c140.cpp: Update data bus behavior and support 12 bit PCM, Verified from assault(namcos2) schematics

[cam900]

Add/Update notes, Use shorter/correct type values, Fix address/data bus bits, Modernize save states, Fix spacing, Reduce unnecessary lines, duplicates, Move little endian interface in c140.h, Move bankswitching and address/data decoding behavior in each drivers

namcos2.cpp, namcos21*.cpp: Fix c140 ROM loading, mirrored maps
namcos2.cpp: Implement c140 ROM address/data decoding logics

vgmplay.cpp: Fix C140 compatibility

[superctr]

Should the sample ROM data bus just be treated as 16-bit? Afaik (from Assualt schematics) the least significant 4 bits of each sample are stored interleaved in a "third" ROM bank with some decode logic sending the correct nibble to the chip. Since this rom have been unpopulated in every PCB I have seen, I'm not sure if it matters but it might be good to be correct from the beginning and emulate this decode logic.

[smf-]

The banking should be moved out of the sound chip. Any address bit twiddling or nibble swapping should be done with an address map/rom load.

…

On 04/05/2020 17:30, superctr wrote: Should the sample ROM data bus just be treated as 16-bit? Afaik (from Assualt schematics) the least significant 4 bits of each sample are stored interleaved in a "third" ROM bank with some decode logic sending the correct nibble to the chip. Since this rom have been unpopulated in every PCB I have seen, I'm not sure if it matters but it might be good to be correct from the beginning and emulate this decode logic. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#6638 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACGVYYRLKBTUSX5I4IALW4DRP3URJANCNFSM4MYYJ4DQ>.

[cam900]

TODO: It has noisy sound after this PR, I will fix later and reopen the review for this PR.

Edit: Regression is fixed, so I reopen this.

[superctr]

I think in the long run C219 should be split from C140. Btw, the C219 uses the same "mu-law" algorithm as the C352, it's extensively used by Knuckle Heads.

[p1pkin]

is there was made any attempts to test this PR / code at practice, or it was pure blind written ?
I see a bunch of compiler errors

Compiling src/mame/drivers/namcos21.cpp...
../../../../../src/mame/drivers/gal3.cpp: In member function 'void gal3_state::gal3(machine_config&)':
../../../../../src/mame/drivers/gal3.cpp:683:14: error: 'class c140_device' has no member named 'set_bank_type'
683 | m_c140_16g->set_bank_type(c140_device::C140_TYPE::SYSTEM21); //to be verified
| ^~~~~~~~~~~~~
../../../../../src/mame/drivers/gal3.cpp:683:41: error: 'c140_device::C140_TYPE' has not been declared
683 | m_c140_16g->set_bank_type(c140_device::C140_TYPE::SYSTEM21); //to be verified
| ^~~~~~~~~
../../../../../src/mame/drivers/gal3.cpp:688:14: error: 'class c140_device' has no member named 'set_bank_type'
688 | m_c140_16a->set_bank_type(c140_device::C140_TYPE::SYSTEM21);
| ^~~~~~~~~~~~~
../../../../../src/mame/drivers/gal3.cpp:688:41: error: 'c140_device::C140_TYPE' has not been declared
688 | m_c140_16a->set_bank_type(c140_device::C140_TYPE::SYSTEM21);

[rb6502]

The compile issues with this were blocking top of tree. Please compile and run changes before submitting them.

[cam900]

Fixed in #6659.