This is an interview with the Development Team of Allegiance. I have to thank them for taking time out of their day to answer the questions for this interview. Read and enjoy!
Executive Producer: Joel "solap" Dehlin
Lead designer/art Director: Rob "ribski" Girling
Dev Manager: Curt "dontmess" Carpenter
Gameplay Developer: David "tkela" Pugh
Product Planner: Jon "riobravo" Grande
How many people are currently working on the game?
[Joel] The number of people on the team fluctuates widely, depending on the
cycle. Currently, we have me, rob (the designer/art director), one 3d
artist, 4 testers, 6 developers and one summer intern. in the early days, we
had like 3 people. at one point we had an additional 5 or 6 artists. For the
vast majority of the project, we've had only 4 developers.
What is your average day like working on Allegiance?
[tkela] I normally get in around 8 AM and start by reading the news groups and trying to figure out what to nerf next.
Once I've laid my evil plans, I start my real job: fixing bugs and adding features. By this time, the testers have normally found 2 or 3 problems with the morning's build and are playing developer tag to get them triaged. If I'm lucky, they miss me and I can continue working on bugs while fiddling with balance while the compiler churns.
At least once a day, there is a 3-way meeting (between me, rob & joel) to discuss balance issues. On good days, this takes 15 minutes to sanity check the changes I've worked out above. On other days, it can be as much as an hour or two trying to figure out how to deal with the latest discovery of the beta testers.
After that, it is back to bugs, entering the balance changes and trying to verify that everything works as intended.
As the BETA progresses Allegiance is becoming a much better game, with new features added every patch. Can you tell us features are in development? What features are planned for the future?
[joel] To be honest, not many new features are coming from here on out. we're mostly done. The main things people haven't seen are the ratings & ranking system and squadron support. Both will be in by the end of the year. that is not to say many things won't change. We make balance changes all the time (much to some beta testers' dismay) and tweak & polish things all the time, but there won't be any major, new features.
Since this is the first game of it's kind what was the hardest aspect to nail down during creation?
[joel] No question--the vision. Before i ever came on board, Rob had planned this incredible game, a huge superset of what it is now: rpg, fps, rts game, adventure, space action, co-op missions, etc, etc, etc. all mixed together. Incredibly cool. but about two years ago, we had to sit down and focus ourselves: "What do we really have time and resources to build?" That's when the game you see now was born. It took a while to get there, but it was a fun process. One of the most difficult things was designing by committee. We wanted everyone to have a say in the design process, but Rob was responsible for design. So Rob and I would brainstorm, write specs, brainstorm, write specs, brainstorm & write specs. then we'd have these HUGE, week-long meetings with the entire team where we'd discuss the suggested game features and things would just mushroom & blossom all over the place. You'd be amazed at how different the ideas were. Sometimes people agreed. Sometimes people disagreed. Sometimes civilly, sometimes not so civilly. :) Rob had the arduous task of taking all of the input in, digesting it, making the hard calls & decisions, and then reforming the specs. He was the vision holder. ugh. Just thinking about the process makes me weary. The design meetings were super fun, but it was hard work. That said, I absolutely loved that part of the project. (But i'm a masochist).
From our personal Allegiance combat experience with voice communication in game we can say that it offers a huge advantage, without any drain on gamespeed, are there any plans to add an in-game option for this?
[joel] Nope. There are lots of cool 3rd party packages that do this so we're content to let people just use those. Maybe in a future game. :) Note from ShadowLight: Ooooo, I like that last sentence!
PC Gamer mentioned that the different factions would be dramtically different from each other, can you elaborate on that and when those changes will hit the beta?
[joel] We don't know yet. We're trying to get the basic balance stabilized. Once that happens, we'll start to introduce differences in the factions.
Latency (Lag) in Allegiance is rarely, if ever a problem. Can you explain what measures were taken to ensure that lag wouldn't become a problem?
