trunk/src/emu/sound/tms5220.c
| r249066 | r249067 | |
|---|
| 48 | 48 | |
| 49 | 49 | TODO: |
| 50 | 50 | * Ever since the big rewrite, there are glitches on certain frame transitions |
| 51 | | for example in the word 'robots' during the eprom attract mode, |
| 51 | for example in the word 'rid' during the eprom attract mode, |
| 52 | 52 | I (LN) am not entirely sure why the real chip doesn't have these as well. |
| 53 | 53 | Needs more real hardware testing/dumps for comparison. |
| 54 | | * Ever since the timing rewrite, the above problem is slightly worse. This |
| 55 | | time, however, it is probably a 'real' bug, which I (LN) am in the process |
| 56 | | of tracking down. |
| 57 | | i.e. the word 'congratulations' in victory when you get a high score. |
| 58 | 54 | * Implement a ready callback for pc interfaces |
| 59 | 55 | - this will be quite a challenge since for it to be really accurate |
| 60 | 56 | the whole emulation has to run in sync (lots of timers) with the |
| r249066 | r249067 | |
|---|
| 62 | 58 | * If a command is still executing, /READY will be kept high until the command has |
| 63 | 59 | finished if the next command is written. |
| 64 | 60 | * tomcat has a 5220 which is not hooked up at all |
| 65 | | * Is the TS=0 forcing energy to 0 for next frame in the interpolator actually correct? I'm (LN) guessing no. The patent schematics state that TS=0 shuts off the output dac completely, though doesn't affect the I/O pin. |
| 66 | 61 | |
| 67 | 62 | Pedantic detail from observation of real chip: |
| 68 | 63 | The 5200 and 5220 chips outputs the following coefficients over PROMOUT while |
| r249066 | r249067 | |
|---|
| 267 | 262 | //define INTERP_SHIFT / (1<<m_coeff->interp_coeff[m_IP]) |
| 268 | 263 | |
| 269 | 264 | /* Other hacks */ |
| 270 | | /* HACK?: if defined, outputs the low 4 bits of the lattice filter to the i/o |
| 265 | /* HACK: if defined, outputs the low 4 bits of the lattice filter to the i/o |
| 271 | 266 | * or clip logic, even though the real hardware doesn't do this, partially verified by decap */ |
| 272 | 267 | #undef ALLOW_4_LSB |
| 273 | 268 | |