Closed meithecatte closed 4 years ago
Thank you for your contribution! The swap
optimisation makes a world of difference! 🔥
I'll have a quick look at the other changes as well soon :)
On Fri, May 15, 2020 at 08:19:09AM -0700, Tijn Kersjes wrote:
@tkers commented on this pull request.
@@ -413,7 +413,7 @@ instruction add, ~n ~A ~~> %11 %000 %110 op, n, 2 cycles :: ~(HL) ~A ~~> %10 %000 %110 op, 2 cycles :: ~ss ~HL ~~> %00 ss1 %001 op, 2 cycles ::
- ~e ~SP ~~> %11 %101 %000 op, e, 4 cycles ::
- ~e ~SP ~~> %11 %101 %000 op, e_, 4 cycles ::
I'm very curious how you managed to find this bug 😄 Hats off!
Thankfully, the routines got assembled at an address which made $0002
out of range of a jr
.
"10 print" now runs much faster :D