EPROM and EEPROM Programmers

On Apr 30, 2012, at 4:06 PM, Richard wrote: Sure. To be clear, what I meant initially by "one way" is that USB transfers occur when the host requests them (i.e. a device cannot say "Here, host, have some data!", but rather the host has to poll the particular device looking for data). That's what I meant by "asymmetric" (it's not a bus where all devices have equal standing as targets or initiators, like SCSI or PCI or Ethernet or 1394), and what someone else (Cameron?) meant by its "polled nature". Half-duplex is generally true of USB, but it's not what I was talking about. There's only one (differential) data line in the USB bus, so communication is necessarily half-duplex, but that's not really material to the problem I was getting at. USB 3.0 may have extra lines in its socket (I haven't really looked) to enable full-duplex communication, but I don't know what that entails since I've never looked into USB 3.0. There are "host to host" USB cables available (probably the same devices that Glen Slick was talking about), but they require an active device in the middle to store and forward the data from both hosts as if they were devices on the other end. They were used for file transfer a long time ago when the serial port was going away and people were bemoaning the loss of the null-modem cable transfer; I recall someone made a homebrew PlayStation2 app which could use that to load homebrew code over USB instead of burning lots of discs (this was before the Ethernet attachment for the PS2 was popular; the "slim" ones have it built in, so now Ethernet is the medium of choice). - Dave