After spending the last month staring at and comparing Jak 3 and Jak 1
versions of a bunch of `target` code for my jetboard mod, I figured this
would be a good opportunity to revive this ancient PR #1714 along with
some other small misc fixes/improvements.
Instead of directly replacing the old fields, I decided to opt for using
overlay fields to maintain backwards compatibility with existing manual
patches, files without ref tests and mods that might use these fields.
Looking at the balance-plats, they showed an immediate visual clue,
which was the stuttering of Jak's feet on their surface.
This is precisely what happens when you don't include sticky on a
collision; it doesn't "pull" Jak along with it, and he repetitively
"falls" onto it's surface over and over as it moves downwards.
This is a mystery, since this actor does have sticky, in fact, it's not
even possible to detect riders without sticky, which these certainly do.
This same appears on PS2, so I knew fixing it was not an option.
I had an immediate theory of what was happening: since Jak is
continuously falling onto it as it moves downwards, he's only riding
some frames, and skipping others. A higher frame rate = more moments
he's touching.
This means that basically, the balance-plat's entire speed is determined
by when his feet happen to micro-stutter. This is the hardcoded scalar
for it's entire speed.
This means I needed to emulate the 60 FPS "falling" at different
framerates. I tried adding an fps counter to the actor so that the rider
check only ran once every 60th of a second, but I knew adding a variable
to the actor just for fps wasn't a good idea.
However, I was able to confirm that would work. Luckily, this
[comment](https://github.com/open-goal/jak-project/blob/0fa93ce7b83d1621f3fd040625bfa865eadaeef0/goal_src/jak1/engine/game/game-h.gc#L116)
gave me an idea, which meant I could use a variable of process drawable,
avoiding adding a variable.
This unique problem needs a unique solution, I gate rider checks so they
only pass if one sixtieth of a second `(seconds 0.01666666)` has passed
since the last ride. This makes it act like his feet are stuttering like
they do at 60 FPS, avoiding the extra frames where his feet are touching
at high framerates that add more speed.
I also added the same for the send-to messages, since at high frame
rates the passive return to center that runs every frame was stomping
the send-to-next velocities which run less often.
Then, I managed to refactor it down to a tiny edit, a guard in one place
that gates on 1/60th of a second, framerate already 60, or, self grow
(since you don't want to double-gate these).
```lisp
(when (or (= (-> *pc-settings* target-fps) 60)
(time-elapsed? (-> self state-time) (seconds 0.01666666))
(-> self got-grow))
```
It's just: this guard, `(set-state-time)` on ride, and
`(set-state-time)` on grow (limits increased-rate return to center
stomping grow).
Lastly, there was an issue with the existing multiplies by FPS RATIO,
which was a clear double scale of the acceleration (once when setting
accelerate, then again when adding the accelerate). I'm guessing since
the extra riding frames added so much speed, this double scalar helped
slow it down, however when corrected the double scalar made it too slow
since it was slowing down for framerate twice.
There are two issues here:
1 is the "punch in" where the logo punches inwards towards the camera on
first appearance, this effect reveals the edges of village1
significantly and happens for all users even on regular 16:9.
2 is on aspect ratios wider than 16:9, you see even larger areas of
village1 on either side.
The blackout is an artgroup that breaks apart and is set to the shape of
the Jak 1 logo. It has a baked in animation and is a set size, so it
cannot simply be scaled or adjusted to be wider.
Seeing the sides of village1 before seeing the logo is not the immersive
experience this opening was going for, where the game "bursts" into the
screen.
My first thought to fix was to simply call `set-letterbox-frames` until
the right moment. This worked, however, it didn't produce an ideal
result simply because the letterbox also went on top of the light blue
"logo volumes" starburst.
Having experimented with the range of edits possible here, I decided the
best approach is actually to simply delay displaying village1 ever so
slightly. This avoids editing the very particular blackout art group,
with set sizes of chunks that break apart. This also avoids covering the
blue volumes with a side letterbox/covering.
This scales with any aspect, supports both different language logo
blackout shapes, and fixes both wide aspect issues as well as the
regular 16:9 punch in issue. This is an important edit as it's the first
thing people see.
<video controls
src="https://github.com/user-attachments/assets/9e5bfd39-4189-4c7c-bf76-e7ea0df0464f"></video>
---------
Co-authored-by: Hat Kid <6624576+Hat-Kid@users.noreply.github.com>
Migrates all code, there should be no change in the compilation output
(linter should check this)
At first i was considering making these builtins, which short of a bunch
of code generation, would require some sort of dynamic definition in
`Atoms.cpp`. This isn't hard but, i figured it would be better to keep
it simple and just generate the OG macros.
