AlmostDecentGames

Weekly Updates on the Progress of the Game I'm making!

[almostdecentgames-blog]

What’s been going on?

I’ve neglected to update on all the changes going on with this project mainly due to stressing over getting some separate work done for my Games AI module, and the constant bugs Steam VR loves to throw at me. Anyway onto the progress!

The Good: I have a full level design in VR now, complete with a properly rendered Sniper Rifle gun puzzle! Getting the gun to import from cinema4D was hell on earth though: At first neither the textures would import to unity, nor would the shapes as they’d instead be jagged, spiky versions of their original form. After looking into the problem I found the solution to be rolling the fbx file format back to 6.1, and setting the location to be at the root of the original cinema4D file. I also managed to find a unity package for a cityscape that adds buildings in the distance for the horizon, though the horizon aspect doesn’t seem to work in VR. I also made a building in cinema4D with window spaces that I can put assassination targets in for the player to shoot, which I then melded with one of the buildings in the cityscape to keep with the surroundings in the scene.  I also added the diegetic UI I planned on doing to the scene, and have it toggle on different instructions based on if the player has assembled the gun puzzle yet. The Bad: Time sunk into trying to fix the snapping is gone and I wish I could have it back. The latest Steam VR is a badly documented mess, which meant I found no solutions online to my problems and the posts I made sank into the void with no answer. The time lost to pursuing answers on this and testing my own ideas could have been used on other areas of the project, sigh I also still need to finish implementing raycasting for the shooting mechanic, though I don’t know if I will have the time. The Ugly: Snapping. It sort of works, but it’s not what I wanted and it doesn’t allow for the freedom of play I desired in this game. In order for what I have currently to function the player must not attempt to grab an object that has been parented to another, and instead move only the parent object out of the pair. This also means you must snap objects with only one controller, as if you hold both pieces you are trying to snap Steam VR’s controller logic stops the snapping from happening and subsequently breaks it permanently.  So the result is a less than ideal solution, but at least it’s something I can use to an extent. 

[almostdecentgames-blog]

Overdue Update

While looking into my snapping problem I realised I was sinking far too much time into it at the moment and have widened my priorities to reading papers on puzzle design and on pacing challenges, as the design of the puzzles is as important as anything else in my project.  I’ve also been modelling gun parts for the game in between trying things out for the snapping problem, and have finished the first gun puzzle’s parts entirely. The final step for this is the texturing!

In reference to the everlasting snapping problem I believe I might have a better solution than detecting what objects are in what hands; by instead having the objects snap when their invisible ‘snap point’ colliders enter one another. The next step for me now is testing this!

[almostdecentgames-blog]

Possible fix for adaptability in snapping script

After taking a while to think about how to solve my previous problem of how to solve the snapping script not being flexible for more than 2 specific targets, and that it initially involved me dragging objects into its public variable slots, I think I have got a solution.

I think if I have a script on my objects that detects when they are grasped by one of the 2 controllers, and then changes the variables in the snapping script to themselves depending on what hand they’re in, it’ll work. Fingers crossed!

[almostdecentgames-blog]

With progress comes more obstacles...

I’m now tasked with getting the script to work properly in vr with every object I pick up, but since the test snap script was designed to be put on 2 specific objects it means I need to do some remodelling.

[almostdecentgames-blog]

A Solution!

After consulting a friend we came to the same conclusion, that using a timer to stop the objects from resnapping for a certain amount of time after detaching was the best bet.

And it turned out to work! I added a float to store the last time the objects decoupled, and only allowed snapping to happen if some time had passed since they had last decoupled

[almostdecentgames-blog]

Development Update!

I’ve run into both solutions and problems with the developing the snapping mechanic for my game.  While I now have rudimentary snapping working in a non-vr test scene, the object refuses to decouple from the attached position unless I move it away at a high speed.  At the moment I’ve no clue what is causing this but I am looking into it!

Video of the bug happening

[almostdecentgames-blog]

2D Prototype Feedback

After testing my 2D prototype on 6+ people I’ve found that the most common mistake when assembling the gun was orientation of the pieces. I believe this problem will be less prevalent when the puzzle is translated to 3D, as the sections that attach will be protruding outward, but I also plan to make the attachable areas glow in some way to make sure this problem doesn’t transfer. 

I think it might be an idea to highlight the areas that attach in some way when making my next 2D paper prototype; to see if it helps at this stage.

[almostdecentgames-blog]

Week 2: First Prototype!

