[SlimDevices: Audiophiles] MEASUREMENTS: Raspberry Pi 3 & HiFiBerry DAC+ Pro...

[Archimago]

Hey guys. Just posted some measurements on the Pi3 + HiFiBerry DAC
combo. I suspect many of us here are already doing this kind of thing
with Pi computers and various other SBC's. Certainly makes an
inexpensive system for a Squeezebox replacement with piCorePlayer
installed...

http://archimago.blogspot.com/2016/10/measurements-raspberry-pi-3-hifiberry.html



Archimago's Musings: (archimago.blogspot.com) A 'more objective'
audiophile blog.
------------------------------------------------------------------------
Archimago's Profile: http://forums.slimdevices.com/member.php?userid=2207
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[Jeff07971]

Post by Archimago
Hey guys. Just posted some measurements on the Pi3 + HiFiBerry DAC
combo. I suspect many of us here are already doing this kind of thing
with Pi computers and various other SBC's. Certainly makes an
inexpensive system for a Squeezebox replacement with piCorePlayer
installed...
http://archimago.blogspot.com/2016/10/measurements-raspberry-pi-3-hifiberry.html

Thanks for that, it kind of bears out my subjective view of my DAC+
Pro.

Sounds really good and is very quiet.



*Players:* SliMP3,Squeezebox3 x3,Receiver,SqueezePlayer,PiCorePlayer
x3,Wandboard
*Server:* LMS Version: 7.9.0 - 1453293926 on Centos 7 VM on ESXi 6 on HP
N54L
*Plugins:* AutoRescan/BBCiPlayer/PowerSave/PowerSwitchIII/Squeezecloud
*Remotes:* iPeng8/Orangesqueeze/PC/Jivelite
*Music:* 383GB,1269 albums 17756 songs 4381 artists mostly FLACs

*Want a webapp ?* See
http://forums.slimdevices.com/showthread.php?104305-Webapp-for-LMS
------------------------------------------------------------------------
Jeff07971's Profile: http://forums.slimdevices.com/member.php?userid=49290
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[sckramer]

Post by Archimago
Hey guys. Just posted some measurements on the Pi3 + HiFiBerry DAC
combo. I suspect many of us here are already doing this kind of thing
with Pi computers and various other SBC's. Certainly makes an
inexpensive system for a Squeezebox replacement with piCorePlayer
installed...
http://archimago.blogspot.com/2016/10/measurements-raspberry-pi-3-hifiberry.html

The D+PRO is not my daily driver, and I'm more interested in the digi+,
digi+pro, and i2s-- where I find bypassing power is a fundamental
improvement. So really can't say if stock USB power vs. 5V/3.3V LPS
improves on the D+PRO...Buuut I've set this up just in the last few
days, for another project involving making a good a/b testing platform
for the Kali FIFO buffer (long story). So maybe I'll listen, then go to
USB & see if it's diff.

However, I noticed in your testing, you always go through USB (not
GPIO)-- On the digi+ using linear power vs the stock USB power wart
*thru the USB* didn't audibly change anything... also Hans mentioned
this in his vid re: (usb/GPIO) whereas thru USB made no change.

You can see my setup somewhere in the linux section, Kali... there's a
pic.. I'm borrowing my 4x LPS's (LT3042 based, and cheap) supplies to
play with this--

I'm cool if it measures the same, but would really like to see you
re-run the tests taking it all the way thru GPIO, see my setup -- Then
maybe play with the digi+/digi+ pro also :D



'AudioSystem' (http://bit.ly/1ruj0Zw) - 'YouTube'
(http://bit.ly/1ULq4fT) - 'Twitter' (http://bit.ly/1q8dTgc)
------------------------------------------------------------------------
sckramer's Profile: http://forums.slimdevices.com/member.php?userid=20311
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[Archimago]

Post by sckramer
The D+PRO is not my daily driver, and I'm more interested in the digi+,
digi+pro, and i2s-- where I find bypassing power is a fundamental
improvement. So really can't say if stock USB power vs. 5V/3.3V LPS
improves on the D+PRO...Buuut I've set this up just in the last few
days, for another project involving making a good a/b testing platform
for the Kali FIFO buffer (long story). So maybe I'll listen, then go to
USB & see if it's diff.
However, noticed in your testing, you always go through USB (not GPIO)--
On the digi+ using linear power vs the stock USB power wart *thru the
USB* didn't audibly change anything... also Hans mentioned this in his
vid re: (usb/GPIO) whereas thru USB made no change.
You can see my setup somewhere in the linux section, Kali... there's a
pic.. I'm borrowing my 4x LPS's (LT3042 based, and cheap) supplies to
play with this--
I'm cool if it measures the same, but would really like to see you
re-run the tests taking it all the way thru GPIO, see my setup -- Then
maybe play with the digi+/digi+ pro also :D
Thanks for the great tests as usual!

