I still can't believe it's really happening. World of Warcraft Classic! Blizzard has remade WoW as it was all those years ago when it launched (don't look at the dates, it's a bit depressing - where did all that life go eh? Nervous laugh). But here we are, one month to go. One month until we can play old WoW anew. Will the experiment work? Will we actually want to play it? Will all the yearning and petitions and hoo-ha have been worth it? Or will we discover it was a bit boring after all?
As the doors close on the beta, and as the team knuckles down in preparation for launch, I take lead software engineer Ryan Birmingham, and senior software engineer Omar Gonzalez, aside for a chat.
Thank you for joining me. I've just been watching videos of the final moments of the WoW Classic beta, of Ragnaros stamping on everyone in Orgrimmar, and it was joyous! What a great idea to spawn him in from Molten Core - who came up with it?
Brian Birmingham: That was one of our live-ops producers - it was a great idea of his. We really liked being able to go back to the feeling of small-community - where everybody knows each other - that we had in the original launch of World of Warcraft. We did crazy things on the servers as they were shutting down during the [original] beta test so we wanted to do some kind of throwback to [them].
He had the idea of, 'What if we spawned Ragnaros in Stormwind and Orgrimmar?' and, 'What if we follow that up with the Guardians of Blizzard and have them run across the world, chasing everybody down and killing as we wipe out the servers and turn them off?' It was a lot of fun.
Omar Gonzalez: We wanted to create something really memorable. We thought giving everyone Thunderfury as drops they could pick up-
Brian Birmingham: It was a bit harder than we expected for people to kill the Sons of Flame and get the Thunderfuries we put out there, but we did see some people running around with them.
I remember people trying to get Thunderfury for the first time! (It was a very long, convoluted and expensive process, and therefore Thunderfury weapons - one of only a few Legendary weapons in the game at the time - were very, very rare.)
Brian Birmingham: Oh yeah - I was one of those people! I raided Molten Core for so long back in the day. I was in the 'one-binding club' where you got one of them but not the second one [you needed both to complete the quest]. Every time we killed Baron Geddon I was like, 'Please, please let the second one drop!' Never did - it never did!
I was obsessed with Ragnaros in Molten Core all those years ago and I love seeing him come back to prominence - it was bizarre being able to solo him because of all the level-cap increases! But those old raids aren't new to anyone any more, so what do you do about counting world-first boss kills in Classic WoW? Do they override the original scores? That's not fair, surely - people already know the strategies!
Brian Birmingham: We don't count them - that's a community feature.
Omar Gonzalez: We see they happen so we're aware of them...
Brian Birmingham: It's certainly still a race for people who want to do that, but I would encourage people to do it at their own pace. There's still a challenge there - we've all seen it, we all know how to do it but there's still a long road to go. Which strategy for optimising your gear? Which levelling routes? How to invest in, and gear up, each character? There's still going to be a bit of a debate. We've seen a lot of surprising results where people thought they knew how things worked back in Classic - but don't.
What kind of thing?
Brian Birmingham: We made a post on our forums about exact hit-rate calculation. Some of the fun is trying to figure out the minutiae of these combat equations. The hit-chance against a boss creature was eight or nine per cent, no one could figure it out, so we confirmed it was eight per cent. But people were pointing out, 'Oh but I have plus-eight per cent hit-chance on my gear and I'm still missing the boss - you guys must have it wrong,' and we dug in there and said, 'Actually, if a boss is three levels higher than you, it ignores your first one per cent bonus to hit, so it's eight per cent chance to miss, and your first point of plus-one per cent chance to hit is totally ignored.
It's really detailed minutiae but it was this edge case that existed back then, and we made sure we went back and restored those things.
Omar Gonzalez: We have a tremendous community of incredibly dedicated fans, especially when trying to discover these combat mechanics. There was a recent issue with crit percentage chance. We didn't release the numbers, the actual math, so they sat there for whole weekends punching turtles! For hours and hours on end! To generate thousands and thousands of these combat log entries, which they did a tremendous amount of maths on and ended up deriving very accurate numbers through sheer force of will. It's that kind of community and devotion that fuels our devotion and passion.
I assumed people would already have those calculations from way back when!
Brian Birmingham: They have them almost exactly right. The mechanics have changed in all these expansions, so as people have gotten better and better at figuring equations out, we've changed them over time as we've made different design decisions.
