On Thu, 2003-02-13 at 19:59, David Brownell wrote:
> Dan Parks wrote:
> >
> > We only had 2 URB's, and that was the problem. The reason why we only
> > had 2, was because the less URB's you have, the lower delay you have. I
> > guess we're going to go to 3 URBs. Thanks a lot for your help.
>
> What delay that would increase with the number of urbs queued? I'd
> think tolerance for irq delays would increase ... you always need to
> keep enough queued to handle your maximum IRQ latency.
Please correct me if I am wrong, but from my reading of how the URB's
work, this situation is accurate:
If you have 4 URB's with an isochronous USB connection. As soon as you
open the connection to the device, there is going to be a set of empty
packets the number of which is equal to the number of URB's. So if you
have 4 URB's, the first four packets sent will be blank. If you're
controlling something....say a robot....where it is important to
minimize the delay of when you write a packet and it is received, you
would want to have less URB's.
number of URB's: 4
delay: 4
e = empty packet
time 1 2 3 4 5 6 7 8 9
_ _ _ _ _ _ _ _ _
//////////////////
dev e e e e 1 2 3 4 5
|
-------- delay of four
\/
our 1 2 3 4 5 6 7 8 9
code
If this picture is wrong please correct me.
Dan
-------------------------------------------------------
This SF.NET email is sponsored by: FREE SSL Guide from Thawte
are you planning your Web Server Security? Click here to get a FREE
Thawte SSL guide and find the answers to all your SSL security issues.
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
