Review and Measurements of PiFi DAC+ Pi Sound Card

[xxhh1234]

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?

I have not questioned the authenticity of the results, but if the contrast products are very different, the results must be very different, the author will give you misunderstanding in the analysis. For example, when you compare the data with pifi DAC + for $13 and hifiberry DAC + Pro for $40, you say that pifi DAC + is much worse than hifiberry DAC + pro. The original two products used internal clock and external clock. The test results will certainly be different. After all, the price difference is more than double. The price difference between the two products is great, and the way of product realization is also great, so the test results must be different.

 

[xxhh1234]

Yes, but most of the players are not wealthy. People prefer to buy products with high cost performance, low price and acceptable performance. Of course, players are willing to spend more money to bring better performance, which is also the law of the market.

 

[Eirikur]

Can I ask why all this interest on these boards instead that "real" desktop DACs?

Fair question: I want to make a truly portable all-in-one system: Pi bolted to a touchscreen, and a clean enough DAC screwed to the Pi. The Pi4 even supports dual-screen HDMI!

 

[CRKebschull]

I have not questioned the authenticity of the results, but if the contrast products are very different, the results must be very different, the author will give you misunderstanding in the analysis. For example, when you compare the data with pifi DAC + for $13 and hifiberry DAC + Pro for $40, you say that pifi DAC + is much worse than hifiberry DAC + pro. The original two products used internal clock and external clock. The test results will certainly be different. After all, the price difference is more than double. The price difference between the two products is great, and the way of product realization is also great, so the test results must be different.

No, sadly the price difference (in this case what, $26? Right, that's 100%, but of course 100% of a small price is still a small sum) does not mean anything in terms of performance, nor does it allow you to draw conclusions about the implementation.

The Soekris DAC1421 measures slightly worse than the Pifi DAC+, and it costs 32X more (EU780 vs. EU24 for not one bit better performance.)

 

[SmackDaddies]

Yes, but most of the players are not wealthy. People prefer to buy products with high cost performance, low price and acceptable performance. Of course, players are willing to spend more money to bring better performance, which is also the law of the market.

Which is exactly what Amir said. And his tests show that the cheap one is no better than your average cell phone or computer dac, so there is no reason to spend any money on the inferior one (my addition, not Amir's). That way you really save money.

 

[mt196]

Fair question: I want to make a truly portable all-in-one system: Pi bolted to a touchscreen, and a clean enough DAC screwed to the Pi. The Pi4 even supports dual-screen HDMI!

I'm learning that this little boxes can be used for multimedia purposes and since their little dimensions they are very useful.

Well I never needed something like that (I live with my parents so I don't manage the living room and TV), so probably that is why I was unaware of all this interest about them

 

[janderclander14]

I have a few more boards to go.

Just to have a reference, could it be possible to do some basic measurements of the 3.5mm audio output of the Raspberry itself? The latest driver integrated into Raspbian claims to improve the audio quality a lot (which is done in a pwm pure digital way) and be near cd quality. Maybe you will find that these simple dacs are not needed anymore.

 

[CRKebschull]

Just to have a reference, could it be possible to do some basic measurements of the 3.5mm audio output of the Raspberry itself? The latest driver integrated into Raspbian claims to improve the audio quality a lot (which is done in a pwm pure digital way) and be near cd quality. Maybe you will find that these simple dacs are not needed anymore.

Unfortunately a DAC HAT (some of which like the ALLO Katana or the ApplePi DACs perform in the top Tier of all DACs tested here on ASR) of any kind is still needed if you don't want really horrible audio performance. I do not know if the new RPi 4B has better measurements, but the RPi 3 A/B + have very poor performing on-board audio.

 

[xxhh1234]

I think since it is a comparative test, we should choose the same product test, either at the same price or with the same product plan. HiFi DAC + and hifiberry DAC + Pro are quite different in these two points, so the significance of comparative testing is not significant. Hifiberry DAC + Pro is not comparatively tested with es9038q2m products with similar price. Obviously, the index and function of pcm5122 are not as good as es9038q2m. According to this comparative test, can we conclude that hifiberry DAC + Pro is poor?

 

[SmackDaddies]

I think since it is a comparative test, we should choose the same product test, either at the same price or with the same product plan. HiFi DAC + and hifiberry DAC + Pro are quite different in these two points, so the significance of comparative testing is not significant. Hifiberry DAC + Pro is not comparatively tested with es9038q2m products with similar price. Obviously, the index and function of pcm5122 are not as good as es9038q2m. According to this comparative test, can we conclude that hifiberry DAC + Pro is poor?
View attachment 31783View attachment 31784

Amir welcomes reader submissions of equipment to review. Feel free to purchase and send it to him and I am sure he will do so. It's not like Amir gets paid to do this.

 

[xxhh1234]

In addition, I also asked that since it is a comparative test, we should test the validity of the data of the comparative test. Are there many test factors considered consistently?
1. To test the reliability of product samples, at least multiple test samples should be used.
2. The hardware environment of the test, including whether the test instrument has passed the calibration test, whether the tested cables are consistent, whether the test power supply is consistent, and whether the power supply indicators are consistent.
3. The test software environment, including raspberry pie configuration, system version, kernel version, test software version, audio file, since the chip is the same, it also includes whether the configuration registers of sound card are consistent, whether the registers are dump out to compare consistency, driver differences and so on, which all affect the test results. Important factors. Perhaps different filter configurations and different jitter printing configurations will lead to different results.
We all hope that the website can be better and more professional.

