has traveled much of the world and lived among people of different cultures, whose favorite band is Mumford and Sons, who enjoys cassoulet au canard as much as jerk chicken, and is as much photographer as I am pharmacist. However, I don't think I'd truly experienced music until I stumbled across the head-fi thread that led to my purchase of the Shure SE-215 years ago, followed by the Vsonic GR07. I remember laying in bed with the lights off, when the first few bars of Mumford and Sons' Sigh No More album started playing. I remember feeling the walls of the studio around me, the slow building of the music, the first lines of the song, followed by some of my favorite lines in song history... 
Love; it will not betray you​
Dismay or enslave you, it will set you free​
Be more like the man you were made to be​
There is a design, an alignment to cry​
Of my heart to see,​
The beauty of love as it was made to be​
I had heard this song a few times prior to this moment, but this was different. With my new earphones I felt like I was living the music, the addition of gear seemingly removing a thick shroud that previously prevented me from truly connecting to my music. ​
After this I went on a mad rampage of rediscovering all my music, something I'm sure most of you can relate to. Vivaldi finally spoke to me, and Elvis seemed new again. I was firmly in honeymoon territory.​
After a while I decided I wanted more, and that feeling never went away. Since then I've owned and extensively auditioned dozens of IEMs, headphones, and DAPs, some of the setups being the equivalent of a semester of my college tuition. Each successive "level" of gear I ascended to left me more critical of the one I just left. Treble that seemed sparkly to me before now appeared excessively peaky and grating. Bass that pounded excitement into my skull now appeared boomy and  overshadowing of the midrange. However, I did gain an appreciation for critical listening and being able to pick out what I wanted in my listening experience.​
In between the GR07 and where I am now, I went through quite a few phases and even more reviews. I've dedicated at least 1,000 hours to A/B'ing little 5 to 10 second song clips to check the differences between different IEMs, cables, DAPs, and amps. As colossal a waste of time as it may seem, it's allowed me to sharpen my ear and listen more carefully to the tiniest differences, and it's given me a good deal of confidence in my testing process. ​
So, before I speak about the AK320 I feel I should disclose a few things about the audio biases that I have acquired over these last 5 years since the night I fell in love with audio.​
  1. Volume matching is everything when evaluating gear. Listening to one piece of gear at a louder volume immediately confers advantages in clarity, dynamics, sound stage, extension on both ends, and general PRaT.
  2. As a general rule, lossless is a waste of time and space. Years of transcoding FLAC to MP3 and A/b'ing the original with the conversion has convinced me of it. This, for me renders using FLAC PCM/DSD files as useless to me when testing. I simply find the best mastered version of the files I need, convert them to MP3 320 if necessary and delete the larger FLAC file. Further convincing me of this is the our very limited human hearing. Unless a million years of evolution happened since my birth, there's no human being alive who can hear past 30Khz, nor can the human body handle up to 16 bits of dynamic range without certain death. I've come to see "hi-res" as yet another audiophile marketing gimmick. 
  3. I conducted a multi-DAP review a couple of years ago, where I compared the HM-901, ZX1, X5, Studio V, and Sansa Clip+ to my iPhone 4 and found absolutely zero significant difference at the time. matter of fact, the ZX1 and my iP4 were completely indistinguishable, not to say I could find a difference between the others worth making a serious note of. That experience left me severely disenchanted with DAPs in general. However, Astell and Kern products weren't available to me for testing at the time.
You can get an idea of my testing methodology here: 
The AK320
As I mentioned before, I extensively compared a few DAPs 2 years ago. At the time, the AK240 was Astell and Kern's flagship product, and I was raging against the pricing trends in personal audio. Since that time, the Ak380 has arrived, almost doubling the cost of the AK240.
Many have hailed the Ak380 as the ultimate DAP, but interestingly the paradigm is that its cheaper sibling, the AK 320 is every bit as good "when dealing with non-DSD files."
Because of my experiences with lossless and DSD files, I'm going to assume AK380 and AK320 are virtually identical sounding for my purposes.
I've been listening to the AK320 for over a month now, learning every little nuance of its sound profile. Next, I spent spent close to as much time comparing it to my other DAPs, primarily my iPhone 6S. My testing methodology for this sit down review has involved very meticulous volume matching. I queued up the same 5 second clip of the same file on all players to ensure the SPL leve was similar. 
