Who can provide comparative DAC-listening experience ?

Dear all,
I would be really keen on reading some comparative listening impressions of the sound quality from different DAC’s that are frequently discussed in this forum. Often we see qualifications such as “very good”, “clear”, “fabulous”, etc. But I would be happy to hear an opinion of those who have the possibility to directly compare two- or more different DAC’s, then telling what they prefer and why, and if the additional cost/hazzle is worth it.

Let’s say the following DAC’s are frequently named:

  • Hifiberry
  • Wolfson Audio Card
  • IQ Audio Pi-DAC
  • Musical Fidelity, VII or V90 DAC
  • HifimeDIY Sabre USB DAC (one of the plenty)
  • Shiit Modi
  • E-DAC / O-DAC (NwAVguy…)

Anybody has any direct listening experience for a nice comparison? (Well, Michelangelo does have for sure… :wink: )
I know it’s all super subjective, but that’s what we like about music, right? Let’s gather, not argue about some strictly subjective impressions!

Best regards, stereo

Seriously, just pick an I2S DAC on ease of set up, support or price.

Alternatively just buy several, make your own choice and sell the one(s) that you don’t find as good (even purchase multiple Pis so you can compare directly). You could sell them on as a kit.

Don’t waste money on a USB DAC if you have a Rasperry Pi. I2S provides much less chance of skips or stuttering.

Check the Wolfson, although someone said it sounded very good, I believe it has a lower output, so if your amp and speakers aren’t so powerful you might struggle to get it loud enough.

I have an ESS9023 I2S DAC which sounds fine to me.

I doubt you will hear differences, and moreover even if you do, you wouldn’t know which one is better and more close to the original, you would just notice they are different.
Pick one compatible with your target price.


It’s just a toss up. I’d go for either the HifiBerry or the IQ Audio Pi-DAC. Both have very helpful developers who post on here and are involved with other Pi music players (RuneAudio, Pimusicbox etc) as well. They are even designing amplifiers and cases to go with their DACs!

You’re better off saving your money and spending it on music or an amp or speakers. Like Lag-na said, there’ll be hardly any difference in sound quality between them, and if you don’t audition them together, you’ll never know anyway.

I agree totally with the views expressed by twerp and Lag-na.

While we would all wish to listen to music like it is performed in the real, however, there are are several major obstacles which prevent us from realising this wish.

For instance, no matter how pure and clean the sound leaving the loudspeaker, the sound falling on the listener’s ears
can be quite different. That sound is borne from the loudspeaker to the listener by the acoustics of the room.
Out in free space, sound travels on and on - unimpeded, unreflected, unrefracted, undefractered and unabsorbed.
The addition of walls, floor and ceiling introduces a whole new set of rules for sound in an enclosed space.

Every home listening room is beset with sound reflected from walls, floor and ceiling. These reflections affect the quality
of the music heard, depending on subjective opinions or on such objective factors as the loudspeakers and listener’s
positions, the size of the room and the reflective properties of the room surfaces.

For example the effect of spaciousness has been variously described as ambience, envelopment, apparant source
width, subjective diffusion, spatial diffusion, etc…
The microphones recording a music event, however, are not placed at the listeners position in the audience and will
record a different spaciousness. The reflection effects picked up by the microphones will differ from those perceived by
the audience.

One could go on considering such effects, parameters, etc. where the sound heard in the listening room from the best
equipment is not the sound that is heard from a life performance.
Under such circumstances is it really meaningful to be unduly concerned about the small differences in the playback sound of different equipment!!!

Which one would have a good gain factor. I intend to use a passive preamp with it.

I’ve got a pidac into the kitchen with a tpa3116 amp and jbl speakers in the kitchen, and a hifiberry with a tripath 2020 amp in my bathroom. Both sound great and I’d happily buy either again. I prefer the pi with the pidac on but that’s only cause its got the better amp and speakers on it

What will the DAC be driving? Hifi amplifier? Powered loudspeakers?

ES9023 based DACs can sound brilliant considering the low cost, but they have a relatively low voltage and high impedance output which means that audio quality can be compromised if connected to an unsympathetic amp (making things sound a bit flat). There are some buffered ES9023 DACs out there that side-step this issue by adding another circuit after the DAC chip that is better at ‘driving’ the next component in the signal chain.

If the input characteristics of your amp are well suited to the raw output from an ES9023 DAC then I’d reckon this would be an excellent choice. If not, then I’d look at buffered ES9023 options as well as other DAC chips with output stages.

Long story short, you’re going to struggle to beat a decent ES9023 DAC if this suits your system.

I would be interested too in audio comparisons of the listed DAC’s in the post.

I also think it would be interesting to compare with some of the “offboad” DAC’s out there too. for example

  • Audioquest Dragonfly (Sabre DAC)
  • Cambridge Audio DACMagic
    -TEAC UD-H01 (Burr Brown DAC)
    -Audiolab Q-DAC (Sabre DAC)

I now have a Wolfson Audio Card so I will be able to make some audio comparisons as I have access to the above DAC’s. I’ll leave that for a rainy day sometime in the future.

In the meantime I attach a comparison that I did which was more based on the capability of the Synology NAS & RPi to playback music with an Audioquest Dragonfly.

nwt-stuff.com/onewebmedia/Bencht … gonFly.pdf

Thanks for this…very interesting.

Dear All,

I’ve been looking at some of the budget DAC’s available and it’s a real minefield out there. It would be an impossible task to do any sensible Audio comparison between all this stuff.

