Posts: 349
Threads: 66
Joined: Apr 2007
Hi,
While debugging a program on the 34S side-by-side with a 15C, I noticed that in the 34S, currently CLx leaves stack lift enabled, which is not the way HP calcs work. In the HP machines, the purpose of disabling stack lift with CLx was so if an erroneous number was entered, CLx would allow entering a replacement value without affecting the stack. For consistency, they kept this behavior the same while running in a program as well.
Hopefully, it makes sense to change this so it works in the 34S like all other RPN machines.
Thanks,
Jake
Posts: 1,545
Threads: 168
Joined: Jul 2005
Agreed.
5 ENTER 6 CLX 8 + should return 13, not 8 as it does now.
Good catch, Jake!
Posts: 3,283
Threads: 104
Joined: Jul 2005
I think this can be done. Thanks for testing.
Posts: 3,229
Threads: 42
Joined: Jul 2006
Definitely a bug, I've got two fixes. One is much cleaner, a tiny bit larger and changers many opcodes. The other is ugly.
- Pauli
Posts: 3,229
Threads: 42
Joined: Jul 2006
And the ugly but smaller fix is applied.
- Pauli
Posts: 1,545
Threads: 168
Joined: Jul 2005
Wow. That was fast. So the CLX + bug is fixed as well as the stack lift issue with PI or a constant?
Very nice!
Posts: 3,229
Threads: 42
Joined: Jul 2006
Hopefully they are both fixed. So are complex constants and complex RCL which were also broken by recent changes.
At least, they will be when we do a new build.
- Pauli