September 8th, 2009
We all had a blast at PAX this weekend, but the fallout is rampant illness tearing through the Live team. With so many people out sick, we have decided to delay the scheduled skill balance until next week. Sorry guys! *heads back to bed*
-brought to you by the sniffling, fevered, coughing mess that was once called Linsey
August 28th, 2009
Live Team has been pretty quiet for the last few months, so we’ve built up a sizable list of things we want to talk about with the community. This is going to be a long one, but bear with me; I'll be discussing some important topics in detail. We’ve got a lot to cover, so let’s get started!
Welcome Robert Gee
First off, I am delighted to announce that we have filled the open design position on the Live Team! Robert Gee comes to us from Big Huge Games and the Guildhall at SMU. He is a big Guild Wars fan, having played since the first BWE's. He rivals even me when it comes to game play hours and general Guild Wars knowledge.
In addition to his deep understanding of the game, Robert brings his programming skills and a fresh new perspective to the team. Izzy is convinced that Robert is actually a robot from the future that is perfectly programmed to infiltrate our defenses. Unless (or until? dun dun dun) this proves to be true, I'll stick to just being wicked excited to have
a partner in crime another designer to bounce ideas around with. I hope you all give Robert a warm welcome to the wiki once he gets his page up and running.
The Anniversary Update – lessons learned
Back in January we talked a bit about our hopes for the year. Now, eight months later, I want to let you know how it's going so far. The Anniversary Update was a huge success, but it was clear to us that we had bitten off more than we could chew. Once the update went Live, we stepped back a bit to regroup, assess our process, and seek another designer. This took some time away from development, but it was important, and we learned a lot from the experience.
Here’s the biggest takeaway: we just don't have the manpower to produce maintenance builds, skill balance updates, festival builds, and large content updates all at the same time.
That was a pretty big revelation for us. It meant some big changes were needed. We started working on streamlining our update process with the hope that extra work now meant less work later. We began working on automating the base functionality of holiday festivals, weekend events, the monthly tonic changeover, and the monthly tournament map rotation. All of these things have previously been done manually every time and required rigorous testing to make sure no numbers were off and no files were missing. You’d be surprised at how much work each festival needs every time it runs even when we aren't adding new content to it. Automating these systems will free up time for other projects, reduce the number of maintenance builds, remove the need to do a build to start holiday festivals, and simplify the upkeep of GW for its caretakers. It's an exciting project for us because of the effect it will have on our day-to-day work.
Maintenance and Content Updates – new process, new schedule
We’re going to slow things down by switching to a longer development cycle for each skill balance. We'll be shooting for a skill balance every other month instead of every month.
Skill balance updates take a surprising amount of development time, but because much of this is necessary to the process, simply cutting back on how much time we devote to them isn't a good option. It takes a lot of research to properly craft a skill balance update. A lot. We read the forums, read the wiki, review the meta, watch in observer mode, discuss skill balance with players, and most importantly, play the game ourselves. It could easily be someone's entire full-time job or the work of a whole team of dedicated balancers. Trying to do a skill balance every month, along with everything else, rushes the process more than we'd like, and it doesn’t give us enough time to truly polish each update before it goes Live.
The switch to a bimonthly schedule allows us to maintain the quality of our work, and—this is important—it allows us to gather community feedback about the skill balance updates before they go Live.
To do this, we’ll try to post the update notes/dev update for each skill balance about a month before the update actually goes Live and ask you for your feedback. The Test Krewe will be able to try out some of the changes discussed by the community in the testing environment and report their findings. This process will give the community the opportunity to give us feedback and identify problems before the final changes are made. We'll take your feedback, make adjustments where necessary, and release a revised set of update notes/dev update when the build goes Live. Our hope is that by taking some extra time to polish and get feedback from the community, our skill balances will be better tuned to smooth out any issues without creating new ones.
At first, we thought that releasing larger yet more infrequent content updates would allow us to tackle the bigger projects that didn't seem possible on a faster-paced schedule. We’ve found that our schedule is still as fast paced as ever. That doesn't stop us from taking on bigger projects; we just aren't releasing the updates as quickly or as often. So we’re going to ease off that kind of schedule in favor of something a little more fluid. We will still prioritize our projects with a content update in mind, but they will be generally smaller in scope than the Anniversary Update, and we won't be afraid to do them in close succession should it be warranted. For instance, we would like to release a content update in the next month and a half, but we'd also like to add new content to Halloween and/or Wintersday.
Xunlai Tournament House Update
We now have an accurate list of XTH winners from the May monthly predictions and are in the planning stages to distribute those points. We’ll need to deploy some new code to upgrade the system, which requires a bit of delicate timing. Part of this deployment must occur through a build, but some of it will happen automatically on the servers. During this time, if anyone speaks to Kun Shao, it could disrupt the system and garble up the list.
