I'm going against the grain here.
First off, it completely depends on how much awareness of the launch you're really going to generate with press releases, blog postings, etc.
When I get back from vacation, I simply delete most of what's in my RSS reader -- a huge pile of news builds up, and I just don't have the time to read it all, so why try. I think many other people do the same. If you generate a lot of press in the first weeks of August, you're wasting it. Wait until at least the second half of August, possibly the last week of August.
So I'd propose something like this: Make a critical estimation on how much high-quality coverage (influential bloggers, mainstream media) you're really going to get. If you think you're going to get significant attention (many startups don't), then postpone the official launch until late August, but start a "private" beta now. Your "private" beta should be private by name, but really you should try to get as many people in as you can. And then "launch" in late August / early September with as much publicity as possible.
If you don't have something unique to interest the media with (rockstar team, seriously cool and innovative app), then you're unlikely to generate substantial hype. In that case, just launch now, and build awareness bit by bit.