Exploring The Studious World Of Adventure Game Studio

Adventure! Game!
Adventure Game Studio is a free engine and toolkit for the creation of point-and-click adventure games. Since its appearance in 1997, it’s grown in scope to the point where it’s now utilised in many commercial titles. Lewis Denby put on his fictionsuit and plunged face-first into the world of AGS – taking a look at the games and speaking to some of the developers who swear by it. Let’s see what it means to them.

Last summer, I set myself a project: to learn Adventure Game Studio and try my hand at making a couple of small games. I knew that it was a free engine and toolkit for the development of classic point-and-click adventures, and had the idea that it was a stripped-back, basic engine-in-a-can that was capable of some interesting stuff in the right hands, but mainly targeted at amateurs. Even knowing this, what I found surprised me.

AGS is, in fact, an enormously capable toolkit, brimming with possibilities. It’s true that it can’t handle 3D graphics or much other fanciness, but with just some basic scripting knowledge you can create surprisingly complex adventure game logic. It’s aimed at amateurs, but that doesn’t always restrict what you can do with it.

I’ll wager that quite a few people reading RPS have played AGS games without knowing it. You might have even bought one in a shop. Bold and intricate games like Gemini Rue – which Comrade Cobbett called “one of the best commercial [adventure games] in ages” – were built using this toolkit. I’d be fascinated to know what AGS creator Chris Jones would have thought if you told him this back in 1997…

It began life as a personal project for an ambitious young man who wanted to make an adventure game, but didn’t have access to an engine in which to do so. But Jones quickly realised it was the development of the toolkit, rather than something using it, that truly excited him. To this day he’s still not made a game of his own. He is, like so many people in the games industry, someone who helps other people do the final order game creation. He is one of the grand facilitators.

What he has done is update AGS beyond recognition from that early plan. The more I delved into the software, the more I realised how much work must have been done to deliver something so powerful. Quite easily, and with a bit of creativity, you can make some startlingly good games – as many have done, both as freebie releases and as professional products. Among my favourite are those in the free Technobabylon series, created by James Dearden, also known as Technocrat.

“I wasn’t big on using anything like C# or Java when I started making games, but I’d been using things like Game Factory,” Dearden tells me. “AGS had the right balance of ‘simple’ and ‘powerful’ to keep me going with it.”

Technobabylon – a powerful story-driven series set against an uneasy sci-fi backdrop – wasn’t the first AGS game Dearden made. But, oddly, it was the first adventure game. With a splash of creative coding the toolkit can be employed for a range of different genres, from strategies to RPGs, and it was those that he began with.

It wasn’t until 2008 that Dearden became involved in the passionate community that’s sprung up around Adventure Game Studio. It’s a broadly helpful and pleasant place to spend your time, but I’ve also always found it to be strangely insular, sitting far away from the usual indie development scenes and generally keeping itself to itself. And, just as it often seems to struggle to reach out, it has a slight reputation for being difficult to penetrate for the outsider.

I’ve certainly got that impression when approaching it as a member of the press. Usually if you drop a name like RPS indie devs are very interested in having a chat, but this doesn’t always seem to be the case with AGS community types. Recently, an impressive-looking commercial title turned up, and I thought I might like to give it a bit of attention. After an odd, suspicious back-and-forth between myself and the developer, the conversation simply went quiet, and repeat emails enquiring about the possibility of review code were ignored. I’ve yet to see the game reviewed anywhere, even though it costs about $15 to purchase, and word on AGS Street is that it’s pretty damn good. Strange. Why?

“The idea of promoting your game to the press is a little strange within the community,” prolific AGS developer Ben Chandler tells me. Chandler has made more games with the toolkit than anyone else I’ve come across (he admits he’s lost count, but thinks it’s somewhere around 30), but even he says it’s only recently that he’s learned the value of looking beyond the AGS boards. “The idea of mailing journalists a copy of one’s game before release was something I’d never heard of until a writer asked me to do it one day. Then it made sense to me.” But he says many developers simply release within the community, and make no effort to promote elsewhere.

This, says Chandler, is in a large part because the AGS website and forums triple up as an arena for gaming discussion, development help and marketing. “Being such a versatile community means that a lot of people don’t feel the need to engage in other communities across the web,” he says, “and therefore newly released games only get announced here by these developers.”

