Fiio K13 R2R - Owners Thread

[antcollinet]

On paper, the R2R has significantly inferior measurements. But I couldn’t hear any difference.

The exact expected result. The inferior measurements are not sufficiently inferior to make an audible difference.

 

[antcollinet]

I appreciate the technical breakdown, but I have to respectfully disagree based on my own measurements and hands-on testing.

I’m pretty meticulous about my chain. I run Adam D3V speakers EQ’ed to flat from 50Hz to 10kHz using a measurement mic, and I’ve EQ’ed my LCD-X (and even the DT 1990 for kicks) to match that same flat response using pink noise as a reference. I’ve even gone the extra mile of building a DIY in-ear microphone to measure exactly how my own ear geometry and headphone placement affect the frequency response in real-time.

With all that data in front of me, I still find the NOS setting provides the widest perceived soundstage. In NOS, the instruments—especially high-hats—don't feel like they are playing "inside" my head; they sit just outside the cups with a slightly "fuzzy" or airy quality that feels more natural to my ears.

As for the "treble roll-off" argument: I don't actually see a significant drop-off in the highs on my measurements specifically due to NOS. In fact, moving the headphones even 1 cm on my head changes the frequency response more than switching the filter does. If there is a roll-off, it’s negligible compared to the physical fit and seal of the pads.Short story is that it does something, I just can measure it...

At the end of the day, I’m taking a chance on the K13 for the R2R implementation, but even if the "magic" is subtle - of any, the onboard EQ options and the remote are worth the admission price alone.

If you are really interested in finding out if you are hearing is really due to the filter - and not perceptive bias, the only way is to set up a properly controlled DVM level matched blind test (you'll need a friend to help - and you'll need to spend some effort to make sure your controls are adequate - for example not being able to interact in any way with your friend during the test). See if you can spot the same sound stage differences when you have no idea what filter is being used. See if you can repeat that achievement 10/10 times.

 

[AndyBell]

The exact expected result. The inferior measurements are not sufficiently inferior to make an audible difference.

Yup. I did my tests *before* visiting this forum. I had got the R2R DAC because an AI recommended it for a different sound.

I’m a c++ programmer in an engineering field where measurements are *everything*. Before that I was in the medical industry, dealing with drug trials. So I understand that measurements must be blind and human biases are powerful - really powerful.

My tests weren’t as stringent as a medical one. After all, this is only audio. But it didn’t take long to realise I could hear no difference - both blind and sighted. The same applied to upsampling.

The K11 R2R and K13 R2R’s NOS mode still features a low pass filter - so I could not (and still cannot whenever I try again) hear a difference when I enable it.

To be honest, I like the K11 R2R. It’s a handsome unit and sounds fine. But so does the DX3 and all the other DACs I tried.

I’m glad I found ASR - I now spend my money on music and my time listening to it.

 

[Opalius]

If you are really interested in finding out if you are hearing is really due to the filter - and not perceptive bias, the only way is to set up a properly controlled DVM level matched blind test (you'll need a friend to help - and you'll need to spend some effort to make sure your controls are adequate - for example not being able to interact in any way with your friend during the test). See if you can spot the same sound stage differences when you have no idea what filter is being used. See if you can repeat that achievement 10/10 times.

Fair point on the perceptive bias—it’s the reason I actually sat down with my wife yesterday for a blind test.
I had her switch the filters while I stayed completely "blind" to the settings. Interestingly, I was able to distinguish between Filter 2 and NOS every single time.

However, I'll be honest: switching between the other oversampling filters and NOS was a total hit-and-miss for me. There's definitely a specific delta between Filter 2 and the NOS mode that my ears (and my music—mostly dense symphonic metal and "fantasy" tracks) pick up on immediately in terms of that spatial "fuzziness" I mentioned.

To put some data behind my madness, I’m going to spend tonight finalizing the graphs. Tomorrow, I’ll post the measurements from the K11 using my DIY in-ear mic across the different filter settings.

As I said before, I’m seeing more variance in the frequency response just by shifting the headphone cups a few millimeters than I am from the filters themselves, but the measurements should at least give us a baseline for what’s happening at the eardrum level. Either way, the K13 is on route!

 

[antcollinet]

Fair point on the perceptive bias—it’s the reason I actually sat down with my wife yesterday for a blind test.
I had her switch the filters while I stayed completely "blind" to the settings. Interestingly, I was able to distinguish between Filter 2 and NOS every single time.

However, I'll be honest: switching between the other oversampling filters and NOS was a total hit-and-miss for me. There's definitely a specific delta between Filter 2 and the NOS mode that my ears (and my music—mostly dense symphonic metal and "fantasy" tracks) pick up on immediately in terms of that spatial "fuzziness" I mentioned.

