mirror of
https://github.com/holub/mame
synced 2025-04-22 16:31:49 +03:00
commit
6000634b89
@ -42,7 +42,7 @@
|
||||
report can be identified by a fixed pattern in the five most
|
||||
significant bits.
|
||||
|
||||
Reports are five bytes long. The first byte contains the the button
|
||||
Reports are five bytes long. The first byte contains the button
|
||||
state; the second and fourth bytes contain X delta; the third and
|
||||
fifth bytes contain Y delta. The two delta values for each axis
|
||||
should be summed. Delta values range from -120 to 127 to prevent
|
||||
|
@ -42,15 +42,14 @@
|
||||
************************************************************
|
||||
|
||||
The device has active high(!) SO strobes triggered by
|
||||
read accesses, which transfer data from the the 8 SI
|
||||
lines to the bus. Logically SO0-7 and SI0-7 ought to
|
||||
be hooked up to the same input matrix, but this only
|
||||
appears to be the case with the Astrocade home systems.
|
||||
The arcade games instead channel the SI inputs through
|
||||
a quartet of MC14539B (pin-compatible with 74153) CMOS
|
||||
multiplexers and connect the SO strobes to unrelated
|
||||
outputs which generally use the upper 8 address bits
|
||||
as data.
|
||||
read accesses, which transfer data from the 8 SI lines
|
||||
to the bus. Logically SO0-7 and SI0-7 ought to be hooked
|
||||
up to the same input matrix, but this only appears to be
|
||||
the case with the Astrocade home systems. The arcade
|
||||
games instead channel the SI inputs through a quartet of
|
||||
MC14539B (pin-compatible with 74153) CMOS multiplexers
|
||||
and connect the SO strobes to unrelated outputs which
|
||||
generally use the upper 8 address bits as data.
|
||||
|
||||
***********************************************************/
|
||||
|
||||
|
@ -233,8 +233,8 @@ public:
|
||||
/// \brief Attempt discovery
|
||||
///
|
||||
/// Concrete derived classes must implement this member function.
|
||||
/// Should return false if the the object is required but not found,
|
||||
/// or true otherwise (the report_missing member function can assist
|
||||
/// Should return false if the object is required but not found, or
|
||||
/// true otherwise (the report_missing member function can assist
|
||||
/// in implementing this behaviour).
|
||||
/// \param [in] isvalidation Pass true if this is a dry run (i.e. no
|
||||
/// intention to actually start the device), or false otherwise.
|
||||
|
@ -36,24 +36,24 @@ WRITE8_MEMBER(laserbat_state_base::csound2_w)
|
||||
The Laser Battle/Lazarian sound board has a SN76477 CSG, two TMS3615
|
||||
tone synthesisers, and a TDA1010 power amplifier. It receives
|
||||
commands from the game board over a 16-bit unidirectional data bus.
|
||||
The CPU cannot write all sixteen lines atomically, it write to lines
|
||||
1-8 as one group and 9-16 as another group.
|
||||
The CPU cannot write all sixteen lines atomically, it writes to
|
||||
lines 1-8 as one group and 9-16 as another group.
|
||||
|
||||
The game board makes the the audio output from the first S2636 PVI
|
||||
(5E) available on a pin at the sound board interface connector, but
|
||||
it isn't routed anywhere, so you won't hear it.
|
||||
The game board makes the audio output from the first S2636 PVI (5E)
|
||||
available on a pin at the sound board interface connector, but it
|
||||
isn't routed anywhere, so you won't hear it.
|
||||
|
||||
The TMS3615 at 05 is clocked at 250kHz (4MHz crystal oscillator
|
||||
divided by 16), and its divide-by-two output is used to clock the
|
||||
TMS3615 at 04. This gives a base 16' note of C3. The combined 8'
|
||||
or 16' outputs are selectable by jumper, allowing board to be
|
||||
or 16' outputs are selectable by jumper, allowing the board to be
|
||||
switched between two octaves. There's no indication of which octave
|
||||
would have been selected in the Lazarian manual.
|
||||
|
||||
There's a filter network between the TMS3615 outputs and the power
|
||||
amplifier with several parameters controllable from the game board.
|
||||
|
||||
The audio output of the SN76477 isn't actually used. Rather the
|
||||
The audio output of the SN76477 isn't actually used. Rather, the
|
||||
signal from before the output amplifier is taken and used to gate
|
||||
distortion elements in the analog filter network.
|
||||
|
||||
|
@ -36,7 +36,7 @@
|
||||
|
||||
P2U
|
||||
===
|
||||
For paging via port 0x78A, a 16K RAM card with RAM at 0x0000 and 0x3fff and the banking logic (see above) is added to the the standard 48K memory card.
|
||||
For paging via port 0x78A, a 16K RAM card with RAM at 0x0000 and 0x3fff and the banking logic (see above) is added to the standard 48K memory card.
|
||||
P2S, P2U: 2x 320K, double sided, 40 tracks, 16 sectors/track, 256 bytes/sector floppy disk drives
|
||||
|
||||
P3, P4
|
||||
|
@ -4,7 +4,7 @@
|
||||
*
|
||||
* Ericsson Information Systems PC "compatibles"
|
||||
*
|
||||
* The Ericsson PC was the the first original Ericsson design for the office PC market replacing the
|
||||
* The Ericsson PC was the first original Ericsson design for the office PC market replacing the
|
||||
* Step/One which was an OEM:ed clone of the Matsushita Mybrain 3000 (see myb3k.cpp driver).
|
||||
*
|
||||
**************************************************************
|
||||
|
@ -250,7 +250,7 @@ WRITE16_MEMBER(seta2_state::vregs_w)
|
||||
int vpos = m_rasterposition;
|
||||
|
||||
// in the vblank it specifies line 0, the first raster interrupt then specifies line 0 again before the subsequent ones use the real line numbers?
|
||||
// It seems more likely the the raster IRQ stays asserted for the entire line, thus triggering a second interrupt unless the line number is changed?
|
||||
// It seems more likely that the raster IRQ stays asserted for the entire line, thus triggering a second interrupt unless the line number is changed?
|
||||
if (m_rasterposition == m_screen->vpos()) hpos = m_screen->hpos() + 0x100;
|
||||
//logerror("setting raster to %d %d\n", vpos, hpos);
|
||||
m_raster_timer->adjust(m_screen->time_until_pos(vpos, hpos), 0);
|
||||
|
Loading…
Reference in New Issue
Block a user