Dave Gilbert, creator of the Blackwell series, says not falling into this trap is essential – especially if you want to sell your game. There are minimal restrictions to doing this (basically: replace all the default art, don’t use MP3 audio), yet relatively few AGS developers actually do – and there’s sometimes a little resistance to the idea within the community.

A thread exists where players and developers discuss the notion of ‘going pro’ with AGS. There are a range of thoughts on the matter, but the most prevalent one seems to be: “Go for it if you want, but don’t expect everyone to be happy about it.”

“AGS is a community that is founded in the tradition of freeware,” explains Gilbert, “and there are so many free AGS games available that most wouldn’t see a reason to pay money for one. Of the several thousand members of that forum, only about 50 have actually purchased my games.”

But reaching out is a daunting task, he adds, especially if all you know is AGS. Perhaps all this accounts for the community’s reputation for being insular. “This conversation comes up every couple of years,” Gilbert says. “Honestly, nothing could be further from the truth. The one thing that surprises me about the AGS community is how there’s a regular stream of newbies coming in. It’s a very welcoming group.”

Dearden agrees, and – contrary to my assumptions – thinks commercial releases are especially well supported. “In the community, I think [going pro] is a goal of serious developers, and among other watchers it’s something they’ll commend, as long as it’s realistic,” he says. “If someone leaps in with a Final Fantasy knock-off that they insist is going to sell for $10 plus DLC, there’ll be a few eyebrows raised. Otherwise, there’s a lot of encouragement and support from the community, with so many people willing to lend a hand.”

“If you want to make an old-school point-and-click adventure game, there is no better choice than AGS,” concludes Gilbert. “It is specifically designed with the assumption that you are making that kind of game, and it takes a ton of the gruntwork out of the process. Not only that, but the support community is one of the best I’ve ever seen.”

Ben Chandler – who, as well as his free games, has collaborated with Gilbert on the Blackwell series – is similarly full of praise for what Chris Jones has engendered over the years. “It often amazes me that the thing I always wanted to do as a child is achievable without any training or education about the topic whatsoever,” he says. “The fact that one can, in the privacy and comfort of their bedroom, create the games they wish to without the need for any specialist equipment, and then share it with players all over the world, is pretty incredible. Being able to do this has changed computer game development from something I always held as an unachievable dream into something I can’t imagine not doing. Tools like AGS give common people like ourselves the ability to realise a dream – and that is something I am very grateful for.”

If there’s a message in all this, it’s this: You’ll probably need to be good with logic, be prepared to read some manuals and learn some simple coding, but that patience will be rewarded. I can’t think of another development toolkit that offers more flexibility considering the minimal specialist knowledge required. You can make a simple, functional game in half an hour, or spend years fine-tuning your latest masterpiece. If you have an idea for a 2D game that’s not enormously mechanically ambitious, the chances are you’ll be able to hack it together in AGS.

And the community continues to thrive. Isolated it may be, but it’s a dedicated group of people following their game development dreams, always eager to help others along the way. At the moment there’s talk of iOS and Linux ports of the engine, monthly design competitions, and even entire game worlds that have been collaborated on and shared between different developers’ games. A teenage Chris Jones wanted to make a point-and-click adventure, but he’s arguably delivered something far greater: the chance for anyone with an idea to make it real, and a forum full of the people, knowledge, and ideas required to succeed.


  1. Kid_A says:

    Wot, no mention of the Chzo Mythos?

  2. Ice-Fyre says:

    RPS should do a review of some of the free game engines out there. I know of Unity and this, but there must be loads out there

    • robkun says:

      Like the Visionaire Game engine used in the stunning-looking Stasis which was featured on RPS yesterday.

    • frightlever says:

      In light of the “They Might Be Giants” post – Inform 7. It has the best graphics.

    • Zanchito says:

      Very much interested in this idea!

    • Dizzard says:

      I’d like to see more of this too.

      Found the article about AGS really interesting.

    • LionsPhil says:

      SDL should be enough for anyone! *shakes fist*

    • eks says:

      RPS aren’t developers so I don’t think they have any place reviewing developer tools. Perhaps a series of interviews with various developers who have been using the different tools would be a better option.

  3. rustybroomhandle says:

    AGS has a great creation toolkit, easy to learn and with agood productive workflow. The runtime engine itself is orrible though.