To address the issue, we’ll be doing this in three stages: first, a Live build to remove Kun Shao from outposts to prevent him from interfering; then, a second build to deploy the upgraded code and distribute the points; and finally, a third build to reintroduce Kun Shao so that you may redeem those points. As a thank you for all your patience, we’ll be giving everyone on the distribution list a free bump to the next-highest reward bracket.
At this point you may be saying, "Cool, thanks for the extra points, Linsey, but when will the XTH come back online so I can make my predictions?"
Well, that’s still up in the air. We believe that the bug which caused errors in the distribution of points has been identified and fixed, so we’re now preparing to move into the testing phase. We’ll be simulating the Live environment internally and putting the system through rigorous stress testing to see if the fix took. If it did, then yay happy fun time, the XTH will go back up, and hopefully we'll never have to touch it again.
However… (You knew there would be a "however," didn’t you?) Should it fail the stress tests, it would mean the fix didn’t take because the bug goes deeper than we hope, and that would be Very Bad. If that is the case, we will be faced with a high probability that this bug is not something we can fix. Nor can we redeploy the Xunlai Tournament House with this database bug outstanding. So, let’s all keep our fingers crossed, and I'll keep you updated on how the testing is going once it gets underway.
The PvP Love Update
Meanwhile, work continues on our second content update for the year. Right now, we have a list of over a dozen projects that we'd like to put into this build, though I expect this list to shrink considerably as we get closer to lockdown.
Here are some of the projects on our list:
- The Henchmen Project - Restrict heroes from PvP and replace them with an array of new henchmen using skill bars that have been chosen from builds submitted by the community. We’re also looking at the PvE henchmen and revising their builds as needed.
- PvP Skins - Reduce the TRP cost of PvP skin upgrades, make them purchasable with Zcoins, and (fingers crossed) fix the graphical glitch that prevents us from releasing the Eye of the North skins.
- GvG Tiebreaker - We've been watching and discussing the Mind Blast ele build for some time now. We could just nerf Mind Blast to get rid of it, but the Ranger or Paragon alternatives would then step into its place and nothing will actually be fixed. The tiebreaker mechanic is the heart of this problem. We’ll continue to work on adjustments to the tiebreaker so that these solo-plinkers won't be so powerful.
- New Schedule for Automated Tournaments - The current schedule's automation isn't finely tuned to provide tournaments during the peak hours of both Europe and North America. It also doesn't really provide an opportunity for the two continents to face each other in a tournament outside of the Monthly. Happily, the new schedule will be more accommodating to both player bases.
- Bug Fixes for Fort Aspenwood and Jade Quarry - We’ve identified a number of AI bugs in both formats—it's just a matter of getting the time for Joe to dig into them and see what can be done to fix them.
- The Hall of Heroes - We have worked on a new Hall of Heroes map that is tailored to its three rotating objectives. This map still needs to be tested for bugs as well as play balance, so we’ll probably wait until the Test Krewe is in place to conduct this testing before releasing it.
- Syncing Random - Our server programmer has been investigating how to fix this age-old problem of organized teams appearing in Random Arena, and we’re hoping to have the results in time for this build.
(TA + HB)^Sealed = WIN!
This is a pretty big one, so we really wanted to warn everyone of this in advance. The plan is to remove both the Hero Battles and Team Arena PvP formats and replace them with a 4v4 Sealed Deck format.
Let's talk a bit about the reasons for this. We recognize that the Hero Battles format has reached a state of acute distress. It has always been a niche format with a player base comparable to that of the Ascalon Academy, despite its tournament support. This is largely due to flaws in the core mechanics of the format, compounded by years without skill balances to keep it in check.
Under these circumstances, we’ve seen this player base react in extreme ways to show us the flaws of the format. We’ve seen a metagame in which rolling dice or playing rock-paper-scissors has determined the outcome of a match. The final round of the Monthly Tournament was purposefully thrown month after month to send us a message.
After the Anniversary Update, we decided to take on the problems of Hero Battles for the next content update. Many different options were discussed, but unless we were talking about making major changes to the format, every option seemed like mere band-aids that wouldn’t truly fix the core issue. As a small team, it's not healthy or wise for us to invest our limited time in projects which just create more work for no substantive effect. With that in mind, we focused our attention on ways to change the format and generated a bunch of ideas for this. A simple truth emerged: to fix Hero Battles, we would need to virtually redesign the entire format.
We’d be spending a huge amount of time creating a new format to replace Hero Battles, with no guarantee that it would actually fix anything or have a player base to support the new format. That just doesn't sound awesome to me. People who frequent my wiki talk know I am uncomfortable with the idea of adding any new formats which could further fracture the community. If we’re going to invest a lot of time into designing a new format, I want it to be self-sustaining (going back to that whole "not creating more work" thing) and sure to have a player base.
