Another test to try when you get a round tuit
(http://en.wiktionary.org/wiki/round_tuit): Record the analog audio
into the sound card with nothing connected to the inputs, to see if any
of the noise being measured is coming from the computer or the sound
card.
--
TiredLegs
This might be the most interesting thread in this forum ever. Thanks for
your work so far Phil. Looking forward to reading about anything you
might discover in the future.
--
paulduggan
paulduggan's Profile: http://forums.
NewBuyer;525745 Wrote:
> Hey Phil, just wondering/nagging: Any progress? :)
Not yet - sorry. Won't get time to do this until the Easter break...
--
Phil Leigh
You want to see the signal path BEFORE it gets onto a CD/vinyl...it
ain't what you'd call minimal...
SB Touch Beta (wired) - TACT 2.
Phil Leigh;515454 Wrote:
> I'll look into it at the weekend...
>
> SB3 not faulty. Will retest with Touch and SB3.
Hey Phil, just wondering/nagging: Any progress? :)
--
NewBuyer
NewBuyer's Profile: http://forums.slimde
I've been dealing with LF noise on my latest DAC design for a while. It
turns out that measuring LF noise is not so easy, there is so much
floating around in the environment its tough to separate external noise
from device noise.
I finally wound building a very low noise amplifier powered by
bat
Themis;515466 Wrote:
> I appreciate very much these tests, Phil... :)
> I'm waiting the results with great interest. These tests are a very
> neat approach, imho: they might not tell -where- there might be a
> difference, but the tell you -if- there is one. It's very important to
> eliminate fals
I appreciate very much these tests, Phil... :)
I'm waiting the results with great interest. These tests are a very
neat approach, imho.
--
Themis
SB3 - North Star dac 192 - Croft 25Pre and Series 7 power - Sonus Faber
Grand Piano Domus
--
NewBuyer;515449 Wrote:
> Hey there Phil,
> Did you ever fully get to the bottom of those repeating-interval LF
> artifacts you measured on your SB3 analog outs, and on the difference
> between your measurements and JA's measurements? Is your SB3 unit
> possibly faulty, or is it possibly resultin
Hey there Phil,
Did you ever fully get to the bottom of those repeating-interval LF
artifacts you measured on your SB3 analog outs, and on the difference
between your measurements and JA's measurements? Is your SB3 unit
possibly faulty, or is it possibly resulting from a ground loop through
the i
DCtoDaylight;492264 Wrote:
> You using coax or toslink for the spdif connection? If coax, there's
> another potential ground loop to consider
Hmmm - the ground of the spdif and analogue outputs are connected (with
0.8 ohm resistance between them) and both are connected to the ground of
my P
DCtoDaylight;492264 Wrote:
> You using coax or toslink for the spdif connection? If coax, there's
> another potential ground loop to consider
Coax - I'm going to get my Fluke...
the soundcard doesn't have a toslink input :-(
--
Phil Leigh
You want to see the signal path BEFORE it gets o
You using coax or toslink for the spdif connection? If coax, there's
another potential ground loop to consider
--
DCtoDaylight
Audiophile wish list: Zero Distortion, Infinite Signal to Noise Ratio,
and a Bandwidth from DC to Daylight
---
DCtoDaylight;492093 Wrote:
> How is this done? I guess you connect the s/pdif from the SB3, as well
> as the analog, and somewhere in the soundcards setup, you select the
> s/pdif as the clock source?
> I'm thinking of duplicating your tests, and playing along!
Yes that's exactly how to do it.
Phil Leigh;491604 Wrote:
> NOTE: the soundcard is locked to the s/pdif clock from the SB3. There is
> consequently no sample rate drift between DAC and ADC (well, it's
> reported as 0.0009ppm!)
How is this done? I guess you connect the s/pdif from the SB3, as well
as the analog, and somewhere i
Once you iron out your test procedures, it will be great to test those
hi-end DACs out there for "linearity", starting w/ the TP? Of course,
for CD or WAV recorded at 16bit/44.1Khz, it is still fairly coarse and
low rez results.
--
empty99
---
DaveWr;491912 Wrote:
> Hi Phil,
>
> Well looking at JAs stuff he did get a peak at 30 Hz. I think maybe he
> was averaging your 22 and 44 Hz by his smoothing settings for
> measurement. Both graphs definitely show LF stuff that's not mains
> hum!
>
> I think, levels not withstanding, we have
NewBuyer;491910 Wrote:
> Could there possibly be a ground loop between your SB3 and your computer
> via these unbalanced analog connections? Perhaps the analog outputs of
> the SB3 and Touch have different potentials for ground loops(?)
Good point - I'll get my meter out later and check.
--
Hi Phil,
Well looking at JAs stuff he did get a peak at 30 Hz. I think maybe he
was averaging your 22 and 44 Hz by his smoothing settings for
measurement. Both graphs definitely show LF stuff that's not mains
hum!
I think, levels not withstanding, we have to agree that what you
measure is ther
Could there possibly be a ground loop between your SB3 and your computer
via these unbalanced analog connections? Perhaps the analog outputs of
the SB3 and Touch have different potentials for ground loops(?)
--
NewBuyer
N
This is silence out of the SB3.
Ignoring the dB scale, which may be misleading, there are clear
harmonic peaks at 22/44/66 & 88 Hz...