HI screamer.
I don't have the Digi models so would not be able to speak about that.
Would be very interested of any blind A/B testing results you may have.
Of course for this post, I was just looking at ethernet streaming to the
DAC.

For the most part, I have not used SPDIF of any sort other than as
convenience (like TosLink out from my TV) for years on the audio side.
Objrctively, coax and TosLink have more objective jitter and I find USB
more consistent from a sound quality perspective. This is why I've
measured USB or ethernet more over the last few years and used USB for
playback where available. Of course that's not to say I believe I can
clearly hear a difference...



Archimago's Musings: (archimago.blogspot.com) A 'more objective'
audiophile blog.
------------------------------------------------------------------------
Archimago's Profile: http://forums.slimdevices.com/member.php?userid=2207
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[arnyk]

Post by Archimago
Hey guys. Just posted some measurements on the Pi3 + HiFiBerry DAC
combo. I suspect many of us here are already doing this kind of thing
with Pi computers and various other SBC's. Certainly makes an
inexpensive system for a Squeezebox replacement with piCorePlayer
installed...
http://archimago.blogspot.com/2016/10/measurements-raspberry-pi-3-hifiberry.html

Interesting data point. As the report mentions, one possible area of
concern is the device's approximate -80 dB suppression of spurious
responses and noise. This is about 16 dB poorer than CD quality (-96
dB) but about 20 dB better than a typical real world threshold of
audibility (-60 dB).

These days true CD quality performance (ca. -96 dB) is available at the
analog outputs of an < $100 BD player which usually includes a network
(LAN+WAN) media player with considerable power and flexibility.

Not exactly an apples-to-apples comparison, but not too far off when all
costs are included would be the performance of an Intel NUC micro-PC's
the Realtek ALC283 on-board chip audio system:

http://stn-bc.ru/?q=node/835

"
To assess output sound path designed for headphones or external
speakers, we resorted to the instrumental testing using an external
sound card Creative E-MU 0204 USB and utility RightMark audio Analyzer
6.3.0. Testing was conducted to stereo mode, 24-bit / 44.1 kHz.
As a result of testing the audio section in the laptop Acer Aspire
R7-371T was rated "very good." Full report with the test results in RMAA
6.3.0 software on a separate page, the following is a brief report.

Frequency response (in the range of 40 Hz - 15 kHz), dB
0.02, -0.09 Excellent

Noise level dB (A )
-86.4 good

dynamic range, dB (A)
86.2 good

THD,%
0.0057 Very good

THD + noise, dB (A)
-74.2 Mediocre

Intermodulation distortion + noise,%
0.030 good

Crosstalk dB
-85.8 Excellent

Intermodulation at 10 kHz,%
0.017 Very good

Overall very good
"


------------------------------------------------------------------------
arnyk's Profile: http://forums.slimdevices.com/member.php?userid=64365
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[Wombat]

There is a follow up:
http://archimago.blogspot.de/2016/10/measurements-hifiberry-dac-pro-pcm5122.html
Nice that you extended the test and have some words about how alone
"less ringing" in the name of the filter may trigger the audiophile
brain most certainly more as any technical or even audible benefit.
Guess Linear Power and alike terms do the same.



Transporter (modded) -> RG142 -> Avantgarde Acoustic based 500VA
monoblocks -> Sommer SPK240 -> self-made speakers
------------------------------------------------------------------------
Wombat's Profile: http://forums.slimdevices.com/member.php?userid=4113
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[Archimago]

Post by Wombat
http://archimago.blogspot.de/2016/10/measurements-hifiberry-dac-pro-pcm5122.html
Nice that you extended the test and have some words about how alone
"less ringing" in the name of the filter may trigger the audiophile
brain most certainly more as any technical or even audible benefit.
Guess Linear Power and alike terms do the same.

