• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Review and Measurements of PiFi DAC+ Pi Sound Card

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
This is a review and detailed measurements of itead's PiFi DAC+ Version 2 Raspberry Pi sound card (DAC). It was kindly purchased by a member and drop shipped to me. It costs US $40 but company has it on sale for US $36.50. It competes with the HifiBerry DAC+ whose pro version I recently reviewed. Unlike that unit, the PiFi DAC+ does not have independent clock source and must be using I^2 S clock. It is said that this clock has more jitter so will be interesting to see if that is the case in measurements.

Here is what it looks like:

iTead PiFi DAC+ Raspberry Pi DAC Audio Review.jpg

The common TI PMC5122 DAC chip is used which retails for $2.40. So not hard to imagine how these things sell so cheaply. There are a few extra bits on this such as a post regulator an IR input port and a 3.5 mm jack. The latter is just mirroring RCA connectors and as such, has high impedance. While it will make sound, don't be tempted to use it as a headphone jack as it will likely have anemic bass and low power levels.

DAC Audio Measurements
As with the other Pi DACs I have tested, I used Ropieee OS image to stream content to the DAC over the network using Roon media player. Here is our dashboard view when playing a 24-bit, 1 kHz tone:

iTead PiFi DAC+ Raspberry Pi DAC Audio Measurements.png


The output level is a hair higher than HifiBerry DAC at a little over 2 volt which is fine. What is not so fine is the much lower SINAD (signal over noise and distortion). While the HifiBerry DAC+ PRO achieved 92 dB, we have a shortfall of 7 dB. This places the PiFi in our bottom tier of all DACs tested to date:
Best Audio DACs reviewed and Measured 2019.png


Multitone test likewise shows a step down in performance:

iTead PiFi DAC+ Raspberry Pi DAC 7-tone multitone Audio Measurements.png


I like to see 95+ dB here and we are way short of that at just 78 dB.

Jitter response while again worse that HifiBerry DAC+ Pro, is still quite good:
iTead PiFi DAC+ Raspberry Pi DAC Jitter Audio Measurements.png


So not sure what all the screaming is about regarding the I^S Clock. The DAC chip's PLL is likely cleaning it up good.

Filter response is different flavor of bad from HifiBerry:
iTead PiFi DAC+ Raspberry Pi DAC White Noise Filter Response Audio Measurements.png


As with Jitter, probably not an audible concern though.

Conclusions
Using the same recipe as the HifiBerry DAC+, the itead PiFi DAC+ falls short in measurements. We were barely good enough with Hifiberry but now are behind the game with some 7 dB higher amount of distortion. Seeing how itead PiFi DAC+ is more expensive anyway, I see no reason to recommend it over HifiBerry offerings.


------------
As always, questions, comments, recommendations, etc. are welcome.

I know it is still summer but the panthers are already making their Christmas list and there are some really high dollar gifts in there! No way would I want to spend so much of my own money on them. So please donate generously to make sure we don't have angry panthers come end of the year! https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 

Eirikur

Senior Member
Joined
Jul 9, 2019
Messages
318
Likes
509
Thanks Amir, disappointing outcome though...
I'd love to see a cheap Pi DAC breaking the 17-18 bits "barrier" of clean output.
 
Last edited:

tw99

Senior Member
Joined
Mar 13, 2019
Messages
467
Likes
1,069
Location
West Berkshire, UK
Since this board is simple, and similar to the Hifiberry, can we deduce what's causing the big difference in performance ? E.g. is it board layout, the clocking scheme, setup of the DAC, inferior analogue components... ?
 

CRKebschull

Member
Forum Donor
Joined
Feb 7, 2019
Messages
87
Likes
94
Location
Germany
I have a few more boards to go. :)

I think it's great that you're looking at various PI DAC HATs. I have a HifiBerry DAC+ Pro XLR (which you recently tested -- very cool) and I very much would like to see the real performance of more HATs.

