On Mon, Apr 25, 2005 at 12:24:43PM +0200, Frederick Ros wrote: > Quoting "Farahmand, Farinam" <[EMAIL PROTECTED]>: > > > I think there is a speed limit with bulk mode ( not sure! ) I will try to > > find the answer. > > AFAIR the bulk mode can be referred as "best effort" : it tries to send URB as > fast as it can, but depending on the load of the hub the throughput can be > reduced.
Well, "best effort" term looks more appropriate for the isochronous mode, where data integrity is sacrificed for latency. Quoting from the USB spec v2.0 (chapter 5.6): > In the USB environment, requesting an isochronous transfer type > provides the requester with the following: > * Guaranteed access to USB bandwidth with bounded latency > * Guaranteed constant data rate through the pipe as long as data is > provided to the pipe > * In the case of a delivery failure due to error, no retrying of the > attempt to deliver the data Therefore isochronous transfers are typically used for audio and video. > For information, ISO allocates some of the bandwidth to be sure the desired > thoughput will be reached. OTOH bulk transfers can use up all the available bandwidth, so on a free USB bus they should be at least as good (purely from USB POV, the ADI device may treat them differently for whatever reason). However, according to Matthieu, when in bulk mode on a high speed DSL link, there are massive ATM cell losses, and many USB transfers complete with -EOVERFLOW, which corresponds to babble. Hence the question. Roman.
