From c7038b3977a3c588d648373e01f47d1c74782499 Mon Sep 17 00:00:00 2001 From: Lord-Nightmare Date: Tue, 7 Mar 2017 05:22:14 -0500 Subject: ...and again (nw) --- src/devices/sound/tms5110r.hxx | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/devices/sound/tms5110r.hxx b/src/devices/sound/tms5110r.hxx index 6aea0f8c7e4..7b185fa98e2 100644 --- a/src/devices/sound/tms5110r.hxx +++ b/src/devices/sound/tms5110r.hxx @@ -35,7 +35,15 @@ */ /* quick note on derivative analysis: - Judging by all the TI chips I (Lord Nightmare) have done this test on, the first derivative between successive values of the LPC tables should follow a roughly triangular or sine shaped curve, the second derivative should start at a value, increase slightly, then decrease smoothly and become negative right around where the LPC curve passes 0, finally increase slightly right near the end. If it doesn't do this, there is probably a wrong value in there somewhere. The pitch and energy tables follow similar patterns but aren't the same since they never cross 0. The chirp table doesn't follow this pattern at all. + Judging by all the TI chips I (Lord Nightmare) have done this test on, + the first derivative between successive values of the LPC tables should + follow a roughly triangular or sine shaped curve, the second derivative + should start at a value, increase slightly, then decrease smoothly and + become negative right around where the LPC curve passes 0, finally + increase slightly right near the end. If it doesn't do this, there is + probably a wrong value in there somewhere. The pitch and energy tables + follow similar patterns but aren't the same since they never cross 0. + The chirp table doesn't follow this pattern at all. */ /* Chip types based on die marks from decap: -- cgit v1.2.3