summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorAndre Brait <andrebrait@gmail.com>2022-12-04 07:56:04 +0100
committerGitHub <noreply@github.com>2022-12-03 22:56:04 -0800
commitf32070bfef50a51300d41bbd1f5732a64055fd5a (patch)
treeb1179520714b0a791cfe44cf6548554e2f997421
parent004ca5501167d00cb9a6f40fb7bfbc0ba901f4b1 (diff)
[Keymap] Change DEBOUNCE_TYPE to sym_eager_pk to reduce latency (#19153)
-rw-r--r--keyboards/gmmk/pro/rev1/ansi/keymaps/andrebrait/rules.mk19
1 files changed, 19 insertions, 0 deletions
diff --git a/keyboards/gmmk/pro/rev1/ansi/keymaps/andrebrait/rules.mk b/keyboards/gmmk/pro/rev1/ansi/keymaps/andrebrait/rules.mk
index 8983ba87c3..d5942a2e90 100644
--- a/keyboards/gmmk/pro/rev1/ansi/keymaps/andrebrait/rules.mk
+++ b/keyboards/gmmk/pro/rev1/ansi/keymaps/andrebrait/rules.mk
@@ -2,3 +2,22 @@
# Disabling MouseKey because it breaks my KVM switch
MOUSEKEY_ENABLE = no
+
+# Cherry MX-style switches and diodes are not susceptible to noise, no need for noise-resistant algorithms.
+# This significantly reduces latency.
+#
+# The matrix scan frequency seems to be around 1820 Hz, so even sym_defer_g would perform ok,
+# but the "defer" part would mean we would wait DEBOUNCE ms before sending any events.
+# Using "asym_eager_defer_pk" does not seem to benefit us in anything.
+# The GMMK Pro has more then enough system resources for a per-key algorithm.
+# Using an "eager" algorithm leads to extremely low latency while also reducing the chances of chattering
+# due to it's "post-event" debouncing (of sorts).
+#
+# I have observed zero chattering or double-keypress issues on my Gateron Yellow switches.
+# Most chattering issues on the GMMK Pro seem to be related to its proprietary hot-swap sockets anyway.
+DEBOUNCE_TYPE = sym_eager_pk
+
+# Useful for debugging
+# CONSOLE_ENABLE = yes
+# DEBUG_MATRIX_SCAN_RATE_ENABLE = yes
+# DEBUG_MATRIX_SCAN_RATE = yes