Thanks Wombat.

Indeed, we have to be careful I think about the wording people use and
recognize that with these time domain impulse response graphs, there's
always a price to pay in the frequency domain... Advertisers and
magazine writers interestingly almost never point this out. Of course
this is inconvenient for folks who want to show their beautiful
non-ringing impulse images and sweep the fact that the DAC has all kinds
of ultrasonic aliasing products!



Archimago's Musings: (archimago.blogspot.com) A 'more objective'
audiophile blog.
------------------------------------------------------------------------
Archimago's Profile: http://forums.slimdevices.com/member.php?userid=2207
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[Archimago]

Post by arnyk
Interesting data point. As the report mentions, one possible area of
concern is the device's approximate -80 dB suppression of spurious
responses and noise. This is about 16 dB poorer than CD quality (-96
dB) but about 20 dB better than a typical real world threshold of
audibility (-60 dB).
These days true CD quality performance (ca. -96 dB) is available at the
analog outputs of an < $100 BD player which usually includes a network
(LAN+WAN) media player with considerable power and flexibility.
Not exactly an apples-to-apples comparison, but not too far off when all
costs are included would be the performance of an Intel NUC micro-PC's
http://stn-bc.ru/?q=node/835
"
To assess output sound path designed for headphones or external
speakers, we resorted to the instrumental testing using an external
sound card Creative E-MU 0204 USB and utility RightMark audio Analyzer
6.3.0. Testing was conducted to stereo mode, 24-bit / 44.1 kHz.
As a result of testing the audio section in the laptop Acer Aspire
R7-371T was rated "very good." Full report with the test results in RMAA
6.3.0 software on a separate page, the following is a brief report.
Frequency response (in the range of 40 Hz - 15 kHz), dB
0.02, -0.09 Excellent
Noise level dB (A )
-86.4 good
dynamic range, dB (A)
86.2 good
THD,%
0.0057 Very good
THD + noise, dB (A)
-74.2 Mediocre
Intermodulation distortion + noise,%
0.030 good
Crosstalk dB
-85.8 Excellent
Intermodulation at 10 kHz,%
0.017 Very good
Overall very good
"

Hi Arny,
Not sure which part in the testing you're referring to with -80dB. If
it's in reference to the "digital filter composite", I think that's a
result of the limitations of the digital antialiasing filter in handling
that extremely high amplitude wide bandwidth white noise and the
occasional inter-sample over when doing the upsampling.

Overall though from the Rightmark results, the HiFiBerry DAC+ Pro will
*easily* beat that NUC! Plus this device will be able to handle 24-bit
audio with noise floor lower than -105dB in the audible spectrum...



Archimago's Musings: (archimago.blogspot.com) A 'more objective'
audiophile blog.
------------------------------------------------------------------------
Archimago's Profile: http://forums.slimdevices.com/member.php?userid=2207
View this thread: http://forums.slimdevices.com/showthread.php?t=106213

[arnyk]

Post by Archimago
Hi Arny,
Not sure which part in the testing you're referring to with -80dB.

I'm referring to either of the THD+n or IMD+N measurements shown in the
Audio Rightmark results at
http://archimago.blogspot.com/2016/10/measurements-raspberry-pi-3-hifiberry.html,
either sample rate.

Either THD and IM are commonly called "Spurious Responses".

In one THD test the THD is 0.0047% and in the other it is 0.0049%. In
my book, six of one, half dozen of the other or about 0.005%

In one IMD+N test the IM is 0.0077% and in the other it is 0.0066%. In
my book, also about six of one, half dozen of the other or about 0.0075%


0.005% about the same as 86 dB down, while 0.0075% is more like 82 dB
down. I approximated both quite roughly as -80 dB.

The DAC chip in the Hi Fi Berry has a noise spec that is somewhat better
than its THD+N or IM+N performance. While playing music and other
signals, the THD+N or IMD+N numbers are probably more representative of
its actual performance than the number for background noise.


------------------------------------------------------------------------
arnyk's Profile: http://forums.slimdevices.com/member.php?userid=64365
View this thread: http://forums.slimdevices.com/showthread.php?t=106213