Samplair
I would expect it to make some difference, if you do any processing in cubase (moving a fader is processing too in this respect).
32bit floating or 32bit integer whats purest for recording?
Hello! 
-No. Anything you do inside Cubase will render the audio to 32bit flt.What if I don't fader move but put 32bit effects in insert on Cubase channels will the 32bit fixed still exist ?
Dimitrios
- ASIO1 Flt 64 (for 32bit flt recordings), or ASIO2 Flt 64 if you have a Sync Plate and want a sample accurate synchronization with an External Digital Device.On 2004-02-11 08:16, braincell wrote:
I always wondered about which ASIO module I should use.
Totally agree!On 2004-02-11 03:00, garyb wrote:
everything in the sfp realm is 32bit which allows a bit more accurate math(headroom),however,sx is 32flt which is 24bit with room for overflow.you sound files will be 24bit at best.there is no problem with this. a well recorded track will sound good enough to win an award
By the way is there any app that can record 32bit fixed except for VDAT ?
Also what's the difference that will be if I use ASIO 32bit fixed for recording into cubase instead for 32bit floating?
I will use 32bit floating for mixing back to SFP (I will use Cubase inserts for Powercore and UAD-1 vst effects)
Dimitrios
Also what's the difference that will be if I use ASIO 32bit fixed for recording into cubase instead for 32bit floating?
I will use 32bit floating for mixing back to SFP (I will use Cubase inserts for Powercore and UAD-1 vst effects)
Dimitrios
samplitude is 32bit fixed afaik.On 2004-02-12 09:23, musurgio wrote:
By the way is there any app that can record 32bit fixed except for VDAT ?
Also what's the difference that will be if I use ASIO 32bit fixed for recording into cubase instead for 32bit floating?
I will use 32bit floating for mixing back to SFP (I will use Cubase inserts for Powercore and UAD-1 vst effects)
Dimitrios
as far as cubase goes,probably not much audible difference between 32flt and fixed although if you were using cubase in 32flt i would expect the flt module to be a better match.
really a non-issue.there are many better ways to spend your energy in order to get a great sound.
with this I completely agree! last times i just tweak around all this non-issues and I can not get any ready song at all...On 2004-02-12 12:48, garyb wrote:
as far as cubase goes,probably not much audible difference between 32flt and fixed although if you were using cubase in 32flt i would expect the flt module to be a better match.
Considering I'm in XTC and I don't want to edit that *.pro often I found that 32flt asio charge the system harder than 32-source64 and noticeable when I record the mixdown of "heavy" projects (several channels with VSTis and FXs both native and XTC). "Any audible diference" only in clips&pops sooner or later
But comes the point where I find myself confused around 32flt or not (SX talking).
In fact with asio2 dest-64 the SX setup shows asio 24 bit but if I rec with 32flt in project setup the resulting *.wav file gets larger and in the properties its specified as 32 bit file. Are this just issues?
really a non-issue.there are many better ways to spend your energy in order to get a great sound.
-
Nestor
- Posts: 6683
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
The matter, instead of choosing which ASIO to use, would be to know exctly, WHICH ASIO to use in relation to WHICH audio edditor or sequencer. This is the matter I understand. Correct me if I'm wroing.
Perhaps, those more knowledgable, could give a clear direction to how the whole "chane" should be done.
For instance: if you use Cubese 5.1 or 32, there should be a more convinient way of working than in SX, using a mere 24 bit resolution ASIO driver... What do you want 32 bits for, if you cannot handle it in Cubase? But, if you use SX, you should search perhaps for the ASIO1 flp.
Am I wroing thingking about this as a combination matter?
Perhaps, those more knowledgable, could give a clear direction to how the whole "chane" should be done.
For instance: if you use Cubese 5.1 or 32, there should be a more convinient way of working than in SX, using a mere 24 bit resolution ASIO driver... What do you want 32 bits for, if you cannot handle it in Cubase? But, if you use SX, you should search perhaps for the ASIO1 flp.
Am I wroing thingking about this as a combination matter?
