A simple room

libellen’s framebuffer subsystem seems to be working fine, needs a lot of work, though.

Besides work, where I’m probably tackling the most challenging task of my career (sorry, can’t give any details), I’ve started investigating the DS homebrew scene. A few years ago I got myself a DS Phat mostly because it was a fun toy to hack on and, truth be told, I never got around to do it. There was a reason, though: Most DS emulators sucked and copying stuff to the flashcart every freaking time was tiring and cumbersome. Moreover, writing DS code at work and home was a bit too much. Oh, most DS emulators still suck balls.

But since I’m working on a different platform now (and a fairly tricky one, by the way) I started messing with my DS again. For those not in touch with it, let me tell you that the DS homebrew is a very healthy one, probably because flashcarts are dirt cheap right now, it doesn’t require any kind of hardware modding and the development tools are in fairly good shape.

The compiler (devkitARM) and the low level library (libnds) are quite good, but if you don’t feel like writing your own high level library there is pretty much only one option: PAlib. And here is where things get messy. It’s very easy to use (that’s why it’s so popular among the sceners), but it’s huge, messy and poorly mantained. Looking at their community forums, it looks like a few coders stepped up to clean it up, but it doesn’t look like it’s getting much better.

To make a long story short, I ended up using another, much smaller and tidy library: libellen by LiraNuna. It only handled sprites and backgrounds, but it was a much saner start point than PAlib. Eventually, I grew tired of the little project I had started (yet another top-down shooter, nothing fancy) and decided to re-implement another nice toy I coded a few months ago (a Wolf3D-style raycaster). Unfortunately, implementing such a thing using palettized tiles is kinda tricky so I decided to add direct framebuffer drawing routines to libellen. LiraNuna was kind enough to give me commit permission to his SVN repository and I’m now commiting the first bits to my own libellen branch.

I went as far as actually implementing a very basic, brute-forced raycaster, but it’s not ready for public consumption yet. I need to sort out some fixed-point math stuff first. Also, I need to add double-buffering to libellen. I’m a little short on time, as always, so please bear with me. Yes, the random image links are back :)

This is pretty much all I wanted to say. I guess I’ll write a more lenghty post about the expo later when I’m a little less sleepy and less tired, this year there was a lot to see :)

It's finaly here

So, it finally happened: Yesterday the first batch of Atmosphir closed beta invites went down the tubes to some lucky people who get to play the game before anyone else. If you received your invite then congratulations, and if you didn’t, don’t panic! There’s enough game for everyone, I’m sure it’s just a matter of time :)

Just like any other developer, I’m very happy to see this project finally getting to the players. It’s just the begginning, but knowing that there’s a bunch of people having fun with something you worked on feels just great. On a sidenote, Mazes of Fate DS is also waiting for you in the shelves, so if you want to make me feel even happier you can also give that a try.

But, as always, I digress. Today’s topic is about a few bumps I ran into while putting together the GNU/Linux binaries/installer of this game, an EPIC FAIL and how the community came to my rescue.

Hit the jump for the details.

The challenge

As much as I like GNU/Linux, deploying closed-source apps to a random GNU/Linux (Screw this, just Linux from now on) has always been tricky for a a few reasons:

1) An insanely heterogenous ecosystem. Even though each distro is more or less the same than any other, each one is almost a universe in itself at the same time. There’s no way to be sure what libraries the target system will have or what version of that linbraries you app will be running on. Since Atmosphir relies in a few third-party libs, I had to do my best to make sure the users would be able to just download, install and run. Sure, one can try to link everything statically, but in this case it was impossible for the following reason.

2) Licensing. Some of the third-party libs Atmosphir depends on are licensed under the terms of the GNU Lesser General Public License. I don’t want to get into details regarding this, telling you that I couldn’t just bundle everything into the binary. It was possible, however to distribute those libs as external objects (Which I finally did).