When we roll things all the way back to Classic, they have the best guesses they had at the time, and as they keep working on things and testing things, they are going to get closer and closer. They have better tools today. I'm sure they will eventually figure it out.
But going back to the original question - how that affects the race to world first - there is still a race there if people want to have it. We will have a phased content roll-out so Ragnaros will be available at the beginning, on August 27th, when we release, but the later raids like Blackwing Lair, Ahn'Qiraj and Naxxramas, we'll release in later phased content roll-outs.
OK so let's make a Ragnaros prediction! How long until someone downs him?
Brian Birmingham: Ha ha! Oh I'm sure it will take years...
Omar Gonzalez: Ha ha - so hard! Well, we actually have a little pool going on around the office. I'm not going to say who picked what but it's definitely of interest to us as well - a little bit of fun we like to engage in.
Brian Birmingham: I think I have the longest number! I'm betting on Rag - Omar and [the others] are betting on the players!
So, with the closed beta just finished, what have you been noticing about how people have been playing it? A lot of time has passed since the original launch. Are people's habits different, or are people slipping easily back into rhythms they used to have? What have you seen?
Brian Birmingham: One thing I noticed was just how passionate people were about the beta. Our expectation for betas is generally we'll get a lot of people come in, try things for a little bit, and then they'll say, 'OK, I'm going to wait until release now - I've helped you guys test and now I'm finished.' We were really thrilled and blown away by just how much people wanted to play, night after night, even though they know these characters are going to get wiped out before we release. It really blew us away how dedicated people were.
Omar Gonzalez: I was really impressed with the varied ways players would engage with the game. A lot of our fans are very dedicated mathematicians or statisticians, and they sat there and figured out all the combat math. Other players were very exploratory and wanted to engage in the world, so they wandered into higher level areas to poke and prod. Some players wanted to just hang out in the capital cities and meet other players. Others levelled up repeatedly, or helped other player level up. There was this great variation even though this was temporary, which we did not expect but we love very much.
Has there been any unexpected feedback from the beta - any issues you've seen and want to tell people you're aware of and are addressing? For instance, I've seen people talking a fair bit about layering and opinions are divided on whether it's working.
[Layering is a new way for Blizzard to spread player load. Imagine a glass of water on a tray: the tray is the server, the glass is the layer. When the glass fills up, another glass can be added but the water pools remain separate. Layering allows WoW servers to hold multiple pools of players, allowing them to scale up and down invisibly. But how many players each pool holds remains a bit of a question mark. And there appears to be potential for exploiting layer-crossing, not that I understand it completely. There's a lot more information in a thread about layering on the World of Warcraft forum.]
Brian Birmingham: The only thing I really want to call out on layering is a lot of people were wondering if the population thresholds in the beta were accurate, and they're not. There were some people who said, 'Did they just turn this on to test it functionally to make sure it worked?' That's accurate. We set the thresholds much lower on the beta than we would for live because we wanted to make sure we were actually testing the feature. There were other things people pointed out as bugs we were not expecting.
Omar Gonzalez: There were hunter animation issues. There have been a lot of changes we had to undo so hunter gameplay was accurate to how it originally was - not just in combat mechanics but movement aesthetics and animation.
Brian Birmingham: There are a lot of subtle things our hunter community has been very helpful to point out!
Good old hunters.
[They both laugh.]
What do you do now between beta closing and WoW Classic launch, on 27th August?
Omar Gonzalez: Bug smash!
Brian Birmingham: That's exactly right! We've collected all these bugs so now what we're doing is prioritising and crushing them as quickly as we can. We're trying to make sure we get the most impactful and obvious bugs - the most combat-affecting and aesthetic-affecting - and make sure we push those to the front and focus on those, and get them crushed so we can get a stable and, hopefully, authentic-feeling release.
Funny you should mention 'authentic-feeling'! One thing I remember about original WoW launch was queues, lots of queues! That's the authentic experience. It would be fitting, wouldn't it, if WoW Classic had the same?
Omar Gonzalez: Ha ha.
What are you doing to prepare for launch load - do you actually expect any queues?
Brian Birmingham: Our goal is to get as many people into the game and playing as possible. That's what we want - we don't want people to wait in a queue. But part of this is a problem of prediction - how many people are going to show up?
We're trying, just like we did in 2004, to make our best guesses in terms of what we're going to see and make sure we have plans in place to prepare for those so we can minimise the queues. If players blow us away again, there will be queues until we can fix something.
