#infernobomb

So I have finished the very short “Story Mode” for Party Animal. 

Trick-shotting it up in the story mode

To be more specific, I finished designing and creating the maps, drawing the backgrounds, and implementing the game-play features/bug fixes to allow the Story Mode to work properly. If this is the first post you’re reading, this “Story Mode” is not intended to be the main aspect of the game. In fact, the only reason I created it was so that if you were unable to experience the multiplayer properly, you would still have something to enjoy. It is a very short experience, averaging around 50-60 minutes long, but I am still happy with how it turned out.

The big reason that this is so good is that now I can finally turn my attention fully onto the multiplayer of Party Animal, which is the main, and hopefully the most enjoyable aspect of the game. Of course this will not be some massive ground-up implementation of multiplayer, it is actually fairly close to being finalised, but there are still quite a few features I think are necessary to include in order for the finished game to feel like a good product;

Multiplayer with the ‘Prop Hunt’ gamemode

Things such as Steam API integration and a larger variety of gamemodes are left to do before I am happy to release the first version. That and bug fixes...lots and lots of bug fixes.

That’s about it, you can now feel updated,

Infernobomb

here’s a bonus picture, because I worked hard on drawing the map backgrounds

Hey guys,

As I stated in the last post, I have spent the majority of this week working on multiplayer. I almost completely reworked the previous system in order to reduce any lag caused when several people are on a server at a time. 

I’ll explain what I did as quickly as I can;

The previous multiplayer system would work by sending all of your player’s information to the server each step (ie. position in room, current action, health etc). The new system will only send this information when it changes. The reason why I hadn’t previously done this is that you run into the problem where a new person joins the server, and needs to receive all of the player’s information, not just what has recently changed. Implementing this system took quite a few days, and i’m not sure how effective it has been at reducing lag...

On the side of more interesting developments, I have added the small feature of being able to see other players walking around when you are both spectating;

He is ‘ghosting’ he he...

Nothing major, but it should make the experience more fun when you can hang out with other dead people.

While we are talking about small features, I implemented a “water height” system to rooms;

The dreaded water level

This feature is also not really a ‘huge’ deal, but it should allow some more interesting maps, and personally I really like how it looks.

Finally, I added in the option for rooms to have ‘foregrounds’. This allows you to design more detailed rooms, without having to create multiple objects for every minor detail;

This week I’m planning to hopefully fix AI objects in multiplayer, as they still do not sync up locations or actions properly.

Thanks for reading :)

Hi everyone,

This is a small post just to give you all a few free games to play!

Firstly, we recently made a short game prototype over a weekend in a small ‘game jam’ style. In it, you play as an organ surgeon who specialises more on the side of collecting the organs than in giving them:

His method of collection is very messy

Take a look at it, and let us know what you think of it!  https://incinious.itch.io/dr-transplanter 

Another announcement, is that we have also recently released a new mobile game for Android, called Splat Tennis (I mentioned it some time ago, if anyone remembers). It is a fast paced, arcade game, with several unlockables and incredibly accurate tennis physics*. 

Also take a look at this one, I tried my hand at making some music. Not sure how good any of it is though... https://play.google.com/store/apps/details?id=com.incinious.SplatTennis

I hope this helps reach the itch that only free games can satisfy, and that it doesn’t feel like I am bribing any of you...

Until next time

Hey all, this post is a bit late. Life has been generally busy this time of year, but it’s starting to ease up again, so I thought I should put out a short post, just to keep up my monthly posting streak...

Not a bunch has been worked on, but I can still show some things;

When developing the payload gamemode, I discovered the issue of trying to sync up the coordinates of AI objects, namely the payload itself. In order to improve the accuracy of the payload, I modified how AI objects are displayed, so that their positions are updated based on the nearest player, rather than just having the Host calculate their positions. This way, the AI object will always appear most accurately to the player who is more likely to be interacting with it.

Think of the payload as an AI object that runs away from the player

A less interesting part, which took me a while because it was so boring to work on, was implementing a full Options menu. The way I programmed it made it easy to also add access to it while in-game.

I’m proud of some of these GUI elements, they’re very smooth to interact with

Some other big-ish additions are;

- Gamemodes are now retrieved from the Host, so a player can no longer be have a different Mode to the other players

- Explosions are properly implemented, with damage / knockback falloff working

- Players can no longer get stuck inside each-other when solid. If they do end up getting stuck, they will stop detecting collisions with each-other until they are no longer near one another.

