Review and Measurements of PiFi DAC+ Pi Sound Card

[xxhh1234]

He doesn't recommend it because of it's poor performance. I am at a loss as to what is so hard to understand the review

Pifi DAC + is already a very cheap product five years ago, and it is not appropriate to compare it with the new product. Why not compare RPI-ES9038Q2M and hifiberry dac+pro, which are both new products at the same price? Do you think why the products of flagship chip ES9038Q2M are so cheap and have good performance and support DSD /DoP, while the products of PCM5122 are so expensive, who should be recommended?

 

[guenthi_r]

Filter response is different flavor of bad from HifiBerry:
View attachment 31769

@amirm dont forget, this DAC have selectable filters!
Take look @Archimago site for the PCM5122
I preferred the sound of the ringing-less low latency FIR filter.

 

[amirm]

@amirm dont forget, this DAC have selectable filters!
Take look @Archimago site for the PCM5122
I preferred the sound of the ringing-less low latency FIR filter.

I can't change any filters with the software I have.

 

[tw99]

There are some clues in this repair video of it:
It references a website that talks about the necessary mods.

Following the link, the main problem is supposedly with a 5V-5V converter, which allows noise from the RPi supply to pollute the DAC, plus some other smaller issues.

It's interesting, as I'd always assumed that these Pi HAT devices were much of a muchness.

 

[BDWoody]

Pifi DAC + is already a very cheap product five years ago, and it is not appropriate to compare it with the new product. Why not compare RPI-ES9038Q2M and hifiberry dac+pro, which are both new products at the same price? Do you think why the products of flagship chip ES9038Q2M are so cheap and have good performance and support DSD /DoP, while the products of PCM5122 are so expensive, who should be recommended?

He measures what he gets.
He doesn't do 'shootouts.'
It is up to you to review and/or compare the numbers to interpret what they mean.
Draw whatever conclusions you like.

You keep making it about cheap vs expensive or new vs old... It isn't any of that.
You seem to have some kind of block here. It really isn't that hard.

If you think you've got a better way to do it, the world awaits your version. Have at it. Buy yourself an analyzer and show amir how it's done. Otherwise trying to make a pointless point is getting old...

 

[guenthi_r]

The Price is not correct, the PIFI DAC 2.0+ is about 13 EUR incl shipping from China.

 

[AudioSceptic]

The Price is not correct, the PIFI DAC 2.0+ is about 13 EUR incl shipping from China.

Please read the thread. Amir quoted the "official" price and, as he said in reply to my question, does not do price searches. In any case, he considers the performance to be inadequate, regardless of price. Can't we just move on?

 

[xxhh1234]

He measures what he gets.
He doesn't do 'shootouts.'
It is up to you to review and/or compare the numbers to interpret what they mean.
Draw whatever conclusions you like.

You keep making it about cheap vs expensive or new vs old... It isn't any of that.
You seem to have some kind of block here. It really isn't that hard.

If you think you've got a better way to do it, the world awaits your version. Have at it. Buy yourself an analyzer and show amir how it's done. Otherwise trying to make a pointless point is getting old...

For the testing of audio products, there are only a few test concerns, price, performance and function. It's better not to use the wrong price as the introduction of the product at the beginning of the test, which will make the reader feel more expensive, but it's not so. Since website testing is mainly about testing performance, please ignore function and price.
In addition, since it is a separate test, do not compare, after all, the two products are years behind, just like you can not use the iPhone 7 and iPhone X to compare performance. If you need to compare, then it is better to use the products of similar age and price to compare. For example, RPI-ES9083Q2M
Different opinions about the test results are also for the website to be more professional and better. Don't you think so?

 

[xxhh1234]

He measures what he gets.
He doesn't do 'shootouts.'
It is up to you to review and/or compare the numbers to interpret what they mean.
Draw whatever conclusions you like.

You keep making it about cheap vs expensive or new vs old... It isn't any of that.
You seem to have some kind of block here. It really isn't that hard.

If you think you've got a better way to do it, the world awaits your version. Have at it. Buy yourself an analyzer and show amir how it's done. Otherwise trying to make a pointless point is getting old...

I have never questioned the test results. I believe the test results are real and effective. I also thank the website and Amirm for their time and energy. I also hope that the website can be more professional and better, so I put forward my different views. Since the website mainly tests the performance of these audio products rather than the price and function, then don't mention the price in the test, and give some background information about the product before the test, such as what chip scheme to use and what kind of chip scheme to use. When the product, whose product and other product information, since the test, then we do more professional, better, which will help more players understand the product, understand the quality of the product, this is not better?

 

[BDWoody]

Different opinions about the test results are also for the website to be more professional and better. Don't you think so?

I think you don't understand the site enough to be able to have valid input on its professionalism.

