6502/CLK
1
0
Fork 0
mirror of https://github.com/TomHarte/CLK.git synced 2024-12-13 00:29:14 +00:00
Code Issues Projects Releases Wiki Activity
11,869 Commits 10 Branches 121 Tags 500 MiB
268842681a
Commit Graph

171 Commits

Author SHA1 Message Date
Thomas Harte
8a11a5832c Uses GI::AY38910::Utility far and wide. 2021-03-26 23:19:47 -04:00
Thomas Harte
7e39550fc0 Attempts to make JustInTimeActor sequence-point aware. 
With the objective of chopping out a lot of future boilerplate.
 2020-11-15 21:58:18 -05:00
Thomas Harte
5dcf720bb5 Extends list of BusOperations. 
Now to retest, widely.
 2020-10-15 21:35:01 -04:00
Thomas Harte
2477752fa4 Adds further [[fallthrough]] attributes. 2020-06-19 23:36:51 -04:00
Thomas Harte
495024d6fe Cleans up all redundant lock/unique_guard declarations. 2020-06-15 00:24:10 -04:00
Thomas Harte
48afc54af6 Cuts down unused parameter warnings to just a few that may well indicate implementation errors. 2020-05-30 01:06:43 -04:00
Thomas Harte
267006782f Starts to add Qt target; resolves many build warnings. 2020-05-30 00:37:06 -04:00
Thomas Harte
512a52e88d Increases const correctness, marks some additional constructors as constexpr, switches std::atomic construction style. 2020-05-20 23:34:26 -04:00
Thomas Harte
66c2eb0414 Further tightens const and constexpr usage. 2020-05-12 22:22:21 -04:00
Thomas Harte
25996ce180 Further doubles down on construction syntax for type conversions. 2020-05-09 23:00:39 -04:00
Thomas Harte
31c6faf3c8 Adds a bunch of consts. 2020-05-09 21:23:52 -04:00
Thomas Harte
f417fa82a4 Splits 'CRTMachine' into three parts: ScanProducer, AudioProducer, TimedMachine. 
Simultaneously cleans up some of the naming conventions and tries to make things a bit more template-compatible.
 2020-04-01 23:19:34 -04:00
Thomas Harte
ec6664f590 Takes steps to guarantee property naming; reintroduces Electron runtime options. 2020-03-17 23:52:55 -04:00
Thomas Harte
394ee61c78 Starts a switch to reflectable-style runtime options. 
The Amstrad CPC and ZX80/81 have made the jump so far, subject to caveats. The macOS build is unlikely currently to work properly.
 2020-03-16 23:25:05 -04:00
Thomas Harte
8e3bf0dbca Starts moving towards a Deflectable-based system of runtime options. 2020-03-15 23:48:53 -04:00
Thomas Harte
b2c07b3110 The Atari ST doesn't offer quick loading. 2020-03-01 22:10:41 -05:00
Thomas Harte
50d356be2f Ensures all audio sources, including compound sources, announce whether they're stereo correctly. 2020-02-16 18:31:45 -05:00
Thomas Harte
e02d109864 Nudges the LowpassSpeaker towards supporting stereo generation. 2020-02-15 18:03:12 -05:00
Thomas Harte
9ca2d8f9f2 Tried to be less lazy with lambda captures. 
This is primarily defensive.
 2020-02-14 23:39:08 -05:00
Thomas Harte
91a3d42919 Ensures no DMA clocking whatsoever when asleep. 2020-02-12 23:23:42 -05:00
Thomas Harte
05bcd73f82 Attempts to pull drive ownership into DiskController. 
For the sake of being more intelligent as to drive clocking, hopefully. And, eventually, to support multiple drive selection.
