mswlogo;543330 Wrote:
...At 80 (on the new Transporter scale) you are not attenuating a
normal range for volume control... Dropping down to 80 is not really
a legit range to test... with that test you probably only lose a little
over .5 bits... If you only drop 10dB it would be difficult to
I think that when used as digital transports, any audible differences
between the Touch and SB3 must certainly reflect the jitter-sensitivity
noise-susceptibility etc of the DAC(?)
For instance, I recently had a chance to play with a Touch alongside an
SB3 and a separate transport - all into a
I thought we had established that we are not losing detail until
you've shifted over 4 bits but raising the noise floor out of the DAC
(so the quietest sounds get closer to it or even in extreme cases,
below it - although you can still hear them). Personally I reckon you
can lose 4 more bits
Hi there,
i have a problem. I crashed my new amp (creek). i just have a sb3
classic with a creek destiny. The destiny crashed after two hours. Then
we tried it again with a diffent device (creek destiny) the new device
crashed also. Now..The advice was to lower the output signal of my SB,
mswlogo;543050 Wrote:
Correct in that example Volume 20 has has 75.8dB of dynamic range.
They also note that Volume 20 is 50%. I believe translating that,
that would mean 50% of the 0-96dB scale. Or in other words 48dB of
attenuation. 8bits of attenuation.
For the Volume 30 case. They
johannes;543568 Wrote:
Hi there,
i have a problem. I crashed my new amp (creek). i just have a sb3
classic with a creek destiny. The destiny crashed after two hours. Then
we tried it again with a diffent device (creek destiny) the new device
crashed also. Now..The advice was to lower the
Phil Leigh;543499 Wrote:
I thought we had established that we are not losing detail until
you've shifted over 4 bits but raising the noise floor out of the DAC
(so the quietest sounds get closer to it or even in extreme cases,
below it - although you can still hear them). Personally I
darrenyeats;543576 Wrote:
I think I see where you're coming from.
If you look at the tests volume 30 is 12.5db of attenuaton but the
result is only 7.1db reduction in dynamic range from 40 (max). This is
noticeably different to stepping from volume 30 to volume 20, a further
12.5db
mswlogo;543590 Wrote:
Correct. At least some folks are reading.
The DAC is not an absolute finite device. It's not 24bit, it's not
20bit, it's barely even 16bit.
As you shift down. You're not losing discrete WHOLE bits. You're just
pushing them into a range of the DAC where it performs
mswlogo;543590 Wrote:
Correct. At least some folks are reading.
The DAC is not an absolute finite device. It's not 24bit, it's not
20bit, it's barely even 16bit.
As you shift down. You're not losing discrete WHOLE bits. You're just
pushing them into a range of the DAC where it performs
mswlogo;543590 Wrote:
The DAC is not an absolute finite device. It's not 24bit, it's not
20bit, it's barely even 16bit.
mswlogo,
In principle, I agree. But it's a question of how good the particular
DAC is whether this effect impinges materially.
Have a look at
mswlogo;543590 Wrote:
The DAC is not an absolute finite device. It's not 24bit, it's not
20bit, it's barely even 16bit.
mswlogo,
In principle, I agree. But it's a question of how good the particular
DAC is whether this effect impinges materially.
Have a look at
darrenyeats;543596 Wrote:
mswlogo,
In principle I agree. But it's a question of how good the particular
DAC is whether this effect impinges materially.
Have a look at
http://www.stereophile.com/mediaservers/207slim/index4.html . These are
measurements on the Transporter. Look at figure 5
mswlogo;543600 Wrote:
I have a Transporter, great unit. But I don't use it's DACs. I use
Meridian DACs. I can hear loss with 6dB attenuation. And so can 22
other Meridian owners. Use analog volume.
Perhaps the reason is the following:
mswlogo;543588 Wrote:
Because they do crossover in
johannes;543568 Wrote:
Hi there,
i have a problem. I crashed my new amp (creek). i just have a sb3
classic with a creek destiny. The destiny crashed after two hours. Then
we tried it again with a diffent device (creek destiny) the new device
crashed also. Now..The advice was to lower the
mlsstl;543615 Wrote:
What do you mean by crashed your amp? Is it physically burnt out or
non-operational and needs repair?
Or is it just not functioning correctly and needs to be powered down
and restarted (as one might think of a computer crash)?
