Definitely different. And I would hesitate to blame Ableton here. Very odd.
I don’t know Ableton very well but I would check to make sure there is no EQ or compressor on by default on the master track. That would be weird though.
Wait, how does that come in to the picture at all? I thought this was all through the monitors? Or are you going DI out from it to the DAI or something?
Confirmed.
Nothing on master track at all.
The track that I export out of Abelton matches the playback vs. the live monitoring from my experience. Which leads me to think it is how the Focusrite is doing something, but why it would do different things in record vs. playback is odd unless it has to do with their ASIO driver.
That definitely seems odd.
What is the sampling rate? 96KHz? Shouldn’t matter (48 vs 96 vs 192 should not have nearly this kind of effect) but if it is less than 48 that might be an issue.
44100, the default in the driver config.
There is another ‘switch’ in the Sample Rate area of the Focusrite driver config called "Default SR & Pitch Conversion with two settings, Normal and High Quality…High Quality is selected.
This is Abelton’s help box on this setting…
Shouldn’t be an issue.
Here it is again with the High Quality switch off…
You tell me what you think before I tell you what I think…
Sounds much closer. I wonder if the interpolation is adding some kind of compression effect.
Yes I agree it sounds much closer as well.
So it is either the driver or Abelton, depending on where that setting lives.
Try going to 96KHz or higher.
I don’t think any potential coloring from the Focusrite should necessarily be attributed to its price.
I have upgraded to this DAI a while ago:
and it has a JFET instrument input (for guitar, bass, etc) that is actually designed to add harmonically rich, tube-like color to the input signal from the instrument.
Of course, it also has a more standard (“clean”) line/mic input (with additional phantom power).
What I especially like about this interface is the monitor mix/blend dial (that @howard also referred to in one of his replies) to allow you to adjust what you hear while recording (and not just with an either-or switch).
I realize this doesn’t solve your current problem, and I also don’t want to give the impression that just buying new hardware makes all problems go away, but I did not regret upgrading to the Audient from my previous DAI (a Roland Rubix, which, I think, is comparable to the Scarlett in price and performance).
I am not as concerned about the unit adding color or not, this I can deal with.
My issue is that during monitoring of recording and then listening during playback, the sound is markedly different, and it should not be. If I am monitoring while recording and ‘dialing in’ my tone, then I should hear that same tone profile when i hit play for playback. The tone coloring of an input should show in both of the above.
Based on the soundcloud samples above, it does appear that the “high quality” setting does something on one vs. the other (monitor while recording / monitor playback) through both the headphone and direct monitor outs. When switched to “normal”, it gets a lot closer, so much so I would say ‘close enough’. I have not yet dug in to see if this setting is in the driver or Abelton. Either way, don’t like it so will try the next recording in normal mode and see how it goes.
I did some Googling and found a few other folks who rose a similar issue and were told it was perception. Based on the sound samples above this is clearly not the case. As an engineer who needs things to make sense before accepting them, I at least had to know what was the root cause. Now I know how to avoid it (hopefully).
Did you try increasing the sample rate?
I understand and I realize my reply wasn’t super helpful in that regard 
So, as you guys have been inching closer to, the problem may lie with whatever signal processing is going on in Ableton… one quick try could be to use either GarageBand or Reaper instead and see whether you experience similar issues.
If you want to dive deeper to satisfy your engineering curiosity, you might have to look more into “sampling” (including over- and undersampling), “Nyquist limit”, “anti-aliasing” and similar stuff that are somewhat out of my pay grade 
(It is usually here where I’d hope that @DaveT would chime in 
)
Not yet, was at my buddy’s house last night working on testbed bass and then stuck in construction traffic on way home, will try it though.
Good idea to try another DAW. I am starting to get a little pissed that I splurged for Abelton. I only pulled the trigger cause a guy that works for me uses it and figured he could be my tech support. There are several things I don’t like about it.
Was looking for an excuse to give Reaper a try…
This wouldn’t be caused by a 44.1 sampling rate. I was following initially and support all the great advice given.
There’s a low frequency EQ bump. It sounds like the kind of “enhanced bass” that PCs use to make crappy speakers or headphones sound better. I’d hunt to make sure no audio enhancement settings are on anywhere on the computer control panels.
To narrow things down, I support recording a track in Audacity. It’s simple and free. Then you know whether to hunt in Ableton or in the computer.
My first thought was Ableton had some EQ presets on the master, yeah. Probably a scoop. Sounds like he checked that though.
If this were the case wouldn’t this be present in both samples? Nothing PC wise changes between the two.
Maybe. Assuming the direct monitor from the DAI takes the same path through the OS that Ableton does. For stuff like this I tend to try any experiment that’s easy even if it I can’t explain it. Sometimes I learn an explanation I didn’t know about.