 2020-02-11 21:59:13 -05:00
Thomas Harte
24340d1d4f Resolves fetch errors. 2020-02-09 17:04:49 -05:00
Thomas Harte
085529ed72 Makes the shifter behaviour conform to its documentation. 2020-02-02 17:26:39 -05:00
Thomas Harte
019474300d Centralises responsibility for picking irrelevant numbers for a computer-style monitor. 2020-01-30 23:26:02 -05:00
Thomas Harte
af976b8b3d Eliminates modulus operation per ROM access. 2020-01-30 23:09:24 -05:00
Thomas Harte
f3db1a0c60 Eliminates ad hoc scheduling for delayed DE -> LOAD. 2020-01-29 22:50:22 -05:00
Thomas Harte
ce28213a5e [Mostly] unifies deferral process. 2020-01-29 22:46:08 -05:00
Thomas Harte
f9ce50d2bb Adds some debugging `asserts. 2020-01-29 22:45:44 -05:00
Thomas Harte
f0a6e0f3d5 Splits out the queue management stuff from queue+action. 
Temporarily breaks ST video in the endeavour.
 2020-01-29 22:18:41 -05:00
Thomas Harte
0e29c6b0ab On further reflection, I think events should occur after the running period. 
I'm testing this now for sanity in 2/4bpp mode.
 2020-01-28 23:26:37 -05:00
Thomas Harte
c5edc879b6 Switches back to testing the monochrome monitor. 2020-01-28 22:12:57 -05:00
Thomas Harte
65309e60c4 Corrects sequence point generation by allowing for hsync_end != end of line. 2020-01-28 20:38:20 -05:00
Thomas Harte
2c0cab9e4d Adds line length latching as a line event. 2020-01-28 20:22:37 -05:00
Thomas Harte
b1ff031b54 Fixes runtime test. 2020-01-27 23:41:08 -05:00
Thomas Harte
7e8405e68a Makes 72Hz horizontal sync independently relocatable. 
... and moves and shortens it, based on my guesswork as to requirements.
 2020-01-27 23:40:01 -05:00
Thomas Harte
c8fd00217d Resolves loss of horizontal resolution in 1bpp mode. 2020-01-27 23:08:28 -05:00
Thomas Harte
9d340599a6 Towards ST 1bpp support: puts vsync in an appropriate location, starts experimenting with proper CRT timings. 2020-01-27 23:00:30 -05:00
Thomas Harte
294e09f275 All these 'override's can be 'final's. 
At least for the purpose of being communicative. I doubt there's much to gain in terms of compiler output — the DiskImageHolder can avoid some virtual lookups but nothing else leaps out.
 2020-01-23 22:57:51 -05:00
Thomas Harte
6802318784 Removes audio_queue_.flush() calls; I don't think I really need to block. At least, not usually. 2020-01-23 20:13:16 -05:00
Thomas Harte
428d141bc9 Factors out the logic behind the Atari 2600's frequency switching. 2020-01-23 20:12:44 -05:00
Thomas Harte
8404409c0d Causes the Atari 2600 to obey normal flush semantics. 
This stuff is going to become more important with run_until.
 2020-01-22 22:05:51 -05:00
Thomas Harte
a71c5946f0 Ensures proper manipulation of scan_statuses, leading to the correct result out of a CRTMachine. 
Possibly with the exception of the TMS, as I appear to have uncovered an unrelated issue there.
 2020-01-21 22:28:25 -05:00
Thomas Harte
d97a073d1b Adds the necessary routine for all machines to be able to respond to get_scan_status. 
They all just as the CRT, as all are currently based on the CRT. Which doesn't currently know the total clock rate it would need to in order properly to scale the answer to the question. Further thought coming.
 2020-01-20 21:45:10 -05:00
Thomas Harte
c1bae49a92 Standardises on read and write for bus accesses. 
Logic being: name these things for the bus action they model, not the effect they have.
 2020-01-05 13:40:02 -05:00
Thomas Harte
e4349f5e05 Slightly clarifies logic. 2020-01-04 21:32:34 -05:00
Thomas Harte
7b2777ac08 Sorts cases into order; adds copious audio mirrors. 2020-01-04 21:06:21 -05:00
Thomas Harte
0fbcbfc61b Switches to more idiomatic address listing. 2020-01-04 20:35:47 -05:00
Thomas Harte
3ab4fb8c79 Enables an assumption of partial address decoding at the ACIA and PSG. 2020-01-04 17:27:55 -05:00
Thomas Harte
19ddfae6d6 Adds Joystick key code mode, ensures events aren't posted in interrogation mode. 
