On Jun 11, 2009, at 4:50 PM, Thomas Charron wrote: > On Thu, Jun 11, 2009 at 3:25 PM, Lloyd Kvam<pyt...@venix.com> wrote: >> On Thu, 2009-06-11 at 13:41 -0400, bruce.lab...@autoliv.com wrote: >>> I bet it won't do 480Mbits sustained. >>> (For a 10 GB file write) However, I just might try it for the heck >>> of >>> it! >> I use external USB and firewire drives fairly regularly. USB does >> not >> come close to achieving 48 MB / sec. Firewire (400 mb flavor) does >> noticeably better than USB. > > There are SO many variables in the case of USB, you can't just > blankey statement that USB doesn't come close to achieving it.
I'm fairly confident that you can too make such a blanket (and/or blankey ;) statement. Its not so much variables as it is that USB's design simply isn't at all efficient for high throughput I/O. Its a master-slave mode, dependent on host resources to get things done. Protocol overhead, latency dealing with having to use system resources, etc., effectively cap the actual data throughput potential well below 480Mbps. I'm reasonably sure I've never seen better than around 30MB/s with a USB drive and random google hits on the 'tubes seem to back that up. -- Jarod Wilson ja...@wilsonet.com _______________________________________________ gnhlug-discuss mailing list gnhlug-discuss@mail.gnhlug.org http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
