About Thermite


Ludum Dare 31
Ludum Dare 27

Thermite's Trophies

Thermite's Archive

‘SoundWaves’ – LD31 Jam – Post Mortem

Posted by
Wednesday, December 10th, 2014 2:21 pm

So, our second jam entry is posted and we’ve regained enough sleep hours to interact with the world again :p

Laz0rs! And balls of death!

There’s a lot of things we didn’t have time to implement, and the game balance sometimes works and sometimes doesn’t, depending on which speaker(s) you shut down and when.

Here’s a short list that looking back we would’ve put in given more time:

– More levels / track options (we were a bit constrained by the file sizes on the tracks)
– Music could have been specifically written to fit the game better OR Pre-analyse tracks to enable music imports (that would take us a while and would be hell to balance)
– Smooth screen transitions
– Powerups (the ship we made actually had 5 guns, but it was a bit too easy so we cut it back)
– Enemy design
– Background
– Smart bombs
– Different game modes (Survival, Time Attack etc.)
– High Score / Leaderboards
– Better UI
– Persisting Upgrades
– More varied particle effects
– Shields

So, in the last LD post-mortem we posted, we specifically said we’d use tools we were familiar with. Well, we didn’t. We decided to switch up to Unity to see what all the fuss was about. And it’s pretty awesome, if not a bit contradictory to our usual programming styles (public variables EVERYWHERE!?).

Next time we will be using Unity again, and now we’ve got used to it hopefully come up with something more polished, or something more adventurous that we can struggle to get working :p

We had fun making it, and fun playing it, hopefully you will too, check it out:


We’re In!

Posted by
Friday, December 5th, 2014 10:20 am

The gruesome twosome that is Thermite Games will be joining in the jam for LD31 :]

We last took part for LD27 and this time we’ll be using Unity, probably SFXR and GraphicsGale and/or Blender.

Looking forward to the lack of sleep and hopefully a half decent game!

Mechanical Madness LD27 Jam Post-Mortem

Posted by
Thursday, August 29th, 2013 2:47 pm

Title Screen

Well our first LD is over, we’ve made our first ever game together as a team, and we’ve got the obligatory platformer out of our systems :)

Technical stuff:  Made with Haxe 3.0, Flixel, Flashdevelop, GraphicsGale, sfxr, Autotracker, Excel.

The level that no one can complete!

The level that no one can complete!

What Went Right

  • We finished a game in 72 hours without killing each other!
  • The toolchain worked really well, Haxe, Flixel & FlashDevelop felt very familiar despite not using any of them before the warmup.
  • Use of Microsoft Excel (hardcore mode!) as a level editor.  There are obviously some very good tile map editors out there, but learning them would have taken way more time than setting up a spreadsheet with some conditional cell colouring.
  • Bringing in a third person for brainstorming and powerup graphics on the first day.  Thanks Graham!
  • We anticipated that the main source of bugs would be unexpected interactions between multiple powerups, and set time aside to get this working properly.
  • Almost all of the powerups we came up with initially made it into the final game.
  • Staging the powerups and level progression to ease the player into mechanics without explicitly telling them. Combining different powerups gave us a clear sight of what obstacles a level needed to have, and a convenient “to-do” list for the 20 levels we built.
  • Targeting web – easier for people to play off the bat, rather than having to install or download or compile, it just works on most platforms.  Having never built a Flash game before, this was surprisingly painless.
  • The platformer controls and the level design we thought went well, it felt polished and enjoyable to play, even when getting squished every 5 seconds 😀


Cryptic instructions hidden in the levels

Cryptic instructions hidden in the levels

What Went Wrong

  • Difficulty with scaling the character & collisions, as well as the timestep changes interacting poorly with the collision logic in Flixel.  Getting it running at 60 resolved most of this, but there are still spots where a random bit of wall will just make you explode.
  • Lack of experience with the IDE & HaxeFlixel meant the initial setup wasted about an hour.
  • Music was a bit of an afterthought, we tried a few different packages to create the music eventually settling with Autotracker.py.
  • Collisions. The collision logic wasn’t quite doing what we expected it to, and we wasted time re-implementing certain collision features that were already present in Flixel.
  • We had smooth interpolation for the scaling of the character, but had to take it out as the player kept getting stuck in walls during the scale change. In the end we had to bodge in level-specific fixes.
  • Not understanding transparency in GraphicsGale – a lot of time was spent sucking the backgrounds out of sprites using Photoshop.
  • While we’re pleased with making a platformer that feels nice to play, the genre is obviously well-worn as a Ludum Dare standard.  We defaulted to this because we knew we could get the game finished in the 72 hours, but it’s kind of old hat to people who’ve been doing LD for a while.


What We’ll Do Better Next Time

  • Familiarise ourselves more with the tools, as well as deciding in advance what software packages we are going to use rather than flailing around!
  • Artwork – practice creating artwork for next time, it had been a while since Bob had done any serious artwork and the simple 5 minute sprites that we knocked out were OK, but could have been better. Paul plans to learn some pixel art techniques too!
  • We need to have alternatives for different game types, HaxeFlixel was good for rapidly building a 2D game, but it would have been nice to have the option of 3D.
  • Work out how to use Flixel properly, rather than having to hack bits of code together to bend it to our will.
  • Get together some flexible game ideas that we can adapt to the theme, instead of just defaulting to a platformer.



Please play our game, and let us know what you think!

Watch our timelapse video!

Play these awesome games that we’ve tried over the last few days!

[cache: storing page]