The IEM used was what I consider my endgame, the Aurisonics Rockets. I chose it specifically for several reasons. Of all the IEMs I've listened to (an incomplete list can be found in my profile), I've found the Rockets possess the most pinpoint imaging I've heard, the best timbre and tonality I've heard, the absolute best vocal reproduction I've come across and the driver is incredibly quick. I just can't fault this earphone. Also, it is slightly insensitive, so it should be a good driving challenge for a DAP.
I kept the track list short for the review as well, whittling my library down to tracks that would test very specific DAP qualities.
  1. Act 1: Largo al factotum. Il Barbiere di Siviglia. Melbourne Symphony Orchestra. Performed by Jose Carbo. I chose this track because it's a classic example of great performances hampered by shoddy recording techniques. The classical music genre is rife with examples like this. It requires a good deal of power to squeeze out enough dynamic range from the tracks, especially if using less than sensitive phones.
  2. Reminder. Babel. Mumford and Sons. I chose this track specifically for the detail in the vocals, and the timbre of the simple background acoustic instruments.There's a little excerpt between 1:04 and 1:08, where he says "oh my love don't fade away." You can hear the little breaks in his voice where he sounds close to tears. It's little details like that which make the music experience more tactile, and is only accessible to gear that can extract it. I know the Rockets can, so I'm testing to see if the DAPs hinder the Rockets ability to pull the details.
  3. How. What We Saw From the Cheap Seats. Regina Spektor. This is a track for both details, tonality and imaging. It's a piano heavy track, with a little instrumental bits dispersed around the soundscape like the cello that plays simultaneously with the lower piano registers. Plus, Regina's voice is as emotive as ever.
  4. Groove [Binaural]. Explorations in Space and Time. Lenny White, Jamey Haddad, Mark Sherman. This is hands down my favorite track for testing soundstage and imaging, as well as instrumental timbre. I remember Jude making a video about this album years ago, and being amazed at the technology behind it. Since acquiring my Rockets, my appreciation for this album has grown exponentially. I'm literally surrounded by sound, while being able to virtually reach out and touch each instrument. So precise is the imaging. I wanted to use this track to test how each DAP throws out cues while preserving black space.
The testing method involved queuing shorts portions of each track, volume matching them across DAPs, then A/B'ing for differences in how the DAPs rendered the files. I set out with a null hypothesis that I would find no blind-testable differences, and then tried to disprove this null hypothesis. I did this for an average of 30 minutes per track.
I mainly tested between my iPhone 6S, and the AK320, and  the end result was that I could not reliably differentiate the players. Again, this is the AK320 that should sound perfectly identical to the AK380 "with PCM files." I'll leave the conclusions to the readers, along with examples of tests done by others.
  Dear All,
Following a conversation I had on this thread, I decided to test a bit between some of my DAPS, namely my Ipod Classic 6th Gen, my RWAK240, my Cowon P1 and my HM901 Minibox card. I did not include the Tera because it only pairs well with the SE5 and it is a PITA to navigate in a test.
I spent roughly an hour doing this, using FLAC on all DAPs, except 256 VBR MP3 on the IPC. The two iems I used are Spiral Ears SE5 and Hidition NT6. I was very careful to do some volume matching, which was no so easy on the 901, and preferably with low volume, because this is often how I listen.
The results are quite simple: the differences are quite small. I doubt I could differentiate in a blind test. The most obvious was 901 vs. IPC, more crisp sound, but I am still not sure I could differentiate in a blind test.
The 901 is a bit more powerful than the others, and thus drives even iems with slightly more authority.
The 901 and P1 vs. the IPC have slightly better dynamic range.
The boutique all have very slightly better clarity, and a deeper soundstage (main difference overall).
The top performer is still is the 901, but by a very small margin.
As a side note, on my iems, I hear absolutely no difference between the balanced and unbalanced mode of my RWAK240.
Conclusion, if I had to rate sound, from this test I would say:
HM901 Minibox :95/100, Cowon and RWAK240: 93 and IPC : 90.
If I rate them by value (price vs. SQ and ergonomics) then IPC > Cowon P1 > HM901 > RWAK240 (HM901 is less ergonomic but much cheaper than RWAK240 and slightly better sounding).
I buy boutique DAPs also for other reasons than SQ, which is why I am not so shocked : capacity, getting out of itunes, and mostly just because I am a geek who likes new pretty and nicely sounding stuff.
People do not yell at me:
  1. What I consider insignificant differences might be huge to you
  2. I do not listen to "audiophile" music on HD or DSD. I just listen to music, so it might be different for you
It has been a while since I wanted to clear some gear....well I suppose the HM901 might go. I was waiting for the 901S but I do not like the look. I like the P1 and the 240 is nice looking to me, plus it pairs very well with the REF1.
But honestly if there was a 256GB Flac playing IPC, I might let them all go. And I might still buy the Lotoo because I like it :D

