Thanks for all the input folks. It definitely isn't a need, since apparently Digital Audio is good enough now that keeping the recording level lower no longer has a negative impact.
 
I agree, based on that, when compression is necessary (just by virtue of time constraints and wanting to keep it moving forward without doing retakes due to clipping) I'll go the external compressor route.
 
Thanks again all!

When you're recording, assuming you're working with 24 bit  just make sure your meters stay around -9dB to -12dB and you'll be golden come mixdown.

Before I started using the Roland Studio Captures, I always had a dbx compressor inserted into the channel on my mixer to catch those peaks before they even went in to Sonar. With the Studio Captures, 12 of the channels have compressors built in the interface. I no longer even use the mixer! This sounds like a commercial for Roland! YIKES!

I would still suggest, if you're frequently noticing the compressors kicking in, to turn down the input gain.

As someone said in the thread above, you can go through a preamp, connected to an external compressor (I use an ART PRO MPA II Reference series 2 channel preamp and an ART PRO VLA II 2-channel compressor) before going D/A via line level into your audio interface to the DAW.
 
Certainly some very mild compression, on some lead vocal, backing vocals, guitar and bass guitar can produce a great digital track to still be able to do a lot of things with if needs be for engineering or artistic reasons ITB (in the box) You also get a predictable character and tone which can differentiate your studio sound too if you are prepared to tweak and listen enough. A little is a good thing, and if you overdo it, you are stuck with what you have got unless you retrack. YMMV
 
Garry Kiosk

If 20+ years makes you "old skool", then what does that make me? Prehistoric School? Yikes.
 
So speaking as someone who's also from an old school, I gotta say that the ability to NOT commit effects when recording is the greatest advancement in recording since, well, digital recording itself. Once upon a time we printed effects out of necessity, but now have no reason to do so because we've got unlimited devices and oodles of headroom.
 
If you're worried about clipping while recording, that's going to happen in your audio interface and must be dealt with on the analog side of things. A compressor/limiter plugin won't see the signal until it's too late to mitigate problems. What you need is an outboard hardware compressor in front of the interface. Many interfaces have them built-in, but if yours doesn't there are plenty of inexpensive compressors or mic pres with compression.
 
If you can't afford an outboard compressor, or want to keep your setup simple and compact, you can always use the really-old-school method of closely watching recording levels. In the digital world, we've got so much headroom to play with it's no longer necessary to record as hot as possible. Set your analog levels low enough that there's no danger of clipping and do a test recording. If your peaks are between -30dB and -12dB, you're good to go.

I agree with bit to a certain extent. You will probably find that a hardware compressor or one built into the interface makes your job a lot easier. Even if you make sure your level is low enough not to clip without a compressor, an inexperienced singer who shoves the mic down his throat one minute and stands 3 feet away from the mic the next is enough to put you over the edge sometimes. A more experienced singer will be less the problem. 
And as he mentioned, always remember that -12 to -15 is really the equivalent of hitting zero on an analog desk.

This is top on my list to hook into my saffire pro for this very reason, nice discussion:
http://uk.focusrite.com/m...s/octopre-mkii-dynamic

I go through a RND Portico channel strip when overdubbing or recording a single channel.  It has parellel compression built in, which is usually what I use.  I'm not worried about overs - just keep the levels sane (as per above) - but I like how the compression firms up a sound before hitting conversion and how it evens out the sound a bit.  It doesn't do much for a screamer as described above, who swollows the mike at full volume but backs off for the softer parts.  On regular dynamic tracks it does shave the amplitude a bit and the noisefloor is not a problem w/in my system.  It makes it much easier to mix later and I don't have to digitally squish the sound so much.
 
For stereo, I use comps too, usually.  Same result, even tho they don't do internal parellel compression.  Just judicious use again to preshape the dynamics a little and run the audio through some nice electronic components before digitalization (I don't use a mixer here at home).
 
@

The reason for recording post compression in times past was to keep the levels up.. away from the noise floor. Now however, given that the noise floor is significantly lower than in days past, that rationale is defunct so unless you have outstanding hardware gear that you want to use in your signal chain i don't see any reason to add a destructive element to the source material