of the project setting…I thought it did…if it doesn’t I owe Foolomon an apology, my projects are set to 16 bit but I thought /think the audio engine works at 32 bit always… a recent conversation with a fellow muso who is cleverer than me on puters made me think
so basically I’m thinking that the audio engine is working at 32 bit float but my setting of 16 bit means it is rendered within the project to 16 bit…which means the project is actually 32 bit float…
I’m just a simple photographer…music is a hobby,I really don’t like the tech side of it, can some one clear this up for me in simple terms please…
wow…what a quick reply…so your saying that even if my project is set to 16 bit it’s actually using a 32bit float engine then rendering it to 16 bit within the project…
it’s 32fp throughout up until the fader on the 2bus. some older plugins may still not internally process with a fp resolution (like older waves plugins i believe; in that case you may need to use gainer plugins in front of such plugins in case you’re running too hot a signal into them). some, on the other hand, may upsample internally etc. but that’s out of control of cubase’s processing, which, as said before, is internally 32fp.
Cubase processes in 32bit-float internally. Your project is 16bit in that when you record, files will get created in 16bit reaolution (it is not possible to record in 32bit-float).
Exactly. Due to the IEEE specification on how floating point numbers are stored in memory, it’s possible to get a much higher degree of precision in 32-bits using floating point than if Cubase used integer or fixed decimal arithmetic.
so if i import a file into cubase and it gives me the option of converting it to 32 bit what is it realy happening ? and if i set a project to 32 bit also what is happening to the files being created,
also i have seen audio files people have uploaded to the net which are supposedly mp3 32 bit,
you’re setting the bit depth for recording / resampling that way. if you’re doing a lot of recording /resampling of low-level material, higher bit depth of the project may be favourable. once cubase grabs these files and passes the audio through its engine, everything is floating point. setting the project to 32fp is also good if you’re resampling/bouncing internally signals that are very hot (above 0dBFS). you then retain resolution above 0dBFS. try making a 32fp signal that goes, say +10dB over the 0. then re-import that sound. you’ll see the waveform is clipped. but lower the event volume by 11dB and you’ll start seeing the original shape of the waveform, ie. not hard-clipped – it has retained the resolution thanks to the floating point representation.
Hey everyone. This topic would seem to possibly answer a question I’ve had since I upgraded from Cubase 4 (just last July - yeah I know ) to Cubase 11.
So while I’m working on my project (24bit 48k) everything sounds fine and clear, no squashed-to-death pumping, harshness, etc. But on export - it sounds like I’ve put the whole mix through a distressor or something (okay mild exaggeration)!
Might this be the reason? That the internal 32bit FP processing allows for all that extra headroom that cannot be accommodated when mixing down into a 24/48 wav file?
Thanks for all of your time and I’m looking forward to learning a lot from everyone here! (First time poster!)
Tony
) to Cubase 11.