3) If something fails, the only way to know what’s going on is to drop to the freaking command line and watch the output provided by the game starts up. That is, unless you write platform-specific code to handle that and pop up some kind of message box. In the Linux case, this was impossible without adding yet another dependency (GTK or QT, both massive).

The epic fail

So, I managed to get things working. It isn’t magic, I just bundled the required libraries into the installer package including their respective licenses. Unfortunately, due to a stupid mistake I made, the game erroneously asked for libpq (the PostgreSQL library) rendering the game non-functional because of a library it didn’t even need. Fortunately, I was saved by…

The community

Yes, I bet that if you ever heard about Linux, you’ve heard about the so-called community. It’s a bunch of people that write code, write documentation or simple help users in need. I know it sounds cheesy, but they’ve saved my many, many times whenever I ran into some kind of issue with free (as in free speech, not free beer) software.

Regarding Atmosphir for Linux, two users helped us. The first one went so far to ever run the game through a debugger (yes, your eyes are working well) to provide us more information about the incident. We’re already looking into it, by the way.

The second one ran into some problems with dependencies (he found out about my libpq screw up), solved them and then decided to write a tutorial on how to get the game running (he also solved another dependency problem which didn’t arise during my tests). Kudos to you Jalada, and many thanks :)

Having all this info at hand will allow us to address the issues faster and better.

Conclusion

Even though it ended up being a little more difficult than I expected, I think that releasing the Linux version was the right idea. Not only it adds a few more players to the Atmosphir playerbase, but this is our little grain of sand towards gaming on alternative platforms.

And this also shows how important and useful are communities. It’s not about deploying a box and countng money anymore, we have to encourage participation and, between some boundaries, drive our efforts towards the direction our players point.

After all, it’s always been about the players, hasn’t it?

Oh, and my colleague Daniel Benmergui goes to Japan. Must be something about some kickass games he made or something. A recurring visitor once told me that Japan doesn’t actually exist, that it’s just a myth. Sounds like a cool myth though.

Today’s all about links. Farewell.

As probably most of you already know, the Dreamcast wasn’t as successful as Sega expected (mostly due to their own screw ups) and the white fun machine would be pulled off production just two years after (screwing the fans once more).

Fortunately for me, I had a Dreamcast while it lasted, so I was able to enjoy classics like Soul Calibur, Power Stone, Shenmue, Sonic Adventure, Phantasy Star Online, Ikaruga and many more firsthand. Heck, now that I think of it, I was playing online games in a home console back in the year 2000. One day I’ll talk about this console, I have very fond memories of it.

For the time being, Happy Birthday to you Dreamcast! :D

In other, pseudo-related news, Sega and BioWare have published a developer diary video about Sonic Chronicles: The Dark Brotherhood. Yes, BioWare. Yes, Sonic. And yes, a RPG for the Nintendo DS. Here’s the video for those who haven’t seen it yet:

In my humble opinion, it looks good. As a few colleagues already pointed out, it remains to be seen how the Sonic character cast and story plot fare for a role-playing game, but I’ll give them the benefit of doubt at least.

Regarding the video, the fact that one of the developers states that the game is pretty much finished but not yet up to BioWare’s standards talks quite some about who’s the project boss, doesn’t it?

Only time will tell I guess, September 30 isn’t that far away.

But I digress. In that specific article I’m pointing you to, Ryan states that, while Japan might be the center of the gaming world money-wise, it’s certainly no longer in the development front. As basis of that statement is the apparent disinterest in HD gaming shown by the japanese gamers. Again, the base behind this statement is the less than impressive hardware sales for the Xbox360 and PlayStation 3 consoles.

If we read a little further, we find the following quote:

“It’s not about the tech in Japan, it’s about the games, it’s about nostalgia.”

This is the interesting part for me. Let’s get this straight: Japanese gamers have a historic, well-known aversion to western gaming hardware and software. It wasn’t until the Tales of Vesperia launch (a classic japanese-style RPG for those who haven’t heard of it) that the 360 saw real sales. That game alone moved thousands on consoles off the shelves, something not many games were able to do. So, finally a true japanese game arises and the western, HD-enabled console, sells out. Seems like the japanese didn’t hate western consoles nor HD graphics that much after all.

