Julf wrote:
> For general measurements (speakers etc.) I use a cheap Behringer UCA202,
> but it is only 16 bit/48 kHz, so somewhat limited. For high-frequency
> stuff I use a 'redpitaya' (http://redpitaya.com/), and for high-SNR
> stuff I have been using an E-MU 0204 (24/192).
>
> I also have a
Julf wrote:
> For general measurements (speakers etc.) I use a cheap Behringer UCA202,
> but it is only 16 bit/48 kHz, so somewhat limited. For high-frequency
> stuff I use a 'redpitaya' (http://redpitaya.com/), and for high-SNR
> stuff I have been using an E-MU 0204 (24/192).
>
> I also have a
Julf wrote:
> If they actually cared about real differences - but that would scare
> away advertisers. Remember it is the advertisers, not the subscribers,
> that pay for the audiorags.
>
>
>
> Indeed. And as I keep saying, having the headroom (given the word length
> of modern DSP architectur
sckramer wrote:
> What's the soundcard/software you use?
For general measurements (speakers etc.) I use a cheap Behringer UCA202,
but it is only 16 bit/48 kHz, so somewhat limited. For high-frequency
stuff I use a 'redpitaya' (http://redpitaya.com/), and for high-SNR
stuff I have been using an E
Julf wrote:
> I am asking you to verify 1) that the things you are hearing correspond
> to objective differences, and 2) if so, what kind of differences.
That's cool, plan on doing more videos on some things related to this,
I'll be keeping certain pieces around as reference etc, as I learn more
sckramer wrote:
> So are you asking me to prove to myself I'm not hearing anything?
I am asking you to verify 1) that the things you are hearing correspond
to objective differences, and 2) if so, what kind of differences.
"To try to judge the real from the false will always be hard. In this
f
Mnyb wrote:
> Yea but is it ? that would be something for audiorags to actually test?
If they actually cared about real differences - but that would scare
away advertisers. Remember it is the advertisers, not the subscribers,
that pay for the audiorags.
> But imho stuff should have healthy head
Julf wrote:
> Not really. Just measure noise and jitter (using the 2-tone
> intermodulation test). Again, doable with any decent sound card and some
> software. Just look at the measurements 'Archimago'
> (http://archimago.blogspot.nl/) has published.
Ok, I'll check it out--
So are you asking
sckramer wrote:
> To prove if noise is effecting the digital, you'd need to compare 2
> signals & look for bit errors, or differences (and depending on where
> the bit error happens in the "16-bit" binary word... can translate to
> big audio blip, or nothing at all)
>
> ..then on top of that jit
Julf wrote:
> Indeed, and yes, intersample peaks are an issue with older DACs, but I
> would expect modern designs to deal with it - it is pretty easy to do,
> after all.
Yea but is it ? that would be something for audiorags to actually test ?
there are actually some old AES pappers on it on the
Julf wrote:
> Would love to see your double-blind ABX logs... :)
>
>
>
> Oscilloscopes are good for fairly high frequency work, but their
> linearity, resolution and signal-to-noise ratio are not very good. For
> audio work, you are much better off with a good sound card and software.
Yeah, w
sckramer wrote:
> The fact that it sounds better (and not in a "I just bought some audio
> jewelry, it must sound better :D) --> compels be to like to be able to.
>
Would love to see your double-blind ABX logs... :)
> I'm researching how good an oscilloscope I should get... (of course I'm
> g
Julf wrote:
> Have you measured the difference in jitter or noise?
The fact that it sound better, compels be to like to be able to.
-- very convinced you have to get this initial critical stage right...
else you tune the rest of the system to cover up the problems.
I'm researching how good an
sckramer wrote:
> There are no analog stages
Ah, should have re-read the thread. Was assuming that there had to be at
least a DAC involved, otherwise a linear +5V supply makes no sense.
"To try to judge the real from the false will always be hard. In this
fast-growing art of 'high fidelity' t
Julf wrote:
> Ah, should have re-read the thread. Was assuming that there had to be at
> least a DAC involved, otherwise a linear +5V supply makes no sense.
