Chris Park from Arcen Games joins us this week to talk about independent game development and how his sci-fi RTS AI War: Fleet Command saw the light of day. Park talks about the design process, the challenge of modeling risk and reward and the difficult part of difficulty levels. It’s also one of our longest podcasts ever.
Listen here.
RSS here.
Subscribe on iTunes.
Arcen Games
Tom’s column on AI War: Fleet Command
Feature Series: The Decade
Quarter to Three, where you can find Tom’s movie podcast
Send your questions for Episode 40 to troy.goodfellow AT gmail.com
Phew! An epic podcast for me, I’ve been championing AI War since the early days of the Impulse release, and I’ve been impressed by Chris’ dedication to his fans, but also to his approach to the Real Time Strategy game and the unique features of Co-operative play against an asymmetrical complex AI with an overlay of Grand Strategy decisions to be made, centered around real time tactical combat.
It’s obvious from the podcast that Chris has thought through the design process in an evolutionary way, and is extremely enthusiastic about whats possible in the future. I’m still staggered by the pace at which AI War progresses, and the way the same game we bought in May has become something so much more in such a short space of time, distilled and whittled by players suggestions and Chris’ vision at what will make the game enjoyable for the widest spread of strategy players.
The trouble is, there is so much to say about AI War, that even an hour plus long TMA podcast isn’t going to cover some of the complexities and decisions that are made tactically and strategically. For instance, I think there was only a brief hint at the fact that some of the structures you can capture in systems actually help reduce the AI aggressiveness towards you, something key to maintaining an AI horde you can handle as you expand and evolve your fleet. I would urge everyone to take a look at some of Chris’ articles on his blog site[1], especially the AI ones, just to delve into the logic and processes behind the game.
I think I would echo the statement that if indie developers took on board the Arcengames method of game evolution by such close and immediate contact with its players, there’d be a lot more games that grow into playing communities, where the players have a vested interest in the development and the gaming itself. To think AI War is now out there on almost all digital distribution channels, with a thriving and eager community supporting future projects, it makes me almost glow with satisfaction that this unique gem of a game, has been discovered by so many strategy gamers, and hopefully its design and lessons learnt will be taken into future strategy projects.
Proud to be an AI War player and I wish Chris all the best for the future of Arcengames. I couldn’t have asked for any more, my favourite co-operative strategy game on my favourite strategy podcast. I’m hoping the TMA team will one day take AI War for a co-operative spin, and give us an after action report to drool over.
[1] http://christophermpark.blogspot.com/
The problem with AI-Wars is that the UI sucks to much for a commercial game.
With a freeware game like DF you can take a weekend and try to find the good game that makes up for the bad UI.
But AI-Wars demo is not long enough for this. You are left with a feeling of: “How could I ever even try to run more than 10 ships in this UI”.
Maybe there is a good game behind this, I can’t tell in the Demo.
Until the UI is at least on nethacks level I am not planning to buy this.
All I can say to help Tobias is that the majority of the game’s controls are fairly standard RTS, there are some esoteric key shortcuts to help with management, and the UI is laid out to allow you to see the maximum of action as it happens.
I didn’t really have any problems with the UI as such, at the beginning the zoom was a bit flaky, but thats been honed to something very usable now.
Theres a Key Command printable reference sheet available from the Arcengames forums[1], thats very useful to have. Also take a look at the AI War Wiki[2] for further information to help acclimatising with the controls and UI.
[1]http://arcengames.com/forums/index.php/topic,1247.0.html
[2]http://www.arcengames.com/mediawiki/index.php?title=AI_War:Fleet_Command
I can’t think of anything that AI War’s interface lacks compared to other RTS games. It even has some unique features that make controlling large numbers of ships easier, like a filter to select only ships of a particular tech level, or only ships damaged below 33% health, or select half of the currently selected ships. There’s free-roaming-defender mode, and you can tell your ships which enemy ship type is a priority target.
Tobias, you mentioned two other games in your post: Dwarf Fortress and NetHack. So please excuse me if this question is ridiculous, but have you tried playing AI War using a mouse?
Anyway, AI War is great. My favorite thing about it is that it’s a 4X game with no snowball effect, so if you choose challenging settings, then the final battle is also the climactic battle. This is because the AI doesn’t play like a human who’s trying to win at all costs, instead it reacts to how much of a threat you appear to be. The AI isn’t pursuing total war against the humans, an analogy that Chris has used is that the humans are ants invading the AI’s house.