 

[SmackDaddies]

In addition, I also asked that since it is a comparative test, we should test the validity of the data of the comparative test. Are there many test factors considered consistently?
1. To test the reliability of product samples, at least multiple test samples should be used.
2. The hardware environment of the test, including whether the test instrument has passed the calibration test, whether the tested cables are consistent, whether the test power supply is consistent, and whether the power supply indicators are consistent.
3. The test software environment, including raspberry pie configuration, system version, kernel version, test software version, audio file, since the chip is the same, it also includes whether the configuration registers of sound card are consistent, whether the registers are dump out to compare consistency, driver differences and so on, which all affect the test results. Important factors. Perhaps different filter configurations and different jitter printing configurations will lead to different results.
We all hope that the website can be better and more professional.

Your a "new" member. Before you start impugning the character and professionalism of Amir, you should take Mark Twain's advice.

 

[SmackDaddies]

In addition, I also asked that since it is a comparative test, we should test the validity of the data of the comparative test. Are there many test factors considered consistently?
1. To test the reliability of product samples, at least multiple test samples should be used.
2. The hardware environment of the test, including whether the test instrument has passed the calibration test, whether the tested cables are consistent, whether the test power supply is consistent, and whether the power supply indicators are consistent.
3. The test software environment, including raspberry pie configuration, system version, kernel version, test software version, audio file, since the chip is the same, it also includes whether the configuration registers of sound card are consistent, whether the registers are dump out to compare consistency, driver differences and so on, which all affect the test results. Important factors. Perhaps different filter configurations and different jitter printing configurations will lead to different results.
We all hope that the website can be better and more professional.

Are you the manufacturer or rep for the manufacturer of the board? If so, you should identify yourself as such.

 

[CRKebschull]

In addition, I also asked that since it is a comparative test, we should test the validity of the data of the comparative test. Are there many test factors considered consistently?
1. To test the reliability of product samples, at least multiple test samples should be used.
2. The hardware environment of the test, including whether the test instrument has passed the calibration test, whether the tested cables are consistent, whether the test power supply is consistent, and whether the power supply indicators are consistent.
3. The test software environment, including raspberry pie configuration, system version, kernel version, test software version, audio file, since the chip is the same, it also includes whether the configuration registers of sound card are consistent, whether the registers are dump out to compare consistency, driver differences and so on, which all affect the test results. Important factors. Perhaps different filter configurations and different jitter printing configurations will lead to different results.
We all hope that the website can be better and more professional.

I don't think it makes any difference. Amir uses his RPi with various DAC HATs and measures the results. The THD, IMD, SINAD, etc., are as they are. Either the RPI supports the DAC HAT or it doesn't. All these HATs seems to be using Ti PCM5XXX DACs, which are supported in the ARM6/7 Linux Kernel, and by ALSA. The Hardware outputs what's measured, and the results are plain to see.

 

[xxhh1234]

Amir welcomes reader submissions of equipment to review. Feel free to purchase and send it to him and I am sure he will do so. It's not like Amir gets paid to do this.

Yes, these test samples should be provided to the tester for free. If the tester purchases the product at his own expense, it may have a certain subjective impact on the test results. For example, in this test, if Amir bought PIFI DAC + for $13, compared with hifiberry DAC + Pro for $40, it might not be so harsh.

 

[xxhh1234]

Are you the manufacturer or rep for the manufacturer of the board? If so, you should identify yourself as such.

No, I have bought this pifi DAC +, built roon for my friends, and bought hifiberry DAC +. I think the price of hifiberry DAC + is a little expensive, and the voice doesn't surprise me. So I feel that the comparison should at least be similar. It's unfair to not compare products with big differences.I personally work in the EMC testing laboratory. Mainly dealing with testing

 

[BDWoody]

I have not questioned the authenticity of the results, but if the contrast products are very different, the results must be very different, the author will give you misunderstanding in the analysis. For example, when you compare the data with pifi DAC + for $13 and hifiberry DAC + Pro for $40, you say that pifi DAC + is much worse than hifiberry DAC + pro. The original two products used internal clock and external clock. The test results will certainly be different. After all, the price difference is more than double. The price difference between the two products is great, and the way of product realization is also great, so the test results must be different.

Price has zero to do with anything.

He showed how the $9.00 apple dongle outperformed a $13,000 disaster.

He doesn't do comparison tests...he just tests. He may reference something else, but these aren't like head to head matchups. That's up to you.

 

[BDWoody]

Yes, these test samples should be provided to the tester for free. If the tester purchases the product at his own expense, it may have a certain subjective impact on the test results. For example, in this test, if Amir bought PIFI DAC + for $13, compared with hifiberry DAC + Pro for $40, it might not be so harsh.

Are you kidding? His tester will know what he paid?

 

[xxhh1234]

Are you kidding? His tester will know what he paid?

Of course he knows the price.Maybe he was impressed by the $40 price.

 

[BDWoody]

Yes, but most of the players are not wealthy. People prefer to buy products with high cost performance, low price and acceptable performance. Of course, players are willing to spend more money to bring better performance, which is also the law of the market.

Price and performance in this realm is almost completely disconnected... Competence with regard to a DAC is a commodity, and has been for 20+ years.

Elegance on the other hand is rare. The new little balanced headphone DAC being an example.

Won't sound better, but it's slick as hell.