*MUSIC* The most Powerful Language in the world! *INDEED*
-
powerpulsarian
- Posts: 136
- Joined: Tue Dec 30, 2003 4:00 pm
you've had almost one and a half year to complete KimGR's homework... 
imho it's all rather esotheric unless it comes to a world-class recording session in a world-class concerthall with... you name it... gear.
And even in those cases I remember a quote from my favourite violinist, who said that todays recording systems are so damn unforgiving that a 'regular' performance is almost impossible for the artist - she referred to a 20bit Sony direct stream thing btw.
as a rule of thumb you could analyse your signal path and detect it's weakest points - how much distortion would be introduced respectively.
in 99.9% of the cases it will be less than the difference between 24/32 bit or the difference between float and fixed.
then each audio engine/application does it's own processing of which only the programmer (if he or she's interested at all) might know the details.
a bad fixed algorithm will sound worse than a good float - and vice versa
then there's the inevitable drifting of clock (aka jitter) which will introduce another level of signal degradation
then most of us will have to live (and mix) with way less than optimal room conditions and this will lead us to false perceiption which WILL spoil the result more than any bit error ever could.
and last but not least our ears - who's had his (or hers) checked by a physician the last 12 months ?
aside from the fact that anyone (by being used to it) considers his hearing perfect and objective (often simultaneously taking personal flaws into account, if known) - the hearing is a dynamic process.
You get used to whatever - your listening adapts.
Enter a room with horrible speakers, recognize the crappy sound - then (try to) forget about it and spend (half) an hour in this environment doing whatever you like.
If you turn to the 'music' again after that period, it will probably sound not as bad anymore, as it did when you entered the room.
imho the best format is the one that doesn't annoy you and is most convenient
cheers, Tom
imho it's all rather esotheric unless it comes to a world-class recording session in a world-class concerthall with... you name it... gear.
And even in those cases I remember a quote from my favourite violinist, who said that todays recording systems are so damn unforgiving that a 'regular' performance is almost impossible for the artist - she referred to a 20bit Sony direct stream thing btw.
as a rule of thumb you could analyse your signal path and detect it's weakest points - how much distortion would be introduced respectively.
in 99.9% of the cases it will be less than the difference between 24/32 bit or the difference between float and fixed.
then each audio engine/application does it's own processing of which only the programmer (if he or she's interested at all) might know the details.
a bad fixed algorithm will sound worse than a good float - and vice versa
then there's the inevitable drifting of clock (aka jitter) which will introduce another level of signal degradation
then most of us will have to live (and mix) with way less than optimal room conditions and this will lead us to false perceiption which WILL spoil the result more than any bit error ever could.
and last but not least our ears - who's had his (or hers) checked by a physician the last 12 months ?
aside from the fact that anyone (by being used to it) considers his hearing perfect and objective (often simultaneously taking personal flaws into account, if known) - the hearing is a dynamic process.
You get used to whatever - your listening adapts.
Enter a room with horrible speakers, recognize the crappy sound - then (try to) forget about it and spend (half) an hour in this environment doing whatever you like.
If you turn to the 'music' again after that period, it will probably sound not as bad anymore, as it did when you entered the room.
imho the best format is the one that doesn't annoy you and is most convenient
cheers, Tom
-
powerpulsarian
- Posts: 136
- Joined: Tue Dec 30, 2003 4:00 pm
I don't know - I use VDAT to make 16 bit recordings... 
Of course it's much more convenient (for processing) to have an increased headroom of 24 or even 32 bit - but the sound itself doesn't get (fundamentally) better (that's wishful thinking imho) if the target is an audio CD.
with my sources 16 bit is ok (the old A16 has 18 bit converters anyway)
Aside from that I don't think that the byte-padding will put much (extra) load on current CPUs
cheers, Tom
Of course it's much more convenient (for processing) to have an increased headroom of 24 or even 32 bit - but the sound itself doesn't get (fundamentally) better (that's wishful thinking imho) if the target is an audio CD.
with my sources 16 bit is ok (the old A16 has 18 bit converters anyway)
Aside from that I don't think that the byte-padding will put much (extra) load on current CPUs
cheers, Tom
Quite A Thread Here.
I just recently began exploring Scope VDAT. I remembered a live recording last year which floored me. Wasn't sure if it was the FOH console or VDAT that made such a high quality recording. For live recordings always suck IMHO for control due to Ambient leakage. Theyr'e great for capturing superb performances that can't be duplicated in the studio.
