Raj's WIP Thread

[Raj]

Hello gamers,


I'm going to be keeping notes here on all my work, and its completion status. Keep in mind I have a job, and a few side projects, and I'm a gamer, so these things may come along slowly at times and very quickly at others.

If anyone has development experience and would like to help with small subproblems, let me know.

So here's a big list. I will periodically update this list as things get done. There's a few features I'm working on which aren't included here, either because adding them would be a chore relative to the substance of the changes, and also because there are things which simply can't be known until the day they're released because of the merchers.

Color codes:

Spoiler: show

Feature color codes:

• In progress
• Beta
• Submitted for review

Subtask codes:

• Todo
• In progress
• Done with minor todos, or todos dependent on other subtasks
• Done

Combat Achievements

Tombs of Amascut

Spoiler: show

Architecture/framework:

• Instance abstraction (thank Mike)
• Path abstraction
• Action handling backend improvement: giving easy access to arbitrary object state for object/NPC click processing
• Invocations/scaling
• Party system

Puzzles:

I have deliberately started with the most difficult problems, and I'm quite upset with myself for it.

Het:

• Light beam/mirror reflection
• Light beam collision
• Puzzle layout generation
• Seal behavior
• Dark orb things
• Pickaxe storage

Crondis:

• Spikes
• Poison
• Water mechanics (waterfalls/tree)
• Crocodile NPC

Scabaras:

• Puzzle 1
• Puzzle 2
• Puzzle 3
• Puzzle 4
• Puzzle 5

Apmeken:

• Baboon NPCs
• Waves
• Repair/DD mechanics - Someone with experience with team ToA, please message me. I only know the solo mechanics for this room.

Boss fights:
All (this section is itself a todo)

[Raj]

Made big progress today on the light puzzle generation. Here's the closest anyone will get to a rendering of it until tomorrow (minimum) when I finish and tweak it.

Spoiler: show

Screenshot 2023-03-14 090020.png

Spoiler: show

Screenshot 2023-03-14 085959.png

Spoiler: show

Screenshot 2023-03-14 085332.png

Basically, the solution was to generate a guaranteed possible light beam path (indicated by L), and then place barriers (indicated by B) around it in a way that they don't block the guaranteed solution. The block of X's is the unwalkable area in the middle of the room reserved for the Caster and Shielded Statues, and the Protected/Weakened Seal.

Still need to add breakable barriers and random mirrors, but it's something.

[Raj]

Light beam collision
Puzzle layout generation

[Raj]

Hello all, so I am indeed making progress on the Tombs, but slowly. Past few days I’m mostly grinding my own project(s), since I’m currently jobless as of a few days ago and I now have the ability to do that without worrying about intellectual property and conflicts of interest and non-competes and whatnot.

I found out the animations in our cache for Akkha and co. (his shadows) are just broken. The data for the animation frames simply isn’t there. So I’m needing some help from Mike on that. But I’m still progressing through the puzzles for now. The last thing I did was set up mining for Het’s Seal, triggered room finish on its death, and set up the events that need to happen when the room is finished (for raid progression and whatnot).

The biggest time sink is that I’m rewriting a bunch of old systems so that they’re less bug prone and have robust and intuitive APIs. Like I had to make heavy modifications to the World Object system to allow for moving the mirrors with the right click “push” option, updating the position both on the server and on player clients by just calling object.setX(x). I also needed the java objects for world objects to be a bit more mutable so I could just keep a list of puzzle objects without needing to modify that list and add new objects every time a player moves a mirror or breaks a barrier. Previously these things would make a new java object with its own reference. I believe I fixed a bug where objects which would disappear after some time would be unwalkable, despite being invisible, in the process.

Essentially, things will substantially increase in speed as I go along due to having progressively better tools to use. However, initially writing those tools is going to be a time sink.

[Raj]

So ToA progress has ground to a halt. If anyone has info about loading the new OSRS animation system on a 317 client, send it my way. For now, I feel the time investment of the reverse engineering I'll have to do is simply not worth it.

While investigating that, I got a bit sidetracked and partially integrated our client with RuneLite. Most of the things needed for the GPU, stretched mode, and HD plugins are at least partially hooked into our client. I am running into an issue related to the way we do rendering. Specifically, RuneLite and OSRS seem to share an AWT Canvas object, whereas we draw onto the Graphics object provided by a JFrame. I think this is somewhat solvable with very little work, but it's unfortunately not as simple as drawing onto the canvas.getGraphics() Graphics object, for reasons I'm not quite sure of yet. AWT has never really been my thing. But that is a thing that may at some point happen. I'm not sure yet of what I need to do to get it working, so putting it in the actual thread seems wrong, but I'm sure some of you find this interesting.

In terms of what I'm actually working on, I can't really say, but big things soon™

[Raj]

If anyone is wondering, I am currently cranking out the rewards for the Combat Achievements. There's about 200 combat achievments already implemented. The rewards are going to be a bit of work though.

After that, I will be returning to TOA very likely. Because the animations for the TOA bosses simply don't work without an update to the way we convert the OSRS cache to the 317 format, the planned solution is to replace the TOA NPCs with equivalent counterparts. So, instead of Akkha having Akkha's model, he will still have all the same mechanics, but his model will likely be a recolored Vet'ion. At least that way we can get the content out, since most people probably don't care much for the lore anyway, and when we have a solution for the new animation system, we can restore the original models and update the animations used.