[curt] Glad you noticed! In most multiplayer games, there is a fair amount of coordination that has to go on between all the machines to keep the game "in sync". If machine B needs information from machine A before it can go on, then typically machine B (or worse--everyone) will pause until it gets what it needs. It's these waiting periods that typically create the lag. Obviously there's nothing anyone can do to make a byte get from machine A to machine B any faster, but we specifically designed our communication layer to minimize (and really, nearly eliminate) any situations that would require one machine to have to wait for another. Like most massively multiplayer games, Allegiance is server-based, but again, the client (Allegiance) is never forced to wait for the server. Not all server-based massively multiplayer games out there do this, and if you've played many, you know what I'm talking about. By never waiting on anything, this means that everyone can run the game at the max frame rate that their hardware will allow. In order to make this possible, we have to do a lot of prediction of the movement of all the ships out there. So every ship you see is actually in a predicted location and orientation. Unfortunately, however, with inherent Internet latencies as they are, nothing is perfect. In our case, warping is still a possibility, since we're predicting movement.
Fortunately, noticeable warping has proven to be pretty minimal and only occurs when the server doesn't get a update from a player for some long amount of time, like say 2 seconds, because their computer burped, net spike, or whatever. I think the worst form of warping is when your own player gets warped, so we decided to never do that--out of courtesy to those with lame Internet connections. :-) So if you're flying around with a bunch of ships and just one of them warps, that's a pretty good indication that their connection burped, but if all the ships warp, that's a pretty good indication that your connection burped. In either case, the game won't lag, as you noticed.
How much are the premuim servers going to cost per month? What advantages does that offer over the free servers?
[jon]
Short Answer: Cause it frickin' ROCKS!
Long Answer: The Allegiance Zone Club will cost $10 per month and will give players access to a whole host of new content, additional features, and special events for Allegiance. The new features will include the ability to play in large-scale (~200 player) arenas (some persistent. some custom-crafted events). In addition, the Club will give users access to new content, initially including 2 additional races with their own complete tehcnology trees. Over time, we'll continue to introduce new content in the Allegiance Zone Club, including new ships, weapons, tech, etc. All of the content additions in Allegiance will be tied to special events that are run exclusively in the Club.
I've seen some great ideas posted on the Newsgroups what's the process you guys go through to reject or accept these ideas?
[rob] (long) Its great to see peoples ideas about what they think would make the game better, balance issues, gameplay tweaks, new units etc. I've been very impressed with the quality of the suggestions we have received during the BETA....(actually its kinda scary how good some of the ideas are). When we see suggestions and feedback on the newsgroups the first thing Solap, Tkela and I do, typically is ask the following;
- Does this idea solve a real gameplay problem?
- What problems does this idea introduce?
- Does this fit into the higher gameplay goals for the game?
- If appropriate, how can we implement this?
1. Lets discuss a real example.. A problem we are concerned about right now is the scout rush. -- (Early on players send all players to enemy sector to kill the miners, the opposition never gets a foothold, this ends in a short frustrating game). We've seen a number of suggestions from the newsgroups on this including; anti-scout missiles, player controlled mining ships, faster miners, smarter miners, more miners, mining tenders, guns on miners, slowing down scouts etc. At a superficial level any of the ideas suggested could be a good potential fix, but we have to ask ourselves what is the real gameplay problem; is it that people don't know how to defend miners?, or perhaps Scouts are just too darn deadly against miners?, scouts are too tough? too agile?, miners are too wimpy?, or when actively trying to defend against scouts I cant protect them? all of the above?. A good solution to this problem will not be an isolated fix, and will not nerf down the scout role so much that it removes the fun of scout based miner harassment.
