USB "Loop" tester?

other things contains a 4 port USB 2.0 hub. What I'd like to do is have a PC USB host generate a series of test patterns through the hub, read them back from the downstream side of the hub into a second port on the PC and do a binary comparison, logging any errors. I could write some code to do this myself, but I was hoping to save time and just buy or rent something to do this. Anyone have any ideas? I sort of assumed that this would be a common type of testing for USB cable manufacturers and such, but haven't been able to find anything.

Read and write files to a memory stick?

You won't have much luck with a pure software solution, because the USB cables are "directional". The PC is always the master and others are slaves. With a hub in between it look to the PC as a slave and for the connected devices as the PC. Therefore the PC cannot read the data from the other side of the cable. You would need something that has two upstream ports and plays the role of two slaves reading one side and perhaps echoing this data on the other port.

g other things contains a 4 port USB 2.0 hub. What I'd like to do is have a= PC USB host generate a series of test patterns through the hub, read them = back from the downstream side of the hub into a second port on the PC and d= o a binary comparison, logging any errors. I could write some code to do th= is myself, but I was hoping to save time and just buy or rent something to = do this. Anyone have any ideas? I sort of assumed that this would be a comm= on type of testing for USB cable manufacturers and such, but haven't been a= ble to find anything.

two usb-rs232 adapters and a nullmodem cable?

send and recieve file with terminal program, compare sent and recieved

-Lasse

ng other things contains a 4 port USB 2.0 hub. What I'd like to do is have = a PC USB host generate a series of test patterns through the hub, read them= back from the downstream side... logging any errors.=20

Good idea, but (1) the protocol for block-devices like a disk-emulating flash drive will hide errors (and retransmit),=20 (2) most memory sticks have slow write rates and you'd get less test per second than some other possibilities.

Maybe the USB bridge-two-computers devices (which I presume is dual-port RA= M) would be faster. I'm not sure how you'd log errors, but one could check da= ta transfer rate and see if the process gets slower, even if block parity/CRC DOES mak= e errors disappear.