Slashdot Mirror
Super Mario Inspired SuperTux Issues Its First Official Release In 10 Years (phoronix.com)
An anonymous reader writes: SuperTux, the free software game inspired by Nintendo's Super Mario Brothers, has put out its first stable release in a decade. SuperTux 0.4 rewrites the game engine to make use of OpenGL, SDL2, and other modern open-source game tech. SuperTux 0.4 additionally features a lot of new in-game content, an in-game download manager, and support for translations. SuperTux 0.4 can be downloaded for Linux, Windows and Mac via GitHub.
I prefer TAGAP
http://www.tagap.net/
I enjoyed playing Supertux with a younger family member some years back and have some good memories of the game. This includes some of the addons and the early developments into the forest levels with advanced features that were fun.
Two things I'd like to ask if anyone knows:
Firstly, why can't the development team put together a single website with up to date info about this game? There was a move to Berlios De and git from SVN if I remember from lethargik, and now it seems to somewhere else. However, no information is left on the other websites that explains where the current stuff is happening or which pages are now obsolete. It would make a lot of sense to clarify this even for people just downloading the game.
Secondly: What the hell happened to Nolok? Did he really get replaced by a yeti as the main evil guy (as also in the unrelated game SupertuxKart)?
actually, it's worst than that. The source of the thing is ~480MB...
45 MB of music, 27MB of graphics, and 5MB of sounds. Beyond that largely portability, hardware agnosticism, and security.
I used to get high on life, but I developed a tolerance. Now I need something stronger.
You included the .git tree in your count, so you are counting everything twice. For the rest you can blame git as well, it stores and downloads the complete history, so everytime a music file gets updated a little, you end up with a duplicate music file in the project history, same with the graphics. Git isn't very good at dealing with changing binary files and none of the half dozen workarounds (annex, large file, etc.) are worth the effort to save a hundred megabytes.
45 MB of music
Would be 4.5 MB if it were MODs, or .45 MB if it were MIDs
27MB of graphics
This, not the music, is actually what pisses me off. Use low-res graphics to make them look like low-res graphics. Then use a fancy scaler to make them smooth when scaled way up. We all have supercomputers on our desks and laps now, there is no need to ship high-res textures just to save some CPU. Using low-res graphics with Quincunx or similar actually produces a better retro look than using high-res graphics anyway, so if your goal is to look all retro, that's a better way to do it.
and 5MB of sounds
That's bigger than the whole original game!
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Yet, SuperTux looks worse than Super Mario World for the old SNES in terms of style and taste. Yeah, it's higher resolution and might contain more colors on screen, but it's rather ugly as most FOSS games go.
MOD need specialist support to play now. Plus they need to be made by the original music creator - they capture each instrument individually and, as such, many things you might want to do you can't, and the mod player is responsible for decoding, timing, etc. which is an overhead.
MID technically needs hardware support or software instruments to play - and never quite sounds the same. The storage size doesn't include the software instruments - It was 45Mb last time I downloaded a soundfont for a soundcard. Plus they need to be made by the original music creator, and can't have various effects and changes done to them.
MP3 etc. are just sound recordings, not music formats, and are the final composition, layered with other instruments, effects, etc. Library support is pretty universal.
Graphics don't scale as nicely as you might think. Like fonts, you can't just scale up or down unless the original is vector. And then you have to process them and people complain that a game with a few boxes slows to a crawl when a lot of enemies are on-screen or requires a long startup time to rasterise them all in the right size first, or requires a 3D card with hundreds of megs of texture memory to hold them all.
Don't forget that nowadays, just the SDL library is several megabytes. It's supports all kinds of things that didn't even exist back in the 8-bit days. Hell SDL_TTF rendering requires a large library, plus FreeType, plus a font (the DejaVu fonts are 600Kb each or thereabouts). Sure, this is all "wastage" and you could just encode a bitmap font. For every possible screen resolution. Or run in fixed resolution. Like the 320x240's (or even half that) of the 8-bit era. Everytime you double a resolution, you QUADRUPLE the storage size required. So today's 1900x1200 screens require a lot more sprites to fill them and a lot more detail in those sprites to not look shit, and a lot more storage to hold it all.
There are reasons that things grew. I grew up in the 48Kb era. Have you looked at things like the Skool Daze disassembly. Fuck spending all that time squeezing that stuff into individual bits and still ended up with a ten-screen game because of memory restrictions.
Don't forget the amount of libraries that are sucked in to any simple program now. Dozens of megs for something as simple as calculator. Everything comes back to MSVCRT and a ton of Windows DLL's. On Linux, everything needs libc, and a bucket of support libraries and devices.
The reason it's so much more is because computers do so much more. And in terms of programming, I'd rather they spent time on making the game rather than pissing about optimising the graphics format for a 2D platformer. As it is they are short of people, short of code, lagging in development - and you want them to spend an age pissing about prematurely optimising shit using obsolete formats for the sake of some bit-level purism? That's a sure way to lose every developer on the project.
Especially compared to "#include " and just getting started straight away, even if that drags in megabytes of libraries that almost EVERY game written today uses.
This is an example of why Linux fails as a desktop.
One guy in the middle of some podunk area of South Asia who doesn't want to make any money at all can turn out an insanely stupid and insanely popular game in a couple weeks by himself.
You've got '10 years' and its on 0.4 and hardly any fun at all ...
Open Source games are going to fail repeatedly. Artists who do the actual meat of games are not stupid enough to work on YOUR GAME for free. Without artists and story tellers guiding your way, your design by committee games are going to continue to suck. You can't just hack shit together to make a work of art.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Now we just need a final release of Grand Theft Gentoo and it will be the year of the Linux gaming desktop.
Mod parent up!
He explains well why simple games like Tetris and Candy Crush will never be popular. No story line, no 3D graphic artists, no big explosions, etc.