this is mabye a boring question but what´s the expierence of other experienced users? i think if you record from the a/d (which are 24Bit) it is enough to record in 24bit but if you record the pulsarinstruments then it is maybe better to record in 32 bit.
all i want is to achieve the best quality that´s possible.
32bit recording or 24bit in Cubase?
-
King of Snake
- Posts: 1544
- Joined: Fri Apr 13, 2001 4:00 pm
- Location: the Netherlands
- Contact:
keep in mind that, even with a signal from a 24 bit input, working in 32 bit will still give you higher resolution when using software fx and such. Wether the difference between 32 bit and 24 bit is significant enough though is a question only your ears can answer. Personally I think 24 bit is a good compromise between quality and filesize, but I'm really just a hobbyist producers at the moment and not a professional.
The resolution of 32 bit floating point files is somewhat the same as 24 bit integer files. The 32 bit FP format uses 24 bits to store its data and the remaining 8 bits contains information about where the decimal point needs to be placed.
So what does this mean? As long as your record level peaks stay close to -0 dB, there shouldn't be any difference. The only advantage of 32 bit FP over 24 bit integer is when you record material with a low level. Because of the floating decimal point, there always will be full 24 bit resolution between the loudest peak and silence, even if this loudest peak is far away from -0 dB. But as long as there aren't any 32bit AD converters, this advantage is useless.
In practice this means that there are few situations where 32 bit FP recordings are really usefull. One situation where 32 bit FP can be an advantage is when you internally bounce or render tracks with FX to new files in Cubase, because the internal resolution of Cubase is also 32 bit FP. Another situation is indeed when recording Pulsar synths, although you have to be sure you have loaded the ASIO2-flt (=32 bit floating point) drivers in your project. Still, if the output of the synth is close to -0dB, the advantage is negligible. So at the end I would advise to just use 24 bit as it results in files that are 1,5x as small as 32 bit files and it uses less bandwith of the PCI buss too.
cheers,
vincent
So what does this mean? As long as your record level peaks stay close to -0 dB, there shouldn't be any difference. The only advantage of 32 bit FP over 24 bit integer is when you record material with a low level. Because of the floating decimal point, there always will be full 24 bit resolution between the loudest peak and silence, even if this loudest peak is far away from -0 dB. But as long as there aren't any 32bit AD converters, this advantage is useless.
In practice this means that there are few situations where 32 bit FP recordings are really usefull. One situation where 32 bit FP can be an advantage is when you internally bounce or render tracks with FX to new files in Cubase, because the internal resolution of Cubase is also 32 bit FP. Another situation is indeed when recording Pulsar synths, although you have to be sure you have loaded the ASIO2-flt (=32 bit floating point) drivers in your project. Still, if the output of the synth is close to -0dB, the advantage is negligible. So at the end I would advise to just use 24 bit as it results in files that are 1,5x as small as 32 bit files and it uses less bandwith of the PCI buss too.
cheers,
vincent