I feel like I say this every time, but it is so close to being done. I am aiming to have a few gameplay videos / trailers releasing soon, so look out for those.

I have realised that enough time has passed for another post to be in order, so buckle yourselves in because this post is definitely going to be super exciting, and not be another one of those boring technical posts...

To start off with, I have recently created a new map to be part of the initial release. This one has a focus around a “payload escort” game-mode, where each team has to move the “cart” from one teams side to the other. The cart is moved when someone stands next to it, and the game is won when a team pushes the cart into the other teams goal 3 times:

The setting is in the sky, for some reason

The reason for adding this map, and for adding a few other maps that I have planned, is firstly because these game-modes are fun to play, but secondly as a way of demonstrating the types of game-modes that can be created yourselves, using the in-game editors. And now that I’ve hooked you with an interesting gif, it’s time to talk about technical things...

Now, while I was testing the new map in multiplayer, I came across a variety of issues. Players were not able to shoot each other, people would be disconnected at the end of the round etc. I worked out that each of these bugs all stemmed from the same issue, and that was the method that client’s were being ID’d by wasn’t working properly. Let me elaborate...

When someone joins a game hosted on a server, that player, or “client”, needs to receive a unique ID number. This ID number is used to differentiate players from one another, and needs to be the same value on each players computer. The way these unique ID’s were generated previously was using a “stacking” method. When a new player joins, they receive the largest ID number not currently taken, and when a player disconnects, all other player’s ID’s update accordingly;

Player with ID 2 gets new ID 1

Now, this ended up causing a few issues. While it felt like the most efficient method to provide ID’s to players, problems occurred when step 3 was occurring at different times on different Client computer, for example if a client still has the ID 2, and sends it’s actions to the server, who has that client stored with an ID of 1, then those actions would not be performed.

In order to fix this, a new method of providing ID’s is now used, which follows a more basic approach. Instead of reshuffling the player ID’s when a client disconnects, the ID the player had instead remains available, so that the next Client that connects will be given the first available ID to use;

The new client takes the first available ID, which is 1

This new method means that Client ID’s are no longer varying throughout the game, making multiplayer interactions as a whole a lot more stable.

Another major change that was made was in regard to optimising the actions system. Each object can be given a list of actions that can be performed, based on test-cases. For example, a door can be opened when within range of the player, or a bomb can explode when a timer runs out. These actions are all saved in an objects file, and are loaded in as a string during the game when that object is created. Now this may sound pretty stupid, but previously these actions had to be decoded each time they were performed by the object. This means that every-time a player went near to the door, for example, it would have to go through each loaded action, and decode the ‘string’ that they were saved as in order to work out what kind of action it was.

Naturally, this was extremely inefficient, and caused noticeable lag when an object had a larger number of actions to perform. The change that I made was to optimise this action system, making actions to be decoded only once, when the object is created. These decoded actions can then be easily accessed later on, without having to go through the loaded string to get the saved values.

That should be the super technical changes out of the way, so here are a few “smaller” improvements I have made;

- A kill-feed/alert system has been implemented, detailing when someone has been killed, or when players join and leave a game;

That Old man really likes leaving

- The Point system in multiplayer is now purely server-side, meaning that it is a lot more reliable keeping track of these scores across all the connected players

- Support for controllers has started, with movement and basic aiming implemented;

Like I said previously, the game is getting very close to finishing, and I plan to have it done by the end of this year. 

Hopefully this post has been an interesting read, unless you hate technical stuff, in which case my apologies,

Hey all,

So that game we made a while back is alive and kicking once more. I have pushed out an update for Squidge on the google play store that includes more characters to unlock, and a High Scores table. Take a look at it if you haven’t before! 

https://bit.ly/2HhvuLS 

Oh, and I also put together a little promo video for it:

so professional...

And now back to your regularly scheduled program. I have managed to get my current development process for Party Animal down to a fine art. I work on a story level, and fix any bugs I find along the way. At the moment I am working on the obligatory underwater levels. Of course everyone hates them, but if I have put water into my game, than you are going to have to enjoy it!

wow, look at all those pretty corals!

Currently I am up to implementing the Giant fish boss you will have to run away from, but there are naturally so many bugs to get past to do with swimming, AI, and physics in general.

