Claude Jones wrote: > On Sunday August 20 2006 8:33 am, Tom Horsley wrote: >> The motherboard BIOS never seems to have a problem seeing the >> keyboard via the hub so something about the linux usb >> bootstrapping must have some kind of timing problem. >> Grrr.... > > My day job is to manage about 50 Windows machines in a software company. Don't > think that it's a Linux-only problem. I've wasted a lot of time on this > particular problem. There's not much rhyme or reason to it - sometimes USB > works, and sometimes it doesn't, in many different scenarios, not just the > one you're describing. I've gone through many different theories, but I don't > have a definitive answer - the protocol is imperfect, the manufacturers have > imperfectly implemented it - those are the two I tend to suspect now. If the > powered hub didn't fix it, I'd say you're going to have to live with it, or > go with a PS2 solution. I keep several PS2 keyboards around our facility for > this very reason. Things are improving, and USB has gotten a lot better, but > I still encounter the occasional problem on new equipment. > > That's been my experience - I hope someone comes up with a definitive solution > and proves me wrong. To make things more interesting, sometimes the problem is a USB 1.1 device on a USB 2.0 controller. I can't remember the devices off hand, but I have run into a couple of USB devices that said they would work at 2.0 speeds when they really couldn't. They would work fine when hooked to a USB 1.1 controller or hub, but not when hooked to a 2.0 controller or hub. There are also some devices that will not work correctly with some hubs. But using a different hub solves the problem. It all depends on how well the specs are follower, I guess. At times, I think it is just how the different manufacturers shortcuts work together. Mikkel -- Do not meddle in the affairs of dragons, for thou art crunchy and taste good with Ketchup!