+---+
|Filename: silence.jpg |
|Download: http://foru
DaveWr;491892 Wrote:
> Hi Phil,
>
> The 16 bit - 96 dB is the dynamic range of a 16 bit signal (smallest
> discernible to maximum signal). That is not the same as the noise
> floor. The SB3 is a 24 bit DAC product. JA is measuring digital
> silence, therefore the SB3 DAC gets a zero signal at
Hi Phil,
The 16 bit - 96 dB is the dynamic range of a 16 bit signal (smallest
discernible to maximum signal). That is not the same as the noise
floor. The SB3 is a 24 bit DAC product. JA is measuring digital
silence, therefore the SB3 DAC gets a zero signal at 24 bit input, even
though origina
DaveWr;491868 Wrote:
> Hi Phil,
>
> I understand it may be confusing, but since you are showing a noise
> floor that is 30db worse than JA measurements, there must be some
> extraneous signal getting in.
>
> I originally thought power supply, but you said same on both, any
> chance of an eart
Hi Phil,
I understand it may be confusing, but since you are showing a noise
floor that is 30db worse than JA measurements, there must be some
extraneous signal getting in.
I originally thought power supply, but you said same on both, any
chance of an earthing problem?
Or is your 0db line not
Changing PSU makes no difference to the SB3 noise floor. Turning the VFD
display off reduces the noise by maybe 25%...
--
Phil Leigh
You want to see the signal path BEFORE it gets onto a CD/vinyl...it
ain't what you'd call minimal...
SB Touch Beta (wired) - TACT 2.2X (Linear PSU) + Good Vibrat
DaveWr;491771 Wrote:
> I don't understand your result on the SB3, it doesn't compare with John
> Atkinson results in Stereophile, you seem to be measuring significantly
> more noise.
>
>
>
> Dave
Well, I can't answer that since we are clearly using different
measurement methodologies so it wo
I don't understand your result on the SB3, it doesn't compare with John
Atkinson results in Stereophile, you seem to be measuring significantly
more noise.
Dave
+---+
|Filename: 906SB3FIG3.jpg
Here is the Touch noise floor playing digital silence...
(also shows that the soundcard is very, very silent!)
There's a small peak at 50Hz which is mains breakthrough from
somewhere.
+---+
|Filename: Touch Noise floor].jpg
100Hz sine wave generated in Audacity - compared to same file played
back on SB3...
The 100Hz sine is almost nulled out completely, but notice the
(low-level) LF crud. There is some interaction between this residual
noise and the remnants of the 100Hz sine wave...
This is coming from the SB3 DAC
SuperQ;491712 Wrote:
> Is this the audio board you're using for ADM work?
>
> http://www.m-audio.com/products/en_us/Audiophile2496.html
Yes - that's the one!
--
Phil Leigh
You want to see the signal path BEFORE it gets onto a CD/vinyl...it
ain't what you'd call minimal...
SB Touch Beta (wir
That'd be fun, too.
On 2 Dec 2009, at 17:46, Phil Leigh wrote:
>
> cunobelinus;491666 Wrote:
>> Do you have a TP with which to try the same trick?
>>
>>
>>
>>>
>>> [/color]
>
> Sadly...no. But I do have a Touch!
>
>
> --
> Phil Leigh
>
> You want to see the signal path BEFORE it gets o
Interesting, I don't think I can do it with my M-Audio USB FastTrack Pro
in Linux, but maybe the windows driver for it can set the SPDIF clock
slave similar to the PCI version.
--
SuperQ
SuperQ's Profile: http://forums.sli
Is this the audio board you're using for ADM work?
http://www.m-audio.com/products/en_us/Audiophile2496.html
--
SuperQ
SuperQ's Profile: http://forums.slimdevices.com/member.php?userid=2139
View this thread: http://forums
cunobelinus;491666 Wrote:
> Do you have a TP with which to try the same trick?
>
>
>
> >
> > [/color]
Sadly...no. But I do have a Touch!
--
Phil Leigh
You want to see the signal path BEFORE it gets onto a CD/vinyl...it
ain't what you'd call minimal...
SB Touch Beta (wired) - TACT 2.2X (L
You might just by measuring differnces in low frequency phase effects
caused by LF rolloff differences. Significant effects at 10 times -3db
point.
Dave
--
DaveWr
DaveWr's Profile: http://forums.slimdevices.com/member.ph
Do you have a TP with which to try the same trick?
On 2 Dec 2009, at 16:56, Phil Leigh wrote:
>
> Looking at the spectrum plots of the difference files, it seems that
> actually the big differences are in the lower frequencies. For the SB3
> DAC the correlation above 280Hz is not bad, whereas i
Looking at the spectrum plots of the difference files, it seems that
actually the big differences are in the lower frequencies. For the SB3
DAC the correlation above 280Hz is not bad, whereas it rises (worsens)
sharply below that. On my MF Dac, it doesn't start to rise until below
160Hz.
Taken at
I found this interesting - and maybe opens up a whole new avenue for the
right-brained amongst us :-)
Test setup: 15-second WAV file ripped from CD with AccurateRip.
Soundcard=M-Audio 24/96
Record analogue out from SB3 and compare in ADM against the pristine
WAV rip.
That's a brutal test of the
38 matches
Mail list logo