2. Typically every new idea breaks some previous assumption or idea. One suggestion weve had was to introduce more miners into the game..but this creates new problems for example; each new ship in the game consumes some amount of the data bandwidth available, so the more miners you have the less players you have.... [technical constraint] How do you keep track of more miners? [interface problem]. What implications does this have on the number of Helium rocks in a sector? How much helium in each rock? [balance problems]. Is managing more miners, more or less fun? [gameplay problem] . Obviously the less problems you generate with each proposed solution the better.
3. The higher goals for the game are things like "Best multiplayer space action game ever", "Action with Strategic elements'. If an idea comes along --- another example; player controlled mining ships, we have to ask ourselves does this satisfy all of the design goals for the game? Would our time be better spent implementing a different feature?. Player controlled mining ships might solve some balance problems in the short term, (players can at least scramble to get resources and use skill to avoid getting killed), but we want players to be engaged in combat with other players, not spending large amounts of time drilling in front of asteroids. (without alot of new work its going to be pretty dull).
4. On the implementation front. (this may be obvious to some of the coders out there) your suggestions and ideas tend to fall into one of two broad categories; 'new code' and 'new data'. New code is typically a new capability for a class of object, for example allowing a ship to dock with another ship (e.g.: to create a Carrier class ship.), (or allowing player controlled ships to mine.). New data is an adjustment to the loadout of a specific ship, or the basic attributes of some class of object, mines, weapons, missiles that do more/less damage, scouts that turn quicker, chaff that is more effective. etc. At this stage of the development process even though the product wont be on the shelves till Q2 00, we are anywhere from 95-100% done with all the code (depends on who you ask ;-). Adding 'code' or gameplay features such as the Carrier above is very hard. (not saying it wont happen). However changing or adding data (new toys) is significantly easier from a process perspective.
Bottom line -- Keep your ideas coming, this is only meant to help you understand the design process not to stop your creativity and ideas.
What games most influenced the design of Allegiance?
[rob] I think the biggest influences for me are Battlezone and AOE. For me BZ introduced the "Straction" concept that allowed players to play a great mix of strategy and action.. I just loved that sometimes you would be concerned with resources, sometimes considering strategic choices.. and other times just jumping into the cockpit and blasting stuff to bits. That flavor is certainly in Allegiance.. but of course now you can let your buddies handle specific aspects of the job whilst you focus.
AOE had a more subtle impact on the design through its consumption of team bandwidth. ;-) Although not many of the ideas in Allegiance owe anything specifically to Age, it was one of the most commonly sighted games in our design brainstorms... 'sure its just like the "foobar" in Age" people would suggest, hoping to add some cache to a design idea. Plenty of other games deserve a mention, X-Wing vs. Tie, Privateer, Starcraft and Wing Commander X amongst them. Although Tribes didn't influence our design much, we have plenty in common with them. Mostly Tribes helped demonstrate how much fun team based games could be, genuine thanks to the Dynamix guys for paving the way in this emerging genre.
[joeld] I'll add two cents here. Like Rob said, we were playing a huge amount of Age of Empires during the time when this game went from a "persistent universe rpg" to a "real time strategy game in space where humans play the units." Age was (I believe) a big factor which swayed us in that direction. But i'd also have to give a nod to Command & Conquer and Warcraft, both games which defined real-time strategy games and which turned made many of us into the RTS-heads that we are today.
Battlezone was an absolutely beautiful game. A wonderful game. We were psyched to see it come out, but pretty anxious because we knew we were both going for the action/strategy meld and we realized that if they didn't succeed, we also stood a chance of not succeeding. It was hard for people who never played Battlezone to understand what it was all about. We'll suffer a similar situation if we're not careful. One key thing we learned from Battlezone is that we had to separate the Commander & the pilot roles so that you didn't have to do both at the same time. We did this by specializing the roles: one dude plays commander, the rest play pilots, turret gunners, bombers, etc, etc...
We loved the notion of " quick voice chats" that tribes did so well. We used a similar approach but integrated the "quick comms" with our command infrastructure so players could send "quick comms" which could be either canned voice chats, or actual "commands" or both.