  I agree with Mims findings, especially when using iems, which are sensitive enough to not trouble a mobile phones amp, the differences in actual sound quality between a high end dap and a good mobile phone like IP are extremely minimal. Most people just hear the small sound sig difference between the devices rather than any quality difference, then the rest is their "mind" telling them one is better than the other, and mostly because of different bias' more than their actual preferences (which are influenced by bias lol).
Honestly I hate using anything apple though, makes me feel like a sheep/prisoner for some reason lol. So I'll just stick to my ZX1 for iems, and DX90 for slightly more demanding portable HP's which like the extra power. Plus I'm a tech head and enjoy buying stuff for the hell of it 


Agreed too all that you wrote! The only thing I think is missing is an idea of roughly how loud you listen. I've always said that, other than for functionality, unless one wants to get the best out of high-quality recordings with high-end headphones or IEMs, the very expensive DAPs wont be worth it, unless you're doing something like buying an AK240 as a complete desktop/portable solution with all accessories. Also that the primary benefit of the better DAPs is that they drive full-sized headphones much better than, say, an X5 does -- vastly better in fact, and that's before we get into UI differences and storage.

Thanks for reading.
Apr 20, 2016 at 8:38 PM Post #2 of 23
My one question is why you gave it a five star review if it does nothing different from an iPhone in the sound department at more than double the price.
Apr 20, 2016 at 10:19 PM Post #3 of 23
My one question is why you gave it a five star review if it does nothing different from an iPhone in the sound department at more than double the price.

Because it's a perfectly good DAP. It's no worse than the iPhone, which I find totally transparent. Plus the design and UI are flawless. A&K knocked it out of the park there. I can't rate the iPhone at 5 stars and still knock the AK320 down a peg. I'd be a hypocrite. It's not my business what others spend their funds on.
May 31, 2016 at 3:41 PM Post #4 of 23
I enjoyed your review and find it amazing how the iphone 6 is so good-My son and daughter have them and i will be getting their casts off in due course

Now a piece of advice in relation to your words
"There are very few things about my life that I would consider certainties. Skepticism is my natural reaction to information that is not founded in some sort of established fact. Amplifying this behavior is the last several years of post grad work that has bent my mind into doubting anything that isn't peer-reviewed and based in solid science. As a result, I'm quite reluctant to believe many things".
That makes perfect sense to me.But to add.Never believe a dr,policeman or a professional in any field INITIALLY.Always get a second or third opinion and cross reference the facts.I have personal experience of being lead up the garden path by so called experts.
There are about 10 billion galaxies in the observable universe! The number of stars in a galaxy varies, but assuming an average of 100 billion stars per galaxy means that there are about 1,000,000,000,000,000,000,000 (that's 1 billion trillion) stars in the observable universe!-Nice fact that one and you and me get so uptight about silly things when we should be amazed at lifes creation and our purpose.Have you ever branched out into the ambient genre of music?
Jun 10, 2016 at 5:23 AM Post #5 of 23
  1. Volume matching is everything when evaluating gear. Listening to one piece of gear at a louder volume immediately confers advantages in clarity, dynamics, sound stage, extension on both ends, and general PRaT.

