I make games http://www.farmergnome.com
About farmergnome (twitter: @farmergnome)
Archive for the ‘LD #25’ Category
So this is my second Ludum Dare, and I think for the most part it went a lot better than last time (in case you are wondering – my last entry… http://www.ludumdare.com/compo/ludum-dare-24/?action=preview&uid=10234 ) mostly down to the technical details and sleeping better (or worse?)
So this game is called HAPPY LITTLE MURDER FRIENDS ( http://www.ludumdare.com/compo/ludum-dare-25/?action=preview&uid=10234 ). It is about serial killers going on a homicidal rampage. You control one of 5 killers and each town you enter you must kill your targets that you have marked out – all while avoiding the law and killing too many people who aren’t on your hit list.
What went Right…
Box2d was quite possibly the best decision I made for this project, a lot of my time on Soul Jar was spent fixing weird platforming glitches (and they still weren’t entirely fixed) – it saved me maybe 6-8 hours of messing around and headaches. Totally worth it. I think MMF2 was an excellent choice as far as tool-set also, I can’t seem to move as quick with Unity (Though I am still learning it!) My familiarity with it definitely speed things along.
Lack of sleep due to being well rested before was probably the second biggest factor in adding the content required to make this game happen, I got tired on day two but I tricked my body into waking up again (tons of coke/coffee right before a power nap did the trick) – I tried this when developing Soul Jar but I ordered my tasks wrong, I was tired when I was coding and fresh doing the art and music. I focused on nailing down the programming elements while I was fresh this time, and as soon as my mind wandered I fell back onto doing the art/music- productivity was definitely better.
I knew I would need TONS of variation in a game like this, not only in the backgrounds, but the characters also. I wrote a really basic R,G,B randomizer and did a variation version of it to be used in contrast, then I applied this to everything without mercy. This saved a lot of time rather than painting variations by hand, as they were done in Soul Jar, and looks quite good also!
Basic Art Style:
Everything in Murder Friends is extremely low res, but more importantly than that I focused on broad colors rather than details, this allowed me to block out entire landscapes in a matter of minutes rather than hours which was something that drastically slowed me down last time. Probably the only thing I put some focus into as far as pixeling was the UI and main characters – but that was kind of unavoidable as they are onscreen constantly and need to stand up.
All the effects are done with a similar method to the color swaps, just tinting a certain color channel rather than animating it by hand, definitely saved some time on this one. I don’t think I will be hand pixeling fades glows much anymore, thanks to MMF2′s rather powerful set of effects with the hardware acceleration turned on.
What went Wrong…
I spent a fair amount of effort designing features that never made the cut, and while it doesn’t chip into my design time (I designed the specifics while I mocked up art/music – so it really didn’t cut into my time too much), I think I have a habit of not spending enough time sitting down with the design which leads to a lot of cut features. Most notable that got cut was the ability to drive cars and hide in houses (I ended up adding the later, though much more ghetto than you may think)
I wasted a solid hour or two this entry having to pixel over most of the fonts in-game, was absolutely a giant waste of time and tedious to boot, I kind of left this till last as it wasn’t core to the game working, but next time I think I will pick a good pixel font that scales well, rather than using Impact font and touching it up.
This was the constant weak spot of the code, probably broke more than anything else across the 48 hours, I think if I redid this I would find a way to cheaply animate the characters without having to have separate animations for each character, not only does it take time to pixel in, but managing animations is extremely expensive time wise.
The win screen was in concept going to be a newspaper that had the recent massacre on the front page, and act as a win screen/score screen of sorts. But I had to throw it out after day one as I hadn’t even done the main UI yet, let alone the win screen. The win screen that I went with was just a almost direct copy of the credits screen, just with the items swapped.
Well that about covers it, thanks for reading. If you would like to play it head over here : http://www.ludumdare.com/compo/ludum-dare-25/?action=preview&uid=10234 and say hi!