A local production facility here in town has a Neve 53 ( circa 1971 ) console, which my friend is the house engineer, and has recorded on for 20+ years.
He prefers Pro Tools as his choice of weapons. Afterall, he has recorded with it for 14 years. So of course he will never stray. But I asked him to let me a/b my beloved Scope VDAT against it, using the same source material. He said that he loved the quality of VDAT, but of course began pointing out why he uses Pro Tools. All valid responses,i.e. better GUI, more power,etc. But my ears hear no difference in quality. Until we burned both to CD I was bloviating.After listening to the CD the quality of both seemed compressed and equal. There is nothing like good nearfield monitors and a kick ass Neve IMHO. I just wish that there was a better media than CD/DVD to listen to. Well actually there is. I still prefer live. But it is so nice to put a staunch Pro Tools engineer on the defensive. I am more confident now about my investment.
Jimmy V./ Punisher of VSTi's
<font size=-1>[ This Message was edited by: scope4live on 2006-03-08 17:25 ]</font>
I just recently began exploring Scope VDAT. I remembered a live recording last year which floored me. Wasn't sure if it was the FOH console or VDAT that made such a high quality recording. For live recordings always suck IMHO for control due to Ambient leakage. Theyr'e great for capturing superb performances that can't be duplicated in the studio.
A local production facility here in town has a Neve 53 ( circa 1971 ) console, which my friend is the house engineer, and has recorded on for 20+ years.
He prefers Pro Tools as his choice of weapons. Afterall, he has recorded with it for 14 years. So of course he will never stray. But I asked him to let me a/b my beloved Scope VDAT against it, using the same source material. He said that he loved the quality of VDAT, but of course began pointing out why he uses Pro Tools. All valid responses,i.e. better GUI, more power,etc. But my ears hear no difference in quality. Until we burned both to CD I was bloviating.After listening to the CD the quality of both seemed compressed and equal. There is nothing like good nearfield monitors and a kick ass Neve IMHO. I just wish that there was a better media than CD/DVD to listen to. Well actually there is. I still prefer live. But it is so nice to put a staunch Pro Tools engineer on the defensive. I am more confident now about my investment.
Jimmy V./ Punisher of VSTi's
<font size=-1>[ This Message was edited by: scope4live on 2006-03-08 17:25 ]</font>
-
Music Manic
- Posts: 1743
- Joined: Wed May 15, 2002 4:00 pm
- Contact:
Re:
I know this was posted a while back, but wanted to confirm if it's true. I thought it was the other way around.kimgr wrote:The difference between using asio-32(& 24) or asio-flt is: Where's the conversion done ?
If you use -32, the host does the conversion to float.
If you use -flt, the conversion is done by the Creamware drivers.
Todays homework: What sounds best on the various hosts ???
Kim.
Aren't the algorithms of conversion the same? It was posted that the -flt ones were faulty!
Thanks
Re: 32bit floating or 32bit integer whats purest for record
that quote by kimgr certainly was correct 7 years ago...
things have evolved since then, in particular DAW software - Cubase was an audiophile's nightmare back then
Sonic Core has thinned out the Asio mess in Scope 5 anyway reducing in to Asio 32 (integer) and Asio float.
Both in version one and two flavour.
As a Scope system has superior monitoring capabilities (imho), there's no explicite need for version 2, which just adds the 'direct monitoring' in software part.
With good sources there still is a slight difference between the (Scope) monitor mix and what most(?) DAWs do.
Possibly related to conversion of numeric formats, but it's an academic discussion.
You CAN'T take that Scope mix to 99.9% of your listeners as they simply lack a Scope system for playback.
So the quality argument is moot.
You'd rather focus on the ENCODING process for the audio target.
Wether it's a CD or any of the digital online formats, that process (the encoding software) makes the audible difference.
A much(!) bigger difference than any rounding error could produce.
If you're not apple-phobic some browsing in the iTunes store 'preview' examples can be a nice hands-on experience of the bandwidth the results show even within a single format
cheers, Tom
things have evolved since then, in particular DAW software - Cubase was an audiophile's nightmare back then
Sonic Core has thinned out the Asio mess in Scope 5 anyway reducing in to Asio 32 (integer) and Asio float.
