// --------------------------------------------------------------------------- // This file is part of reSID, a MOS6581 SID emulator engine. // Copyright (C) 2004 Dag Lem // // This program is free software; you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation; either version 2 of the License, or // (at your option) any later version. // // This program is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with this program; if not, write to the Free Software // Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA // --------------------------------------------------------------------------- #define __ENVELOPE_CC__ #include "envelope.h" RESID_NAMESPACE_START // ---------------------------------------------------------------------------- // Constructor. // ---------------------------------------------------------------------------- EnvelopeGenerator::EnvelopeGenerator() { reset(); } // ---------------------------------------------------------------------------- // SID reset. // ---------------------------------------------------------------------------- void EnvelopeGenerator::reset() { envelope_counter = 0; attack = 0; decay = 0; sustain = 0; release = 0; gate = 0; rate_counter = 0; exponential_counter = 0; exponential_counter_period = 1; state = RELEASE; rate_period = rate_counter_period[release]; hold_zero = true; } // Rate counter periods are calculated from the Envelope Rates table in // the Programmer's Reference Guide. The rate counter period is the number of // cycles between each increment of the envelope counter. // The rates have been verified by sampling ENV3. // // The rate counter is a 16 bit register which is incremented each cycle. // When the counter reaches a specific comparison value, the envelope counter // is incremented (attack) or decremented (decay/release) and the // counter is zeroed. // // NB! Sampling ENV3 shows that the calculated values are not exact. // It may seem like most calculated values have been rounded (.5 is rounded // down) and 1 has beed added to the result. A possible explanation for this // is that the SID designers have used the calculated values directly // as rate counter comparison values, not considering a one cycle delay to // zero the counter. This would yield an actual period of comparison value + 1. // // The time of the first envelope count can not be exactly controlled, except // possibly by resetting the chip. Because of this we cannot do cycle exact // sampling and must devise another method to calculate the rate counter // periods. // // The exact rate counter periods can be determined e.g. by counting the number // of cycles from envelope level 1 to envelope level 129, and dividing the // number of cycles by 128. CIA1 timer A and B in linked mode can perform // the cycle count. This is the method used to find the rates below. // // To avoid the ADSR delay bug, sampling of ENV3 should be done using // sustain = release = 0. This ensures that the attack state will not lower // the current rate counter period. // // The ENV3 sampling code below yields a maximum timing error of 14 cycles. // lda #$01 // l1: cmp $d41c // bne l1 // ... // lda #$ff // l2: cmp $d41c // bne l2 // // This yields a maximum error for the calculated rate period of 14/128 cycles. // The described method is thus sufficient for exact calculation of the rate // periods. // const reg16 EnvelopeGenerator::rate_counter_period[] = { 9, // 2ms*1.0MHz/256 = 7.81 32, // 8ms*1.0MHz/256 = 31.25 63, // 16ms*1.0MHz/256 = 62.50 95, // 24ms*1.0MHz/256 = 93.75 149, // 38ms*1.0MHz/256 = 148.44 220, // 56ms*1.0MHz/256 = 218.75 267, // 68ms*1.0MHz/256 = 265.63 313, // 80ms*1.0MHz/256 = 312.50 392, // 100ms*1.0MHz/256 = 390.63 977, // 250ms*1.0MHz/256 = 976.56 1954, // 500ms*1.0MHz/256 = 1953.13 3126, // 800ms*1.0MHz/256 = 3125.00 3907, // 1 s*1.0MHz/256 = 3906.25 11720, // 3 s*1.0MHz/256 = 11718.75 19532, // 5 s*1.0MHz/256 = 19531.25 31251 // 8 s*1.0MHz/256 = 31250.00 }; // For decay and release, the clock to the envelope counter is sequentially // divided by 1, 2, 4, 8, 16, 30, 1 to create a piece-wise linear approximation // of an exponential. The exponential counter period is loaded at the envelope // counter values 255, 93, 54, 26, 14, 6, 0. The period can be different for the // same envelope counter value, depending on whether the envelope has been // rising (attack -> release) or sinking (decay/release). // // Since it is not possible to reset the rate counter (the test bit has no // influence on the envelope generator whatsoever) a method must be devised to // do cycle exact sampling of ENV3 to do the investigation. This is possible // with knowledge of the rate period for A=0, found above. // // The CPU can be synchronized with ENV3 by first synchronizing with the rate // counter by setting A=0 and wait in a carefully timed loop for the envelope // counter _not_ to change for 9 cycles. We can then wait for a specific value // of ENV3 with another timed loop to fully synchronize with ENV3. // // At the first period when an exponential counter period larger than one // is used (decay or relase), one extra cycle is spent before the envelope is // decremented. The envelope output is then delayed one cycle until the state // is changed to attack. Now one cycle less will be spent before the envelope // is incremented, and the situation is normalized. // The delay is probably caused by the comparison with the exponential counter, // and does not seem to affect the rate counter. This has been verified by // timing 256 consecutive complete envelopes with A = D = R = 1, S = 0, using // CIA1 timer A and B in linked mode. If the rate counter is not affected the // period of each complete envelope is // (255 + 162*1 + 39*2 + 28*4 + 12*8 + 8*16 + 6*30)*32 = 756*32 = 32352 // which corresponds exactly to the timed value divided by the number of // complete envelopes. // NB! This one cycle delay is not modeled. // From the sustain levels it follows that both the low and high 4 bits of the // envelope counter are compared to the 4-bit sustain value. // This has been verified by sampling ENV3. // const reg8 EnvelopeGenerator::sustain_level[] = { 0x00, 0x11, 0x22, 0x33, 0x44, 0x55, 0x66, 0x77, 0x88, 0x99, 0xaa, 0xbb, 0xcc, 0xdd, 0xee, 0xff, }; // ---------------------------------------------------------------------------- // Register functions. // ---------------------------------------------------------------------------- void EnvelopeGenerator::writeCONTROL_REG(reg8 control) { reg8 gate_next = control & 0x01; // The rate counter is never reset, thus there will be a delay before the // envelope counter starts counting up (attack) or down (release). // Gate bit on: Start attack, decay, sustain. if (!gate && gate_next) { state = ATTACK; rate_period = rate_counter_period[attack]; // Switching to attack state unlocks the zero freeze. hold_zero = false; } // Gate bit off: Start release. else if (gate && !gate_next) { state = RELEASE; rate_period = rate_counter_period[release]; } gate = gate_next; } void EnvelopeGenerator::writeATTACK_DECAY(reg8 attack_decay) { attack = (attack_decay >> 4) & 0x0f; decay = attack_decay & 0x0f; if (state == ATTACK) { rate_period = rate_counter_period[attack]; } else if (state == DECAY_SUSTAIN) { rate_period = rate_counter_period[decay]; } } void EnvelopeGenerator::writeSUSTAIN_RELEASE(reg8 sustain_release) { sustain = (sustain_release >> 4) & 0x0f; release = sustain_release & 0x0f; if (state == RELEASE) { rate_period = rate_counter_period[release]; } } reg8 EnvelopeGenerator::readENV() { return output(); } RESID_NAMESPACE_STOP