PBPL 2.50b (possible) For/Next ?BUG? or just the way it is...

[skimask]

While working/testing/failing those souped up divide routines, I came across something.
Don't know if it's a bug, or just plain unavoidable...

mn = -2147483600 'almost max'd out at 31 bits
mx = 2147483600 'almost max'd out at 31 bits
st = 1234567 'step size can vary quite a bit
x var long

For x = mn to mx step st
     lcdout $fe , 1 , HEX4 x.word1 , ":" , HEX4 x.word0
Next x

The loop never stops! Ever!
It seems that when the step size is larger than the difference between 2^31 -1 and mx, x will roll over from a positive number to a negative number (high bit set) and the loop will keep running. Conversely, if using a negative step size, x will roll over from a negative number to a positive number and the loop will keep running as before.
Looks to me like it's an unavoidable situation...
Any thoughts?

[mackrackit]

If the "step" is to big no matter what size of variable, (byte,word) wouldn't the same thing happen, a roll over?

[skimask]

If the "step" is to big no matter what size of variable, (byte,word) wouldn't the same thing happen, a roll over?

I would think so. But it seems that with PBPL that internally everything is dealt with as though it were a LONG (case in point, those division routines I've been monkeying with). I'm fairly sure the rest of the math is dealt with in the same fashion. Therefore, a byte or word, might not act the same way. Not sure. It's just something I noticed yesterday.

[mackrackit]

It will be the weekend before I get back to the bench. I am going to play with it.

I was playing with a taylor series awhile back and gave up, thought it was me and my bad coding... probably was but maybe it is something with PBPL?

[skimask]

I was playing with a taylor series awhile back and gave up, thought it was me and my bad coding... probably was but maybe it is something with PBPL?

Not a Darrell Taylor series I hope!!!

[mackrackit]

Not a Darrell Taylor series I hope!!!

Nope. But might be some relation.

This Taylor
http://en.wikipedia.org/wiki/Brook_Taylor
any resemblance?