I find the PI + DAC HAT combination to be astoundingly compact, simple, and affordable and easy to use (with Roon in my case) and the entire package is so small, that the only thing you can see of my stereo system is the NC400 mono blocks and the speakers, the rest (PI, DAC HAT, PSU) is invisible behind the books of the book shelf.

Here is another new PI DAC HAT (DACBerry Pro+) which looks rather interesting, with a PCM5242 DAC and a TPA6120A2 headphone amp, with XLR daughterboard option.
 
Last edited:

Ralf Stocker

Active Member
Joined
Jan 22, 2019
Messages
147
Likes
90
I would have a question: How should the filter response of a DAC filter actually look like? You can see very often in app notes, that the -3dB point is at 100 kHz or more. This is justified, that the 20 kHz should not be influenced.
 

mt196

Active Member
Joined
Apr 7, 2019
Messages
229
Likes
156
Location
Rome, Italy
Can I ask why all this interest on these boards instead that "real" desktop DACs?

I know you can find the good one among them, like the khadas tone board, but it seems likely the exception rather than the norm
 

CRKebschull

Member
Forum Donor
Joined
Feb 7, 2019
Messages
87
Likes
94
Location
Germany
Can I ask why all this interest on these boards instead that "real" desktop DACs?

I know you can find the good one among them, like the khadas tone board, but it seems likely the exception rather than the norm

Because the RPi is an excellent streaming platform, running anything from Squeezeplayer to Roon endpoints, stable 24/7 with very little energy consumption, connected via ethernet or WiFi.

One way to go is to connect an external DAC to your RPi streamer - via USB for example, or other methods. These (often but not always) require their own PSU. And interconnects, and space.

If you want to stream to your Amp, and you have chosen RPi to do that, you maybe don't want any more external appliances to produce the analog signal. You maybe want to have a DAC right on your RPi, ie. a HAT. In one box, with SE or balanced output.

The costs run from minimal to still very affordable for SINAD in the same league as top-end "real" desktop DACs, because ofcourse Pi HAT DACs are just as real.

[PS, the Khadas Tone Board is not a Pi HAT DAC. ;) ]
 

miero

Active Member
Joined
Aug 1, 2018
Messages
241
Likes
289
@amirm can you make a dashboard for both 44100fs and 48000fs? I guess one of them will be worse due to a shared clock.
 

xxhh1234

Member
Joined
Aug 21, 2019
Messages
19
Likes
1
I feel that there are some problems with this test and comparison. Assuming that all test environments and configurations are consistent, the test data is also true and valid. I think the choice of test objects is problematic. First of all, we should use the same method for comparison, such as Hifiberry DAC + and PIFIDAC + test, instead of PIFIDAC + without crystal oscillation and hifiberry DAC + pro with two kinds of crystal oscillation. This is biased in selecting comparison objects. Secondly, the price differences of hifiberry DAC + ($28.90), hifiberry DAC + Pro ($39.90), and pifi DAC + ($13.6) , and the price of pifi dac+ advantages are very obvious.
If you compare indicators, I recommend using the same scheme of hifiberry DAC + and pifi DAC +,and the results are more interesting.
1566388954(1).png
1566388842(1).png
1566389405(1).png
 

CRKebschull

Member
Forum Donor
Joined
Feb 7, 2019
Messages
87
Likes
94
Location
Germany
I feel that there are some problems with this test and comparison. Assuming that all test environments and configurations are consistent, the test data is also true and valid. I think the choice of test objects is problematic. First of all, we should use the same method for comparison, such as Hifiberry DAC + and PIFIDAC + test, instead of PIFIDAC + without crystal oscillation and hifiberry DAC + pro with two kinds of crystal oscillation. This is biased in selecting comparison objects. Secondly, the price differences of hifiberry DAC + ($28.90), hifiberry DAC + Pro ($39.90), and pifi DAC + ($13.6) , and the price of pifi dac+ advantages are very obvious.
If you compare indicators, I recommend using the same scheme of hifiberry DAC + and pifi DAC +,and the results are more interesting.

