Is DP 32 bit floating point?

[cboy1]

Ok so I read this weekend about floating point sounding betterr for a wider stereo image, and couldn't remember if DP was floating or not.
The point was that you could mix in the box if the software you had was 32 bit floating like Nuendo...... Hmmmmmmm? Being a die hard DP user that got me wondering like I'm sure it does for so many in any platform they use ( is mine better or at least comparative) but that got my wheels turning fast since I mix OTB could it really be that different to mix in the box using floating point?
Any thoughts or comments, someone engage this one who might mix in or out since I haven't mixed in for a couple of years now my ears and perception are trained to here my mixes a certain way.

[Armageddon]

I'm not sure, but 32-bit is grayed-out as an option on my system. It may be because I don't have a 64-bit machine or that my audio interface doesn't support it, although I'd assumed the 32-bit option was contiguous to DP and the MOTU Audio System, not the machine/interface it was running on. If it's just my interface, I suppose I could just unplug it and try running DP through my built-in audio when I go for a mixdown. Though, if the 32-bit float is just happening during the mixing stage, then, again, it shouldn't matter what I'm running on (DP isn't outputting 32-bit files in any case, as far as I know, so there has to be a down-conversion back to 24-bits for the final output).

[Dan Worley]

Yes, DP is 32-bit float. Always has been, as far as I know.

Arm, I don't know why it's grayed out on your machine. Never heard of that. It will be interesting to find out why. But that setting is not keeping DP from running 32-bit float. That's just for writing files. I've never used it.

c-ya,

Dan Worley

[carrythebanner]

If the file format is set to Sound Designer II, you can only choose 16- or 24-bit integer; 32-bit floating point is not available for SDII. If you want to use 32-bit floating point for the audio file sample format, choose AIFF or WAVE as the file format.

[Phil O]

Be sure to make the distinction between file format and internal engine. DP's internal engine is 32 bit floating point, but the file format for projects is selected under Prefernces (or in the control panel). You can select 32 bit floating point as a file format if you like, but as carrythebanner points out you must select bwav or aiff as well.

I've been using 32bit bwav as a project format for a while now and unless something better comes along, I will be for some time to come. Some here will argue that it's overkill, but I've been getting great results and the only disadvantage, as far as I can see, is slightly bigger file size. It's a format that the mastering engineer that I send most of my clients to accepts, and it's the standard that I've adopted (for now anyway).

Phil

[Prime Mover]

Short answer: Yes

In-depth answer: Yes

next question...

[Dan Worley]

Be sure to make the distinction between file format and internal engine. DP's internal engine is 32 bit floating point, but the file format for projects is selected under Prefernces (or in the control panel). You can select 32 bit floating point as a file format if you like, but as carrythebanner points out you must select bwav or aiff as well.

I've been using 32bit bwav as a project format for a while now and unless something better comes along, I will be for some time to come. Some here will argue that it's overkill, but I've been getting great results and the only disadvantage, as far as I can see, is slightly bigger file size. It's a format that the mastering engineer that I send most of my clients to accepts, and it's the standard that I've adopted (for now anyway).

Phil

Phil, you hear a difference?

Thanks,

Dan Worley

[daniel.sneed]

Keep in mind that DP stock plugins ARE 32 bit float, but many third party plugins are NOT.
Among them: iZotope, Waves... Sigh!

I wish they would all go the 32 bit float road.
This would free us from avoiding clipping anymore, except on the very last DAW stage before DA converters.

[Dan Worley]

Keep in mind that DP stock plugins ARE 32 bit float, but many third party plugins are NOT.
Among them: iZotope, Waves... Sigh!

I wish they would all go the 32 bit float road.
This would free us from avoiding clipping anymore, except on the very last DAW stage before DA converters.

I don't know about the other iZotope plug-ins but Ozone 4 does 64-bit calculations internally, according to their literature. Doesn't that mean something good? Come to think of it, I've never heard Ozone introduce clipping distortion. I've heard it not sound good (mostly because of me), but never with clipping distortion.

c-ya,

Dan Worley

[FMiguelez]

Keep in mind that DP stock plugins ARE 32 bit float, but many third party plugins are NOT.
Among them: iZotope, Waves... Sigh!

REALLY?!?!?!

I could've sworn Waves' plugs processed the data at at least 32 bit floating point...
I remember reading in their manuals that NATIVE plugs processed everything at 32 bFP (but II could obviously be confused... Please clarify).

So how are they processed then?

[FMiguelez]

.

I couldn't DREAM of recording my audio tracks at 32 bFP... not a chance in hell with my G5.

BUT, if one records everything at 24 bits, and bounces or prints the final mix at 32 bFP just before mastering it, that should suffice, me thinks. That would be "pure" enough for the ME, yes?

Certainly kids with their MP3s, iPods and earbuds won't complain about it

[daniel.sneed]

Hey FM and Dan, I didn't say they don't sound good. The whole thing is about fixed versus floating point.

No clipping occurs at plugin stage with floating point plugs, such as DP stock plugins.

AFAICR, waves plugs are 64 bit fixed point.

[Tritonemusic]

I don't know about the other iZotope plug-ins but Ozone 4 does 64-bit calculations internally, according to their literature

I just wanted to add this quote from MOTU's website:

64-Bit mastering tools for crystal-clear sound

The master fader is just the beginning. Digital Performer ships with mastering tools including MasterWorks™ Limiter, Multi-Band Compressor, Leveler and Parametric EQ. Bring out the ultimate detail from full-bandwidth program material. With 64-bit, floating-point precision, these tools preserve and enhance the natural sound and fidelity of your 24-bit audio with no added noise.

I wonder if MW EQ is included with those...

[daniel.sneed]

If you don't take it, please, make a simple and fast test.

On an audio track, insert two DP trim plugs.
- first set to +40dB
- second set to -40dB
- listen to the track while looking at the fist trim plug. It will certainly clip hard red, but the track will sound totally clean.

Fixed point plugins do not behave like that.

I usually track hot, so clipping may often be a concern to me.
I wish all developers go the floating point road.

[FMiguelez]

Hey FM and Dan, I didn't say they don't sound good. The whole thing is about fixed versus floating point.

No clipping occurs at plugin stage with floating point plugs, such as DP stock plugins.

AFAICR, waves plugs are 64 bit fixed point.

But I never said you said they didn't sound good!

It's just that now I'm confused. I always thought Waves' NATIVE plugs processed stuff at 32 bFP. And I thought their TDM versions processed stuff at something else I don't care about (because I own native). Probably 64 fixed point?

I'll see if I can find about this info at Waves' site.