For example, one issue I resolved had to do with swimming specifically. I mentioned in a previous post that I implemented the feature of sticking to slopes when you walk down them. Now this came back to bite me, as it only worked on slopes underneath you, but when you swim upwards into slopes, you would get stuck. Adding the ability for the player and ai objects to slide along roof slopes is the kind of thing that no-one appreciates, but is annoying to implement:

Featuring scary fish monster

One other feature that I implemented was a smaller one, but I think it is one of the nicer looking ones. It is as simple as being able to choose where a player will carry an object from when they pick it up. But it adds so many options; You can carry weapons from the handles, hold a swinging lantern, carry a bottle by the neck etc.

Almost finished with the Story mode now, and then it will be on to some final multiplayer maps and features, and then hopefully I can look forward to releasing it!

Thanks for sticking with us all these years, it really means a lot!

Hey all,

I have finally been making some headway on the “Story Mode” maps, and I think the short adventure is over halfway done. I recently added in a cat-boss:

Smart kitty

My ‘development process’ while doing this has been to work out what I want to do, and implementing it if it’s not currently possible with the game. The idea is if I add enough variety of features, it will be useful for someone who may want to make their own “Story Mode”.

More often than not however, I end up discovering that features I thought worked don’t actually. The most recent example of this was room ‘warping’. One of the actions that objects can perform is to ‘warp’, or teleport, the player to a different level. This is quite a commonly required action in the Story mode, and it seemed to work fine with single player.

It was when I tried using it with multiplayer that things started acting weird:

No, this ground obviously can’t be stood on

The way loading up rooms works is that if there is someone in the room that the player is currently ‘warping’ to, the game will request that players version of the room. If there is noone in the room, it will instead ask the host to send the file version to them, and use that instead.

To begin with, the game was simply crashing when you attempted to join a room that the host was not part of. I fixed that by working out that the request for the room was never being sent properly, but then the real fun began. I ran into a variety of issues which meant I had to change how ai positions are sent, change how rooms that have been sent over multiplayer are loaded, and make client objects leave and rejoin different lists. To give an example of what it’s like fixing these issues, one of the bugs I found was that sometimes when a player shot another player, they would damage them twice instead of once;

notice the two damage numbers (4)

Now this took a good hour and a half to work out what was happening, and it ended up requiring two lines of code to fix. TWO!

So that should hopefully explain why this game is taking so long to finish. There’s so many features that it takes forever to work out what is causing a particular issue. It hasn’t just been bug fixing and story mode-ing. I did add a few other things;

- I made walk speed animations dependent on speeds ie. moving really slow means you look silly

- I gave AI and Client objects a different walking sound to the player, so it’s less confusing

- Added these sweet flips

Trick shot

But yes, mainly bug fixing and Story mode-ing. I’ll try to keep the next post shorter.

So my main focus for the past few weeks has been getting some headway on “Story” maps. Prepare yourself for another ‘technical’ post.

My past experience has shown me that it is best when releasing a game to have an option for a “solo” experience, even if the game is designed to be multiplayer. Because of this, I am trying to include a short but hopefully interesting ‘Story Mode’ in the final release, if nothing else but to demonstrate the capabilities of the map-making tools.

To make it easier to create these maps/levels, I spent some time improving the current map-editor. The first of these changes was to do with adding objects to a map. When placing objects onto a map, you can choose from a list of available objects found in the resources folder. Previously, this list was only a list of object names. While this works, it sometimes made it difficult to find the object you were looking for, especially if you forgot the name of the object:

Wow so much variety, I bet all those objects 100% useful

In order to improve this, the map editor now also imports the image for each object found in the resource folder, and then displays that image alongside it in the list.

The next change I made was to the way the room size was set. Previously the width and height of the map could be manually defined within the editor, however once objects had already been placed on the map, resizing it would only increase the far right and bottom sides of the map, without changing the location of the objects accordingly.

In order to make this process more efficient, the editor now allows resizing of the map by clicking and dragging it’s borders inside the editor. This drastically improves the efficiency of map making, as the user no longer has to guess what size the map should be, and instead can easily change it on the fly:

What a visually pleasing map editor...

I have also recorded a much too long video of me using the new map editor to create a level for the “Story Mode”. Hopefully one of you will find it interesting to watch:

Aside from these changes, I also worked on some improvements to gameplay such as improving how npc’s weapons work, making climbing ladders better, and adding ‘random item’ objects for maps, which can allow for a map to be different every time you play it.

Thanks for reading,