Hi, Roger,
I found now the problem of the data missing happens BETWEEN the groups of
irps. In my case a group of irp are sent each 20ms. But the data missing
does not happen for each 20ms. It happens irregularly, sometimes at 620ms,
840ms, 1160ms, etc. It looks like that after a while, asynchsubmit cannot
keep up with the speed. It is my guess only.
I also found that if I set the length of the queue[] to 50 instead of 2, no
data missing is seen in my 2-second long recording. But it is certainly not
a good work around.
Any ideas?
Thanks
From: Roger Lindsjö <[EMAIL PROTECTED]>
To: Charles Chen <[EMAIL PROTECTED]>
CC: javax-usb-devel@lists.sourceforge.net
Subject: Re: [javax-usb-devel] speed of submitting UsbIrp to an iso pipe
Date: Thu, 14 Jul 2005 09:27:24 +0200
Charles Chen wrote:
Roger,
Follwoing your suggestions, I did more investigations on the jitters, and
I found in each 200 bytes that I received, there are 24 bytes zero. I used
my endpoint' wMaxPacketSize(200 bytes) to set irp, so it seems data are
lost between irps. Also I tried on windows and the recorded curves are
smooth.
Here are some outputs. Can you see what problem might be? I really
appreciate your help.
Can you send the data for 100 IRPS or so, on format:
20, 50 ,70, 100, -120, -90 ....
...irp 2...
...irp n...
I really don't see why there would be hitter between IRPs unless it is
between the group of IRPs. Each group of IRPs are handled by the kernel and
is not interferred with by the java code.
Some more thoughts:
Are you setting the length back to 0?
(could there be an issue with the length setting in javax.usb, try zero out
the array before resubmitting).
Should the stream just contain the audio, or are there packet headers etc?
(there is in my video stream).
Have you tried logging the data sent in Windows? I think there are free
loggers available that should be able to trace the data sent and received.
(look here http://www.linux-usb.org/tools.html ).
I'll not be available for a couple of weeks, Am getting married this
Saturday and then away on some holliday.
//Roger Lindsjö
-------------------------------------------------------
This SF.Net email is sponsored by the 'Do More With Dual!' webinar
happening
July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
core and dual graphics technology at this free one hour event hosted by HP,
AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar
_______________________________________________
javax-usb-devel mailing list
javax-usb-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/javax-usb-devel
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
javax-usb-devel mailing list
javax-usb-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/javax-usb-devel
