The few boards that have come out with usb 2.0 have also had 1.1 ports. Why? Its my understanding that 1.1 devices will work in a 2.0 port it just steps-down to the 1.1 spec. Is that not true?
its because usb 1.1 is built into the chipset/motherboard
the boards with 2.0 have had a second controller added on (basically its just like they added a usb 2.0 pci card, but just mounted it right on the board)
the next generation of chipsets will probably have usb 2.0 built right into the southbridge, and then you wont see 1.1 any longer
(i think the next via socketA chipset is supposed to have usb 2.0 support)
Thanks for the info that makes some sense. But why not make all of them 2.0 and just drop the 1.1 altogether. Would it be to much overhead for the built-in 2.0 chipset and they figure some devices like keyboards and mice don't need 2.0?
It's because every kind of peripheral technology is allowed to mature a bit as a separate PCI device before being pulled into the chipset.
Just look at the history of integrated components - IDE, USB 1.x, sound, LAN, softmodem, FireWire ... they all have been existing as separate entities for a while before they got assimilated into a chipset south bridge.
Another reason is that I have read that if you run any USB 1.1 devices on a USB 2.0 channel, then any subsequent USB 2.0 devices you add on to that channel will have to run at USB 1.1 speeds. (most USB 2.0 controllers have 4 ports running on 2 channels: 2 ports per channel).
Another thing that I'm not sure if is that I heard when you have multiple devices on the same USB 1.1 channel, it splits the bandwidth evenly between them (ie a mouse and a printer would each get 600Kbps max), and USB 2.0 might act this way also (meaning a mouse and printer would each get 240Mbps). As I said, this may not be true, so take my words with a grain of salt.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.