To put some data behind my madness, I’m going to spend tonight finalizing the graphs. Tomorrow, I’ll post the measurements from the K11 using my DIY in-ear mic across the different filter settings.

As I said before, I’m seeing more variance in the frequency response just by shifting the headphone cups a few millimeters than I am from the filters themselves, but the measurements should at least give us a baseline for what’s happening at the eardrum level. Either way, the K13 is on route!

If you are going to do more testing, you need to check accurately (to 0.1%, with a test signal, and DVM at the speaker or analogue output) the levels. It is possible with all the noise and distortion the NOS mode introduces, that it could make a level difference. This will impact your perception in a way. that is simple level related.

BTW - is it just NOS (but still with an analogue reconstruction filter on the output) - or is it NOS filterless?

 

[antcollinet]

As I said before, I’m seeing more variance in the frequency response just by shifting the headphone cups a few millimeters than I am from the filters themselves,

That is not a surprise. And it invalidates your tests if the cups move even a little bit between comparisons. Even tilting your head might make a difference that swamps filter differences.

 

[RandomEar]

NOS mode can be slightly louder than filtered modes. The frequency response drops off around 0.7 dB by 9 kHz and around 3.5 dB by 20 kHz. That should be audible for most people in controlled testing. One measured example can be found here, but sadly I couldn't find any response curve for the Fiio.

However, the drop in the frequency response for NOS is just the average effect. The actual effect on individual samples differs.

I don't think the K13 is a bad choice. I'm sure it measures well enough and features like the EQ and a good amount of power are definitely useful.

 

[TaxTime]

There, fixed that for you. It's a large part of what this forum is for.

Turns out, when it comes to DACs, measurements tell us everything we need to know about how they sound. Ownership has no value exept for telling how good it feels to touch.

Master Thread: Are Measurements Everything or Nothing?

Seems like so many review threads get challenged with: 1. Measurements are not everything. 2. You all never listen. 3. I trust my ears, not graphs. 4. I don't listen to graphs. I listen to music. 5. You all must not listen to music at all. 6. Why don't you all buy the best SINAD gear? 7...

www.audiosciencereview.com

I didn't say I was looking for how it SOUNDS. I know what this forum (and a majority of members) are about.

I cam here looking for problems, glitches, things that work well (and things that don't), user EXPERIENCES, etc.

Thank you to those that ACTUALLY contributed with user experience, rather than simply bashing.

 

[xlinknz]

I didn't say I was looking for how it SOUNDS. I know what this forum (and a majority of members) are about.

I cam here looking for problems, glitches, things that work well (and things that don't), user EXPERIENCES, etc.

Thank you to those that ACTUALLY contributed with user experience, rather than simply bashing.

The glitch that has been mentioned here earlier, is it has a LO mode that is 100%, if you connect the K13 to a power amp and use that mode on the unit or accidentely on the remote it will send 100% to your speakers.

Less an issue its Bluetooth has LDAC but no Aptx. Also frustratingly (to me) the mobile app cannot do what the remote does e.g. volume which annoyed me as I expect a mobile app to do what a physical remote does

Everything else works really really well no software issues for me so far at all (although I have not used PEQ yet which can be on the unit, via the mobile app or a browser)

A 2nd USB C on the side is nice as is 2 RCA outs

 

[antcollinet]

Thank you to those that ACTUALLY contributed with user experience, rather than simply bashing.

If the statement "the inherent nature of R2R DACis is that they are not as accurate as delta sigma dacs" meets your definition of "bashing" as opposed to "a summary of reality" - then go right ahead.

 

[IsaacOscar]

I didn't say I was looking for how it SOUNDS. I know what this forum (and a majority of members) are about.

I cam here looking for problems, glitches, things that work well (and things that don't), user EXPERIENCES, etc.

Thank you to those that ACTUALLY contributed with user experience, rather than simply bashing.

I don't have a K13, but I have the Fiio Tiny and Melody, and their PEQ implementations have some bugs, listed here.
I suspect the same bugs may occur on the K13 as well.

I've also had lots of annoying issues with the Fiio control app and website (loading a saved EQ profile incorrectly sets the names of the on device ones, the EQ switch on the website doesn't work, sometimes when playing with the EQ there is some clicking noises added to the sound, but disconnecting/reconnecting my DAC fixes it; sometimes the APP just fails to change anything on the DAC, again a reconnection fixes it, I could go on but those are the most annoying).

 

[Opalius]

Well, I never did get around to measuring the K11 because the K13 just arrived in the mail!