Thinking out aloud I assume that there these are 2 factors that could affect Audio Quality significantly:-

  1. The DAC Chip Design and Manufacture.
  2. The Communications Interface (i2s versus USB).

With this is mind I would propose the following comparisons as potentially interesting/useful.

Comms Interface
Audiophonics DAC Sabre ES9023 vs Audioquest dragonfly (ESS custom DAC Chip ?)

DAC Chip Comparison
Wolfson Audio Card for Raspberry Pi (WM5102) vs IQaudIO Pi-DAC (TI/Burr Brown PCM5122) vs Audiophonics DAC (Sabre ES9023)

Please let me know you’re thoughts as I wouldn’t want to spend a significant period of time on something that nobody would have an interest in. Any suggestions, Other Thoughts ?

I would also be interested if anyone knows:-

  • if the Wolfson WM5102 is based on the Burr Brown PCM5102 ?
  • Is the DAC in the Audioquest DragonFly a custom ESS Chip or does it use a standard one ?

In Rock We Trust - Kevin

Hi Kevin, there are many other items which will effect sound quality - noise floor / grounding issues / interference etc. The DAC chip itself may also be adversely effected by a bad board layout.

You may also want to consider how you want to go about measuring the differences. By ear or by something like an Audio Precision analyser at £15k :slight_smile:

You may also want to consider various use cases? Built in volume control? Is full 2vrms (line level) output needed? Frequency and bit depth support? CPU overhead at different frequencies when using USB.

As you say - a plethora of variables / minefield.

Burr Brown was purchased by TI in 2003 IIRC - don’t think the Wolfson part is Burr Brown derived as the WM5102 is a much more feature rich beast (note I didn’t say better) than the TI PCM5102.

If you don’t have an IQaudio Pi-DAC already drop me a PM and I’ll have one sent to you.

Very best regards,


Hi All!

I really had enough time to compare Hifiberry DAC with BerryNOS1543 from G2Labs.
The conclusion was that BerryNOS1543 delivered clearly more resolution, more intense and precise bass, wider sound stage. But the most important thing for me was the way of interpretation of music focused on rhythm pronounciation.

If someone woudn’t be able to hear the difference between these two DAC’s I would attest nearly deafness :mrgreen:

Measurements to me are only relevant if I can reproduce them audibly;-)

Besides: Both DAC-pi combinations have been comparably upgraded with high performing separated power supply units.

Greetingz, Robert

Here is my listening experience.
On the BBB:

DragonFly much cleaner than HiFiforDIY, and in turn this outshines Behringer

Note to use the DragonFly, one must go into alsamixer and unmute it, then pull up the volume – DragonFly has some internal volume control.
This has to be done each boot, apparently - the setting is not saved.

On the r-pi model B:

The Wolfson is neck-and-neck with the DragonFly.

YMMV, as usual. This is listening to lots of redbook cd’s, flac and alac, and some 96khz files.

Agree with this statement wickrc. The Dragonfly can be used for a lot more things but in a back to back with a RPi Model B as a Music Player it’s a no brainer. IMHO Much more for you’re cash with the Wolfson (lots’ additional I/O options) and a all round better package. i2s is a neat way to go for a fixed piece of equipment in the house.

BTW the Dragonfly is an USB Audio Class V1 product that certainly has some limitations on comms performance.

Audioquest Technical support blamed the Synology Processor for my original issues but I have my doubts !!

nwt-stuff.com/onewebmedia/Bencht … gonFly.pdf

Interesting. I read the pdf which tests the DragonFly at various sampling rates
(using the color light code on the DragonFly that green = 44.1kHz, blue = 48kHz, yellow=88.1kHz, magenta = 96khz

When I play 24/96kHz through my (high end) alpha-dac, which has a digital read out of the bit rate, it says 96.
When I play the same 24/96kHz through the rune on my BBB, the Dragonfly is lit up magenta – ie, 96kHz, as expected.
There is no jitter, no clicking, no popping. I am running over ethernet from my NAS, a mac mini
When I do:
cat /proc/asound/cards

on the BBB, it says that my DragonFly is operating @ full speed
whether that means usb 3.0 or not…I don’t know. But it sure plays hi-rez OK, at least for me.
And it sure says it is working @ 96kHz

Sounds like the BBB might be the next one for me to Investigate :laughing:

FYI USB Speeds are as follows


Yes the Dragonfly is Audio Class 1.1 so will be FULL Speed 12Mbits/s. 24bit 96kHz = 4.6 Mbits/s so in theory no probs :nerd:


Hi Kevin,

Ah, that’s a very useful table, thank you!

So, on the r-pi B+, with the volumio v. 1.4.1, I get decent performance, not quite as good as the BBB to my ears.
(BTW: for others, the Dragonfly has its own (odd) volume control, so it must be unmuted manually by running
sudo alsamixer)

I should try running the BBB and the r-pi-B+ through my other bigger, outboard asynch DACs, to see how much
difference there is – a Benchmark; the alpha dac/usb combo; a PS-audio; and a Cosecant. I’ll report
back how they work.


Look forward to your impressions and results. If you want to know a little bit more about the DAC’s you are using then use this nice simple software tool http://www.thesycon.de/eng/usb_descriptordumper.shtml. This will tell you some secrets about your DAC. :smiley:


For example this tells me that the Calyxx Kong is Adaptive & Not Asynchronous, Supported Bit rates and max power requested. I wouldn’t spend too much time looking at all the config but it might answer the odd question like why the AQ Dragonfly works without Driver Install on Windows. This is because Audio Class 1.x is built into windows. Anything Windows & V2 will require a driver Install.

AH, this is excellent! I did not know such tools were around…thank you!!