Hi,

For some reason I decided to give up on my 32 bits Cubase with Waves V9, and to upgrade to 64 bits Studio One v5 with Waves v12. I have some unfinished projects so I created a song with the exact same settings on all plugins, yet everything sounds different. The big kind of different (harsh, muddy, poor definition).

Did this happen to someone ? Where does this come from : DAW or Waves ?

I'd like to have some feedback before trying to replicate my former sound on each track (wilm be a pain in the...), maybe I missed something

Thx in advance
Max

 

Last edited: Oct 18, 2021

Oh I actually forgot to say that I'm going for 64 bits. But yeah, it sounds well worse since the upgrade. I still have my 32 bits setup up and running, I'm testing the new setup on another PC btw. So I can compare everything in real time.

 

Oh gosh I feel so stupid... I exported my Cubase project in XML then imported it into S1. I didn't check, I did the Cubase project in 16 bits (don't judge please ) and S1 was in 24 bits. I thought everything would be identical.

It did make a little difference, and I identified the culprit for the rest : I went from iZotope oZone 5 to oZone 9, and this one sounds REALLY different.

Thx for your messages guys, I think this thread is closed !
Max

 

While there is a big thing in exporting on OMF, AAF and MusicXML, mainly to keep layout structures and plugins etcetera, honestly, I found only a slightly lengthier process that is far more accurate and prone to zero issues if it is only for you in another DAW and not someone else.
Nearly every DAW has a bulk track export function with the ability to turn inserts and buses and all plugins off on export.
If you take the time in each plugin to save your setting as a name associated with your track, you only have to import the audio files, load the plugins and load the fxp or whatever setting you saved in your plugins. Even if you changed the sample rate, if the new DAW is higher in rate it will import the unprocessed audio correctly.

 

The bits here have no relation with sound quality!
The only difference is the amount of Memory that 64bit software can allocate/address vs 32bit software.

Other than that it has no relation with sound quality, audio bit depth, or sampling quality at all!


-To the OP, if the settings are equal on both DAWs and all Plugs, it shall give pretty much the same sound/result.

Seems you found out XML bit depth culprit..
Still make sure you check/match the Pan Law.. this doesn't have an effect if tracks are not panned.
But if they were panned, the difference would be massive

 

Last edited: Oct 18, 2021

From a purely mathematical point of view, I agree with you.

A mystery:
But how come that so many people, when they switch from a 32 bit to a 64 bit DAW, say oh now the bass sounds better and deeper.
(Subjective or objective) Or is it all nonsense and one imagines it?

 

Well there's different types of Bits.

There's Software Bits = The amount of Memory that can be addressed/allocated. For instance Windows7 32bit vs 64bit

There's Audio Bit Depth = This determines the audio sampling quality, so the more the better!

DAWs Internal Processing bits = A Daw will process the audio Internally, doing it in 64bit allows for more precise manipulation,
adjustment (mixes can be more precise), and also cleaner processing..

So the OP's scenario was more about Software Bits,
unless we were talking about a very elaborate/complex plug with internal mixing, like a Drum VST or something..

 

Last edited: Oct 18, 2021

A very important thing when switching DAWs is the pan law behaviour. Most of the time people complaining about mixes sounding different on different DAWs with same settings didn't check this.

 

Fabfilter pro-q is 32 or 64 bit, but soundwise is 32 bit float only
Voxengo are 32 and 64 bit, but soundwise are 64 bit float all.

Don't mess up binary vs DSP engines

 

I didn't quite get the point. Could you dwell on this issue in more detail. What do you mean by "soundwise".

 

First. Let me say that you have 'switched/changed' DAW platform. Not upgraded.
Second. Cubase handles Pan law differently than Studio One and, generally has more sonic headroom.

If the project in Studio One running at a higher sample rate Example:
48.1kHz, 24 bit, 32 bit float vs 44.1kHz, 16 bit, the possibility of audio artifacts & degradation can occur. This is why expensive external converters exist.

 

Last edited: Oct 18, 2021

The internal math is the same (that would be "soundwise") while the software "interface with the CPU" (for lack of a better term) can be 32 or 64bit.

 

I will say more: some audio software plugins use 80-bit extended float, and they can be either x64 or x86 (32 bit).
Reaper - you can render your audio as 64 bit float, even if reaper and all plugins there are 32 bit (x86) only, but 64 bit DSP engine

 

For the 32 vs 64 thing, I didn't notice anything different regarding bass depth and overall clarity. That said, I'm not a pro engineer, my mixes are maybe not good enough to make it possible to notice.

@No Avenger It may be psychological, but once I passed S1 in 16 bits I think I lost a bit of "harshiness" in the treble area. But my masters are compressed AF (I gave in to the volume war unfortunately...) and that may help this effect too.

@Fuxor My biggest concern was precisely the sonic similary of Waves v9 and v12, which I thought were the culprits. But man, how was I wrong. I'm dumb af, I noticed my Dynamics module on oZone 9 had gain additions on 3 bands, hence the harshiness... It was all clipped before the Maximizer and I wasn't aware. Once back to 0 gain, everything sounded the same as my previous setup on oZone 5.

@BaSsDuDe The Cubase 5 XML export was a failure to me. If you have tips for me, I'll gladly take them. I got all the tracks and volumes, but no inserts, sends nor panning. I ended up adding the plugins 1 by 1 and enter manually the settings (once again, I'm so happy to have a laptop to work on to setup the new rig !). So basically, I did everything from scratch except the track creation !

@quadcore64 Let me rephrase : I upgraded from a 32 to a 64 bits setup, switched to S1, and updated Waves v9 to v12. Pan won't be a problem, since I pan my guitars L/R and Superior Drummer takes care for its own pan. But I think I noticed the lesser headroom on S1.

And once again, thank you people for your lightspeed answers, much appreciated !

 

No need to rephrase. Pan Law is always at play when moving projects between different DAWs. Some reference below.

• Pan Law: https://en.wikipedia.org/wiki/Pan_law#:~:text=Pan law, or pan rule,perfect acoustics in the room.
• DAW v DAW - Part 6: Pan Laws and Pan Curves : https://www.admiralbumblebee.com/music/2019/12/08/Daw-V-Daw-Pan-Curves.html
• How to use pan law: https://understandingaudio.com/how-to-use-pan-law/

For audio sample rates, most modern DAWs will playback 44.1kHz projects in a 48kHz project with little audible difference. It is best to use the same sample rate, bit rate & floating point as the original. this ensures that aliasing is kept under control.

As you have noticed...Studio one has less internal audio headroom compared to other DAWs like Cubase, Pro Tools, & Reaper. You also have to play with drop out protection in the buffer settings in Studio One.

 

@No Avenger I chose 16bits to spare some resources, since I'm doing hardcore metal. There, the dynamic range isn't really a priority :p I did some tests before and didn't notice a difference in noise between 16 and 24 bits recording (on my particular project, 1 hour for 24 songs... quite eating resources)
Regarding oZone, I remember back in the day that I switched to O6 when it went out, and I was SO disappointed. I came back to O5 in days. I didn't have this feeling once I realized what was wrong in O9, now I'll gladly keep it. I didn't even give O7 and 8 a try. But now I'm happy !

@quadcore64 Wow OK, now that's why the volumes are unbalanced with identical faders from C5 to S1. Not a big deal, that's why I didn't mention it until now, but it's true.

 

a quick google search gives me 64 bit float as internal bit depth that would be like 2820dB internal headroom...