This should fix Turrican due to the latter change; I'm not aware of software that uses the former.
 2020-01-04 09:45:59 -05:00
Thomas Harte
414b0cc234 Reintroduces sync write delay. 2020-01-02 23:36:11 -05:00
Thomas Harte
134e828336 Updates note to self. 2020-01-02 23:33:35 -05:00
Thomas Harte
455e831b87 Corrects bug whereby changing pixel mode mid-line will produce an improper amount of data. 2020-01-02 23:18:21 -05:00
Thomas Harte
42ccf48966 Judging by Pompey Pirates Menu 88, vsync should occur a line earlier, ending during line 0. 2020-01-02 20:16:28 -05:00
Thomas Harte
2f8078db22 Switches to should_log as a global when I'm hacking about. 2020-01-02 20:15:48 -05:00
Thomas Harte
23ed9ad2de Corrects application of negative relative scale. 2020-01-01 13:22:21 -05:00
Thomas Harte
017681a97c Now honours permitted mouse range. 2020-01-01 12:48:38 -05:00
Thomas Harte
90b899c00e Attempts to implement absolute mouse positioning mode. 
Along with mouse direction.
 2020-01-01 12:29:33 -05:00
Thomas Harte
c11fe25537 Merge branch 'master' into EnchantedWoods 2019-12-30 23:32:45 -05:00
Thomas Harte
0a12893d63 Shunts vsync back down to top of frame. 
It's guess after guess, basically.
 2019-12-30 23:01:31 -05:00
Thomas Harte
8e777c299f Switches to latching video interrupts until acknowledged. 
Seems to fix Cisco Heat, at least. I have no idea whether I'm latching the correct thing, whether IACK should clear both or only one, etc.
 2019-12-30 23:00:55 -05:00
Thomas Harte
e23d1a2958 Restores vsync active. 2019-12-29 22:03:36 -05:00
Thomas Harte
6449403f6a Corrects pending_events_ test for sequence points. 
Simplifies around as possible.
 2019-12-29 21:53:45 -05:00
Thomas Harte
c8fe66092b Attempts to correct insertion logic (and mostly bypasses it). 2019-12-29 21:42:41 -05:00
Thomas Harte
b33218c61e Fixes reload test, which really needs to sense the CRT-headed vsync output. 
i.e. not the one heading back to the CPU.
 2019-12-29 20:55:34 -05:00
Thomas Harte
8ce26e7182 Adds a delay on vsync visibility (i.e. as to generating an interrupt). 2019-12-29 19:03:08 -05:00
Thomas Harte
47068ee081 Ensures visible hsync end generates a sequence point. 2019-12-29 17:51:50 -05:00
Thomas Harte
214b6a254a Adds a delay on visibility of the hsync signal, and a test on address reload. 2019-12-29 17:37:09 -05:00
Thomas Harte
93f6964d8a Introduces some preliminary line length unit tests. 
Thereby fixes one potential issue with load_ toggling.
 2019-12-28 22:50:34 -05:00
Thomas Harte
13f11e071a Simplifies border colour change propagation. 
I'm not sure it was even technically correct as was.
 2019-12-28 10:45:10 -05:00
Thomas Harte
f7825dd2a2 Pulls out address reload position as a separate constant. 2019-12-28 10:36:50 -05:00
Thomas Harte
a9d1f5d925 Pulls out address reload as something I can position independently. 
Sadly receding it by 3 did not have the effect I was hoping for, of receding Enchanted Land's first register tweaking.
 2019-12-27 23:47:19 -05:00
Thomas Harte
2757e5d600 Removes untrue comment. 2019-12-27 22:51:11 -05:00
Thomas Harte
5026de9653 Rejigs the video stream to ensure shifter really is continuous. 
... and definitively to avoid potential buffer overruns. Or, at least, to have a mechanism in place definitively to avoid them. Which will be tested and debugged as necessary.