Both in version one and two flavour.
As a Scope system has superior monitoring capabilities (imho), there's no explicite need for version 2, which just adds the 'direct monitoring' in software part.
With good sources there still is a slight difference between the (Scope) monitor mix and what most(?) DAWs do.
Possibly related to conversion of numeric formats, but it's an academic discussion.
You CAN'T take that Scope mix to 99.9% of your listeners as they simply lack a Scope system for playback.
So the quality argument is moot.
You'd rather focus on the ENCODING process for the audio target.
Wether it's a CD or any of the digital online formats, that process (the encoding software) makes the audible difference.
A much(!) bigger difference than any rounding error could produce.
If you're not apple-phobic some browsing in the iTunes store 'preview' examples can be a nice hands-on experience of the bandwidth the results show even within a single format
cheers, Tom
-
Music Manic
- Posts: 1743
- Joined: Wed May 15, 2002 4:00 pm
- Contact:
Re: 32bit floating or 32bit integer whats purest for record
Astroman, I'm juggling a few situations in my mind and they are all related to wordlength.
If I just have audio files playing in my DAW (Cubase), without any fader changes then the ASIO flt maths will be done by Scope right?
What would happen with either 32 or flt ASIO when faders are adjusted by the DAW and Scope plus I use VST and VSTi plugins?
How much mathematical processing is going on there and how many times would the wordlength be changing before it reaches Scope out?
I am concentrating on Cubase outputting the audio signal and not so much the recording into it:
Further to what I said, if I have:
1: Multiple audio channels
2: Vst plugins which are 32bit
3: Vsti plugins ,
all of these streamed on their own ASIO channel. What is doing what?
Is Cubase processing the maths before Scope changes it to integer?
I'm a lttle confused about the math process going on.
I get some very high level from vsti plugins in Cubase and wondered if it would help wordlength processing if the levels were lower. The thing is that I always keep the faders in Cubase at zero so Scope can get the "purer" signal.
Much appreciated!
If I just have audio files playing in my DAW (Cubase), without any fader changes then the ASIO flt maths will be done by Scope right?
What would happen with either 32 or flt ASIO when faders are adjusted by the DAW and Scope plus I use VST and VSTi plugins?
How much mathematical processing is going on there and how many times would the wordlength be changing before it reaches Scope out?
I am concentrating on Cubase outputting the audio signal and not so much the recording into it:
Further to what I said, if I have:
1: Multiple audio channels
2: Vst plugins which are 32bit
3: Vsti plugins ,
all of these streamed on their own ASIO channel. What is doing what?
Is Cubase processing the maths before Scope changes it to integer?
I'm a lttle confused about the math process going on.
I get some very high level from vsti plugins in Cubase and wondered if it would help wordlength processing if the levels were lower. The thing is that I always keep the faders in Cubase at zero so Scope can get the "purer" signal.
Much appreciated!
Re: 32bit floating or 32bit integer whats purest for record
yes, and no.
if you just go out the stereo output of cubase, all the math for plugins, "summing", fader changes, everything but the actual rendering of the final audio signal is done in cubase, on the cpu. Scope merely passes the data to the ad/da.
if you use plugins in Scope, the calculations are done on the dsps. if send each track of a multitrack project out it's own asio channel, and mix in Scope, then the math will be done on the dsps.
it's not either/or, any hybrid form works. the math will be done in the system given the task. Scope is really external hardware to the computer, in that it's a seperate processing unit. there's really no way for one to be on the other. they can share info, but not tasks(for the most part).
if you just go out the stereo output of cubase, all the math for plugins, "summing", fader changes, everything but the actual rendering of the final audio signal is done in cubase, on the cpu. Scope merely passes the data to the ad/da.
if you use plugins in Scope, the calculations are done on the dsps. if send each track of a multitrack project out it's own asio channel, and mix in Scope, then the math will be done on the dsps.
it's not either/or, any hybrid form works. the math will be done in the system given the task. Scope is really external hardware to the computer, in that it's a seperate processing unit. there's really no way for one to be on the other. they can share info, but not tasks(for the most part).