Also thanks for the feedback, I guess. Now I have reason to believe that this probably is an issue with all of my similar projects, which I couldn't have noticed, due to a complete lack of someone telling me.
Recent community posts
Yeah, I was lazy and used the standard-exporter. (deadline was 5AM around here, and honestly - I threw all the modules I had built together, and disabled functions until it stopped crashing).
That error is curious too- 3.5 shouldn't even be a dependency.
I'll package it properly, once I'm at my machine again.
That's a good question. I didn't remember I joined this, until I got the announcement, so I'm caught a bit off-guard.
It's pretty useful to have a list of Ideas you want to try, but haven't got around to yet - regarding the BGE and Blender as a whole, mine involves:
- implement a game-of-life style simulation, and make use of it in some way
- improve pathfinding solutions
- figure out blender's GLSL pipeline, and make use of it in some way
- build a proper IK system, along with dynamically figuring out hotspots
- use Blender as a vector-animation tool for 2D
- find a proper use for compound collision meshes that's not Katamari Damacy.
- compile game-functions via Cython for performance reasons
- figure out a proper way of packaging Blenderplayer and it's dependencies, ideally in a multi-platform kind of way
- find an efficient way to re-map controls on-demand
- utilize a hardware-camera (or video-streaming in general) in the BGE in some useful way.
- dynamically generate and modulate sound according to gamestates
- use sound to drive animation
- build efficient lightmaps, for renderbaking without having to ditch realtime shadows
- multilayered 360 videos could be really impressive for a lot of things
- something using sockets for the achievement of having used sockets
- a clever way of utilising that HTML-parser, I've written
- write a dictionary-parser, and use it… ideally not with a chatbot
- make use of that renderbuffer, because it's there and nobody seems to know about it.
- find an entertaining use for a virtual-filesystem within a game
- make use of some economy models to simulate things
I guess, my entry will involve two or more things from that list, but atm, I have no Idea which make sense. It could be fun picking some at random, tho.
well, that's why the current snappy/flatpack discussion is going on. Because someone finally realized that increasing platform fragmentation causes issues.
Of course you can always solve it by shipping the libraries you need… if you do it properly, that is.
No problem. I'm almost done with something that could pass as a game, but realistically, I still have enough on my plate to warrant an entire second entry. ;)
It's really just a question of how "fluid" you are with the tools you use.
Edit: Also, if you're the Alan Pope, I think you are - then maybe I should ask you for advice. ;)
I've just seen a few who are either trolling (no download, but a price tag, offensive thumbnails), or sport far earlier creation dates.
But some of them disappearing hints at active moderation.
It just struck me as unusual for itch.
Actually, submitting the project beforehand is probably a good Idea. What if your upload 30 minutes before the deadline is slow? I had myself almost not making it last time, because of that.
Hmm, there seem to be some shady entries around, tho.
Actually, the API is very well documented, but due to the prevalent tutorial-culture, this is often ignored. Also people tend to claim things are "not suported" because they didn't find youtube-tutorials for them. (like realtime IK for example).
Another big problem is that people, attracted by Blender tend to see logic-bricks and fall prey to the Dunning-Kruger effect (and some features are only accessible via the API). I'd claim, it's very much as capable as other engines (I'd consider the editors more powerful, tho), but - very much like using Blender as a video-editor - there are a lot of convenience features missing, that one has to implement first. Of course, if you're comfortable with reading manuals and working with APIs, all that could be considered a bonus to the already decently integrated pipeline. - You could in theory do everything but music without ever leaving blender. (and I'm sure people more inventive than me could use it as a sequencer as well ;) )
Edit: Heck, if you're comfortable with using python's OS library as a bash replacement, you could probably replace a whole desktop environment and most of it's software (save a webbrowser) with just blender. :D
I'm making a "newtonic variant of dwarf fortress", and I am sticking to Blender's game engine. I plan on using the Cython compiler to get my extra-libraries in a decent format, but that's on the far end of the list. (Also my Editor of choice is gedit, and I'm working on a DVORAK vortex pok3r… let's not start a holy war about speed vs. extra keystrokes, ok?)
Aside from that… I've made my concept art in GIMP, and depending on if I get far enough to do actual textures / sprites (AO ftw!), I might enlist Krita or Toonboom.
As for music and sound… once everything else actually works™, I'm confident that a solution for this will come along.
Of course it's splitting hairs. Python can (and probably should) be compiled into a native binary, however. That's why I brought the basic/dosbox example up, which describes the essential problem here better. Another issue could be anything web-based.
"Native" is also a generally problematic as a term, especially on the Unixes.
I don't think technicalities like that will disqualify anyone.
GCC (or more clearly, GCJ) can be used to compile Java into a native format, although GCJ got discontinued when Sun open-sourced the JDK.
I doubt that this would be required, but it would be interesting where people draw the line on this. (Basic+Dosbox anyone?)
Given that Linux is the Jam's target, I'd consider it good practice to make Controls rebindable (think customization, also there are probably more users that use layouts such as Dvorak, Azerty or Colemak for example.)
Also it's not that hard to do in most cases, and a big improvement in quality.
If you're anything like me, that's probably a pretty interesting read: https://en.wikipedia.org/wiki/Arrow_keys
The Linux version doesn't run on your current distro? The claim of their being a win64 version is absolute bullshit? Tell me.
I can also deliver a OSX-port, but I'm only having access to Lion, and I'm unsure such a build would be compatible to the latest versions of the Operating system.
A little exercise in delivering the minimum expected framework…
actually, terrain-generation is simpler than SC2000's. Round up the whole thing with buildings made up of tetroid-modules, and connect them via pipelines… hopefully there's a game, somewhere in there.
Tricky, since it seems to be customary to stay silent if you didn't manage to put a game on this.
I started, knowing that tarot is based on the older taroccos card game. Since there already is a game in there (and it's older than the arcana/oracle part), I wanted to explore this particular angle further (also hey, there's already mechanics, right?). Turns out, it's probably the most complex card game of all time (think fizzbin with much more depth than poker) - anyway, after two days wasted on research (and a visit to the city archives, where they keep honest-to-god 500 year old card decks), the only thing I had come up with was the code for shuffling the deck, drawing and rating cards, with no idea that would fit the jam's criteria, besides just implementing the card game.
Anyway - this is the first game jam I see here that had enough criteria to be interesting. The mystery theme also is great thing.
Next time, I guess. :)