I am translating the ehBasic program below into the assembler below that. The Basic program works and blinks an LED on the VIA's PB0 pin. So I typed the assembler into the SBC2.7 machine language monitor. The assembler program runs in about six second, but frustratingly the LED does not blink on the exact same machine!

This is embarrassing because I know assembler, and it's a nearly line by line translation. But I can't figure out what I am doing wrong with such a simple program. Can anyone help?

bne $101D instead of $101F on the second branch?

Thanks, but I don't think that's it. X should contain 0 because the code didn't take the BNE after the DEX. So loading it with zero again shouldn't be needed. I also tested the delay routine separately, and it does delay about a half second at 1 MHz.

At $1000, Y should be #$10 (16 in decimal).
The loop in line 50 FOR X = 0 TO 16, is a count of 17. In assembler listing, line 1016 should be BPL, not BNE.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

@IamRob, good catch.

Thanks Garth. I changed address 1005 to ORA #$01 and the assembly code worked! I knew I was missing something about VIA operation, so my primary goal is solved.

But this raises the uncomfortable question why my ehBasic program worked in the first place! I used it to figure out how to blink the LED, and was led astray by it. To debug I added line 85 PRINT A, reran, and as expected the output was zero, but the LED was blinking! Really weird and now obviously wrong

I then changed line 80 in the Basic program to the correct "A = A OR 1", the LED stopped blinking, and the output changed to 1.

Something mysterious is going on with ehBasic, or my version of it. I am not going to pursue it as I have other things to work on, but it makes me rethink using ehBasic for prototyping.