Well, Amir I think just measures the equipment (DACs, Amps, what have you) in the configuration that he has it. Since the measurements are objective, you can compare them with alle the other equipment, which measures better or worse, and costs less or more.

Separately from that measurement, you can suggest other configurations to measure that are of interest to you, and sometimes Amir even ends up measureing different versions and configurations of the same product.

However, there is nothing fair or unfair about the objective measurements of a piece of equipment, because unless the gear is broken or the measurements have problems, the measurements stand on their own as objective points of data.

We can then make subjective evaluations saying, for example, that for a RPi HAT DAC of price $30 performance of XYZ is acceptable, but we wish to see performance of at least ABC for a HAT DAC of price $200.

You pays your moneys and you makes your choice. And having seen the measurements, you'll be able to make a good informed choice.
 

guenthi_r

Active Member
Joined
May 8, 2019
Messages
130
Likes
103
Location
Europe/Austria
I would have a question: How should the filter response of a DAC filter actually look like? You can see very often in app notes, that the -3dB point is at 100 kHz or more. This is justified, that the 20 kHz should not be influenced.

This DAC has some selectable Filters:
Datasheet page 22+

I have this DAC and it sounds very well.

@amirm maybe this information should be added.
 
Last edited:

xxhh1234

Member
Joined
Aug 21, 2019
Messages
19
Likes
1
Well, Amir I think just measures the equipment (DACs, Amps, what have you) in the configuration that he has it. Since the measurements are objective, you can compare them with alle the other equipment, which measures better or worse, and costs less or more.

Separately from that measurement, you can suggest other configurations to measure that are of interest to you, and sometimes Amir even ends up measureing different versions and configurations of the same product.

However, there is nothing fair or unfair about the objective measurements of a piece of equipment, because unless the gear is broken or the measurements have problems, the measurements stand on their own as objective points of data.

We can then make subjective evaluations saying, for example, that for a RPi HAT DAC of price $30 performance of XYZ is acceptable, but we wish to see performance of at least ABC for a HAT DAC of price $200.

You pays your moneys and you makes your choice. And having seen the measurements, you'll be able to make a good informed choice.
I agree with you, but if the measurements are objective, at least the samples used for comparison are similar, and if not, the results of the tests are less objective.For example, pifi DAC + and hifiberry DAC + both use internal clocks. I think the test results should be similar. And hifiberry DAC + Pro uses external clocks. I think the test results are understandable .As for price comparison, I think the tester got the highest price . Because I have this dac. It's really cheap, sounds good.
 

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,948
Likes
22,625
Location
Mid-Atlantic, USA. (Maryland)
I agree with you, but if the measurements are objective, at least the samples used for comparison are similar, and if not, the results of the tests are less objective.For example, pifi DAC + and hifiberry DAC + both use internal clocks. I think the test results should be similar. And hifiberry DAC + Pro uses external clocks. I think the test results are understandable .As for price comparison, I think the tester got the highest price . Because I have this dac. It's really cheap, sounds good.

He isn't specifically doing comparison tests as such...he tests what he gets and the results are what they are. He has a ratings board that shows them all on a continuum, so you can see where they all stand. Price really isn't relevant to the analysis part...

How can numerical results be less objective? Is that like being less pregnant?
 

CRKebschull

Member
Forum Donor
Joined
Feb 7, 2019
Messages
87
Likes
94
Location
Germany
He isn't specifically doing comparison tests as such...he tests what he gets and the results are what they are. He has a ratings board that shows them all on a continuum, so you can see where they all stand. Price really isn't relevant to the analysis part...

How can numerical results be less objective? Is that like being less pregnant?

Indeed, the Pifi DAC+ doesn't have external clock, HifBerry DAC+ Pro/XLR does (as do many DAC HATs), and it performs better. SINAD is SINAD.

The part where Amir then gives his subjective view of the product is, I think, pretty clearly his view - and he thinks that for the price, you'd be better off getting the marginally more expensive HifiBerry. That view does not change anything about how any of the DACs measured.
 
Top Bottom