aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorKimplul <kimi.h.kuparinen@gmail.com>2025-11-07 17:44:03 +0200
committerKimplul <kimi.h.kuparinen@gmail.com>2025-11-07 17:44:03 +0200
commit2a7b3568792d50e94479298b5d0e5602d4e230f8 (patch)
tree112505488a290b369b5b5fbf726770c19004e471
parent5095d47320c36e2346d41eb7e35dcbd215886557 (diff)
downloadhid-tmff2-2a7b3568792d50e94479298b5d0e5602d4e230f8.tar.gz
hid-tmff2-2a7b3568792d50e94479298b5d0e5602d4e230f8.zip
highlight ramp <-> sawtooth relationship
-rw-r--r--docs/FFBEFFECTS.md12
1 files changed, 11 insertions, 1 deletions
diff --git a/docs/FFBEFFECTS.md b/docs/FFBEFFECTS.md
index a9a313a..87f5514 100644
--- a/docs/FFBEFFECTS.md
+++ b/docs/FFBEFFECTS.md
@@ -289,6 +289,16 @@ magnitude + envelope + duration + offset:
### Notes
+Ramp is not really an effect supported by the hardware in the wheel itself,
+rather it gets emulated by using a single period of a long sawtooth periodic
+wave. It's probably still worth documenting here due to it having to be handled
+separately from a 'generic' periodic effect.
+
+In particular, note how the duration of the effect is specified twice in the
+packet. Interpreted as a packet for a sawtooth wave, the first duration is the
+period of the wave, while the second is the 'actual' duration of the effect.
+This encodes that only a single 'ramp' is played.
+
#### Invert
There is a bug in the Windows driver during the Ramp effect update.
@@ -340,7 +350,7 @@ Valid:
f6 7f - fade_level***
04 - invert (going "down", 05, going "up", 04)
4f
- f7 17 - time again?
+ f7 17 - time again
00 00
00 00 - offset
00 ff ff - end of init