dbreed: fix sprites on scorebar regression [Mamesick]

src/devices/machine/i8251.h

@@ -133,7 +133,7 @@ private:
 
 	/* data being received */
 	uint8_t m_rx_data;
-		/* tx buffer */
+	/* tx buffer */
 	uint8_t m_tx_data;
 	void sync1_w(uint8_t data);
 	void sync2_w(uint8_t data);

src/mame/fidelity/eag68k.cpp

@@ -10,7 +10,7 @@ Excel 68000 I/O is very similar to EAG, so it's handled in this driver as well
 
 TODO:
 - unemulated waitstates with DTACK
-- EAG USART is not emulated
+- EAG USART is not emulated, conflict with input2_r?
 - V10 CPU emulation is too slow, MAME 68040 opcode timing is same as 68030 but in
   reality it is much faster, same goes for V11 of course (see note below)
 - V11 CPU should be M68EC060, not yet emulated. Now using M68EC040 in its place
@@ -77,7 +77,7 @@ to be compatible with EAG though and will load fine with an adapter.
 The USART allows for a serial connection between the chess computer and another
 device, for example a PC. Fidelity released a DOS tool called EAGLINK which
 featured PC printer support, complete I/O control, detailed information while
-the program is 'thinking', etc.
+the program is 'thinking', etc. It can be enabled with POP3 H3.
 
 Memory map: (of what is known)
 -----------

src/mame/irem/m72.cpp

@@ -1932,7 +1932,6 @@ void m72_state::m81_hharry(machine_config &config)
 void m72_state::m81_xmultipl(machine_config &config)
 {
 	m81_hharry(config);
-	/* basic machine hardware */
 	m_maincpu->set_addrmap(AS_PROGRAM, &m72_state::xmultipl_map);
 
 	MCFG_VIDEO_START_OVERRIDE(m72_state,xmultipl) // different offsets
@@ -1940,7 +1939,7 @@ void m72_state::m81_xmultipl(machine_config &config)
 
 void m72_state::m81_dbreed(machine_config &config)
 {
-	m81_xmultipl(config);
+	m81_hharry(config);
 	m_maincpu->set_addrmap(AS_PROGRAM, &m72_state::dbreed_map);
 }