On Tue, 5 Dec 2006 10:46:49 -0500 (EST) Alan Stern wrote: > On Tue, 5 Dec 2006, Oliver Neukum wrote: > > > > Stated differently, USB 2.0 has a theoretical max of 480Mbps, > > > which is 60 MB/sec, but in actual practice only about 50% of that > > > is attainable. > > > > That's not really fair. The protocol itself makes reservations of > > bandwidth for control transfers and real time applications. > > And the protocol itself has bandwidth overhead. The maximum possible > number of 512-byte bulk transfers per microframe is 13, even without > isochronous or control transfers. That amounts to 512*13*8 bytes/ms > or 51 MB/s. usb-storage doesn't make full use of the available > bandwidth, but the largest bottleneck must be the devices themselves > (or their USB interfaces). > > Alan Stern
Since all I had previously seen was the 480Mbps figure, it's been very informative to learn more of the various factors that result in _real_ throughput. Thank you. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
