UberNewbie can't set his clock...

[Byte_Butcher]

Thanks to everyone that responded!

Melanie, I still couldn't that snippet of code you supplied to work right. Using Dig 0 and Dig 1 to pick the 1's and 10's digits out of the BCD didn't give the desired results...

What finally DID work for me was to split the minutes data apart using bitwise AND (&) like so:

   BCDByteH = rtcmin & %11110000 'get the high nibble
   BCDByteL = rtcmin & %00001111 'get the low nibble

That got my upper and lower 4 bits separated out OK so I can increment/decrement them with the pushbuttons.
When I'm done I put them back together into rtcmin with bitwise OR (|)

  rtcmin = BCDByteH | BCDByteL 
      
    gosub settime

That seems to be working so far. More testing still to make sure...

Anyway, thanks to all!

(Oh, and Joe S., you MUST be a married man. You seem to know the code for it pretty well.

Steve

[Archangel]

(Oh, and Joe S., you MUST be a married man. You seem to know the code for it pretty well.

Steve

HE HE HE, YOU GOT IT ! I got a list of honey do's longer than a road across Texas, AND I got another list of Sonny Do's from my mother. Keeps me outta trouble.

[Melanie]

Just for the record, the examples I posted are valid, tried and tested...

Not knowing how you are manipulating your code and Buttons behind the scenes doesn't help... but, if your Seconds Button for example incremented the Seconds every time you pressed or held it...

	While SecondsButton=0
		Seconds=Seconds+1
		If Seconds>59 then Seconds=0
		Pause 250
		Wend

That will give you a variable 'Seconds' which will contain a (Decimal) value in the range 0-59.

Running it through either if the routines I posted previously you get...

DecimalByte Decimal (HEX) = BCDByte Decimal (HEX)

0-9 ($00-$09) = 0-9 ($00-$09)
10-19 ($0A-$13) = 16-25 ($10-$19)
20-29 ($14-$1D) = 32-41 ($20-$29)
30-39 ($1E-$27) = 48-57 ($30-$39)
40-49 ($28-$31) = 64-73 ($40-$49)
50-59 ($32-$3B) = 80-89 ($50-$59)

...which then can be used for the Seconds Register for the timer chip.

Now those routines are only valid for the range 0-99 giving a BCD of $00-$99, which is fine if you are setting clocks as you're only interested in 0-59 worst-case anyway.

The way you've implimented things seems kinda hard work, if individually you are incrementing units seperately from the tens.

Melanie, I still couldn't that snippet of code you supplied to work right. Using Dig 0 and Dig 1 to pick the 1's and 10's digits out of the BCD didn't give the desired results...

Of course it didn't... that routine pulls the units and tens digits out a DECIMAL number to construct (encode) the BCD - not to decode the BCD... you weren't asking for that in your original post! To construct DECIMAL from BCD, you can do something simple like...

	DecimalByte=(BCDByte>>4)*10+(BCDByte & $0F)

[sinoteq]

Melanie and her "one line wonders" of code made me curious. In what order does PBP actually calculate and do things?
Let's use her latest code as an example:

DecimalByte=(BCDByte>>4)*10+(BCDByte & $0F)

Obviously the BCDByte is shifted right by 4 and the top 4 bits are made to 0. After the >>4 the data in BCDByte is "broken" and only the 4 low bits are valid.

Will

DecimalByte=(BCDByte & $0F)+((BCDByte>>4)*10)

give the same results?

When does the shift command actually change the data in BCDByte?

[mackrackit]

When does the shift command actually change the data in BCDByte?

It is my understanding that the shift command in a case like this will not change the data.
It only tells what part of the value to use in the calculation.

[Melanie]

OK, lets break this down...

DecimalByte=(BCDByte>>4)*10+(BCDByte & $0F)

The calculations in the brackets will be perfomed first, and the results stored in a PICBasic variable behind the scenes. BCDByte is unchanged - just as well because we refer to it twice, once to access the top four bits, and once to access the bottom four bits.

Thereafter, PBP follows mathematical rules of priority... ie Multiplication and Division are performed before addition ans subtraction. Because items in brackets are always calculated from the centre out, and each set of brackets is it's own self-contained world, the result of the formula below, will be the same as that above.

DecimalByte=(BCDByte & $0F)+((BCDByte>>4)*10)

However, you do have an extra set of brackets there... but PBP actually resolved the formula without any additional penalty. Might not be so lucky on another occasion.

[Byte_Butcher]

Thanks again Melanie,
Hmmm, I think I almost understand what's going on with that code. I must actually be learning something.

I ended up with this, from the code you supplied and it works excellent.

increminute: 

        rtcmathtemp =(rtcmin>>4)*10+(rtcmin & $0F)  'convert BCD minutes into Decimal minutes
        
        rtcmathtemp=rtcmathtemp+1   'increment by 1
        If rtcmathtemp>59 then      'If minutes exceeds 59
        rtcmathtemp=0               'reset to 0
        endif
  
        rtcmin=((rtcmathtemp DIG 1)<<4)+rtcmathtemp DIG 0 'convert Decimal back to BCD

        gosub settime
       
        return

[Byte_Butcher]

Sigh, well I seem to have a new problem. Although I'm obviously able to read/write to the DS1302 in "burst" mode, I seem to be having a problem addressing individual registers.

For instance, I need to turn the trickle charger on...
I *thought* it was as simple this:

        rst = 1

        'Set Battery charger on and set to "1 diode, 2Kohm"
        Shiftout IO, SCLK, LSBFIRST, [$90, $A5]  ' Set charger   [$90, $A5] 
 
        RST = 0

But that doesn't seem to do it for me.

Anyone got the patience to help me along a little further....

p-p-p-please?

Thanks!

Steve