I have finished a 2D prototype of the sniper gun puzzle to test on my peers as soon as I can!  The design of the prototype is using the TF2 Sniper’s sniper rifle as that is what I had in mind for the gun’s look; simple in terms of design but yet recognisable as what it is.  I am also working on cutting out a 3D version of the puzzle to see if that has any bearing on the difficulty of the assembly when tested on people. But due to it taking longer to make the 3D prototypes, I will first prioritise making all my prototype puzzle designs 2D.

[almostdecentgames-blog]

Update 2: Environment Established!

After some trouble with the latest version of Unity (2018.3) producing 999+ errors with Steam VR for the first environment I set up, I rolled back to an older version of Unity as suggested online (2018.2) and reestablished the basic environment and interactable objects. The result is I can now teleport within the space I’ve defined, pick-up and move objects with my controllers and throw them with the velocity I set in the corresponding component. 

The next step for me now is to figure out connecting objects in VR, so I can create the framework of the first puzzles. I also want to cut out some paper prototypes of my planned puzzle designs to test on my peers; to see how intuitive they are. 

[almostdecentgames-blog]

Update 1

So far this week I have been mocking up puzzle designs for the levels, as the game is centred around assembling your weapon in time to shoot I must make the task interesting.  I’ve decided that the designs should not be all ‘sniper rifle-esk’ to provide more avenues of creativity, I will try to add rocket launcher designs and pistols both of comic proportion/use.  So far I have the simple sniper design of 3 parts, with an alternate design of 5 parts. There is also the ‘stinger’ rocket launcher which I chose as it has lock on capabilities in most games so the player doesn’t need to worry about aiming the projectile, this one consists of 4 parts.  A mini-gun design of 11 parts that will spray over a large area (again to eliminate aiming problems, as I don’t want that to be the focus of the game), a scoped revolver design of 5 parts, and finally a flare gun comprised of 4 parts.  The flare gun could make the building explode, as though a orbital strike had happened, but depending on scaling and time it may just kill with the flare itself.

[almostdecentgames-blog]

The Difficult Gun

I am planning on making a VR game in the genre of creative physical puzzles and assassination. The problem I’m tackling with my game is that assassination or murder games don’t deal with the intricacies of the setup/assembly of the equipment, in my game I plan to make this the focal point of the gameplay! However this doesn’t mean I want the game to feel like a chore, I plan to make it comical rather than a serious simulator. I think that with this combination of a problem that hasn’t been approached and the comical take on it, will make for an enjoyable and unique game.

My game ‘The Difficult Gun’ is about assembling comically complex guns on a rooftop in VR as a hitman within the time limit so you can take out your target, the weapons will start off simple and realistic but progressively contain more parts and consist of more wacky designs. The art style will be simple and cartoony as the game isn’t going for a serious setting, I would hope to come close to TF2’s art style by the end of development.

The experience I’m hoping to make is a humorous yet challenging set of levels, that get harder the further you progress. The main challenge will be the assembly of the gun, and that is what I want to focus on, not worrying too much about making the shot too difficult. The guns themselves may not all be of a ‘sniper rifle’ design, as I’m going for wackiness they may range from rocket launchers to crossbows. I plan on making at least 4 levels, but hope to make many more than that as 4 is really just about a minimal viable product.

[almostdecentgames-blog]

Initial Ideas

‘The Difficult Gun’: VR puzzle/shooter game, you start on top of a roof with a briefcase which contains gun parts. You have to assemble the gun in the time limit and then shoot the target, but as the levels go on the gun becomes more complex/silly to assemble.

Side-Peek: 2 person split-screen FPS where you neither player has a player model, the only way to find where they are is to look at their screen and discern their location from what they are looking at.

Spy-Off: 4 person multiplayer game of spies with various traps and objectives around the map, by default you have no weapon and must pickup the items around the map to have a chance at fighting off your friends. There’s no lives system in the game, you instead win by completing all the puzzles (security rooms) around the map before anyone else.

Educational game set in a dungeon crawler style; before starting you input a list of questions and answers that will serve as challenges along the way. The game will then be a varied experience with randomised encounters that will need you to remember the answers to your questions in order to complete actions (e.g. dodging, clearing a fall, unlocking a chest).

Fighting game with a deck of cards instead of regular controller based combos. The game style will most likely be side-on view akin to Street Fighter, and will be turn based so as to focus more on deck building and comboing cards rather than throwing them out as quickly as possible.