Fixes https://github.com/open-goal/jak-project/issues/233
Found out the hard way that the `abutton` array for pressure sensitivity
uses a different order than the `pad-buttons` enum, so I created a new
enum for abutton indexing.
I replaced any magic numbers across the codebase where it made sense,
e.g.
`(-> *cpad-list* cpads (-> self control unknown-cpad-info00 number)
abutton 6)`
becomes
`(-> *cpad-list* cpads (-> self control unknown-cpad-info00 number)
abutton (abutton-idx x))`
This attempts to do a best-effort quick fix for the sprite alignment in
the menus and first person views on higher aspect ratios. This:
- Hides the binocular borders completely when using a non-standard ratio
- Hides the borders in jak's first person view when using a non-standard
ratio
- Uses a combination of manual alignment and approximation to get the
pause menu closer.
> 32:9 screenshot.
I accomplished the last one by manually aligning all of the core sprites
and text for the most popular aspect ratios. This means that from a
practical standpoint, things should align "perfectly". However, I then
used all of those values to derive a polynomial for each adjustment
based on the aspect ratio. This allows the game to do a half-decent
approximation/interpolation for every aspect ratio in-between the common
ones. It won't be perfect, but it will be better than this:
This PR does two main things:
1. Work through the main low-hanging fruit issues in the formatter
keeping it from feeling mature and usable
2. Iterate and prove that point by formatting all of the Jak 1 code
base. **This has removed around 100K lines in total.**
- The decompiler will now format it's results for jak 1 to keep things
from drifting back to where they were. This is controlled by a new
config flag `format_code`.
How am I confident this hasn't broken anything?:
- I compiled the entire project and stored it's `out/jak1/obj` files
separately
- I then recompiled the project after formatting and wrote a script that
md5's each file and compares it (`compare-compilation-outputs.py`
- The results (eventually) were the same:
> This proves that the only difference before and after is non-critical
whitespace for all code/macros that is actually in use.
I'm still aware of improvements that could be made to the formatter, as
well as general optimization of it's performance. But in general these
are for rare or non-critical situations in my opinion and I'll work
through them before doing Jak 2. The vast majority looks great and is
working properly at this point. Those known issues are the following if
you are curious:
This adds support for generating collide meshes when importing custom
models. A couple of things to keep in mind:
- A single `collide-mesh` may only have up to 255 vertices.
- When exporting a GLTF file in Blender, a `collide-mesh` will be
generated for every mesh object that has collision properties applied
(ideally, you would set all visual meshes to `ignore` and your collision
meshes to `invisible` in the OpenGOAL plugin's custom properties).
- Ensure that your actor using the model properly allocates enough
`collide-shape-prim-mesh`es for each `collide-mesh` ([example from the
original game that uses multiple
meshes](https://github.com/open-goal/jak-project/blob/f6688659f2ef85f5ceaacea6271580c9f4d91ed1/goal_src/jak1/levels/finalboss/robotboss.gc#L2628-L2806)).
~One annoying problem that I haven't fully figured out yet (unrelated to
the actual functionality):
`collide-mesh`es are stored in art groups as an `(array collide-mesh)`
in the `art-joint-geo`'s `extra`, so I had to add a new `Res` type to
support this. The way that `array`s are stored in `res-lump`s is a bit
of a hack right now. The lump only stores a pointer to the array, so the
size of that is 4 bytes, but because we have to generate all the actual
array data too, the current `ResLump` code in C++ doesn't handle this
case well and would assert, so I decided to omit the asserts if an
`array` tag is present and "fake" the size so the object file is
generated more closely to how the game expects it until we figure out
something better.~
This was fixed by generating the array data beforehand and creating a
`ResRef` class that takes the pointer to the array data and adds it to
the lump.
This does a couple of things:
- The `custom_levels` folder was renamed to `custom_assets` and contains
`levels`, `models` and `texture_replacements` folders for Jak 1, 2 and 3
in order to keep everything regarding custom stuff in one place.
- With this, texture replacements now use separate folders for all games
- A build actor tool was added that generates art groups for custom
actors
- Custom levels can now specify what custom models from the `models`
folder they want to import, this will add them to the level's FR3.
- A `test-zone-obs.gc` file was added, containing a `test-actor` process
that uses a custom model as an example.
The build actor tool is still very WIP, the joints and the default
animation are hardcoded, but it allows for importing any GLB file as a
merc model.
Relates to #1353
This adds no new functionality or overhead to the compiler, yet. This is
the preliminary work that has:
- added code to the compiler in several spots to flag when something is
used without being properly required/imported/whatever (disabled by
default)
- that was used to generate project wide file dependencies (some
circulars were manually fixed)
- then that graph underwent a transitive reduction and the result was
written to all `jak1` source files.
