Home » The PARIS Forums » PARIS: Main » Adat cards, lightpipe transfer & dither
|
|
| Re: Adat cards, lightpipe transfer & dither [message #69577 is a reply to message #69576] |
Wed, 21 June 2006 09:23   |
![Deej [1] is currently offline](/theme/PARIS_Forum_Template/images/offline.png "Deej [1] is currently offline")
Deej [1]
Messages: 2149
Registered: January 2006
|
Senior Member |
|
|
After about a year of using a mix matrix that involves streaming audio that
is being processed at 32 bit float in Cubase SX over a 20 bit lightpipe to
be summed in Paris, here's my take on this:
I notice absolutely no sonic hit at all. I think this may be because there
is some incredibly sophisticated stuff going on in the Paris DSP. It's
really strange how this works. Paris can recieve a signal at the ADAT input
that is being processed in Cubase SX at 32 bit float *above* 0dBfs (it's
not clipping in Cubase SX as long as it stays floating and doesn't hit the
mix bus) and though this signal is *techically* being truncated to 20 bit,
there is something going on in the Paris DSP that is apparently allowing the
entire bit depth of whatever audio is received over lightpipe to be retained
while the actual loudness of the signal is attenuated by -24dB. Well, this
gives you an extra +24 dB of headroom to boost the signal again in Paris. It
may be truncated, but it's something inaudible to my ears and what comes out
of the Paris mix bus sounds *HUGE* because it *sounds* like multiple 32 bit
audio tracks being boosted by up to 24dB and then summed analog......at
least, that's my take on it. I certainly notice no distortion/graininess at
all. I did try instantiating a dither plugin on each Cubase audio track a
while back and the cumulative dither of 20+ tracks sounded hissy and
awful.It sounds like voodoo, and I know that bist can't be restored after
truncation, but the extra 24dB of headroom that Paris provides makes it
sound like it's doing exactly this, and then some. I knew there was
something like this happening due to the additional headroom available but I
wasn't sure how much the audible input signal level was being reduced until
I talked to someone (who must remain anonymous right now) who has actually
measured this.
anyway.........whatever is happening, it sounds awesome.
Deej
"Sound Dog" <dogster@tpg.com.au> wrote in message news:44996537$1@linux...
> Say, does anyone know whether you need to dither a 24bit signal to 20bit
> when sending audio back to PARIS via lightpipe (from Logic via and
HDSP962
> for example) ?
>
> Thanks,
> Stewart.
>
>
|
|
|
|
|
|
| Re: Adat cards, lightpipe transfer & dither [message #69628 is a reply to message #69577] |
Fri, 23 June 2006 20:57 |
Sound Dog
Messages: 44
Registered: October 2005
|
Member |
|
|
Hi Deej.
That's seems totally bizarre - I must try that sometime. When you record
audio tracks, do you record directly into Cubase using some other interface
or do you go via PARIS ?
I tried a couple of tests - firstly I extracted a track from CD (Crowded
House "Pineapple Head") into Logic, then recorded that into PARIS via SPDIF
in two different ways:
HDSP9652 SPDIF -> PARIS SPDIF (Logic on Mac sync'd to ADAT signal from PARIS
on PC, using PARIS as master deck with Logic slaved via MTC)
then
HDSP9652 ADAT -> PARIS ADAT (Logic on Mac sync'd to ADAT signal from PARIS
on PC, using PARIS as master deck with Logic slaved via MTC).
I can't hear a difference. There's no level loss or anything. 0dBFS in
Logic = 0dBFS in PARIS.
I also recorded the track from a domestic CD player thru a preamp into PARIS
via an 8-in card, firstly using internal sync in PARIS, then again locked to
Word Clock from an HDSP9652 card installed on the same PC. I think I prefer
the sound of PARIS' internal sync. Seems to be slightly more forward or
something - what should I be listening for ? I recall you saying in one of
your posts that the sound of your PARIS rig improved noticeably once you
started clocking externally.
Hmmm...
Stewart.
DJ wrote in message <44997302$1@linux>...