This is where Sealed Deck came in. We’ve wanted to work the Sealed Deck format that was developed for past PAX tournaments into the game, knowing that a lot of players have been playing it on their own for some time now.
Now set that stuff aside for a moment and let's talk about Team Arena real quick.
Team Arena is another format that has sorely needed some love and attention. Unlike Hero Battles, Team Arena was once very popular, but a competitive atmosphere and a degenerate metagame have caused the player base to dwindle a great deal. We'd like to give the format more support, but we don't have the bandwidth to take on more balance maintenance. Now with Sealed on the table, it seemed like the perfect opportunity to give Team Arena some lovin' and remove its degenerate meta in the process.
Here’s the plan we came up with:
- The Hero Battles Ladder will freeze and there will be no more matches in the format.
- The maps used in Hero Battles will be modified and reused, including the outpost.
- Gaining points in the Commander title will no longer be possible. You’ll still be able to display it, and it will count in the Hall of Monuments and towards the Kind of a Big Deal title track.
- Random Arena will stop promoting to Team Arena after 10 wins.
- Team Arena will be renamed and repurposed to support the new Sealed Deck format.
- Sealed Deck will be implemented with new UI and revised rules aimed at making pick-up group formation a little easier to offset the limited nature of the format.
- There will be a new title for this new format. This title will be the counterpart of the Commander title; either will satisfy a single requirement towards the Hall of Monuments and the Kind of a Big Deal title track, but not both at once. It’s a one-or-the-other kind of thing.
- We hope to have customizable Sealed Deck UI so that players can refine the rule set for scrimmages or player-run tournaments, but this isn’t expected to be done for the initial release of the format.
- If the new format is successful, we’ll consider investing the development time to build a ladder and tournament system for it.
While we’re sad to see Hero Battles and Team Arena fade away, we feel that this plan has a lot of win/win and will ultimately be good for the health of the game. We get to replace two formats that suffer from complicated issues and dwindling player bases with one new format that we already know is fun, popular, and will be easier to support and maintain. Not only that, but because of the random nature of Sealed Deck, it’s a format that has a lot of longevity built into it. A project like this is exactly the kind of thing that Live Team should put resources into, one which eliminates outstanding issues while providing content to increase the game’s longevity.
At this point, you may be wondering just when all this will be happening.
We are going to wait until after September's Monthly Tournament to make these changes. We want the Hero Battles community to get one last full season now that they know it will be shut down. There will still be time to get those last few points for the next rank of Commander, climb the ladder before it is frozen, and shoot for being the last person to ever win a Hero Battles tournament.
A New Approach
I know this is a lot of information to take in, and it comes in a package that is sort of unusual in the game industry. At the beginning of the year, we spoke about wanting to be more open about what we’re working on and set this as a high priority in the months that followed. We had a lot of plans for how to accomplish this, but as they say, "The best laid plans of mice and men often go astray.”
We’ve decided to try a slightly different approach, starting with this post. As you can see, we are relaxing our stance on talking about projects that are in progress and projects that are just getting going. We’ve launched Facebook and Twitter pages to give you quick updates on what we are up to. Taking a less formal approach like this will work easier with our fast-paced schedule and will hopefully increase the flow of information to you.
This degree of transparency is unusual for large game companies, being more common to smaller studios and indie projects. We’ll be sharing information and offering insights into the development process that the public normally doesn’t have access to for AAA games. This kind of stuff goes on behind the scenes anyway; we’re just pulling the curtain back to show you the process. To make this new approach successful, our community needs to meet us halfway. We all have to be on the same side here.
So, in return for this openness, we expect you, our community, to be understanding when projects that we have discussed may be canceled or changed, or a project or build is delayed. We want you to have ambitious expectations for us, but to understand that there are always limitations. Everyone has to deal with finite resources, competing priorities, and unexpected complications, and we’re no different. By embracing this reality, we can work together to bridge the gap between development and community for the betterment of the game.
I’m looking forward to chatting about my work openly, but as always please keep in mind that there will be lulls in activity from time to time.
My hope for the rest of the year is that Live Team will field many of the key outstanding issues that have piled up by the end of 2009, and start 2010 fresh with new challenges and goals. I’m looking forward to the future of Tyria and developing content that helps it continue to grow. That's the fun stuff!
We expect people to have a lot of questions about everything I’ve talked about, so we'll keep our lines of communication open and do our best to answer them. Live Team will be at PAX this year and will have Q&A sessions at the ArenaNet booth on Saturday @1pm and Sunday @Noon. I'm sure most of us will also be hanging out at the booth periodically throughout the course of the convention as well. We’ll see you there!