The end result is that it’s a 4X RTS with additional strategy decisions that you don’t usually get. You need to conquer planets to increase your resource income and form a defensible position around your home planet, but each planet you take will increase the strength of the AI’s attacks on you. Make surgical strikes against valuable targets while ignoring others, and don’t wipe out the AI on one planet if weakening the defenses there is good enough. Blitz past the front-line planets in order to capture an advanced factory before the AI can increase defenses there. But going past the front lines means you’re out of supply range, and if your target is a strong planet, maybe you should conquer an easier planet that’s adjacent first, and use that supply base to help conquer the real target.
So it was a very interesting podcast. I liked the idea that AI War is too big, too complex, and too hardcore for the indie game sites to cover it much! Also, I didn’t realize that its co-op feature is so convenient, and I second the motion for a TMA game of AI War.
I rather enjoyed the podcast, though having not played it, a lot of the monologue was lost on me.
And, sadly, it was a monologue. I get that you guys don’t want to be rude and interrupt your guest, but this was too far in the other direction.
Can’t wait for next week!
Tobias: Sorry to hear you don’t like the interface much at all. I’m curious what you found to be so limiting and frustrating with it, though. While we think it is pretty well geared toward both controlling the game’s massive fleets and presenting useful information to the player about his and the AIs’ worlds, we still really value input on making it better for everyone. Battling an interface is really frustrating, and we don’t want that to be the reason that someone decides they dislike our game.
I’m happy to discuss this with you to try to improve AI War, if you’re so inclined; criticism and suggestions are really useful to us.
@Punning: Well, they asked me to be expansive, although that is quite possibly not what they meant. It’s a challenge to answer thoroughly but also briefly on fairly complicated topics. Or, at least, it’s not something I’ve mastered, and it’s something I’m trying to be conscious about. Never intended it to be a monologue!
Also: I noticed there were at least two instances of where there was some back and forth commentary with Tom and Troy that was cut out, so that it seemed like I was just talking the entire time through those segments. Once was because the call got dropped and there was some discussion when I got back in, but the other was unrelated to that and evidently an editorial decision.
@Chris Park: Oh hey, yours was a good segment, I’m sorry if I gave another impression. Mostly, it was a comment that the guys were treating you differently than other interviewees.
I gotcha. Well, that was probably at least half my thing, not theirs. Like I mentioned, it’s hard to be both expansive and brief, especially in the moment. But, glad to hear you enjoyed the segment!
I get the monologue complaint, but I was impressed that you could speak so expansively without repeating yourself or fumbling. I’d heard of AI War, but hadn’t been interested until heard this podcast. You talk it up well.
Today’s payday. I’ll pick it up. I use Impulse, Steam and Gamersgate –does any one in particular give you a greater cut of the cost?
@ Stormwaltz: Thanks for the kind words, and your interest in the game. Most of the partners we have are pretty similar in the cut we receive, and I can’t really ethically recommend one over another. I’d say use the one you like the best, and I thank you for your business!
Hi guys,
Just listened to this podcast. This sounds like a cool game. Will grab it.
The lore reminds me a lot of the Travellar universe where the Terrans move out into space, hit the Imperium, dont realise how big it actually is and take them on. The Imperium, for various reasons, ignores them until its too late. Though they could have crushed them at any time until the tipping point where it became an even battle.
КраÑавче ! Пиши иÑчё!
A good interviewer knows when to interrupt the guest and when to let him talk. If the guest needs to be guided, then it’s the host(s) role to guide him, if he covers all the points, then let him speak.
For instance, Chris addressed, from the get go, some of the Tower-Defence elements of the game. Had he not do it, it would’ve been Troy, or Tom’s place to ask him about it. Take Tom’s article about AI War, and you’ll find that ALL of the points Tom made were covered here aptly, some because of questions asked by the hosts, and some because Chris mentioned them independently. I think either side (Troy/Tom, Chris, and the listeners) could’ve asked for a better overall interview.
I think this is among the best TMA episodes recorded, and recommend it warmly.
