How good is bad? (Raspberry Pi 4B headphone output DAC review)

[melomane13]

Fairly confident I can hear the difference in this case!

Are you sure?
it's the same speech given by people saying that they hear differences between amps...

 

[antcollinet]

Are you sure?
it's the same speech given by people saying that they hear differences between amps...

Not when the measurements show that the distortion and noise are well above the level of audibility. Not to mention a rather not flat FR.

 

[Gradius]

So the output is just horrible.

What about raspi5 ?

 

[CedarX]

So the output is just horrible.

What about raspi5 ?

Raspi 5 is much better: no noise or distortion… (it has no audio output!)

 

[benanders]

Is there any knowing whether the HP-out was dropped from RPi 5 due to power draw vs. perhaps the Foundation already being aware it was measurably problematic?

If the HP-out was dropped bc of reasons other than measurement, I bet anyone from the Foundation seeing this review would think “Phew!!”

 

[CedarX]

Is there any knowing whether the HP-out was dropped from RPi 5 due to power draw vs. perhaps the Foundation already being aware it was measurably problematic?

If the HP-out was dropped bc of reasons other than measurement, I bet anyone from the Foundation seeing this review would think “Phew!!”

Space constraints on the RPI 5 board is the official explanation.

 

[melomane13]

what is the charge impedency input impedance of audio interface you have used?

 

[antcollinet]

charge impedency

Hi

I don't think that has translated correctly - can you describe what you mean here?

 

[Linards]

what is the charge impedency you have used?

@melomane13 , if you mean input impedance of audio interface in measurements, I used microphone (XLR) input which is 1.5 kOhm.

 

[somebodyelse]

Is there any knowing whether the HP-out was dropped from RPi 5 due to power draw vs. perhaps the Foundation already being aware it was measurably problematic?

If the HP-out was dropped bc of reasons other than measurement, I bet anyone from the Foundation seeing this review would think “Phew!!”

It was never meant to be a high quality output - just something better than a buzzer as part of an educational platform. If anything it's a surprise that general purpose PWM outputs can perform that well as an audio output. It was originally substantially worse - see this thread for the discussion of the improvements made in a driver update some years back.

 

[m8o]

Wow. I didn't even know it existed. I bought an Allo hat for mine.

I have a few. Are they still in business? Or did you buy NOS or used?

 

[Dennis_FL]

I have a few. Are they still in business? Or did you buy NOS or used?

I bought it new from Allo.

 

[m8o]

I bought it new from Allo.

It's a relief to learn they have new stock to sell and still in operation; after more than a year id given up hope.

No Boss2 Acrylic but quite surprised Katana is available to order.

 

[Dennis_FL]

It's a relief to learn they have new stock to sell and still in operation; after more than a year id given up hope.

No Boss2 Acrylic but quite surprised Katana is available to order.

I’m afraid it was a few years ago when I bought it.

 

[m8o]

I’m afraid it was a few years ago when I bought it.

Me too. But then recently had to buy a few Boss v1.2 dacs when i needed them last year from the UK as there was no new stock since the covid shortages and none continued to be found anywhere else; and that e-tailer ran out soon after. But glad to see new stock now.

 

[Blew]

It was never meant to be a high quality output - just something better than a buzzer as part of an educational platform. If anything it's a surprise that general purpose PWM outputs can perform that well as an audio output. It was originally substantially worse - see this thread for the discussion of the improvements made in a driver update some years back.

Given the ~60dB measured here it would appear that these tests were done without setting

audio_pwm_mode=2

It would be interesting to see how it measures with that set. It may just be near to 96dB of CD audio after all!

 

[Linards]

Given the ~60dB measured here it would appear that these tests were done without setting

audio_pwm_mode=2

It would be interesting to see how it measures with that set. It may just be near to 96dB of CD audio after all!

According to documentation online, it seems that audio_pwm_mode=2 is default setting on RPI. With a high probability it was on during measurments. But thanks for the comment. I will keep it in mind if I am to revisit this topic.