I am using Ardour 5.12.0 from kxstudio repos. I have been recording with 32bit float, 48khz because this was recommended to me and/or was default.
I’ve recently read in various sources that my audio interface (Behringer UM2) only does 16bit. The Behringer website doesn’t give bit depth specs. I’ve not had any errors recording at 32 bit but have I just been wasting hard drive space? Is the bit depth a hardware of software thing ie could the um2 drivers be updated to allow 32bit?
I’m very confused, should I just change to using 16bit and get exactly the same result with much smaller file size?
This is strange. Behringer doesn’t specify the bit depth of UM2, as you stated. On bhp’s site is says that it is 24 bit but everywhere else I’m seeing 16 bit…
If it is 24 bit, then 32bit floating is perfect (32f is 24 bit with an 8 bit exponential to avoid clipping in the box with plugins and whatnot).
If it is 16bit, you aren’t doing any harm by using 32f and you still get the advantage of floating point bit depth. You are using more disc space but… are you really that tight on storage?
Thanks for the answer. I saw the bhp site saying 24 bit. Maybe it’s and updated model. I’m not short on disk space but don’t want to waste it unnecessarily.
It’s interesting about the float. I’ll have to do some reading on that.
Does ardour normally only offer the options that are available?
OK so it looks like my interface is definitely 16 bit. So, as ardour processes everything in 32 bit float (I believe) should I just set the session to record in 16 bit?
No, you should leave ardour recording in its preferred disk format (32 bit float). You don’t get any (trivial) access to what Ardour records without export, and that can be whatever format(s) you want, later.
Thank you. I think I’d just about come to that conclusion with the previous answer and some further reading but it’s great to hear from others. I’ll keep doing what I’ve been doing.