>After about a year of using a mix matrix that involves streaming audio that
>is being processed at 32 bit float in Cubase SX over a 20 bit lightpipe to
>be summed in Paris, here's my take on this:
>
>I notice absolutely no sonic hit at all. I think this may be because there
>is some incredibly sophisticated stuff going on in the Paris DSP. It's
>really strange how this works. Paris can recieve a signal at the ADAT input
>that is being processed in Cubase SX at 32 bit float *above* 0dBfs (it's
>not clipping in Cubase SX as long as it stays floating and doesn't hit the
>mix bus) and though this signal is *techically* being truncated to 20 bit,
>there is something going on in the Paris DSP that is apparently allowing
the
>entire bit depth of whatever audio is received over lightpipe to be
retained
>while the actual loudness of the signal is attenuated by -24dB. Well, this
>gives you an extra +24 dB of headroom to boost the signal again in Paris.
It
>may be truncated, but it's something inaudible to my ears and what comes
out
>of the Paris mix bus sounds *HUGE* because it *sounds* like multiple 32 bit
>audio tracks being boosted by up to 24dB and then summed analog......at
>least, that's my take on it. I certainly notice no distortion/graininess at
>all. I did try instantiating a dither plugin on each Cubase audio track a
>while back and the cumulative dither of 20+ tracks sounded hissy and
>awful.It sounds like voodoo, and I know that bist can't be restored after
>truncation, but the extra 24dB of headroom that Paris provides makes it
>sound like it's doing exactly this, and then some. I knew there was
>something like this happening due to the additional headroom available but
I
>wasn't sure how much the audible input signal level was being reduced until
>I talked to someone (who must remain anonymous right now) who has actually
>measured this.
>
>anyway.........whatever is happening, it sounds awesome.
>
>Deej
>
>"Sound Dog" <dogster@tpg.com.au> wrote in message news:44996537$1@linux...
>> Say, does anyone know whether you need to dither a 24bit signal to 20bit
>> when sending audio back to PARIS via lightpipe (from Logic via and
>HDSP962
>> for example) ?
>>
>> Thanks,
>> Stewart.
>>
>>
>
>
|
|
|
|
| Re: Adat cards, lightpipe transfer & dither [message #69656 is a reply to message #69628] |
Sun, 25 June 2006 14:59 |
Deej [1]
Messages: 2149
Registered: January 2006
|
Senior Member |
|
|
"Sound Dog" <dogster@tpg.com.au> wrote in message news:449cb5b3$1@linux...
> Hi Deej.
>
> That's seems totally bizarre - I must try that sometime. When you record
> audio tracks, do you record directly into Cubase using some other
interface
> or do you go via PARIS ?
I record straiught to Paris.
>
> I tried a couple of tests - firstly I extracted a track from CD (Crowded
> House "Pineapple Head") into Logic, then recorded that into PARIS via
SPDIF
> in two different ways:
>
> HDSP9652 SPDIF -> PARIS SPDIF (Logic on Mac sync'd to ADAT signal from
PARIS
> on PC, using PARIS as master deck with Logic slaved via MTC)
>
> then
>
> HDSP9652 ADAT -> PARIS ADAT (Logic on Mac sync'd to ADAT signal from PARIS
> on PC, using PARIS as master deck with Logic slaved via MTC).
>
> I can't hear a difference. There's no level loss or anything. 0dBFS in
> Logic = 0dBFS in PARIS.
I know........it's bizarre. Everything seems to be the same. Now set Paris
to post fader, set the Paris fader to unity, process a track in Logic with a
VST plugin while sending it to Paris.............not start raising the Paris
fader and listen to what happens.
>
> I also recorded the track from a domestic CD player thru a preamp into
PARIS
> via an 8-in card, firstly using internal sync in PARIS, then again locked
to
> Word Clock from an HDSP9652 card installed on the same PC. I think I
prefer
> the sound of PARIS' internal sync. Seems to be slightly more forward or
> something - what should I be listening for ? I recall you saying in one
of
> your posts that the sound of your PARIS rig improved noticeably once you
> started clocking externally.
>
> Hmmm...
>
> Stewart.