Well, the spdif square wave, is an analog signal
CiAudio VDC-SB - raspberry pi b+ hifiberry digi+ (BNC added, removed
TOSlink) piCorePla
sckramer wrote:
> Well, the spdif square wave, is an analog signal
Have you measured the difference in jitter or noise?
"To try to judge the real from the false will always be hard. In this
fast-growing art of 'high fidelity' the quackery will bear a solid gilt
edge that will fool many people
darrenyeats wrote:
> Julf, for me the money shot is here:
> http://www.pinkfishmedia.net/forum/showpost.php?p=1990168&postcount=11
>
> It is far from a single cycle in the example given. It's a macro-effect
> easily seen even at whole-track scale. It also happens to be one of my
> favourite albu
I wonder if the linear 5V is a very pure source for the 3.3/1.8-- as
reading some more on Daniels thread, he could not measure any noise on
the 3.3V on the pi, so it's very well done... not worth trying to
separate it, add a header for 3.3V etc...
Any thoughts?
CiAudio VDC-SB - raspberry pi b
Jeff07971 wrote:
> Daniel (HiFiBerry team)
>
> Keymaster
>
> Note that the new HiFiBerry Digi design uses now the 3.3V from the
> Raspberry. Therefore it is not easy to power the Digi board separately.
>
> Best regards
> Daniel
Bummer, now I'm really gonna bug them about making a digi+ "Pro"
sckramer wrote:
> I figured re: the pi since it still needs the 3V, 1.8V
>
> Interesting on the Digi+ -- they specifically added a place for external
> 5V for this purpose, as I soldered on the header-- suspect that is not
> right on the 3V, I'll trace it & ask them.
The header is so you can po
sckramer wrote:
> I figured re: the pi since it still needs the 3V, 1.8V
>
> Interesting on the Digi+ -- they specifically added a place for external
> 5V for this purpose, as I soldered on the header-- suspect that is not
> right on the 3V, I'll trace it & ask them.
Daniel (HiFiBerry team)
K
Jeff07971 wrote:
> The "Pi B+" itself (the CPU) will still have 2 "switching" PSU's as it
> runs from 3.3V and 1.8V not 5V
>
> U3 on the schematic
>
> Also the later Hifiberry Digi's use the 3.3v from the Pi sourced from
> the Switch Mode PSU onboard the Pi
I figured re: the pi since it still
> This setup the pi's switching supply,fuse,protection is all bypassed
> (you can power the pi from it's GPIO pins) -- in addition to the digi+
> not pulling power from the pi--
The "Pi B+" itself (the CPU) will still have 2 "switching" PSU's as it
runs from 3.3V and 1.8V not 5V
U3 on the schem
BTW, this is not a knock-off chinese stamped out cookie-cutter board
(HifiBerry), it was carefully designed-- Also the guy (Daniel) has nice
blog post documenting some of it development...
In addition HifiBerry facilitated the linux i2s drivers, for this spdif
chip, and a few BurrBrown dac chips
Julf, for me the money shot is here:
http://www.pinkfishmedia.net/forum/showpost.php?p=1990168&postcount=11
It is far from a single cycle in the example given. It's a macro-effect
easily visible even at whole-track scale!
Darren
Check it, add to it! http://www.dr.loudness-war.info/
SB Touch
-
Are you guys familiar with the pi, powering it from the GPIO etc? -- I
might start a thread in DIY or Linux for this... forgot I was in the
crazy audiophile section :D i'm not that wacky I swear :D
CiAudio VDC-SB - raspberry pi b+ hifiberry digi+ (BNC added, removed
TOSlink) piCorePlayer (play
cliveb wrote:
> Please can you speculate as to the mechanism by which a linear 5V supply
> gives an improvement over a SMPS 5V supply?
> Bear in mind that the DIGI+ SPDIF output is transformer coupled and
> therefore galvanically isolated from the DAC it feeds, so noise
> transmission to the DAC
Julf wrote:
> All you need is a computer with a halfway decent sound card.
