I have a question that may not be that clever, but somehow I don’t know exactly and I want to get the best out of the sound.
So if I create a sound in ableton for the OT in order to edit it there, what must be taken into account when exporting if OT is set to 16Bit? Doesn’t matter if I export the signal in 24 bit? The signal should be exported perfectly, but the OT should somehow stay on 16bit, because I have the feeling the sound is so robust … but the sample should be of good quality.
I would only create 24-bit samples unless CF card space is an issue. This way you are not bound to the lower bit depth, but you can choose it for each project differently.
Static machines perform the down-conversion from 24-bit to 16-bit on the fly. Flex machines convert them when the samples gets loaded into RAM.
There shouldn’t be any notable quality difference when using 16-bit samples directly or let the OT do the conversion.
…ot can work both wordwidth at the same time…so in first place, no worries for now…
as long 44k matches all will be fine…
u can also load higher resolutions…but ot will read them as 44k files anyways…which will cause it to play those slower…
16 bit won’t give u full dynamic range…and yes, ur first impression is correct…16bit tends to sound more rocksolid 90ish right in ur face…
but for full resolution and dynamic range experience and since pretty much all daws spit out in 24bit by default, the easiest procedure whenever u start to create/bounce ur own samplechains for later use and abuse in the ot would be to nail them out all in 24bit 44k…
u can always downgrade, respectively push some harshness to the soundexperience within the ot via various fx edit options waiting there for u later on and at any further moment in ur workprocess…
I’m honestly a big fan of setting up my samples on the OT to be as space efficient as possible. I know 24 bit gives you better quality. But I don’t use my OT as a production tool. It may be getting an antiquated practice now, but I like keeping a separate folder. One with an efficient sample library to send out to devices. That way if I’m ever using something with space limitations or load times, it’s not a problem. This also includes having appropriate samples be in mono, unless stereo is required. But I make all of my own sample libraries myself, so it’s a little easier to do that.
I think the worst thing to do on OT is dump long blocks of big loops/stems. It gives you less options for crossfader scenes. Much better to have shorter samples. I always pre-plan out my live sets in a DAW and prep the samples to be Octatrack friendly. This includes truncating loops to get the essential information, and planning what needs to be flex for slicing. Then I name and load my samples into eight folders ( One for each bank ). Makes remixing into the OT so much easier and efficient.
Hi @tnussb.
I read that 24 bit is better for mixing many tracks, but I wonder if it change much with OT’s 8 tracks…
Someone interested to make a mixing comparison in OT using 24 bit samples vs 16 bit ? I don’t use my DAW atm, so it’s difficult for me. Or maybe someone have real* 24 bit loops for tests?
24 bit is recommended for recordings, because of noise floor, but with good gain staging it seems really hard to make the difference.
I use 16 bit only, for space reasons. I may use 24 bit for a serious project recording with OT.
Right. With good gain staging you shouldn’t hear much of a difference just due to the summing (if at all depending on your hearing).
And as you said: it all depends on your use case. My projects are normally not really memory hungry, so 24-bit is baked into my default templates (just because i can). But the world survived long enough with 16-bit and less and songs from that era still sound good …
In the 90’s most people listened to CDs 16/44.1 a lot of samplers being used were still 12bit though there were some 16bit samplers by then of course, but even 8 bit samplers were still being used.
Today most people listen to lossy formats, on earbuds, or phone/tablet speakers, thru algorithms which change the music to suit the platform.
That said I agree with @tnussb if sample time isn’t an issue use 24bit, but 16bit is perfectly good enough if you need more sample time.
Just use 16bits everywhere, technically it allows 96db of headroom, and it’s more than enough. For comparison, a modern dance music has a headroom of 3-4db in the final master.
Increasing to 24bits in a playback/sample is totally useless, 96db of headroom limit is around a noise level of your gear.
It’s not about noise in general (due to noisy electronics), but artificial “noise” introduced due to the rounding of the analog value into a digital value with a fixed number of bits (quantization error / quantization steps).