You continue to argue points that really aren't relevant to the site and the purpose it serves. This has been stated to you repeatedly, yet you continue to repeat your claims of unfairness, completely ignore any answers you may have received, and think that saying the same thing over and over is the key to success.

If you think there are such glaring errors in method and professionalism, head on out and buy yourself an analyzer and have at it. Otherwise, repeating yourself ad nauseum is just...well...ad nauseum.

Discussion of a product in the threads is more about the product, rather than trying to discredit the review or the site if you don't happen to like it.

 

[xxhh1234]

I think you don't understand the site enough to be able to have valid input on its professionalism.

You continue to argue points that really aren't relevant to the site and the purpose it serves. This has been stated to you repeatedly, yet you continue to repeat your claims of unfairness, completely ignore any answers you may have received, and think that saying the same thing over and over is the key to success.

If you think there are such glaring errors in method and professionalism, head on out and buy yourself an analyzer and have at it. Otherwise, repeating yourself ad nauseum is just...well...ad nauseum.

Discussion of a product in the threads is more about the product, rather than trying to discredit the review or the site if you don't happen to like it.

Maybe I don't understand about this website, but I have never belittled it, harmed or defamed anyone or product. I'm just expressing my opinion. Since you and I can't convince each other to accept it, why use these words?So stop it ，hostile friend.

 

[BDWoody]

Maybe I know enough about this website, but I have never belittled it, harmed or defamed anyone or product. I'm just expressing my opinion. Since you and I can't convince each other to accept it, why use these words?So stop it ，hostile friend.

Where did I belittle, harm or defame you? What words have been hostile? Seems as though you are absolutely determined to miss the point.

No hostility, my very dense friend...just letting you know my opinion of your opinions.

Have a nice day.

 

[xxhh1234]

Where did I belittle, harm or defame you? What words have been hostile? Seems as though you are absolutely determined to miss the point.

No hostility, my very dense friend...just letting you know my opinion of your opinions.

Have a nice day.

OK, I'm a fresh , so I salute the old fans of the website. Have a nice day.

 

[somebodyelse]

I can't change any filters with the software I have.

I think you can, just not via the web interface. Ropieee has ssh enabled by default (username and password both root, and no prompt to change it - I wish people wouldn't do things like this...) and once connected you can run alsamixer to get to the settings. I've just tried it with RopieeXL and the hifiberry miniamp, and had to run 'alsamixer --card 1' as there was no card 0. Audiophonics have a screenshot showing filter settings in alsamixer for the ES9038Q2M, and by my reading of the driver source there should be something similar for the pcm512x, assuming it's not overridden elsewhere. The MiniAmp has no settings to change.

 

[SmackDaddies]

It is amazing that such a cheap piece of junk that doesn't perform better than any modern on board dac on a cellphone or computer gets such vociferous defense.

 

[Interference]

It seems to me that the noise floor of this little toy is quite OK, operating it a few dBs below full scale could get it into a sweet spot.

 

[captain paranoia]

I bought one of these for about £10 delivered (PiFi DAC+ v2). I confess that the mention of the reasonable caps swung me towards this than other, equally cheap 5122 DACs.

It sounds okay, but I don't claim to have 'super ears'.

However, this review has prompted me to look a bit closer at the design, given the specified performance of the 5122.

Apart from the observations already mentioned (unnecessary DC-DC converter, ground side inductor), I'm discovering a whole bunch of poor design decisions.

So, those nice Wima caps (assuming they're genuine parts...)? Being used for PSU decoupling. A decent MLC would have been better, and allowed closer location to the chip. The Elna 2u2 electrolytics are used on the charge pump (and they're too far away from the associated pins). Again, MLC non-polarised would have been much more suitable (I've yet to determine if the Elnas are actually polarised correctly in-circuit...). The Wimas may be nice for audio, but for digital decoupling, they're wrong.

There's no signal side ground flooding, as recommended for crosstalk reduction (and general health...).

That 10u 'Nichicon'? Not well placed for good PSU decoupling, and, again, an SMD cap probably better suited.

The audio outs are run under the DAC, thus passing by the charge pump circuitry. Rotating the chip by 90 degrees would have allowed the analogue to be routed straight to the output R/C coupling. Admittedly the 5122 pinout is a bit questionable; one of the analogue outs is next to the charge-pump developed Vneg. I'd have surrounded them by Avdd/Agnd.

Decoupling of the various supplies is inaedquate, or too remote from the pins.

Still to finish tracing the mode selection, and work out if it's using a fixed h/w config, or supports I2C or SPI config. The device is pretty configurable in terms of upsampling/filtering; looks like the filter selection Amir tested was one of the more basic filters, as there are high attenuation filters available that knock the stop band down much more than the results presented here.

It's almost tempted me to fire up my long unused D/A design skills, and knock up a design 'done properly'...