Commit Graph

28 Commits

Author SHA1 Message Date
Tyler Wilding
400ed82715 recording: Many adjustments on VirtualPad from review feedback
recording: Factor in DPIScaling when scaling VirtualPad window


recording: Add a reset button, hopefully better quality backgrounds for lower resolutions


recording: Improved VPad analog alignment

Co-authored-by: sonicfind
recording: Some final VirtualPad coordinate adjustments
2020-10-01 17:52:59 +01:00
Tyler Wilding
14e1ecbbf6 recording: Store VirtualPad window position in .ini file 2020-10-01 17:52:59 +01:00
Tyler Wilding
4519e32586 vs/recording: Remove the committed header files, generate them at build-time
recording: General cleanup, remove obvious code duplication

recording: Improved function arg names in VirtualPad


format: Clang-format all new files
2020-10-01 17:52:59 +01:00
Tyler Wilding
8a968a0c11 recording: set ReadOnlyMode correctly when playing/recording a movie
cmake: Update cmake file lists accordingly


cmake: debugging


cmake debugging
2020-10-01 17:52:59 +01:00
Tyler Wilding
566d0c846f recording: improvements so frame will be less likely to appear un-rendered 2020-10-01 17:52:59 +01:00
Tyler Wilding
32047c6130 recording: Disable widgets in replay mode, simplify ControllerInterrupt
recording: Added some comments to the more complex parts
2020-10-01 17:52:59 +01:00
Tyler Wilding
bbc305d2b9 recording: Linting pass on files involved in this change 2020-10-01 17:52:59 +01:00
Tyler Wilding
d296f0257f recording: Removed redundant #includes and reliance on Common.h
Including Common.h was excessive (for logging) and causing compiler errors on MSVCv142, this bypasses that issue.
The only references to something similar to the errors I was seeing was this https://developercommunity.visualstudio.com/content/problem/845325/stdmake-shared-pch-virtual-method-googletest-cause.html which is resolved in the upcoming VS - 16.6 release (16.5 at the time of writing)
2020-10-01 17:52:59 +01:00
Tyler Wilding
ed829d427c recording: Update controller interrupt to handle both the VirtualPad and the movie
vs: solution updates for new folders / filters / files
2020-10-01 17:52:59 +01:00
sonicfind
11fe2619eb recording: Disable currently active frame advancing when opening a different recording 2020-09-18 23:31:23 +01:00
sonicfind
9936fc7f15 recorrding: Resolve issues with starting a new file 2020-09-18 23:31:23 +01:00
sonicfind
23dba26ba3 recording: Rearrangements + Formatting
* RecordingReset is now a function of the InputRecording class
* TotalFrames is now signed
* InputRecordingControls refactor, IsRecordingPaused -> IsPaused
* Add check to keep frameCounter from overflowing (not like someone is gonna record a year's worth of inputs though).

* Code Formatting adjustments (more to do in a separate PR)
2020-09-18 23:31:23 +01:00
sonicfind
f881bd9bd6 recording: Lock frameCounter incrementation on initial recording file load or during full/fast boot
Ensures that no frames of input are skipped over on the initial load of a recording file or during a fast/full boot when a recording file is active. Improved consistency.
2020-09-18 23:31:23 +01:00
sonicfind
5c5d3a770f recording: Handle mode toggle in InputRecordingControls 2020-09-18 23:31:23 +01:00
sonicfind
975a37f209 recording: Handle Boots & Savestates similarly
* Replaces the savestate load flag with a more general initial load flag to be used by full/fast boots and savestates.
* Have the Pcsx2App::ResetRecordingCounter() method mirror the similar method of the savestate class.
2020-09-18 23:31:23 +01:00
sonicfind
b591c5e9ab recording: Rework Frame Value & Undo Count Functions
* Handle setting frameCounter to 0 in SetStartingFrame().
* FrameCounter, signed -> unsigned.
* Have SetFrameCounter handle the calculations for the relative frame displacement from the startingFrame.
* Optimizes InputRecording::controllerInterrupt.
* Flip the order of execution in inputRecordingFreeze.
* Set framecounter properly on full/fast boot.
* More accurately increment undo count
2020-09-18 23:31:23 +01:00
sonicfind
0304b124ed recording: Recording mode-based refactors
* Recording mode enum, NoneActive -> NotActive
* Changed IsMode method names and added an IsRecording method.
* Add methods designated to setting a recording to a certain mode.

