I don't want to piss anyone off. Please take no offence as nothing I say below is directed at anyone personally.
I think aftermarket chips definitely changes the data map so that some values are different to factory settings. As to how much improvement this change offers, I think it's best to test it with things like the Apexi RSM or an accelerometer like the Tesla. I used to do back and back testing using those when I was tuning my MX6 turbo. It shows the difference before and after intake and exhaust mods very convincingly. The Boxster, being NA, would show less measurable benefits but should still be demonstratable nonetheless if the chip does what it claims. An Apexi RSM can be had for around $200AUD or a Tesla device for around $90AUD on Ebay. In many ways, these offer more real world measurements (0-100kph, 400m) than dynos.
I own both. And intend to do some testing myself when I start to play with the snorkel removal and header installation later on.
On the issue of chips, here are two articles on an interview with the director of Powerchip in Australia you may find an interesting read.
http://autospeed.drive.com.au/A_0402/page1.html
http://autospeed.drive.com.au/cms/A_0408/article.html