The next step will be making this actually produce and use a dependency
graph. Some of the reasons why I'm working on this:
- eliminates more `game.gp` boilerplate. This includes the `.gd` files
to some extent (`*-ag` files and `tpage` files will still need to be
handled) this is the point of the new `bundles` form. This should make
it even easier to add a new file into the source tree.
- a build order that is actually informed from something real and
compiler warnings that tell you when you are using something that won't
be available at build time.
- narrows the search space for doing LSP actions -- like searching for
references. Since it would be way too much work to store in the compiler
every location where every symbol/function/etc is used, I have to do
ad-hoc searches. By having a dependency graph i can significantly reduce
that search space.
- opens the doors for common shared code with a legitimate pattern.
Right now jak 2 shares code from the jak 1 folder. This is basically a
hack -- but by having an explicit require syntax, it would be possible
to reference arbitrary file paths, such as a `common` folder.
Some stats:
- Jak 1 has about 2500 edges between files, including transitives
- With transitives reduced at the source code level, each file seems to
have a modest amount of explicit requirements.
Known issues:
- Tracking the location for where `defmacro`s and virtual state
definitions were defined (and therefore the file) is still problematic.
Because those forms are in a macro environment, the reader does not
track them. I'm wondering if a workaround could be to search the
reader's text_db by not just the `goos::Object` but by the text
position. But for the purposes of finishing this work, I just statically
analyzed and searched the code with throwaway python code.
Removes trailing whitespace from goal_src files, eventually the
formatter will do this as well but it's not ready yet so this is a
decent interim solution.
A competent text editor will also do this / flag it for you.
Major change to how `deftype` shows up in our code:
- the decompiler will no longer emit the `offset-assert`,
`method-count-assert`, `size-assert` and `flag-assert` parameters. There
are extremely few cases where having this in the decompiled code is
helpful, as the types there come from `all-types` which already has
those parameters. This also doesn't break type consistency because:
- the asserts aren't compared.
- the first step of the test uses `all-types`, which has the asserts,
which will throw an error if they're bad.
- the decompiler won't emit the `heap-base` parameter unless necessary
now.
- the decompiler will try its hardest to turn a fixed-offset field into
an `overlay-at` field. It falls back to the old offset if all else
fails.
- `overlay-at` now supports field "dereferencing" to specify the offset
that's within a field that's a structure, e.g.:
```lisp
(deftype foobar (structure)
((vec vector :inline)
(flags int32 :overlay-at (-> vec w))
)
)
```
in this structure, the offset of `flags` will be 12 because that is the
final offset of `vec`'s `w` field within this structure.
- **removed ID from all method declarations.** IDs are only ever
automatically assigned now. Fixes#3068.
- added an `:overlay` parameter to method declarations, in order to
declare a new method that goes on top of a previously-defined method.
Syntax is `:overlay <method-name>`. Please do not ever use this.
- added `state-methods` list parameter. This lets you quickly specify a
list of states to be put in the method table. Same syntax as the
`states` list parameter. The decompiler will try to put as many states
in this as it can without messing with the method ID order.
Also changes `defmethod` to make the first type definition (before the
arguments) optional. The type can now be inferred from the first
argument. Fixes#3093.
---------
Co-authored-by: Hat Kid <6624576+Hat-Kid@users.noreply.github.com>
This change adds a few new features:
- Decompiler automatically knows the type of `find-parent-method` use in
jak 1 and jak2 when used in a method or virtual state handler.
- Decompiler inserts a call to `call-parent-method` or
`find-parent-state`
- Removed most casts related to these functions
There are still a few minor issues around this:
- There are still some casts needed when using `post` methods, as `post`
is just a `function`, and needs a cast to `(function none)` or similar.
It didn't seem easy to change the type of `post`, so I'm not going to
worry about it for this PR. It only shows up in like 3 places in jak 2.
(and 0 in jak 1)
- If "call the handler if it's not #f" logic should probably be another
macro.
Fixes#805
This renames the method object in `defmethod`s to `this` and adds
detection for the `set-time!` and `time-elapsed?` macros.
Definitely my biggest PR yet...
- state handlers that are not inlined lambdas have smarter type
checking, getting rid of 99.9% of the casts emitted (they were not
useful)
- art groups were not being properly linked to their "master" groups.
- `max` in `ja` in Jak 2 was not being detected.
Another huge PR...
Previously, `object` and `none` were both top-level types. This made
decompilation rather messy as they have no LCA and resulted in a lot of
variables coming out as type `none` which is very very wrong and
additionally there were plenty of casts to `object`. This changes it so
`none` becomes a child of `object` (it is still represented by
`NullType` which remains unusable in compilation).
This change makes `object` the sole top-level type, and the type that
can represent *any* GOAL object. I believe this matches the original
GOAL built-in type structure. A function that has a return type of
`object` can now return an integer or a `none` at the same time.