I prefer the sound of Paris clocked externally. The Paris clock isn't nearly
as accurate as a GenX6. This results in some jitter in the D/A's. Jitter
also = distortion. Sometimes distortion sounds good.
Deej
>
>
> DJ wrote in message <44997302$1@linux>...
> >After about a year of using a mix matrix that involves streaming audio
that
> >is being processed at 32 bit float in Cubase SX over a 20 bit lightpipe
to
> >be summed in Paris, here's my take on this:
> >
> >I notice absolutely no sonic hit at all. I think this may be because
there
> >is some incredibly sophisticated stuff going on in the Paris DSP. It's
> >really strange how this works. Paris can recieve a signal at the ADAT
input
> >that is being processed in Cubase SX at 32 bit float *above* 0dBfs (it's
> >not clipping in Cubase SX as long as it stays floating and doesn't hit
the
> >mix bus) and though this signal is *techically* being truncated to 20
bit,
> >there is something going on in the Paris DSP that is apparently allowing
> the
> >entire bit depth of whatever audio is received over lightpipe to be
> retained
> >while the actual loudness of the signal is attenuated by -24dB. Well,
this
> >gives you an extra +24 dB of headroom to boost the signal again in Paris.
> It
> >may be truncated, but it's something inaudible to my ears and what comes
> out
> >of the Paris mix bus sounds *HUGE* because it *sounds* like multiple 32
bit
> >audio tracks being boosted by up to 24dB and then summed analog......at
> >least, that's my take on it. I certainly notice no distortion/graininess
at
> >all. I did try instantiating a dither plugin on each Cubase audio track a
> >while back and the cumulative dither of 20+ tracks sounded hissy and
> >awful.It sounds like voodoo, and I know that bist can't be restored after
> >truncation, but the extra 24dB of headroom that Paris provides makes it
> >sound like it's doing exactly this, and then some. I knew there was
> >something like this happening due to the additional headroom available
but
> I
> >wasn't sure how much the audible input signal level was being reduced
until
> >I talked to someone (who must remain anonymous right now) who has
actually
> >measured this.
> >
> >anyway.........whatever is happening, it sounds awesome.
> >
> >Deej
> >
> >"Sound Dog" <dogster@tpg.com.au> wrote in message
news:44996537$1@linux...
> >> Say, does anyone know whether you need to dither a 24bit signal to
20bit
> >> when sending audio back to PARIS via lightpipe (from Logic via and
> >HDSP962
> >> for example) ?
> >>
> >> Thanks,
> >> Stewart.
> >>
> >>
> >
> >
>
>
|
|
|
|
| Re: Adat cards, lightpipe transfer & dither [message #69712 is a reply to message #69656] |
Tue, 27 June 2006 08:44 |
Sound Dog
Messages: 44
Registered: October 2005
|
Member |
|
|
>> I tried a couple of tests - firstly I extracted a track from CD (Crowded
>> House "Pineapple Head") into Logic, then recorded that into PARIS via
SPDIF
>> in two different ways:
>>
>> HDSP9652 SPDIF -> PARIS SPDIF (Logic on Mac sync'd to ADAT signal from
>>PARIS on PC, using PARIS as master deck with Logic slaved via MTC) then
>> HDSP9652 ADAT -> PARIS ADAT (Logic on Mac sync'd to ADAT signal from
PARIS
>> on PC, using PARIS as master deck with Logic slaved via MTC).
>> I can't hear a difference. There's no level loss or anything. 0dBFS in
>> Logic = 0dBFS in PARIS.
>
>I know........it's bizarre. Everything seems to be the same. Now set Paris
>to post fader, set the Paris fader to unity, process a track in Logic with
a
>VST plugin while sending it to Paris.............not start raising the
Paris
>fader and listen to what happens.
Sorry, you lost me here. Set PARIS to post fader ?
Also, I don't have any VST plugs - do you know if Logic's proprietry plugs
exhibit the same behaviour regarding bit depth ?
Cheers,
Stewart.
|
|
|
|
Goto Forum:
Current Time: Fri Nov 29 23:05:01 PST 2024
Total time taken to generate the page: 0.03134 seconds
|
The PARIS Forums
Search
Help
Members
Register
Login
Home
