Another LCD driver (HD44780)

MagIO2

You can search for "w500ns" in my driver and increase the value there, say to 200 or more. This should also fix the problem. Could you give it a try?

SSteve

Sure. I'll try that. I won't be home from work until fairly late tonight so I might not be able to do it until tomorrow night.

SSteve

I changed w500ns to 200 and now it works perfectly! Thanks for the help.

MagIO2

Thank you for your efford!

As the change fixes the problem I'll spend some more time to run all my test-code with this new setting to see if everything still works fine. Once I found out my password for the ObEx I also might check in a new version which gives the possibility to select FAST_CLK or SLOW_CLK - something like that.

So far tested:
all demos still work even if w500ns is set to 200

SSteve

Should the last line of start be repeat while long[lcdBufAdr]==10 instead of repeat while lcdBuf[0]==10 in case the user passes his own buffer?

MagIO2

Looks like you are the first one who reads the code ;o)

Short answer: yes, you're right

Long answer:
As far as I remember the philosophy changed during coding. First it was "wait until the driver is done" for example after init and after calling exec. Then I changed the philosophy to "see if the driver is busy before giving new commands" - well ... only for the exec as I learn now. BUT ... this bug does not harm. If someone uses the feature of giving the command-buffer (as my CogOS does) the init will not wait for the driver. But even if you call an exec immediately this exec will wait. So, my fix would be: remove the repeat in init.

Thank you again!