/* process the latest information from ALE */
void SpachaseSettings::step(const System& system) {
// update the reward
// update the reward
int score = getDecimalScore(0x9C, 0x9B,0x9A, &system);
// reward cannot get negative in this game. When it does, it means that the score has looped
// (overflow)
m_reward = score - m_score;
m_score = score;
m_lives = readRam(&system,0x9E);
// update terminal status
// If bit 0x80 is on, then game is over
m_terminal = (m_lives <= 0);
}
/* process the latest information from ALE */
void PooyanSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0x8A, 0x89, 0x88, &system);
int reward = score - m_score;
m_reward = reward;
m_score = score;
// update terminal status
int lives_byte = readRam(&system, 0x96);
int some_byte = readRam(&system, 0x98);
m_terminal = (lives_byte == 0x0 && some_byte == 0x05);
m_lives = (lives_byte & 0x7) + 1;
}
/* process the latest information from ALE */
void TutankhamSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0x9C, 0x9A, &system);
int reward = score - m_score;
m_reward = reward;
m_score = score;
// update terminal status
int lives = readRam(&system, 0x9E);
// byte 0x81 is set to 0x84 when the game is loaded, but not reset
int some_byte = readRam(&system, 0x81);
m_terminal = lives == 0 && some_byte != 0x84;
}
/* process the latest information from ALE */
void AsterixSettings::step(const System& system) {
// update the reward
reward_t score = getDecimalScore(0xE0, 0xDF, 0xDE, &system);
m_reward = score - m_score;
m_score = score;
// update terminal status
int lives = readRam(&system, 0xD3) & 0xF;
int death_counter = readRam(&system, 0xC7);
// we cannot wait for lives to be set to 0, because the agent has the
// option of the restarting the game on the very last frame (when lives==1
// and death_counter == 0x01) by holding 'fire'
m_terminal = (death_counter == 0x01 && lives == 1);
}
/* process the latest information from ALE */
void SolarisSettings::step(const System& system) {
// update the reward
// only 5 digits are displayed but we keep track of 6 digits
int score = getDecimalScore(0xDC, 0xDD, 0xDE, &system);
score *= 10;
int reward = score - m_score;
m_reward = reward;
m_score = score;
// update terminal status
int lives_byte = readRam(&system, 0xD9);
m_terminal = lives_byte == 0;
m_lives = lives_byte & 0xF;
}
/* process the latest information from ALE */
void FrostbiteSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0xCA, 0xC9, 0xC8, &system);
int reward = score - m_score;
m_reward = reward;
m_score = score;
// update terminal status
// MGB: the maximum achievable life is 9. The system will actually let us set the byte to
// higher values & properly decrement, but we do not gain lives beyond 9.
int lives_byte = (readRam(&system, 0xCC) & 0xF);
int flag = readRam(&system, 0xF1) & 0x80;
m_terminal = (lives_byte == 0 && flag != 0);
m_lives = lives_byte + 1;
}
/* process the latest information from ALE */
void MontezumaRevengeSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0x95, 0x94, 0x93, &system);
int reward = score - m_score;
m_reward = reward;
m_score = score;
// update terminal status
int new_lives = readRam(&system, 0xBA);
int m_lives = new_lives;
// if( new_lives < 5){
// //std::cout << "new lives" << new_lives << std::endl;
// m_reward = -10;
// }
int some_byte = readRam(&system, 0xFE);
m_terminal = new_lives == 0 && some_byte == 0x60;
}
/* process the latest information from ALE */
void SpaceInvadersSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0xE8, 0xE6, &system);
m_reward = score - m_score;
m_score = score;
// update terminal status
int reset_val = readRam(&system, 0xAA);
reset_val = (reset_val - (reset_val & 15)) >> 4;
if (reset_val == 8) {
// game is not reset yet; not end of game
m_terminal = false;
} else {
// If bit 0x80 is on, then game is over
int some_byte = readRam(&system, 0x98);
int lives = readRam(&system, 0xC9);
m_terminal = (some_byte & 0x80) || lives == 0;
}
}
/* process the latest information from ALE */
void WizardOfWorSettings::step(const System& system) {
// update the reward
reward_t score = getDecimalScore(0x86, 0x88, &system);
if (score >= 8000) score -= 8000; // MGB score does not go beyond 999
score *= 100;
m_reward = score - m_score;
m_score = score;
// update terminal status
int newLives = readRam(&system, 0x8D) & 15;
int byte1 = readRam(&system, 0xF4);
bool isWaiting = (readRam(&system, 0xD7) & 0x1) == 0;
m_terminal = newLives == 0 && byte1 == 0xF8;
// Wizard of Wor decreases the life total when we move into the play field; we only
// change the life total when we actually are waiting
m_lives = isWaiting ? newLives : m_lives;
}
/* process the latest information from ALE */
void QBertSettings::step(const System& system) {
// update terminal status
int lives_value = readRam(&system, 0x88);
// Lives start at 2 (4 lives, 3 displayed) and go down to 0xFE (death)
// Alternatively we can die and reset within one frame; we catch this case
m_terminal = (lives_value == 0xFE) ||
(lives_value == 0x02 && m_last_lives == 0xFF);
m_last_lives = lives_value;
// update the reward
// Ignore reward if reset the game via the fire button; otherwise the agent
// gets a big negative reward on its last step
if (!m_terminal) {
int score = getDecimalScore(0xDB, 0xDA, 0xD9, &system);
int reward = score - m_score;
m_reward = reward;
m_score = score;
}
else {
m_reward = 0;
}
}
/* process the latest information from ALE */
void GalaxianSettings::step(const System& system) {
// update the reward
int score = getDecimalScore(0xAE, 0xAD, 0xAC, &system);
// reward cannot get negative in this game. When it does, it means that the score has looped
// (overflow)
m_reward = score - m_score;
if(m_reward < 0) {
// 1000000 is the highest possible score
const int maximumScore = 1000000;
m_reward = (maximumScore - m_score) + score;
}
m_score = score;
// update terminal and lives
// If bit 0x80 is on, then game is over
int some_byte = readRam(&system, 0xBF);
m_terminal = (some_byte & 0x80);
if (m_terminal) {
// Force lives to zero when the game is over since otherwise it would be left as 1
m_lives = 0;
} else {
m_lives = readRam(&system, 0xB9) + 1; // 0xB9 keeps the number of lives shown below the screen
}
}