The Claw | // Nov 11, 2009 at 3:08 pm
[…] game and the designer, Chris Parks, check out the UKGamer exclusive interview I did in July 2009. Episode 37 of the Three Moves Ahead strategy podcast is devoted entirely to AI Wars and discussion with the developer, worth a listen. Also keep a look […]
As a followup to this TMA podcast, I did quite a lengthy interview with Chris Parks over Google Wave, which I’ve posted here if anyone is interested.
http://blog.theclawproject.net/?p=77
@Tobias
The demo gives you the full tutorial, which includes a campaign with no time limit. Sure, the AI is set to the easiest level, but learning in the interface in the demo should not be an issue.
For the most part I love AI Wars, but having come back and tried 2.0 for the first time I’m having mixed feelings about the graphical overhaul. The screen feels really busy now, and information I used to find easy to tell at a glance I now find myself lost in a sea of colours. My biggest issue is that I really don’t think the new shot graphics match the ship sizes, and as they sort of leap onto their target at the end it’s hard to tell who you’re actually firing on.
On the other hand, perhaps if I’d started with 2.0 and tried to go back to 1.0 I’d have similar complaints ;)
I’d have liked to see someone take Chris to task for his comments on pacing. He brought this up as an issue he intended to tackle, but I think it’s too easy in AI War to get stuck in a cycle of having yourself in a secure setup, build up a fleet behind the lines, use it, then rinse and repeat. Certainly I’d like someone to have asked him how he felt he’d tackled that issue since 1.0 as I believe has has made several strides away from that model.
Can’t wait for the expansion though, I think a third faction is exactly what’s needed to mix things up for those of us playing single-player. I still harbour a hope that one expansion will add bot partners though.
AI War : Fleet Command - The Older Gamers Forums // Nov 15, 2009 at 8:37 am
[…] are many keyboard shortcuts that should be learned). I was sold listening to the developer on Three Moves Ahead Episode 37 – Chris Park and AI War: Fleet Command There are also a lot of interesting videos of game play and strategy on YouTube… […]
@Quitch — that’s interesting about your take on some of the graphics changes in 2.0. In general, I think things are a lot more clear and crisp. The oversized shots are mainly so that you can actually see them at far zoom, so you can tell who is shooting at what. Though, granted, when you have a lot of short-range ships all on top of one another, the carnage is hard to follow. That’s a case where you either need to split forces up and engage groups of guys in batches or at range, or simply largely trust in the unit mix you chose to start the battle.
Regarding pacing, I would posit that if you are finding yourself too secure, you aren’t playing on a difficulty level that really matches your skill. :) Often I get to a point where I FEEL secure, but then the AI does something and I lose a bunch of planets. This effect is exaggerated somewhat in multiplayer, too, because you are forced to have more territory (by nature of having more players to support), and so there are more ingress points.
In the past some players were able to set up extreme-low-AI-Progress setups, but the changes to how energy works, as well as the addition of the Supply mechanic, have largely addressed that; players can still go for that sort of strategy, but it’s not a clear best path for play by any stretch. That’s a great example of players doing something clever that I never thought of, and then having to address that with new game mechanics. If I wasn’t still rebalancing the game based on player findings of that sort, that’s where the game would “die” for those players, since there would be nothing more to do once a best path was arrived at. Some of that is still ongoing, as it will always likely be with a game of sufficient complexity, but there aren’t any unfixed best paths at the moment that anyone has found to my knowledge.
In the expansion, even beyond the third minor faction, it’s really going to shake stuff up just with the new capturables and new AI weapons in general. Making the galactic landscape feel more unique really has a huge effect on the game. Bot partners may be something for a future expansion, but one cool thing is with the new Zenith faction in this expansion, those guys are neutral so they can be a friend or a foe at times. Not quite the same as a full-out bot partner, but sometimes they will fill something of a similar role.
This is an old post but I just wanted to say I picked up AI War during the recent sale and it’s the best strategy game I’ve played in years. I picked up the expansion immediately after completing the intermediate tutorial. I’m saddened that this game hasn’t gotten the hype and recognition it clearly deserves. I can’t wait to play co-op.
Not only is it an outstanding game, but the developer support is the best I’ve seen.
Many thanks, Bertrand — I really appreciate it, and I’m glad you’re enjoying the game so well!
Just bought the game this weekend off of Steam, and I’m really enjoying it. Good job, I like it better than Sins of a Solar Empire. Thanks.