Objective Development Forums

The lenght of the report descriptor changed. Maybe forgot to change the RD lenght in the configuration descriptor ?

Normally that means that the chip has reached end-of-life and its production is about to be shut down. Seems that Atmel changed their mind tho, the last two entries from the ATTiny26 manuals revision history: Rev. 1477K-08/10 Added tape and reel part numbers in “Ordering Information” on page 171. Re...

The 100s/s is a limit imposed by USB low speed (min. 10ms poll rate for INT EPs.) You'd be better of looking at chips w/ a native USB interface that supports full speed like the ATMega32U4 (example dev board).

It's implemented in V-USBs 18MHz modul.

Well, thank you all very much for this fine example of your support for your products. Obviously you can stick your products up your collective arses. Welcome in the real world where you have to think for yourself and actually read documentation. This is a low-frequency forum for a free "produ...

When I said interface, I ment interface as described in USB 2.0 chapter 9.6.5. You need to add a 2nd interface descriptor to the configuration descriptor, followed by another HID descriptor and a endpoint descriptor for EP3. A 2nd report descriptor needs to be connected to that interface as well. I'...

Grendel, I believe the goal is to achieve two devices to show up. This is the acceptable work around others have done and in fact its what Ralh did on his page. Yes and no. The goal is to have two joysticks to show up. Using a single report descriptor w/ two reports seems to work w/ Windows (and it...

Instead of having two reports in one report descriptor you could set up a 2nd interface using endpoint 3 (optional EP in V-USB). Have one stick on EP 1, the other on on 3. This is from the top of my head (getting a bit rusty here ), you'll have to research the specifics.

Changelog.txt wrote: - Set USB version number to 1.10 instead of 1.01.

* Release 2007-06-25

1.01 used to be a bug.

I would try it without diodes 1st. Supposedly 80% of USB hosts will tolerate it and the ones that don't will just refuse service to the device, not burn up

Remove the F_CPU definition from main.c, the definition needs to be global. One way of adding it: in AVRS, load your project then open the menu "Project", select "Configuration Options". In that dialog under "General" fill out the "Frequency:" field w/ 1600000...

RazorConcepts wrote:Additionally, has anyone gotten V-USB software to work with AVR Studio 4? It is very picky, and therefore not able to use the included makefile, which may have something to do with my problems

Used AVRS4 many times w/ V-USB. Just set up the project from scratch and ignore the makefile.

V-USB is all about programming. Even if you use a more hardware oriented chip (like the ATmega32U4 <-- click this ) it's still mostly programming. It's the nature of USB.

Should be possible. You'll have to add some CDC/PSTN subclass specific control transfers, namely SetControlLineState (see PSTN120.pdf, 6.3.12) and GetSerialState (see PSTN120.pdf, 6.5.4). The document refered to is located at usb.org, in this archive.

Check out USB Device Class Definition for Human Interface Devices (HID), pg. 73 "Appendix F: Legacy Keyboard Implementation"