Co-authored-by: Tyler Wilding <xtvaser@gmail.com>
2020-09-18 23:31:23 +01:00
Tyler Wilding
74bba35765 recording: Resolve issues discovered while testing around savestate handling
* Add a flag to explicitly indicate if the initial SS has been loaded.
* Additionally: bracket formatting / spaces -> tabs.
2020-09-18 23:31:23 +01:00
Tyler Wilding
6c80e6b93f recording: Recording controls rename and refactor to attempt to simplify
* RecordingControls - > InputRecordingControls (+ function refactors & simplifications)
* Avoid mutating g_FrameCount - now use isolated frameCounter & startingFrame variables for input recordings
* Pause or resume emulation preemptively based on GUI actions
* Exclusively handle GS window title updates in FrameForGS
2020-09-18 23:31:23 +01:00
sonicfind
4d66818746 GUI/Recording: Resolve issues around playing and creating input recordings under certain scenarios (#3669)
* recording: Only stop the current recording once the menuing hasnt been cancelled

* recording: Avoid changing menu option's state if the recording couldn't be played

* recording: Don't play a recording that uses a savestate, if no game is running

* recording: Don't modify `frameAdvance` flag when explicitly [un]pausing

These functions are only called when creating/playing a recording, and modifying the frameAdvance flag leads to unexpected behaviour (such as the game starting paused if the recording was made from "power-on")

* recording: Refactor and simplify `InputRecordingFile`

InputRecordingFile will no longer be concerned with loading the save-state when playing back an existing recording.  This makes it much easier to only load the save-state if the file is valid and manipulate the emulation state correctly.

* recording: Update play logic with new refactor, resume emulation in the event of a failure

* recording/lint: spaces to tabs

* recording: Properly only examine controller 1A & 2A

The previous controller port checks in place never actually succeeded in their designated task. A new slot check in sio.cpp will perform this task instead.

* recording: Save the savestate in OpenNew() instead of open()

Ensures that the savesate could be saved before trying to create the actual input recording file. It will overwrite any previous backup savestate.
Also, allows for a simplified & easier to read code struture of open().

* Refactor and simplify `InputRecording`

Changes the return type of Play/Create from void to bool.
Optimizes Stop(), Pause(), and Unpause() call placements

Improved handling of emulation pause state, the recording menu on failures, and the conditioning of when a recording file should actually be unloaded.
For example, a currently loaded recording should not get unloaded if a user presses Play *but* chooses cancel in the file browser. However, the emulation should be paused during the duration of this action.
On the flipside, a loaded recording *should* get unloaded if the tools get disabled in settings AND emulation should resume if not already playing.

* recording: Simplify VirtualPad_Open_Click

Co-authored-by: Tyler Wilding <xtvaser@gmail.com>
2020-09-11 21:30:56 +01:00
Tyler Wilding
9d6d7f7f63 pcsx2-tas: Corrections to recording file's header data (#3480)
* The emulator version was hard-coded to PCSX2-1.5.X which is no longer accurate, it is no longer hard-coded and will use the correct version number the same way it is calculated to display in the window title.

* When creating a recording, the game name is preferred over the ISO name. This is determined via the GameDB. When playing back the recording, a simple check occurs to see if the game running is the same one that was used to make the recording. On the playback side, it always only checked with the ISO filename.
2020-08-08 22:25:19 +02:00
lightningterror
62dd5d980d pcsx2-tas: Adjust TAS code to be properly disabled with macro.
VS complains about errors so isolate the code in each file for TAS with macros.

Remove ifndef for m_capturingVideo as it is used for video capture menu.
2020-03-19 20:27:19 +01:00
Tyler Wilding
506ea4c4d2 recording: various formatting and review corrections
Squashed commit:

[7955b42e3] recording: Throw errors on fread/fwrite errors.

[5a2160f9e] recording: Remove function implementation from header files

[f2937ab5f] recording: Fixed UndoCount metadata bug and will gracefully fail if savestate is missing

[d7f4d43e5] recording: Refactored code-style to be consistent

[0f77fbb71] recording: Refactor to use switch statements

[28d7945f6] recording: Resolve CMake warnings and use tagged github links for cross-linking to LilyPad

[7c01c6cb4] recording: corrected disparity between comment and code

[17a8bd8d6] recording: Remove all usages of #define

[3830f5a82] recording: Refactor enums and general cleanup

[569ef7d67] recording: Completely disable new console log sources when recording is disabled
2019-04-21 00:45:44 +02:00
Tyler Wilding
08d923ca6f recording: Append copyright headers
Squashed commits:

[47be08613] recording: Forgot to refactor the usage of std::size
2019-04-21 00:45:44 +02:00
Tyler Wilding
c12c6ed149 recording: Added some more useful logs and cleaned up some TODOs 2019-04-21 00:45:44 +02:00
Tyler Wilding
eb7030cf12 recording: Use conventional savestate functions, save save-state in a separate file alongside recording file
Regressions were discovered after merging with master due to way the save state data was saved within the movie file.
This change uses the same functions used in the GUI to create savestates to create a compressed save-state file.  Eventually this could be re-incorporated back into the recording file and could be backwards compatible.
2019-04-21 00:45:44 +02:00
Tyler Wilding
cb7425c59f recording: removed C++ reserved pattern usages, removed japanese comments 2019-04-21 00:45:44 +02:00
Tyler Wilding
ddd2e3f8f0 recording: main recording functions and input gathering 2019-04-21 00:45:44 +02:00