    For my first ‘real’ game I have opted for OpenSLUDGE. Might not be as easy as AGS but it’s open source. Plus I find the AGS board community a bit too cliquish.

    Also check out Wintermute. Windows only though.

    • TsunamiWombat says:

      Cliquish. For some reason I am now imagining game developers with greaser haircuts and leather jackets. I love you brain.

    • LionsPhil says:

      It bothers me that AGS is not open source.

      IIRC it also uses undocumented, proprietary file format, and has needless cripples in the UI “for your protection”, like not allowing you to delete certain types of object. Goddamnit, this is what I have version control for. (Which works better when not working with odd little binary blob formats.)

      Ooh…SLUDGE is from the guy who made Out of Order (and Ducks). Cheers for pointing it out.

      …gah. Except that the hurr-durr open source community rewrote the backend to output via OpenGL, then found performance issues. YOU DON’T SAY.

      • Rikard Peterson says:

        Hi, LionsPhil!

        The “hurr durr community” you’re referring to is me. While Tobias Hansen has also been doing quite a bit of work on the engine (mainly the Linux port), I’m completely to blame for the porting to Mac OS and the big rewrite of the graphics code, and I’m curious: What performance issues have you found? My tests have shown it to be faster after the re-write. The only downside of the switch to use the graphics hardware that I’m aware of is that it now needs a graphics card with programmable shaders. (Which means that it should still run on anything reasonable modern.)

        / Rikard

        • LionsPhil says:

          I’m going off the site’s own frontpage to some extent:

          Also people with slow graphics cards (like me) for whom SLUDGE slowed down when antialiasing was enabled can hope for some speed increase.

          Why? Why involve the 3D pipeline at all? A Celeron 500 with a completely-ignorant-of-3D integrated graphics chipset can do this in software just fine.

          Was this decision actually supported by empirical evaluation, such as profiling? Was there actually any performance issue with plain old 2D rendering to justify the change?

          • rustybroomhandle says:

            I suspect that the real time scaling, alpha channels and light maps might struggle a bit more in pure software.

          • Rikard Peterson says:

            Good to hear that the performance issues you talked about were only a misunderstanding and not something you had experienced. Note that the thing you quoted was about AA. The new engine has three modes for AA: The decent looking one, linear filtering (i.e. blurring) and off. The quote was about the fancy mode, and it also said that it now was faster than before.

            As rustybroomhandle indicates, one reason that I went the hardware route was that I wanted to add full 32-bit graphics with alpha channels and blending to the engine. (It was previously limited to 16-bit graphics with 8-bit sprites.) It also now (by default – this can be changed) runs at desktop resolution while preserving the aspect ratio of the game. That’s a lot of bytes to push around. (I haven’t been interested in making an 8-bit low res game since the 90s. If that’s what you’re thinking about then you’re right that going hardware would have been overkill.)

            You may be right that a better programmer than me could get it running fine completely in software, but that’s more than I’m capable of. In fact, the original code by Tim Furnish had a simple AA function (in software, and only internal in the sprites – i.e. not touching their edges) that did cause my game to slow down on my computer unless I disabled it for larger sprites. It was very noticeable when a large dragon towards the end of my game almost fills the screen. Now it’s both better looking and faster.

            I would have thought twice about the switch to OpenGL if I’d realised how much work it would be to get it working properly (I’d never done any OpenGL programming before this), and it may not have been a great idea to rewrite that code at the same time as I was rewriting the code to be cross platform. (It took me a while to get anything running at all.) But now that it’s all running fine, I’m happy with how it’s turned out, and I don’t doubt that the hardware route is a good choice. The engine (and most importantly for me – my game) is now in a state where it’s running well on both Mac, Windows and Linux and there are no known bugs.

            That being said, if you (or anyone else) wish to present more arguments against the use of OpenGL that I may be unaware of, or if anyone finds any issue with the engine (bug or otherwise failing to work properly), feel free to contact me. I don’t want to feed e-mail harvesting bots, so I’m not posting my address here, but if you follow the link on my name, there’s an e-mail form on my web site. (We’re getting a bit off topic for this comment thread.)

          • LionsPhil says:

            Ok, thanks for the reply. Sorry if I’m being a bit flamey but the “take everything 2D and shove it through a 3D pipeline” is becoming a bit of a sore nerve for me.

