I read in this forum that the limit for reading samples from the JoyWarrior24F14 is 125Hz, but is this true for reading also from HID interface? I used my code posted here: viewtopic.php?f=3&t=1693
and it reads the HID at about 339 Hz with no repeat values. can anyone confirm this?
thanks,
Ran
HID reading freq
Moderator: Guido Körber
HID reading freq
aut disce aut discede
http://www.bgu.ac.il/~novitsky
http://www.bgu.ac.il/~novitsky
-
- Site Admin
- Posts: 2857
- Joined: Tue Nov 25, 2003 10:25 pm
- Location: Germany/Berlin
- Contact:
Re: HID reading freq
Interesting. This might be due to implementation details of the USB host controller driver used in your configuration.
The USB spec limits low speed devices like the JW24F14 to specify a maximum polling time of 10 ms as the fastest rate. It is up to the host to make sure that this speed is met as a minimum. Basically all implementations do scan the devices every 8 ms if they specify 10 ms. But especially on High Speed hosts (USB 2.0/3.0) the speed can be higher due to the way transactions are handeled.
The USB spec limits low speed devices like the JW24F14 to specify a maximum polling time of 10 ms as the fastest rate. It is up to the host to make sure that this speed is met as a minimum. Basically all implementations do scan the devices every 8 ms if they specify 10 ms. But especially on High Speed hosts (USB 2.0/3.0) the speed can be higher due to the way transactions are handeled.
Re: HID reading freq
I use fedora 17 and libusb for reading the HID interface on a usb2.0 host. I had a problem at first and the code would halt, as I now understand from your reply, due to the speed limitation. The solution was to clear the halt (libusb_clear_halt command) and read the new report. I get good readings with about 3 ms delay (that might be faster with a clearer code).
aut disce aut discede
http://www.bgu.ac.il/~novitsky
http://www.bgu.ac.il/~novitsky