This is just another example of the "specs" game.
First off, forget the PSP - it's Sony, it's closed source and while it will be a solid 3D commercial game system, it won't have aCPU as fast as GP32 so even if emus could be developed somehow and the weird media circumvented, the GP32 is still better for emulation and homebrew.
As for Tapwave, the specs look good. But it's still vaporware. In fact, many systems with wonderful specs by new startups have come and gone over the years, this just being the latest. There is absolutely no guarantee the Tapwave Helix will even happen. Even if it does, many programmers dislike the Palm OS for development, weird memory structure etc. and it will be a while before it has a solid emulation scene etc. in place to compete with GP32. By the time Helix arrives, IF it ever does, the GP32 will have an amazing number of solid emulators and other stuff too.
The real point is that having a real system with real apps in your hand is worth more than a spec sheet. Right now the GP32 has NO competition for what it is - an open source power portable. And it will be at least 6 months or so before that even begins to change.
Let's jump the gun anyway, and say the Tapwave Helix comes out, and turns out as good as the paper specs claim. Let's say they get it right, good controls, screen and all those nifty features. And let's say all the hardcore portable gamers and emulation sceners buy one and development takes off in every direction - commercial games & apps, homebrew games & apps, and of course, emulators for all sorts of consoles and systems. Great. But even then, who's to say Game Park won't react soon after with a GP64? Imagine if a GP64 is backward compatible (it should be) and even faster than Tapwave. The point is any of these things can happen (GP64 is just as "real" as Tapwave Helix at the moment!), and until they do - stick with what you know is real right now. The good old GP32.