However, keep in mind that the return value of `(none)` is still
undefined, just as before. This also makes a cast to `object`
meaningless in 90% of the situations it showed up in (as every single
thing is already an `object`) and the decompiler will no longer emit
them. Casts to `none` are also reduced. Yay!
Additionally, state handlers also don't get the final `(none)` printed
out anymore. The return type of a state handler is completely
meaningless outside the event handler (which is return type `object`
anyway) so there are no limitations on what the last form needs to be. I
did this instead of making them return `object` to trick the decompiler
into not trying to output a variable to be used as a return value
(internally, in the decompiler they still have return type `none`, but
they have `object` elsewhere).
Fixes#1703Fixes#830Fixes#928
This PR adds detection of the `launch-particles` and `seconds-per-frame`
macros to the decompiler, removing a lot of bloat and hiding many
process register uses.
I also added `og:preserve-this` comments to as many manual patches and
comments as I could, which will soon be used in conjunction with CI to
hopefully catch any regressions in future big decomp update PRs.
I have some concerns about the `launch-particles` macro (more details in
`sparticle-launcher.gc`) , but thus far, I have not seen anything break
yet.
---------
Co-authored-by: water <awaterford111445@gmail.com>
Fixes#2167
Reduces test flakiness if ran on multiple threads and gets rid of a few
hundred files from the source tree
I believe this also makes #1434 irrelevant and it can be closed.
---------
Co-authored-by: ManDude <7569514+ManDude@users.noreply.github.com>
This should fix a crash and animation bug in snowy. The way to trigger
the bug:
- go on ice
- move forward slowly
- stop moving forward. Reach zero speed when the frame number isn't
between 30 and 35 of the ice walking animation
- Due to this bug, the animation gets stuck at frame 60
- Take damage (due to normal lurker or ice lurker)
- Sidekick eye animation crashes because a frame number is NaN.
```lisp
(defenum collide-action
:type uint32
:bitfield #t
(solid 0) ;; used for solid things
(rider-plat-sticky 1) ;; used for platforms in rider/platform interactions
(rider-target 2) ;; used for target in rider/platform interactions
(edgegrab-active 3) ;; set/cleared when entering/exiting edgegrab states
(rider-plat 4) ;; used for platforms in rider/platform interactions
(unused 5) ;; totally unused?
(edgegrab-possible 6) ;; used when edge grab checks should be done
(edgegrab-cam 7) ;; set/cleared when entering/exiting edgegrab states
(swingpole-active 8) ;; set/cleared when entering/exiting swingpole states
(racer 9) ;; set/cleared when entering/exiting racer states
(attackable 10) ;; used for something to do with attacking/damaging
(attackable-unused 11) ;; seems to relate to attacking - set in several places but never tested for?
(snowball 12) ;; set/cleared when entering/exiting snowball states
(tube 13) ;; set/cleared when entering/exiting tube states
(flut 14) ;; set/cleared when entering/exiting flutflut states
(racer-grounded 15) ;; set/cleared when entering/exiting certain racer states w/ extra conditions
(racer-unused 16) ;; seems to relate to racer - never set, only cleared in one place?
)
```
Gives proper names to almost every color. It is very apparent that some
colors are context-sensitive/made for a specific purpose, so those
colors were named after that purpose instead of a generic color name.
Updates the decompiler for the new format and there's new macros. This
new format should be easier to read/parse.
Also rewrote `sp-init-fields!` (both jak 1 and 2) from assembly to GOAL.
Hopefully I did not miss any regressions in Jak 1/2 while updating the
files, it's a lot.
Cleans up every `dummy-*` and `TODO-RENAME-*` method up with either
proper names or by renaming them to `[type-name]-method-[method-id]`
similar to Jak 2's `all-types`.
Also fixes the bad format string in `collide-cache` and adds the event
handler hack to Jak 1.
The game boots and runs fine, but I might have missed a PAL patch or
other manual patches here and there, please double-check if possible.
There are *a lot* of file changes and while I have carefully gone
through every gsrc change to fix up manual patches, there might still be
spots that I missed.
I think this got removed awhile back, not sure if it was intentional,
but thought it was nice to have. Feel free to close if it's not
supported for a good reason 👍
And everything else needed for them!
A couple functions are bad currently.
- fixes#1929 - untested on linux
- fixes#1924 - now you need to type `,` before a lambda you want to put
in a pair.
- fix debugger symbol table in jak 2
- made the decompiler output `(meters 2)` instead of `(meters 2.0)`
- fixed a bug with the bitfield enum special -1 case
- made bad game text decomp not exit the decompiler
- added `editable-player` and `script`
Hat Kid