Can't stress it enough, this is important when comparing stuff in audio. 
Jun 10, 2016 at 1:48 PM Post #6 of 23
Amplifying this behavior is the last several years of post grad work that has bent my mind into doubting anything that isn't peer-reviewed and based in solid science. As a result, I'm quite reluctant to believe many things".

My thought exactly, as a researcher leading R&D teams for a living.
Unfortunately, there are ZERO company selling DAPs/Amps around here coming up with decent science behind their marketing...
And company size doesn't matter, they all sell pseudo-science made up for the masses, in the pure "try it, you'll see" style.
And they don't even bother selling better spec sheets that would at least justify these 1000+$ pricings, or UI with new functions.
These days you start to find smartphones under 500$ with 4-6Gb ram and 128Gb rom, but audio companies keep selling you DAPS with 32/64Gb internal rom and one micro-sd that they advertise for Hi-Res music.
Seriously, if you're truely going to sell Hi-Res, anything below 256Gb (internal) is a joke.
Jun 11, 2016 at 7:57 PM Post #7 of 23
So much truth. Let me add that volume matching shouldn't be by ear, it should really be measured. I have a DAP because I like them, and I need something with really low output impedance because of my stupidly low input impedance UM Merlins. Sound quality is solved from e source end. It's easy at this point to develop a reasonably priced transparent dac and amp.

I'm still getting an AK300 mind you. I'm going to try to hack google play music streaming on to that sucker. I'm probably going to end up bricking it, haha.
Jul 27, 2016 at 11:59 PM Post #8 of 23
After a couple of weeks using the AK320 mostly with AKT8IE MK2, I can't help but notice the 3.5MM output is not as good as the 2.5MM balanced output. Clarity, bass slam, and texture is a little better out of the balanced output.
However, I still prefer how the AKT8IE sound with my Mojo. The clarity, texture and imaging is a step or 2 better in my opinion. Even if I deliverately increase the volume on AK320 to make it louder than Mojo, it just highlight how the sound differs from Mojo.
I will eventually do a proper A/B test with volume matched and if possibly blind. But it doesn't look good for AK320 to me. 
Jul 28, 2016 at 10:03 AM Post #9 of 23
  After a couple of weeks using the AK320 mostly with AKT8IE MK2, I can't help but notice the 3.5MM output is not as good as the 2.5MM balanced output. Clarity, bass slam, and texture is a little better out of the balanced output.
However, I still prefer how the AKT8IE sound with my Mojo. The clarity, texture and imaging is a step or 2 better in my opinion. Even if I deliverately increase the volume on AK320 to make it louder than Mojo, it just highlight how the sound differs from Mojo.
I will eventually do a proper A/B test with volume matched and if possibly blind. But it doesn't look good for AK320 to me. 