Here is an on-head measurement using the DT 1990. Please note: these were taken using a DIY microphone inserted into my own ear canal to capture the response as close to the eardrum as possible. For the sake of consistency, I repeated this four times without moving the headphones, and I got the same result each time.

You will see three traces on the graph:

• Yellow: NOS Filter
• Blue: OS Filter
• Red: NOS Filter with a +1 dB High Shelf from 6kHz ($Q = 0.71$)

As the measurements show, the shelf EQ effectively compensates for the NOS treble roll-off. And before the "perceptive bias" questions start: yes, I’ve already done a blind test with my wife switching the filters. I can pick out the difference listening to Evanescence's "Bring Me to Life" 10 out of 10 times.

Even with the EQ filter engaged to level the frequency response, the difference persists. To my ears, the difference isn't tonal—it’s the "soundstage." Simply put, the vocals move slightly forward while the hi-hats feel a bit further away.

I know some will be skeptical, but I have listened to the Fallen album for the last 23 years; I know it beat-for-beat. The change isn't massive, it isn't "3D magic," but it is absolutely there. Anyway, hope this data point is useful for the thread. Cheers!

 

[Pokee1016]

If anyone owning Fiio R2R Warmer seeing this - Does it has pop on power ON/OFF?

This is clearly one of the best stacks available for any price regardless 101,000 I put in some Holland 6DJ8 (Amperex). Silence and I can put this back up against anything under $2000.

 

[Pokee1016]

This is clearly one of the best stacks available for any price regardless 101,000 I put in some Holland 6DJ8 (Amperex). Silence and I can put this back up against anything under $2000.

My god rge spellling. My apologies. Quad of Amperex 6DJ8 is reasonable, NOS- matched. My spelling was awful. Let’s have it

 

[emulle]

I don't have a K13, but I have the Fiio Tiny and Melody, and their PEQ implementations have some bugs, listed here.
I suspect the same bugs may occur on the K13 as well.

I've also had lots of annoying issues with the Fiio control app and website (loading a saved EQ profile incorrectly sets the names of the on device ones, the EQ switch on the website doesn't work, sometimes when playing with the EQ there is some clicking noises added to the sound, but disconnecting/reconnecting my DAC fixes it; sometimes the APP just fails to change anything on the DAC, again a reconnection fixes it, I could go on but those are the most annoying).

I'm not sure if this is applicable to the K13 R2R. I tried to convert two different APO configs, one with low shelf filters and one with peaking filters.

Both of these produces a lot of artifacts/clipping. Whereas when I put them in manually into Fiio Control (not taking your conversions into account) they change the sound more or less the same way as they would do through APO on my other DAC.

 

[IsaacOscar]

I'm not sure if this is applicable to the K13 R2R. I tried to convert two different APO configs, one with low shelf filters and one with peaking filters.

Both of these produces a lot of artifacts/clipping. Whereas when I put them in manually into Fiio Control (not taking your conversions into account) they change the sound more or less the same way as they would do through APO on my other DAC.

I suspect this may be a bug in the K13 R2R.
If you can share your EqualizerAPO filte file, I can see if my Fiio dac's also have this problem, and if there's something wrong with my script.

If you want to be extra helpful, you could use Room EQ Wizard and measure the output of the K13 R2R (connect a cable from an output port to a line in or mic port on your computer), with your converted EQ profile active, and again with your hand made one. Oh and with APO EQ but K13 R2R without EQ as well.

 

[emulle]

I suspect this may be a bug in the K13 R2R.
If you can share your EqualizerAPO filte file, I can see if my Fiio dac's also have this problem, and if there's something wrong with my script.

If you want to be extra helpful, you could use Room EQ Wizard and measure the output of the K13 R2R (connect a cable from an output port to a line in or mic port on your computer), with your converted EQ profile active, and again with your hand made one. Oh and with APO EQ but K13 R2R without EQ as well.

Here's the filters, unfortunately I do not have time to measure anything. I read some reports from other users which states that clipping and artifacts can occur with EQ if the master gain is not matched (in negative) of the highest filter gain. I tried to lower the master gain on the generated EQs and the clipping was gone.

