Now that you don’t manually associate devices with drivers, we rely solely on the contents of the inf file to determine what devices the driver is for and for what OS. So long as the inf file is accurate and well written this is great. Meanwhile, in the real world we have situations whereby sysprep is given a Windows 7 driver to install during an XP installation, or a touchpad driver being installed on a desktop, or certain drivers not being copied at all. This is exactly what appears to be happening at one of our customers right now (all of the above).
Its worth noting that the problem doesn’t seem to be nearly as bad in Windows 7, seemingly because DISM will prevent incorrect drivers being enumerated. With XP though..... it's a different matter.
Has anyone else experienced this or had any thoughts on a viable workaround? For the XP machines we're currently looking at having to drop back to copydrivers.
LANDESK - Why have you took a process that worked well and managed to seemingly break it? On how many models were the 20,000 odd tests done on in provisioning with XP? It very much feels like the test was only done on Windows 7 or on XP with only XP drivers in the HII database on a single hardware model (probably even just virtual)....