That's because the balanced is louder.
Sep 13, 2016 at 5:30 PM Post #10 of 23
[color=rgba(0, 0, 0, 0.870588)]Great post Eke! I remember reading your previous post on comparisons of various DAPs and how they all sounded basically identical. It's refreshing to have somebody such as yourself cutting through the hype and providing an honest opinion. Thank you for this.[/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]I pretty much agree with everything you've written, so we're basically on the same page here, but just for fun, let me play Devil's advocate a bit :wink: I'm always curious to understand how some people hear no differences, while others claim "night and day" differences. (BTW, I hate that expression - it almost never applies to modern audio and it often leaves buyers feeling disappointed when they don't hear the "night & day" differences for themselves.)[/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]I found your post after looking for reviews of the AK320. I'd previously heard all the A&K DAPs via my SE846 and hadn't been particularly impressed. But over the weekend, I got a chance to AB the AK320 against the FiiO X7 and QP1R, in all cases, just using the LO to feed my KSE1500. I thought I heard a clear improvement from the AK320. Now, I'll admit it wasn't easy to volume match, because the line-outs of all these devices were slightly different and I didn't have an SPL meter on me. I know I need to look into this further, but I was seriously considering pulling the trigger on an AK320. And then I saw your post. So here comes the Devil's advocate bit...[/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]I agree with you about it being very hard to differentiate lossless and good-quality lossy formats. When I've tried in the past, I've failed to successfully AB 256 VBR AAC from FLAC. I still have most of my music in FLAC format because I want to be able to ensure no additional losses should I need to use lossy compression for either file transfer to devices that don't support FLAC (like my Gear S2) or via bluetooth, etc. But also consider the following. Imagine you deliberately make a tiny loss in audio fidelity in sound file A - so small that you can't tell the difference between A and the resulting sound file, B. Then, deliberately make another tiny loss in resolution in B to create sound sample C. Again, you'd struggle to hear a difference between samples B and C. Now extrapolate this a bit. You very likely would be able to hear a difference between sound samples A and Z. At least this is my other psychological crutch for remaining mostly lossless and wanting to buy yet another shiny new toy :wink:[/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]Where I'm headed with this is that I think quantifying the small (sometimes very small) differences one hears between DAPs could depend on lossy file formats, transport, RFI/EMI effects, cable impedances, headphones used, etc. The last one is probably the most obvious - my SE846 has very noticeable hiss from many low-priced amps (and some high-priced amps too). That's one very obvious give-away when ABing. The other obvious issue with headphones is that extremely hard-to-drive phones probably stress lower-powered amps, which might make them sound a little thin or distorted. On the topic of headphones, I think perhaps the one and only point where we'd disagree is on the Aurisonics Rockets. I won't rule out the possibility that I had a defective pair, but I sold mine after about 2 months. I loved their ergonomics, but really didn't like their sound. I found their FR quite n-shaped, particularly at the low volumes where I normally hang out, but worse - I also heard a graininess or scratchiness in their sound. The pair I owned just lacked clarity in comparison to all my other headphones.[/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]In the name of science, I guess I now have to get an AK320 and do a proper SPL-matched comparison of its line out :) But if I reproduce your conclusions, there'd be no good reason for me to keep it. I will try to post back at some point and let you know how it goes. Thanks again for your frank and honest review. Keep it up![/color][color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)][/color]
[color=rgba(0, 0, 0, 0.870588)]P.S. Getting off-topic, but also into an interesting area to discuss - the whole hi-res audio debacle. Again, I agree with you. I believe hi-res audio sales have mainly been a tax on stupidity. I've argued vigorously (on headfi and elsewhere) against what I consider audio snakeoil. But since you probably already agree with me on that, let me play Devil's advocate again. Although I can't see a justification for sampling rates > 44 kHz, I think there might - just might - be a case for 24-bit (or even 32-bit float). Consider another wacky thought experiment. Imagine you have a very large dynamic-range recording. (Unlikely in 2016, I know, but just imagine!) Now imagine after the loudest peak of the heaviest symbol crash in the recording, the track goes into one of its quietest moments - and at that point you choose to turn the volume up significantly. Even with an extended dynamic range from a shaped dither, the dither typically pushes that white noise into frequencies we're less sensitive to - not to frequencies we can't hear, just frequencies we're less sensitive to. On the other hand, it's quite possible that the 24-bit version would still have an inaudible noise floor, even with the volume cranked way up on the quietest passage. All that being said, I've yet to successfully AB a hi-res track with its 44/16 equivalent (assuming an otherwise identical master). Others might claim superior hearing, but you'd think that if it were possible for any human being (even just one) to successfully AB hi-res audio vs red-book CD/FLAC, that companies like Pono would have jumped all over them and made commercials/brochures/news articles of that fact. One person is all it would take to validate what companies like Pono and HDTracks do. But I've not seen any scientific study anywhere that's shown anybody being able to consistently distinguish hi-res audio. Sorry for the long rant.[/color]
Sep 13, 2016 at 11:11 PM Post #11 of 23
There will always be reason to keep a DAP over the other, but everyone will have different priorities and opinion. I think nowadays DAPs don't have much difference in SQ. Most of the difference usually is the amount of power output and a slight difference in FR curve (some have slight roll off in one of both ends).
I have tried many DAPs in the past 3 years and I realize that, looks, UI, ergonomics and extra features matters as much as SQ. 
I hate to admit that AK DAPs is a gold standard in UI and IMO have already surpassed the iPod in simplicity. The AK3xx series despite the cost is the most mature and just a joy to use. With AK Connect and the accessories available it can just cover a lot of use case from iem to full-size can. from desktop to mobile.
The AK300 may be the sweet spot. I have opted for AK320 because there was a free Amp which would cost about the same if I go for AK300 + Amp.
Other DAPs I tried always reminds me why I always go back to AK. Fiio X7 while an outstanding DAP, UI/software is still buggy. Hugo/Mojo requires a stack. Sony DAPs just doesn't work for anything than easy to drive phones.
Sep 14, 2016 at 11:05 PM Post #12 of 23