Spoiler: unheardlab

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<FiiO_DSP model="FIIO K13 R2R" version="0.0.1">
<module name="EQ">
<eqGroup>
<eqList>
<eq index="0">
<param name="type">0</param>
<param name="freq">30</param>
<param name="gain">4.0</param>
<param name="q">0.50</param>
</eq>
<eq index="1">
<param name="type">0</param>
<param name="freq">170</param>
<param name="gain">-2.5</param>
<param name="q">2.00</param>
</eq>
<eq index="2">
<param name="type">0</param>
<param name="freq">400</param>
<param name="gain">-1.0</param>
<param name="q">2.50</param>
</eq>
<eq index="3">
<param name="type">0</param>
<param name="freq">700</param>
<param name="gain">-1.5</param>
<param name="q">1.50</param>
</eq>
<eq index="4">
<param name="type">0</param>
<param name="freq">1700</param>
<param name="gain">3.0</param>
<param name="q">1.60</param>
</eq>
<eq index="5">
<param name="type">0</param>
<param name="freq">16500</param>
<param name="gain">3.0</param>
<param name="q">1.00</param>
</eq>
<eq index="6">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="7">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="8">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="9">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
</eqList>
<param name="masterGain">8</param>
</eqGroup>
</module>
<styleName>unheardlabs</styleName>
<description>From convert.js unheardlabs.txt</description>
</FiiO_DSP>

Preamp: -4.0 dB
Filter 1: ON PK Fc 30 Hz Gain 4.0 dB Q 0.500
Filter 2: ON PK Fc 170 Hz Gain -2.5 dB Q 2.000
Filter 3: ON PK Fc 400 Hz Gain -1.0 dB Q 2.500
Filter 4: ON PK Fc 700 Hz Gain -1.5 dB Q 1.500
Filter 5: ON PK Fc 1700 Hz Gain 3.0 dB Q 1.600
Filter 6: ON PK Fc 16500 Hz Gain 3.0 dB Q 1.000

Spoiler: lowshelf

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<FiiO_DSP model="FIIO K13 R2R" version="0.0.1">
<module name="EQ">
<eqGroup>
<eqList>
<eq index="0">
<param name="type">1</param>
<param name="freq">55</param>
<param name="gain">2.5</param>
<param name="q">1.41</param>
</eq>
<eq index="1">
<param name="type">1</param>
<param name="freq">120</param>
<param name="gain">4.0</param>
<param name="q">1.41</param>
</eq>
<eq index="2">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="3">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="4">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="5">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="6">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="7">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="8">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
<eq index="9">
<param name="type">0</param>
<param name="freq">20000</param>
<param name="gain">0</param>
<param name="q">1.0</param>
</eq>
</eqList>
<param name="masterGain">5</param>
</eqGroup>
</module>
<styleName>lowshelf</styleName>
<description>From convert.js lowshelf.txt</description>
</FiiO_DSP>

Preamp: -6.7 dB
Filter: ON LSC Fc 55 Hz Gain 2.5 dB Q 1
Filter: ON LSC Fc 120 Hz Gain 4 dB Q 1

 

[Antagron]

Has anyone experienced the unit having some weird power issues? Sometimes my unit's display turns off but the rights and R2R lights are still on. The only way to fix it is to unplug the cord or use the hard power switch. I had to put mine on a smart plug to reset it easily.

 

[Nabooh]

I just received my FiiO K13 R2R, and I'm very satisfied with it.

For context, it replaces a Topping DX5 II. Given the quality feedback and the issues/failures with Topping devices, I'd rather not play with fire. So, I chose to look for another DAC/AMP with PEQ, and the FiiO K13 R2R came along at the perfect time, as I fully trust FiiO's quality control.

Personally, I couldn't care less about all the intellectual masturbation regarding "analog," "warm sound," or Delta-Sigma vs. R2R. All I want is a DAC/AMP that sounds great and has PEQ. I'm an objectivist first and foremost, and while I appreciate nice numbers and data and spent a long time chasing the "best measurement", I have to admit I struggle to see the point of circlejerking over SINAD, THD+N, and the like, if it remains completely inaudible to ordinary mortals (no offense to the bats out there ).

And even though my habit of seeing huge numbers initially made me hesitate to order a DAC/AMP with a SINAD around 75dB when the Topping flirts with 123dB, the reality is: the difference is simply inaudible/non-existent between the FiiO and the Topping. Between the OS and NOS modes, the difference is also far, far from being as massive as some want you to believe. The sound might be ever so slightly warmer, and even then, that mostly depends on the sampling rate of your files.

Anyway, if I have to choose between a DAC/Amp combo that measures incredibly well but seems to break down quickly with a bunch of issues popping up a few months after release, and a DAC/Amp that measures "average" without it being audible to the listener, features PEQ via a smartphone or Web app, and has no major reliability issues... the choice is a no-brainer for me!

Now, I'm not saying that good measurements aren't a criterion for quality or proof of the care a manufacturer puts into their product. But in this case, where the FiiO's "very average" measurement isn't because the product is inherently bad, but simply due to the DAC's architectural choice—which ultimately changes absolutely nothing for me...

 

[antcollinet]

the choice is a no-brainer for me!

Given that choice - me too.

The choice is not that stark though. There are other brands than Fiio and Topping. Though I am unsure how many offer the features of the K13 at the price.