I have used Cubase for about 8 years. I like the interface, I think it's very intuitive, and I think that it's very easy to use.
I have used Pro Tools for the past 4 years, and I think it wins every single time over Cubase.
One of the greatest benefits of Pro Tools over others, is the way in which it deals with dropped samples. Cubase ignores them; you have NO idea whether or not every single sample has been properly processed, and quite often there are mistakes. Pro Tools gives you the option of turning off these error messages, but they are set to come on as default. This way, if there is a computational error, you know about it, instead of sweeping it under the rug. This was certainly the case the last time I used Cubase anyway.
Secondly, the failure rate with bounces when I used Cubase was phenomenal. I'd sometimes have to bounce the same project 4 or 5 times, until I got a complete bounce. In Pro Tools I think I have only ever had one failure. This is simply not acceptable in what is supposed to be a professional programme. There are also a huge number of bugs (which I have experienced.) and they seem to go unfixed - whereas Digi are pretty quick to give out updates if there's a serious bug.
TyphoidHippo:
I'd like to address a couple of points you made in your previous couple of posts. Firstly, Pro Tools HD does not allow you to record an unlimited number of tracks. There are still a maximum number of voices and tracks which you can have simultaneously. But then I would not want there to be an unlimited number. The more tracks you have running, the more it loads the system, and the more likely it is for you to have errors. The limit is there so that you don't try to exceed your system's capability. The "unlimited track count" of other programmes is a myth; there's still a limit, and the closer you get to it, the more crap your system will run. Unlimited track count is also a marketing tool, make no mistake about that. There's also the issue that for a home recording, if you can't make 48 tracks work for you, you're doing something seriously wrong.
Secondly, your argument about sample rates is so unbelievably flawed, and definitely shows a lack of understanding of the sampling theory. Sure, Nyquist says that you need a sampling rate of at least 40 KHz to sample a 20 KHz signal. True. However you cannot sample anything of greater frequency than 20 KHz with that sampling frequency. "Sure, well we can't hear above that so who cares?" Well... if we allow the input signal to exceed the Nyquist rate, then we'll get aliasing. There are plenty of supersonic signals floating about with just about any instrument you care to mention. So... it becomes necessary to design a low pass filter (anti aliasing filter in the A-D converter and reconstruction/anti imaging filter in the D-A) to remove these supersonic signals. Now you go and design me a filter which can have an infinite slope of cut off above 20 KHz. You can't. You can get fairly close, but you cannot do it, especially in analogue circuitry (which this filter needs to be, being in the analogue domain...). The next issue with these steep filters is that they cause phase-distortions across the entire audio band. So it would be nice if we could design a shallow filter, which would be cheaper, and easier, and leave less phase problems, wouldn't it? So there are two options - band limit the audio further down (say 12 KHz?), or increase the sample rate to leave more room above the audible range for this low pass filter. I know which one I would choose.
I'm not advocating the use of super high sample rates, I'm really not. However it really narks me when I see such a fundamental theory to our field so badly represented in text.
Anyone who's interested in more should read two books:
"The Art of Digital Audio" John Watkinson and "Digital Audio Principles" Ken Pohlmann.
- Google is pretty cool like that. 
. I hope that sums it up.