No need to apologize.
1. I have to ask what you mean by the AK320 sounding "better," especially in the context of it being used as a LO. There are several potential things in play, like the OI of the LO of each device, mixed with the sensitivity of the KSE's input. I don't necessarily agree that more sensitive IEMs require more power. Rather, they require more regulated power, since there's less hindrance in the circuitry. Honestly, my Clip+ could cover all my needs if it was simply internet connected.
2. The great thing about the latest iterations of the LAME codec is that it's designed to fight loss of perceivable quality. The simple fact is that things like noise floor are controlled at the recording and mastering level. Once that final file is "baked" by the engineer, there's nothing the end listener can do. That's why hi-res is such a joke to me. Those files are mostly empty bits that have no use to anyone. The master files are saved in formats untouchable by our players, that have all the recorded channels in their uncombined states, so these 24 and 32 bit files have nothing to do with dynamic range. We can't even physically handle 16 bits of auditory range without permanent damage to our auditory functions. 24 bits would be suicide.
3. Honestly the Rockets have been one of my most transparent listens. I've stuck with it precisely because it has zero grain whatsoever. Yes, it is slightly more mid-forward, but it follows the OW curve with a tilt towards warmth. The imaging is also to die for. What I think happened to you is something that I addressed in my Rockets review. The iems are incredibly small sealed titanium tubes that but a massive amount of pressure on the eardrums if your eustachian tubes cannot properly equalize the space. I use mine with TS-500 tips. With regular silicone tiips, the pressure can hamper rapid movements of the eardrum that translate to treble, and the very slow vibrations that translate to bass. With the foams, the midrange is hands down the best, clearest, most natural I've ever heard. The bass is linear down to about 35Hz, and the treble is well extended without a trace of glare.
Sep 16, 2016 at 2:02 PM Post #13 of 23
Hi Eke,
I feel like I'm just replying to myself here, because I pretty much agree with all your points.  Well, almost all your points - let's just say we both agree that the Rockets are "mid-forward" and I will leave it at that and make no further comment on those headphones :wink:
The reason for me writing anything here was just that I'm a scientist and I want to get to the truth. I'm always curious as to why some hear no difference and some hear "night and day" differences. The truth, I suspect, is either a) placebo/expectation bias or b) very small differences that some people hear and some don't. Point b) is really subjective, because even the tiniest change in sound might bug one person, but another might not hear it, or their brains might not process it or they might find the effect so small it doesn't bother them. I have good friends on headfi whom I trust and respect that don't hear the noise floors I hear on the exact same amp/headphone combinations. I suspect things like this are more than just differences in hearing, but differences in what our brains are looking for in the sound.
I find it fascinating that hi-res audio has garnered such a following and I'm genuinely curious about the issues of pychology and placebo, which of course we're all subject to (and I might be subject to with the AK320 LO). CNET's Steve Guttenberg regularly states on his awfully-named "Audiophiliac" blog that he can hear the improvement in clarity with hi-res audio. This is a guy who looks like he's in his 60s(?), who can't have an upper range anywhere close to 22 kHz, yet alone beyond. But this guy is considered - by some, at least - to be a world authority on audio, and his reviews undoubtably sell products. This is why I feel a need to push back on hi-res - I don't want to rain on anybody else's parade, but I want to protect the vulnerable and the ignorant from being unfairly parted from their hard-earned money. The irony is, those people we want to protect often turn on us for attacking their beloved hi-res formats. It becomes almost a religious debate, and one neither side can win.
Let me go back to one point you may have missed from my earlier post. Imagine you have a recording of a Saturn V rocket firing (these puppies can exceed 200 dB), followed by the gentle crescendo of a distant string quartet, far in the background. Let's assume the recording's dynamic range is in excess of 96 dB, requiring some dithered 16-bit representation. That dither creates noise within the audible range, which we could theoretically hear if we chose to turn up the volume very high during that quiet moment. Of course, you could instantly lose your hearing and/or die, etc., if the Saturn rocket fired again at that point, but let's assume that doesn't happen. An amplifier volume control lets you zoom in to a much tighter dB range, which could expose a noise floor that might not exist had we gone with a 24-bit representation that could have achieved a ~144 dB dynamic range without dither.  You could conduct a similar, but safer, experiment using Audio DiffMaker. Subtract the 44/16 version from the hi-res master and listen to the difference. What do you hear? Dead silence. However, it IS possible to then turn the volume way, way up and hear some noise and artifacts. (I believe those artifacts have to come primarily from the differences in bit depth, not sample rate.) Please don't misunderstand me - for all practical purposes this is all irrelevant and none of this means I believe in hi-res audio. I still think Steve Guttenberg is a naked Emperor.
Back on topic - as to your question about what I heard with the AK320. I thought I heard an increase in resolution. Something akin to when the HD kicks in on your car's FM radio. Not quite as dramatic, but some level of improvement relative to my FiiO X7's LO (which is itself, very impressively clean). Now, I'll admit this was a brief test which was not properly volume matched. What I typically do in situations like that is to try playing sample A at an obviously quieter volume than B and see if I still prefer A. I believe I went into this test with no expectation bias, because I'd previously heard the full range of A&K DAPs via my SE846 and had not been impressed. I didn't expect to hear a difference with my KSE1500, but I thought I did. I hope to have a chance to do some more detailed (and hopefully SPL-matched) listening next weekend and I'll try to report back after that.
One question for the community as a whole. I recently saw a post on reddit that suggested you can get banned from headfi for even attempting to discuss blind AB audio testing. NwAvGuy hinted at that after he was banned. Is there really any truth to this? It sounds like a shady conspiracy theory. Anyway, let's put it to the test...  I plan to conduct a blind, SPL-matched test of my KSE1500 being fed by FiiO X7 and AK320 LOs next weekend and would like to discuss the results on this thread.
Ok, let's see if I get banned... :wink:
Sep 16, 2016 at 3:23 PM Post #14 of 23
If love to see the results of your test.