The problem here aren’t the hardware platforms or their origin per se. As Ryan points out in the previous quote, it is about the games not the hardware. Granted, it might have something to do with nostalgia, but that can be hardly considered a japanese-only characteristic. Chrono Trigger DS will sell a royal ass-ton all around the world and I bet my pants that any game called Duke Nukem will sell a lot too. And Duke isn’t japanese at all. Nostalgia sells all around the world.

Going back to the games, there’s a clear trend among triple-A developments: Blowing up stuff and kicking rear ends in glorious HD sells. It’s a fact, games like Halo or Gears of War are two clear examples of the western gaming tastes or, more specifically, the american gaming taste. Most western gamers will ignore pretty much every other game component if the game looks good too; you just have to read the comments about any Wii game at blogs like Kotaku, Destructoid or 1UP and you’ll find a few dozens of comments about how ugly the game looks just because it doesn’t feature fancy shader effects or doesn’t use the Unreal Engine 3. It’s surprising how many people seems to think they know what a 3D engine actually is…

The fun part is that all those shaders and millions of polygons with hi-res textures are, most of the times, used in yet another FPS or hot-kickass-chick beat’em up game (Boob physics never looked so good, indeed). Or american sports for that matter.

So my question is: That’s the direction the western gamedev is taking? And Japan is lagging behind because of this? This kind of developments are everything but innovative (except for a few exceptions of course): it’s the very same we’ve been seeing for the past years (decade?), but this time bigger, badder,  with more glitz and fireworks. In HD, of course.

Don’t get me wrong, technology is meant to evolve, and the power of the new consoles can, and most certainly should, be used to bring new worlds, new fantasies, new experiences that were plainly impossible with our previous technology. But I don’t see that. The technology is evolving much faster than our game ideas, and that shall bite us back in the not-so-distant future. Fortunately, the independent gamedev scene is gaining momentum and I want to see what can those talented developers bring to the table with proper budgets and development tools and processes.

I don’t want to keep going on because this is turning into a senseless rant, so I’ll shut up for the time being. Japan might be suffering of some nostalgia-induced stagnation, yes. But while our symptoms aren’t the same, we’re not doing much better.

The so-called next-gen hasn’t yet begun.

<offtopic>

Ok, I’m hereby officially stepping down from the One Screen RPG Compo. As poignant as this is, I just don’t have the time to submit something I’d feel comfortable with. Anyway, I hope Lys can someday see the light of the day (kudos for those who get the pun, norwegians excluded), but I won’t make any promises.

This has been a pretty rough week at work and, in fact, I was going to talk a little about that project (based on already public information), but today I learned that everything went zomgconfidentialohnoes!!1shift!1, so I can’t tell you anything. However, if I were you, I’d sign up for the launch.

Without further ado, let’s get to business.

</offtopic>

So, this time I want to talk a little about one gameplay component often misused but (In my humble opinion) fairly interesting: exploration. There are many, many games that feature some kind of exploration activities in their gameplay: from the classic dungeon-crawlers to those shiny new 3D platformers today’s kids like so much, there have been many ways to encourage the player to sometimes step down the path we made for them and take a little trip to the unknown.

One clear example of this behavior are the hidden areas, a commonly used method to reward the players for going a little deeper into the game by giving them extra score, powerful items or similar stuff. This case can be seen on pretty much every 3D platformer out there.

Other games like Final Fantasy IV for the DS, for example, will take a slightly different approach: You will receive a bonus item when you leave a certain zone inside a dungeon if you’ve “cleared” the map (that is, if you’ve explored everything inside that zone). This particular case presents a pretty common problem in japanese RPGs: random encounters.

