I went the other way, started with Ubuntu and dual booted. I never wanted to go back to windows, but sometimes needed to. I installed VirtualBox and got it to boot the other partition. So I can now dual boot, or I can run windows in a VM from Ubuntu.
Not sure if thats the best idea for you, you need to access the underlying hardware I assume.
As CUDA drivers exist for {Linux, Windows}x{32-bit, 64-bit}, I don't have an extraordinary preference for which OS is run natively [other than running games which are, for good reasons, Windows only]. I really wanted Ubuntu available for its rich set of build tools [specifically, gcc 4.2].
That said, one requirement for maintaining GPU VSIPL is having all four combinations of instruction set architecture and OS kernel available and running natively to test a particular build. It would be handy to have a partition for each that may either be virtualized or booted. At the moment, I resort to multiple machines.
2 comments:
I went the other way, started with Ubuntu and dual booted. I never wanted to go back to windows, but sometimes needed to. I installed VirtualBox and got it to boot the other partition. So I can now dual boot, or I can run windows in a VM from Ubuntu.
Not sure if thats the best idea for you, you need to access the underlying hardware I assume.
As CUDA drivers exist for {Linux, Windows}x{32-bit, 64-bit}, I don't have an extraordinary preference for which OS is run natively [other than running games which are, for good reasons, Windows only]. I really wanted Ubuntu available for its rich set of build tools [specifically, gcc 4.2].
That said, one requirement for maintaining GPU VSIPL is having all four combinations of instruction set architecture and OS kernel available and running natively to test a particular build. It would be handy to have a partition for each that may either be virtualized or booted. At the moment, I resort to multiple machines.
Post a Comment