            I’ll shut up until I’ve had a chance to actually look at the code some time. ;)

      • Miyu says:

        The intention is to make AGS fully open source, it’s just that Chris Jones hasn’t had the time to do it.

      • rustybroomhandle says:

        LionsPhil, yeah which performance issues would these be? I’m finding it to be rather snappy, even with the grossly over-sized sprites I am using.

      • snarky says:

        AGS is open source.

      • Gradius says:

        The engine IS open source: link to bigbluecup.com

  4. BatmanBaggins says:

    It sounds great, but I feel like a big barrier to entry is whether or not you have the talent to make your own graphics. Unless you don’t mind using the stock stuff…

    That’s the impression I get, anyway. I want to try it out but haven’t had a chance to yet.

    • rustybroomhandle says:

      I made an AGS game using scenery made in Minecraft. If you just want to learn, there’s always a way.

      • Nathan_G says:

        For someone with absolutely no coding experience at all, is attempting a simple game in AGS viable? Or am I going to have to invest significant time learning the language?

        • Chris Bischoff says:

          I don’t know about AGS, but with Visioniare I managed to create a playable level after about 2 days of messing around when I first started.

        • Premium User Badge

          Hodge says:

          You can do a lot of things through the GUI but to do anything substantial you’ll need to hand code stuff. Even so, I wouldn’t say that it’s super difficult. As with most coding, the main skill you’ll need is the ability to break things down into logical steps and decisions. The rest is pretty much practice.

          I’m a rubbish coder and I’ve bounced off a bunch of things including Unity, but I picked up AGS pretty easily (so far).

        • Lewis Denby says:

          I came to it pretty code-blind about 9 months ago, and now I’m working on a full-on indie game. Take time with the tutorials, read the beginners’ forum, and generally experiment. You’ll pick stuff up.

    • Rikard Peterson says:

      There are at least two solutions to that problem.

      1) Learn to draw (or model in clay or whatever). I don’t believe in “talent” as something you either have or you don’t. (I’m a music teacher, so I know what I’m talking about.) If you put in the time and effort, you will improve.

      2) Find someone to collaborate with. This can be difficult, but it’s not impossible.

  5. Dervish says:

    Why would you include a screenshot of mostly text that’s too low-res to read the text? Or is there an enlarge script that’s just not working for me.

    • Skabooga says:

      That is programming code, I believe. To read such gibberish will only lead to madness, but merely looking at it from far distances will cause mild discomfort at worst.

  6. _Nocturnal says:

    I’d recommend that people interested in AGS also try Ren’Py.

    Like AGS, its main purpose is to help you create games from a specific genre (visual novels in this case), but it gives you the freedom to do much, much more. Personally, I struggled with AGS’s interface and found Ren’Py’s approach way easier. It’s basically just: “Here’s a notepad document, here’s what you type in it to show text, pictures and menus on the screen, here’s how you structure an if/then/else statement, now go ahead and write what you’d like to happen!”

    Adventures and visual novels aren’t really that far apart, too.

    • jrodman says:

      Ren’py also gives you portability for free, while AGS gives you no portability ever.

      Really though, I suspect they’ll appeal to different sorts. Ren’py’s feel of a blank page that you can type anything into is a rather different working mode than AGS’s integrated tools. I prefer the blank page myself.

    • Zwebbie says:

      I haven’t tried AGS, but Ren’Py is absolutely great, if you ask me. You can pretty much master the tools needed to tell a compelling story within an hour. If you’re the kind who can write compelling stories that is, which, sadly, I’m not.
      Another bonus is that both adventure games and visual novels can get away with relatively simple art.

    • Janto says:

      Yeah, I opted for RenPy after seeing what Winter Wolves were capable of getting the system to do, such a full RPG. I’m curious about breaking out of the anime bracket though – I’ve heard rumours it’s tough to sell non anime visual novels, and the style’s not where my art’s at.

    • _Nocturnal says:

      By the way, when choosing between Ren’Py and AGS, one of the most important factors for me was that AGS didn’t really support modern resolutions. It’s an important thing to know about but honestly, a person should simply pick the engine that allows him/her to make games easier.