From my point of view, random encounters are the very nemesis of exploration. Unless the goal of said exploration activity is exposed to the player making it explicit and making it more like a challenge (ie. “You can follow your path or explore The Mighty Cave of the Unknown Filled with Monsters in search for the Holy Sword”), the encounters will be nothing more than annoying, unpredictable and uncontrollable stuff getting in your way, contradicting the idea of encouraging exploration. If the enemy is too weak it’s just annoying, if it’s strong the player will try to get the hell out of the dungeon as soon as humanly possible or, even worse, they’ll die without knowing why they took the risk in the first place. And we don’t want that.

But I digress, in those games the exploration component is offered in the form of secondary features of their gameplay, as stuff that you can do if you want to get the reward but you can also leave aside if you don’t. In fact, this is the case most of the times, often contradicted as I’ve already said for other gameplay components. By the way, random encounters are an entire topic I won’t touch here, some love it, some hate it, such is life.

What I’d like to see is a game where exploration plays a primary role in the gameplay, a game which is actually fun because you have to explore. Pretty much the only games where I’ve seen a solid exploration component are all dungeon-crawlers, but unfortunately these games are usually way too punishing to appeal most gamers except for the hardcore audience and, of course, the dungeon-crawler die-hards (all three of them).

What’s my idea, then? I’m so glad you asked. Well, I was thinking on a semi-online game, which would play mostly like a classic roguelike/dungeon-crawler, where the player would be thrown into a big, mostly unknown world. At the beginning you won’t have any maps and the NPCs won’t be able to tell you much about what’s beyond the initial city’s borders.

An entire world for you to explore, filled with rare items, secret chambers with treasure and, of course, lots of evil creatures to remind you why those lands remained unexplored for so long.

Then why the semi-online thing? With a world filled of secrets and virgin lands to explore, it’s pretty obvious that there will be many adventurers ready to dig out those treasures and secrets, to find (and maybe even defeat) the most vile creatures. Those adventurers would be the players, and each secret found, each hideous monster slaughtered will count as a discovery and those discoveries will be sent to an online database which will show which player accomplished what. I’d even want to make some of those discoveries unique in the game world, hence making use of the online functionality to update the game clients.

Starting to daydream, it would be great to be able to make these discoveries relevant to the game story plot, showing the players as protagonists of the story of the world itself. And, while we’re at it, certain discoveries such as the defeat of a certain extremely powerful monster (like a demon or something) might also affect the weather or the very landscape of a map, changing it forever. It would also be cool to have the forces of nature into the world, so an earthquake might also alter the landscape or kill a few dozen NPCs from time to time.

Apart from the technical ones, this idea has a clear flaw: As in the real life, history is written by the powerful, so every newcomer would have a pretty rough time finding or killing something worth mentioning, they may even have problems finding anything at all. I’m pretty sure this can be solved though, I just need to give it some more brain cells.

Phew, that’s pretty much everything. What do you think?

Ps. On a completely unrelated note, I discovered that if I say java stuff here, a weird google alert will be triggered. So say cheese to our stupid alerts overlords!

I have the idea and the coding is coming as good as my little free time permits. One thing’s for sure, I will send something, no matter what. At least as a proof of concept with little content, I want to keep my brain working so new ideas arise from the dark abyss of my psique.

Anyway, even though I have worked in RPGs before, it never ceases to amaze me how much work these games actually require. The main character, skills, spells, NPCs, dialog lines, it’s a lot of work, no matter how hard you try to simplify stuff. And when you’re done coding all that stuff, you still have a lot of content to add.

Regarding the code itself, I’m trying to keep it fun. How’s this? I’m trying to implement stuff as I see fit, even if it isn’t necessarily in the most academic way, just for the fun of it. Lua is great for this stuff, for most times it’ll work no matter how crazy it seems at first. And thinks are always easier if you put some LÖVE into the mix.

By the way, I’m working in two more “serious” posts: One regarding the exploration component in gameplay and another examining how much did things change in MMORPGs between classics like Everquest and more modern stuff. I still have a few things to say that you might find interesting, don’t give up on me yet :)

Revision 10 commited, Ren logging off