>
> Ok
>
> A linear supply only makes sense if you can feed the analog stages and
> the computer part separately. Linear supplies are notably bad at dealing
> with the rapid switching noise from fast digital circuits.
T
sckramer wrote:
> The 5V supply however *does*
Please can you speculate as to the mechanism by which a linear 5V supply
gives an improvement over a SMPS 5V supply?
Bear in mind that the DIGI+ SPDIF output is transformer coupled and
therefore galvanically isolated from the DAC it feeds, so noise
t
sckramer wrote:
> Yeah no, I'll invite you over, your ears can tell :D don't have that
> kind of measuring equipment.
All you need is a computer with a halfway decent sound card.
> The 5V supply however *does*
A linear supply only makes sense if you can feed the analog stages and
the computer
Julf wrote:
> If you are prepared to go through that effort, how about also measuring
> the difference?
Yeah no, I'll invite you over, your ears can tell :D
CiAudio VDC-SB - raspberry pi b+ hifiberry digi+ (BNC added, removed
TOSlink) piCorePlayer (play song from RAM) - trumps sb touch sound
sckramer wrote:
> Almost want to try & set up a good stereo mic to try & catch the
> difference, even thru video-- I keep the touch around as reference / a
> controller & at my computer desk now
>
If you are prepared to go through that effort, how about also measuring
the difference?
"To try
I recommend skipping the touch and going straight to the
RaspberryPi/Hifiberry Digi+/piCorePlayer
But, you have to use a 5V linear supply-- I stumbled onto this, my plan
was to at least have a replacement for when the touch dies, but this
ended up sounding better than the tweaked sb touch (see my
darrenyeats wrote:
> Julf, I have a DAC1. Set up the right way anyway, it's still a good
> DAC.
> http://www.pinkfishmedia.net/forum/showthread.php?t=137152
I have to agree with the comments from Werner in that thread.
"To try to judge the real from the false will always be hard. In this
fast
Julf, I have a DAC1. Set up the right way anyway, it's still a good
DAC.
http://www.pinkfishmedia.net/forum/showthread.php?t=137152
Check it, add to it! http://www.dr.loudness-war.info/
SB Touch
darrenyeats's Profile: htt
Mnyb wrote:
> Do you remeber our tread about "intersample overs" not many DAC
> especially older ones shave not acounted for this effect
Indeed, and yes, intersample peaks are an issue with older DACs, but I
would expect modern designs to deal with it - it is pretty easy to do,
after all.
"To
Mnyb wrote:
> Do you remeber our tread about "intersample overs" not many DAC
> especially older ones shave not acounted for this effect , which in turn
> is an artefact of modern record producing .
> DAC mfg is often hunting specs so even if some filters alows for gain
> adjustments in theri par
Julf wrote:
> So you suspect your DAC hasn't been designed with enough headroom?
Do you remeber our tread about "intersample overs" not many DAC
especially older ones shave not acounted for this effect , which in turn
is an artefact of modern record producing .
DAC mfg is often hunting specs so
darrenyeats wrote:
> BTW I'd avoid above 90 in any event to give my DAC DSP headroom for
> upsampling/ASRC.
So you suspect your DAC hasn't been designed with enough headroom?
"To try to judge the real from the false will always be hard. In this
fast-growing art of 'high fidelity' the quackery
Mnyb wrote:
> The strength of the TP is its very good built in DAC combined with async
> networking and variable attenuation on the outputs and balanced output
> for those who needs it , it's the whole thing .
Agreed. The TP is effectively a digital preamp with balanced outputs and
a built in squ
Some recordings are very quiet others very loud. Sometimes I don't want
to listen at a loud volume. I tend to use 30-65 most of the time but I
occasionally need say 20-80. I don't want a physical preamp. Therefore I
expect my digital volume to be blameless. 80-90 is not a realistic use
case, far t
Mnyb wrote:
> A guess 80-90 volume is ok for the best recordings ever made?
Pretty much, yes. So far I haven't come across a single recording going
beyond 90 dB or so.
"To try to judge the real from the false will always be hard. In this
fast-growing art of 'high fidelity' the quackery will b
42 matches
Mail list logo