    • Gasmask Hero says:

      Well, I was using AGS for my current project, but Ren’py looks like a much better fit. Plus I see it supports Python…

      …can’t hurt to see what it can do.

  7. MuscleHorse says:

    Wasn’t actually aware of this – and now it looks as if the gaze of the unblinking RPS eye has borked the site.

  8. Stuart Walton says:

    Every now and then I toy with the idea of making a game in AGS. Especially when I play an adventure game that has really shitty puzzles, or is just bad at feedback and hints.

    I need a story first, maybe I should just grab something in the public domain, crib from that, pare it down.

  9. AmateurScience says:

    I periodically try to make something in AGS but am rebuffed by my complete inability to art, or fiction. I do however regularly download and enjoy a number of games on there. It’s a really great community.

  10. povu says:

    I loved Gemini Rue. A nice story, and it’s one of the few adventure games that I’ve managed to solve without ever looking up a guide for anything, without it being too easy.

  11. kentona says:

    This accurately describes the mindset of the RPG Maker community, as well – insular, there are so many free RM games available that most wouldn’t see a reason to pay money for one, the sites triple up as an arena for gaming discussion, development help and marketing, plus there are hundreds of Final Fantasy knock-offs.

    I guess people like their small ponds, and don’t like swimming out into the open waters of the Internet (I know I do!)

  12. celozzip says:

    i’ve tried loads of these game engines and it always comes down to this – i can’t draw. anyone can write code it’s easy especially with all the game engines out there now they make it so simple. but i cant draw so it aint gonna happen. i suppose i could put in my crudely drawn placeholders till i find someone on deviantart or whatever but i just dont have the motivation for that. i’m sure noones missing out anyway.

  13. Lemming says:

    I’d love to make a P&C Adventure, but I have trouble figuring out how exactly to tie a story to things with use x on y moments that a game like this needs without just throwing in arbitrary puzzles.

  14. orient says:

    As someone who can draw and write but can’t do any sort of complex programming, AGS is great. I’ve tried a few different “entry-level” programs — Flash, Game Maker, RPG Maker — but AGS suits my style far better than those. In fact, I’ve started a new project, the first I’m actually going to finish. Shameless plug.

  15. zripgud says:

    I recommend this videotutorials to anyone who wants to start learning how to use AGS: How to use AGS (Adventure Game Studio).

  16. timofonic says:

    Hello to everyone.

    Here are some points I wanted to write for the readers and the article writer, maybe even useful to mention in further AGS-related articles…

    – AGS IS Open Source, just some parts of the editor are missing due to license issues. The community is aware of it, Chris Jones said to release it in the future but there’s still no news about it. So this will be eventually solved, being by CJ or some enough technically skilled community member.

    – AGS runtime is being implemented as an engine for the ScummVM project by Alyssa Milburn (aka fuzzie)[1][2]. The advantages are the extreme portability of this project (it’s already ported to dozens of platforms and the list eventually grows). The ScummVM project is not only Free/Open Source, but it’s a very friendly community of organized and smart developers that even participates on Google Summer of Code.

    – There’s Android and PSP ports available by Jochen Schleu (aka JJS). He even managed to improve the backwards compatibility with the 2.5 version engine.[3]

    – Nick Sonneveld (aka sonneveld) made a fork aimed at improving the code quality and probably other features. He’s into early steps, but doing a great work. He’s also quite communicative with the ScummVM Team in the IRC and such. [5]

    I have hopes in this project, having a second life in the Free Software form and evolve beyond the initial design too. These days most indies and hobbyists in the game world aren’t aware of the power of FOSS and how collaborative communities of these kind work, but they are starting to see the advantages of it and mixing with the FOSS communities like ScummVM. From my geeky point of view, I see it as amazing as the merge of both hacker and UNIX communities in the post-ARPANET era (mostly 90s) up to now.

    And about Wintermute, there’s an implementation named WME Lite and having interest in the ScummVM project itself. It has been a GSoC idea for ScummVM, so maybe we’ll see it part of their familiy soon. [6][7][8][9]

    Remember ScummVM started as a proyect to play Lucas Arts adventure games, then dozens of engines haven been added (and these days there are some work in progress ones to be added) in more than ten years. I think this is a great example of both the technical and creative advantages of Free Software projects.

    That’s all, I hope this can be useful to others ;)