Or does it mean something else?
darrenyeats;543603 Wrote:
Perhaps the reason is the following:
Darren
For some reason quote did not work. But you said the reason I hear
differences is due to digital crossovers.
But that is a VERY good point.
It probably does impact it. But I'm sure it does not account for all of
it.
I just realized that it can't be the digital crossovers.
It was a really good thought though.
It's actually really good example where 24bit helps a lot.
The Digital Crossovers will see every pure bit even if it's attenuated.
The DSP processing is done in 48bit. Then back to 24bit. It not until
mswlogo;543588 Wrote:
I never agreed 4 bits were free. In that test Volume 20 is a 4bit shift.
If 4bits didn't lose detail you'd still have 96dB of dynamic range. But
you get 76 dB. They attenuated 4bits and lost 4bits. I must have said
that 6 times now.
Replay Gain is a slightly
Phil Leigh;543639 Wrote:
Replaygain does not clip. It assesses the peaks.
Cool. Thanks.
--
mswlogo
XP Cat5 Transporter/DuetController SPDIF Meridian G68 DSP6000,
DSP5500HC, DSP5000
XP Cat5 SB3 SPDIF Meridian DSP5000
XP Cat5 DuetReceiver SPDIF Meridian G91 DSP5000
'My
mswlogo;543630 Wrote:
For some reason quote did not work. But you said the reason I hear
differences is due to digital crossovers.
But that is a VERY good point.
It probably does impact it. But I'm sure it does not account for all of
it.
That test clearly shows significant loss due
mswlogo;543633 Wrote:
I just realized that it can't be the digital crossovers.
It was a really good thought though.
It's actually a really good example where 24bit helps a lot.
The Digital Crossovers will see every pure bit even if it's attenuated.
The DSP processing is done in 48bit.
darrenyeats;543653 Wrote:
I can understand how 48 bits helps the DSP to be more transparent (less
rounding errors).
Once the resulting signal is converted to 24 bits and passed to a DAC,
won't it be still subject to the 20 bits of dynamic range problem? Do
you see why I ask? Or perhaps
Phil Leigh;543646 Wrote:
What Sean said was very clear... any attenuation increases the noise
floor/decreases the SNR. That's all. Did he say you could ALWAYS hear
it? - NO. He expressed no opinion on that AFAIK.
You keep talking about loss - loss of what?
Mathematically until you shift
mswlogo;543684 Wrote:
As soon as you shift down (filling the top end with zeros) you are
basically throwing away dynamic range. I call it lost (or under
utilized dynamic range that you have available). Call it what ever you
like.
Correct, Sean does not say how much difference you would
mswlogo;543674 Wrote:
Correct, 48 bit just avoids rounding errors.
Yes the DSP speaker is still subject to the problems discussed here.
Once the DSP is done it's still attenuated digitally by the same amount
before it goes into the DAC.
You can read about Meridian speakers here
darrenyeats wrote:
mswlogo;543633 Wrote:
The DSP processing is done in 48bit.
I can understand how 48 bits helps the DSP to be more transparent (less
rounding errors).
Its not just rounding errors. Nearly all DSP is done by converting the
time-doman signal to the frequency domain using a
pfarrell;543694 Wrote:
DSP is more like dealing with quantum physics rather than Newtonian
physics.
You mentioned the Q-word, so the audiophile-specific version of
Godwin's Law applies, and I declare this thread closed. Thank God :-)
--
cliveb
Transporter - ATC SCM100A
pfarrell;543694 Wrote:
darrenyeats wrote:
mswlogo;543633 Wrote:
The DSP processing is done in 48bit.
I can understand how 48 bits helps the DSP to be more transparent
(less
rounding errors).
Its not just rounding errors. Nearly all DSP is done by converting the
time-doman signal
darrenyeats;543693 Wrote:
Cheers, I've now read that pdf. I don't understand this fully yet.
I was looking for something in the pdf which might explain your
statement above in particular but I couldn't find anything. Why can't
it be the digital crossovers?
Darren
Sorry I'm not going to
Someone forwarded me this thread and asked me to reply, so here goes.
:)
I think the discussion needs to separate the two ways the effect of
jitter is considered.
In one case, jitter on a signal can result in an incorrectly sampled
value. For example, if the sample clock ticks before a digital
31 matches
Mail list logo
