So Windows can now run on Mactels? I wonder what the speed is like compared to VPC on a G5.

Do some research on OpenOSX before you buy. Several people who shelled for the 2.0.0 release were pretty upset - claims of 15-20x improvement over the iMac G5 seemed to be dubious at best. Users were reporting installation times for WinXP of 36 hours in some cases on the new Core Duo iMac. Jeshua Lacock, the owner of OpenOSX, seems to have a questionable interpretation of what "native" or "near-native speed" means. This quote is by Lacock on the MacNN forums:

"And as long as the binaries are native (whatever the program - however slow) it would be running native, wouldn't it?"

OpenOSX has a long history of running afoul with the open-source community, as they (at least, according to accusation) seem to repackage open-source software with the occassional failure to give credit where credit may be due.

If you want to try it out, I'd recommend actually downloading Bochs (the emulator OpenOSX uses) from SourceForge and putting it through it's paces before buying into this product.

Anyways, buyer beware! I'll be waiting for a more reputable company to come out with an emulator before I start spending.

One user on versiontracker reports buying the new version and having it take 17 hours to install Win XP on an Intel iMac. 5 minutes to boot. Now I'm no fan of M$ but Windows isn't that bad. Claims by OpenOSX of near native speed are clearly false if this person is to be believed.

Since OpenOSX is nothing more than a Cocoa frontend for Bochs, perhaps someone with the technical know how would care to get Bochs up and running on an Intel iMac and report on the actual performance. It would cost nothing and would either prove or (more than likely) disprove this organisations claims. They have a bad rep over a long time so I'm extremely doubtful that they are telling the truth.

i bought a copy of their grass packages a few years ago.. kind of a dissapointment and then i bought a copy of mactel last year. it never worked.

Really your time is much valuable than spending hours to even try the product. Bochs is slow to try, I am not talking about an actual usability.

Wait a little longer and we will see something that would be working fast on an Intel Mac like on any other PC.

It's worth noting that none of those reviews are of 2.0.1, the version that they're claiming it's better than.

I don't have a nice shiny new iMac/tel to test it on, though, so I can't comment on its actual performance.

Note that this isn't 'native' in the sense that windows code is running natively. Its just that they've recompiled (at least parts) to be universal binaries. h***, for all I know, they recompiled Bochs itself to be a native app. Regardless of that, though, Bochs converts intel code to the underlying instruction set, so even if it doesn't 'change', its still running the conversion. (Although I doubt bochs was recompiled as a universal binary, it sounds more like they 'optimized' it somehow, but its still running in Rosetta).

Also, they claim that its 2-4 times faster that 2.0. That may be the case. Who knows how slow 2.0 was?

And install times is not an indication of actual usage ability. If it takes 30 minutes to launch Word, that's one thing. Taking thirty hours for a one-time-only install, who cares.

"And as long as the binaries are native (whatever the program - however slow) it would be running native, wouldn't it?"

The product doesn't just advertise native code, it advertises native performance. This product clearly does not perform as native, even if it's code is "native" by Mr. Lacock's definition. This is a clear-cut case of false advertising.