I have never in all these years been tempted to mess with different sample rates. My interface by default comes up in 44k1/24bit and I am happy with what I am getting.
 
But Sonar, up in the blue box, displays 44k/16bit.
 
Why the discrepancy? Am I getting the max out of my interface?
 
(24bit is what I see in the Audio device settings preferences and it is gray. So I couldn't change it even if I wanted to.)

The record bit depth is shown in the Transport module. The default is set in Preferences > File > Audio Data.

Playback is at 24-bit, but recording will be at 16. Check Preferences > File Data > Record Bit Depth.

It does mean you have been recording @ 16 bit rather than 24 -  which of course doesn't matter at all if as you say you are happy with what you are getting and mastering to a medium appropriate for that bit depth. 

Oh, wow. 16bit. All this time. How did that happen - is that the default?

I believe so

I just checked: My current PC is a Win7x64. I got it when Sonar X2 no longer supported WinXP.
I have Sonar X2, X3 and Platinum on it and all are set to record at 16bit!
 
But my old WinXP PC has Sonar 8.5 and X1 on it. They are both set to record at 24bit depth.
 
Could it be the default on the first X2 was 16bit?
And I have just been taking over settings with each upgrade?
This is not something I felt compentent to  mess with. I am sure I didn't change this.
 
I feel like I have been driving a Ferrari with the handbrake on.

@scook: Postings crossed whilst I was checking.
Thank you.
 
I have set them all to 24bit depth now.
 

jpetersen
I feel like I have been driving a Ferrari with the handbrake on.

It's not as bad as you think. Remember, a 24-bit interface doesn't give 24 "real" bits, it's more like 20 bits of resolution on a good day. If you were recording with a 16-bit interface, THEN the handbrake would be on.

Yes, but if I record a very dynamic performance extremely conservatively to prevent clipping at all costs and then I do gain staging/normalization followed by compression/limiting, it can sound like it's been through a bit crusher effect. the level will be pulled up but the resolution will still be that of the original recorded signal.
 
If I normalize a passage of classical music where a pure instrument like a flute is played very softly, I hear this. (not saying I'm doing this, but a female vocal on a soft passage is a real-word example).
 
Anyway, problem solved and so far nobody has complained.
I'll probably get told my new recordings don't have that gritty warmth anymore... :)