As far as the recording issue, human hearing tops out at a useful dynamic range of 60db. The absolute pain threshold is 120db. Even live symphony orchestras use a max of 50db dynamic range for our hearing. Anyone who claims to be able to hear beyond this is a liar or superhuman.

So listening a recording that has a 200db rocket launch encoded, one doesn't have to worry about what comes next. You'll have a very difficult time hearing anything for the rest of your life.

I'm not disputing the fact that it is *possible* to encode more high frequency information into a file. The issue is that human hearing is the limiting factor. Just like how it is so easy to kill cancer and HIV cells in a test tube...the trick is killing them while keeping the patient alive.
Sep 16, 2016 at 3:42 PM Post #15 of 23
I completely agree with you Eke. Only for the sake of science here...  What I'm suggesting is a thought experiment where you boost the lowest dB range in the quietest moment of the track. This is the audio dynamic-range equivalent of putting a hi-def picture under a microscope. Your ears still only need to cope with a limited dynamic range, you're simply using your amplifier to lower the gain during the Saturn V rocket launch and then raise it during the string quartet intro. Theoretically, you should then be able to 1) not destroy your hearing and 2) hear the difference between a 24-bit and 16-bit dithered noise floor. Again, just to be clear, I am not talking about sampling rates here - I'm only considering the effect of bit depth.
If I don't get banned in the meantime, I will post back in a couple of weeks with my blind AB test results :wink:

Users who are viewing this thread