Also simplifies the colour burst and border/pixels selection logic.
 2019-12-27 22:47:34 -05:00
Thomas Harte
5fa8e046d8 It's inaccurrate to call this _the_ shifter. So don't. 2019-12-27 19:03:10 -05:00
Thomas Harte
a9a92de954 Adds a bunch of shout-outs for unimplemented behaviour. 2019-12-25 15:32:33 -05:00
Thomas Harte
a8ba3607b7 Adds (and disables) a minor additional piece of logging. 2019-12-24 21:43:39 -05:00
Thomas Harte
5068328a15 Fixes debugging output. 2019-12-24 19:15:58 -05:00
Thomas Harte
b2bed82da6 Switches to standard logging. 2019-12-23 22:00:40 -05:00
Thomas Harte
0dae608da5 Embraces std::make_[unique/shared] in place of .reset(new . 2019-12-23 21:31:46 -05:00
Thomas Harte
274867579b Deploys constexpr as a stricter const. 2019-12-22 00:22:17 -05:00
Thomas Harte
e5440a4146 Hacks in a colour burst. 
With a major flaw: it's implicit. I think I need a minor rethink of various components here.
 2019-12-20 23:49:38 -05:00
Thomas Harte
47508d50a7 Wires through a composite video option for the ST. 
Which is great and all, except that I've not yet inserted a colour burst. So it's monochrome.
 2019-12-20 20:49:14 -05:00
Thomas Harte
3d83f5ab49 Ensures a proper size handoff and implements a ripple feature I happened to find a forum post about. 2019-12-19 22:58:07 -05:00
Thomas Harte
0007dc23b3 Eliminates bit 0 of the DMA address. 2019-12-19 22:44:21 -05:00
Thomas Harte
ed7f171736 Moves address reload to end of vertical sync. 
I have no information as to when it should be, so this is as valid a guess as any other.
 2019-12-19 22:20:43 -05:00
Thomas Harte
0e066f0f70 Removes 'done' TODO. 
For certain values of done.
 2019-12-19 22:19:59 -05:00
Thomas Harte
d85ae21b2f Adds an explicit declaration of chip type to all AY users. 2019-12-18 19:28:41 -05:00
Thomas Harte
c00ae7ce6a Adds a one-cycle delay on frequency changes. 2019-12-13 19:57:54 -05:00
Thomas Harte
4bcf217324 Ensures delayed loading isn't interrupted by blank, hsync. 2019-12-12 23:20:28 -05:00
Thomas Harte
f6f2b4b90f Removes double DE edge test. 2019-12-12 22:50:35 -05:00
Thomas Harte
95b5db4d87 Tweaks timings yet further, adds a FIFO reset. 
The accuracy of this may require further research.
 2019-12-11 23:22:20 -05:00
Thomas Harte
de4403e021 Corrects blank timing. 2019-12-10 22:17:57 -05:00
Thomas Harte
0a405d1c06 Introduces a latency between DE and load. 2019-12-10 21:24:15 -05:00
Thomas Harte
768b3709b8 Corrects audio clock rate. 2019-12-10 20:25:27 -05:00
Thomas Harte
a7cfb840ef Adds but presently disables a diagnostic for border elimination. 2019-12-08 22:34:42 -05:00
Thomas Harte
c785797da6 Adds a warning for unhandled reset. 2019-12-08 21:01:30 -05:00
Thomas Harte
0408592ada Switches to byte buffers and seeks to reduce unnecessary video flushing. 2019-12-08 20:20:13 -05:00
Thomas Harte
0ed87c61bd Introduces an explicit area of floating bus, starts adding bus errors. 2019-12-08 11:52:43 -05:00
Thomas Harte
08a27bdec7 NTSC frame length is correct; removes TODO. 2019-12-08 11:51:12 -05:00
Thomas Harte
894066984c Moves beginning and end of vertical sync to what I now believe is its proper place. 
At least one demo now successfully opens the top border.
 2019-11-19 20:13:47 -05:00