We are better prepared than we were in 2004 to respond to that situation, so if there are queues, I wouldn't expect them to last for very long - in terms of days or weeks after launch. Launch day is always craziest. Hopefully if you see a queue on launch day, you won't in the next couple of days or weeks.
How big is the WoW Classic launch in comparison to something like a WoW expansion? Is it different internally?
Omar Gonzalez: There's some differences and similarities. We will likely set up a war room - like NASA Mission Control - where we have all hands on deck. On one hand, it's a little bit simpler, because what players will be doing is more known. But on the other hand, it's much more complicated because we've built out, effectively, a whole new infrastructure. That's something we typically don't do for an expansion. An expansion launch is much more of a software upgrade. Our Classic [launch] will include hardware deploys in all of the various regions - something we haven't done in almost 15 years.
Brian Birmingham: Definitely, in terms of rough sizing, we are treating it as the same level of event internally as a major expansion launch. It's all hands on deck. The whole World of Warcraft team is lending a hand, making sure we're good to go.
Make sure you order lots of pizza!
[They laugh a knowing laugh.]
So what happens in terms of scaling up and down after launch? If loads of people flood in, how quickly can you scale up, and conversely, if the Classic novelty quickly wears off, how quickly can you scale down?
Brian Birmingham: That's exactly what makes it a complicated problem - we don't know. We are trying to make sure we have various controls in place to scale in either direction. We might have both things happen: more people show up on launch day, and we have to expand capacity, and then as players distribute throughout the world and play less than every day... We do have controls to go in both directions.
Omar Gonzalez: Our systems are much more evolved than they were when we launched 15 years ago! We have a much more flexible infrastructure able to respond more quickly.
Original WoW had a lot of problems with gold farming [the umbrella term used to refer to the black market practice of people - not Blizzard - selling World of Warcraft gold]. Now, of course, you can legitimately buy gold in World of Warcraft using the Token system. To clarify: is this available in Classic?
Brian Birmingham: We do not have it available, although players have pointed out that since it's a shared subscription, if you wanted to farm a bunch of gold in modern WoW and use that to re-up your subscription, that does get you another month of Classic WoW access as well.
So do you expect gold-farming issues again?
Brian Birmingham: As long as a human is at the keyboard and they're actually running around, doing things... Farming materials for your raids - farming gold, farming herbs - were activities that were OK, they were normal gameplay. Really, where we get concerned, is when someone has got some kind of automation program doing the gameplay for them. That's against our terms of service and we have better detection algorithms and techniques today than we had before, so we hope that will be sufficient to keep it under control.
Going forward, will WoW Classic have a dedicated post-launch development team? How is it carved up internally?
Brian Birmingham: I don't want to get too much into specifics but we are part of the World of Warcraft team. It is the World of Warcraft team shipping Classic but there are dedicated resources.
Ion Hazzikostas [game director of World of Warcraft] said, at BlizzCon 2018, there would be four phases of WoW Classic content patches. But that was back in November - has anything changed?
Brian Birmingham: That has changed - I'm glad you asked. It is no longer four phases, it is six.
Phase one is the launch phase, and Ragnaros and Onyxia are going to be available. Phase two, we're going to unlock Dire Maul and the PvP Honour system - you'll be able to kill people in the world during phase one but there won't be an Honour system tracking it. Phase three is going to be the first two Battlegrounds, Alterac Valley and Warsong Gulch, alongside Blackwing Lair. In Phase four, we release Arathi Basin alongside Zul'Gurub and the green dragons. Phase five is Ahn'Qiraj and then, finally, phase six is Naxxramas.
Part of the reason we changed that is we had a lot of feedback from the community four [phases] was too few, and there were distinct periods of time they wanted to recapture - where certain content wasn't available or certain rewards. It would trivialise the content. For example, people were concerned if you got [Battleground reputation rewards] too early, it would make it trivial to go through Molten Core and Onyxia - and they're right.
What happens when the six phases are done - what happens with expansion roll-out?
Brian Birmingham: I don't know that we have any plans right now. We're focused entirely on WoW Classic and making sure it is going to be as correct and authentic as possible, and we're really excited to see what fans say when they're experiencing it. We started this whole project as a love-letter to the fans - they've been demanding this from us. We wanted to respond and give them what they've been asking us for. We will continue to listen to community feedback, and we want to know what people think and want to see next.