I'm wondering what the properties of thixomolding parts are like compared to casting compared to having the part machined out of a solid piece.
One thing I personally noticed with these types of "cast-looking" parts is that they tend to fail in a brittle way and there's really not much ductility. I've seen many notebooks with Mg alloy casings that shatter when dropped, and I've always preferred the Apple unibody machined aluminum for this reason - if it's bent or drops, it tends to retain its integrity much better. Granted, aluminum doesn't tend to have all that much ductility before it too, breaks, but I've found it more robust.
Mag itself is not very ductile, and more prone to fractures than most AL alloys, with the tradeoff being overall higher yield strength. Depending on the magnesium alloy used for thixo, this can be mitigated somewhat, but you are correct in that almost every "casting" process will give you a less ductile part than the same alloy machined from a solid billet. So much of a metal's material properties come from it's crystalline structure, and how it's melted, formed, and tempered. Thixo tries to get the best of both worlds by using a slurry that is both a solid (containing the proper microstructures for desired material properties) and liquid to ease the injection molding.
Machining from a billet of material guarantees consistent crystal structure, and is usually the go to when strength/appearance is a larger consideration than cost or weight. Billets can be formed very precisely and consistently, without consideration of the final product
One thing I personally noticed with these types of "cast-looking" parts is that they tend to fail in a brittle way and there's really not much ductility. I've seen many notebooks with Mg alloy casings that shatter when dropped, and I've always preferred the Apple unibody machined aluminum for this reason - if it's bent or drops, it tends to retain its integrity much better. Granted, aluminum doesn't tend to have all that much ductility before it too, breaks, but I've found it more robust.