Intel has VTune, AMD has CodeAnalyst. VTune is pretty pricey, while CodeAnalyst is free. If your goal is optimizing for AMD, CodeAnalyst can be quite useful.

While VTune isn't a great "cracking" tool, it's VERY useful in reversing. Lets say your target is a bloated pig-ware program with nearly 100 dlls (not named DLL, but instead goofy extensions that probably had some humorous relevance to the original authors). Now lets say you need to figure out its API so you can perform some cool new functionality. The problem with using Olly on Pig-Ware is that their is just so much code that you would need to be looking at.

In comes VTune.

Launch VTune (with no app, just monitor everything), run target and do action your interested in, and voila, you have the pieces of code to concentrate on. Now you load up Olly and IDA on just that sub-set.

It is worth noting that most protection routines, especially dongle routines, are convoluted and very inefficient. They are written to not be easily understood, and this usually involves lots of busy loops, jumps to pointers, etc., that all appear as hot spots in VTune.

The downside is that VTune dis's from disk image, so if the module in question is encrypted, you'll get pointers to what appears to be garbage.

But for reversing bloated Pig-Ware, it's a god send.