Posted by Ilya Zakharevich on 01/23/08 00:42

[A complimentary Cc of this posting was sent to
John Navas
<spamfilter1@navasgroup.com>], who wrote in article <9mpcp311bpldar5t729kobj8psho28mh23@4ax.com>:
> One more (last) time: USB is a simple cheap bus, very dumb

Sigh; all I asked for was a technical explanation, and all that the
posters provide is belles-lettres. [I explained what a *technical
explanation* of bus performance looks like in another message of this
thread.]

> without the kind of priority interrupts and DMA transfers you're
> assuming.

Sorry, I have no idea what you are taking about here. "Priority
interrupts" and "DMA transfer" are properties of the computer bus [one
"connected" to the motherboard], not of an "external" bus. So it
would be functions of *USB controller*, not of "USB bus", to have
these properties.

So, to enable high-speed transfers, a CONTROLLER should be able to
generate priority interrupts and DMA transfers. Do you say that there
are properties of USB *BUS* which do not allow controller to have
these interfaces ON THE MOTHERBOARD END?

> >DMA transfers still require cooperation. Somebody should set up the
> >buffers in time. (The difference, is, of course, in SIZE of the
> >buffer. So it is quantitative, not qualitative.)

> Not true.

Could you be more specific, please? (Likewise for your other "it aint
so" replies...]

Thanks,
Ilya

[Back to original message]