Andrew, It does basically work, but there are some issues with the lcdproc software which it uses. What is checked in works with the curses driver. I have just recently worked out the bugs with the Matrix-Orbital driver, but it requires patching the MO driver in the lcdproc software.
The main problem is with seeing the key up events, which the driver does not seem to support. Further, the ini file which provides the key definitions does not appear to be case sensitive. The Matrix-Orbital display uses upper case and lower case keys to indicate key press and key release (i.e. 'A' will be returned for key 1 pressed, and 'a' will be returned for key 1 released). Since the ini file is not case sensitive, 'A' and 'a' are treated exactly the same, thus key press is indistinguishable from key release. The key up events are used for functions like continuous mode jogging, where jog starts on key press, and stops on key release. The responsiveness is actually pretty good. I patched the MO driver so that it returns a different value for key press than key release, but did not look into a more generic solution across all drivers which would involve making the key definitions case sensitive (but I suspect that would have other side effects, and therefore the reason it was not done in the first place). I have not at this point contacted the lcdproc developers regarding this problem. If you intend to use something other than MO, then that driver will need some looking into, because it is likely to have the same problem with key up events. There is also a potential issue with other drivers as far as throttling the communications rate with lcdproc (communication occurs over a standard TCP/IP socket). With the curses driver, no throttling was required. But when connected to the MO I had to introduce small delays, particularly at initialization when all of the "screens" load, to eliminate overrun errors. This delay is currently hard coded, and I do not know if this will be sufficient when using different processor speeds or different drivers. At the least I need to make the delay a user definable parameter. The latest code has not been checked in simply because I have not been able to do a developer check-in. I thought this was the same computer I had used previously, but apparently that is not the case. I need to ask Chris to see if he would be kind enough to update cvs with this computers id_dsa.pub file. Getting it working should be fairly straight forward, except you will need to work from trunk, and the possibility of needing to patch the LCD display driver. No patch is required if you just want to use the curses driver. There is also an ini file which needs a small amount of editing. Again very little if using the curses driver, but you must at least add the key definitions when using any other driver. Another annoyance of the lcdproc implementation is that the curses driver will do a straight pass-through for any undefined keys (entered from the keyboard). The other drivers (as far as I know) only pass through the keys explicitly defined in the ini file. At any rate, download and install lcdproc from here: http://www.lcdproc.com/download.php3 There are several options on how to get it running. For development purposes, I start and stop it manually, but it can also be installed as a daemon starting on boot. Once the appropriate driver is configured and lcdproc is running, to get it running you need to add a single line to your hal file as follows: loadusr emclcd Regards, Eric Hello, I have been looking through the archives and have found a brief discussion on 'emclcd' - An informational display, on an external small lcd. Did anyone get this to work??? I would love to build a remote pendant (serial) with all the buttons and mpg's etc etc but have a LCD display with axis positions and any other info that I can dream up. Regards Andrew ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
