 |
| Monsters Loot Swag; Final Playable Character |
As the year 2024 comes to completion, so has the list of playable characters for Monsters Loot Swag. Schiavona is the 8th playable cosplaying catgirl character and fourth and final character who has to be unlocked via achievements. She is a close range, melee based character, with fairly average physical statistics, but is very heavy on inflicting damage.
Cue video of her in action:
All that remains for work to complete the development of Monsters Loot Swag, is now the final level with boss character, and few art related pieces such as an intro containing plot, as if 8 playable cosplaying catgirls battling 8 waifu level bosses needs such flippery as plot!
 |
| Airship Dragoon has been on Steam for 10 years |
Airship Dragoon got two updates at the beginning of December, the first was due to a new Steamworks SDK which had broken some of the 10 year old code for Steam integration, and the second was because, whilst testing the first update on Steam to make sure it actually worked ... I found a bug from 12 years ago ...
Thankfully, 12 years ago me ... who may still have been in his late 30s ... had left a code comment incase something happened that was not supposed to happen - and during this test of the new Steamworks code ... it happened.
The ancient code comment made it relatively simple to track the "something happened that was not supposed to happen" down to a specific function were pathfinding was sent if it completely failed - something so rare I had never had it happen, not anyone report it in the last decade, but I had not added an escape of any kind to this failure state, and so quickly scripted an emergency "move elsewere" command, so not to get stuck.
Rounding out the year work-wise, I finally filed my tax return with His Majesty's Robbers And Crooks (HMRC), something which the gubbermint has gone out of their way to make ever increasingly difficult. Their oh-so-helpful system active attempts to stop me from paying them money and every year I send feedback telling them that their crappy website has been broken for the last 7+ years and tell them how to fix it, every single year.
 | |||
I ran out of characters typing into the complaint box
|
As it's Decemeber, and the time of eat, drink and get indegestion, I bought the biggest steak at a reasonable price that I could get.
 |
| Hello Vitamin B12! |
 |
| Suitable Protective Clothing For Roasting Such A Beast |
 |
| Went For Medium-Rare For Fear Of Some Folks Freaking Out If It Was Still "Bloody As Hell" |
As Monsters Loot Swag comes to the end of it's development cycle, I naturally have one eye on the next project. Previously I had posted some of my test work on a third-person melee combat system. As my previous two projects have taken so long, I had been wondering about how best to make a short form game that could be expanded on if it received due interest from players.
I noticed that there had been quite a lot of complaints on the new Stalker game, about how it's open world design had broken it's previous a-life simulation, and I thought that maybe going down that route could be interesting - both in gameplay but also in coding the concept.
So, something along the lines of a 4km square map (not too big), with bands of enemy warriors from different areas or periods (Saxon, Norse, Magyar, Venetian, Chivalric, Landsknecht, etc) wandering about, picking fights, fleeing greater enemy forces, with a smattering of random loot and treasure chests, maybe some wild life plodding about.
I had a brief test to see the ballpark range of having lots of Ai players visible, having them invisible, not rendering but still having to update physics each engine tick. This was done in debug mode whilst running an IDE so naturally the framerate was terrible, but it did quickly prove what I had suspected, which is rendering 200 skinned meshes is a lot more overhead than just processing the physics for 200 Ai players.
In a player build test without the overhead of debugging, the framerate was 500fps empty, 80fps with 200 characters rendering, and 350fps with 200 characters not rendering but still updating their physics. Not that I would ever consider having 200 Ai players battling it out right in front of the player's camera.
It will be interesting to see what sort of load I put on my processor and gpu if I remove the Ai players when they are nowhere near the player and turn them into scriptable objects who would then approximate their actions using RNG as they move around the map, out of the player/client camera fustrum and view range.
Anyhow, that is all for a later date. I hope you have had a Merry Christmas, and here's to Space Year 2025!
Happy New Year!
