The Big Crunch
As Diablo II's development rolled into 1999, Blizzard North pushed hard to have the game out by the end of the year. But no matter how close the end seemed, it always remained just out of reach. As the months passed, they found themselves mired in what seemed like a neverending crunch.
It was probably June of '99 that we started the crunch in earnest, the real, seven days a week, all waking hours, driving home at midnight and coming back at eight.
With no real deadline, and no real way that we were managing our time, because we didn't estimate any tasks or anything like that, we just didn't estimate our time correctly. And the thing that sucked about that, though, was that we knew we weren't going to make it at our current rate, but we believed that we could do it if we just started crunching, we believed that we could make the end of 1999. And so, we started working really hard on meeting that deadline, and we started working, I was working every day. I took off, the last year of Diablo 2, I took off four days the whole year. I worked every other day, and most days I averaged about 14 hours a day.
So, we worked incredibly hard to try to get this thing out by November or December that year, and so we were working, I don't know, we were crunching for six months, or something like that. We started in, like, May. By October, the end of September, early October, I remember the phone call with Mike saying, "You know, you guys are not going to make it, and we're going to delay this thing to next year." I couldn't accept it, I couldn't believe it, it didn't really register. It was like, "That's impossible, we're going to make it. We're going to do this, we're close." But, somebody who has a view from an outside perspective sees how far away you really are.
So, they said, "We're going to take a few extra months. Three months, four months, six months, whatever. We're going to try and do this, but you guys need to continue and kind of finish it up." So we ended up extending the crunch. I took off Christmas Day and I came back the next day, and things like that. It was really brutal.
We were already going to go overtime. We knew we were taking too long. We were probably already over our estimated time, but nobody really cared, because the game was going really well. But I'm pretty sure it was '99, June, maybe even May, we said, "OK, we got to crunch to get this game done by the end of this year, by the end of '99." And so, we worked for four months, just really crunching too hard, and it took Blizzard South, it took this strike team to say, "You know what, you guys are not going to make it. This is not going to happen this year." That was super depressing, because we had been working so hard for four months, and we argued, but it took a day or two to sink in, that yeah, we're not going to make '99, we were crunching almost for nothing. Sure, we got a bunch of stuff done in this time, but, we're burned out and we're not going to make it, and that was a pretty depressing moment. We decided, OK, we're going to just work normally until the end of the year and get going again at the beginning of 2000. So that was my biggest moment of realizing, "Uh-oh, we're not even close. This is going to go a long time." It was a very depressing moment.
We were doing everything. "We got to do this, that, and the other thing. We have to get this fixed, and this has to happen." So, we were running around with our hair on fire for a year, and we were trying to- There was nothing that was one individual thing that we felt was going to hold us back. "Oh, this would have been complete if it weren't just for whatever. Oh, we were all sitting around just adding more content to the game, because we were all waiting on the Battle.net server to work," or something like that. There was nothing like that. Every part of it was behind schedule. Content was behind schedule. Monsters were behind schedule. Levels, story, cinematics, technology, it was all behind.
"We were already going to go overtime. We knew we were taking too long. We were probably already over our estimated time, but nobody really cared, because the game was going really well." - Erich Schaefer
The game was always playable. So, as we put in new levels and things like that, you could play it every day and iterate on it, start your character out and play stuff. So you could play up to the point where the content kind of stopped, and so it gave us the chance to go back and actually complete the content. It wasn't one thing in particular, it was that the content was only so far. Like, Hell didn't exist, I don't think, by the end of 1999. I think that we put that all in in 2000. So, that whole act, and that whole section there. I don't think that it started going in until February or March, or something like that, if I remember correctly.
It was the assets, and it was the iteration on it to make it play well, and then the balance of all the skills, and of course, now we're in Act 3, and you have to do a whole other part of the skill tree, playing in the actual levels that it's going to be requires going back and revisiting the skill tree again. So every time we'd put something in, you kind of had to do this recursive look through everything you've made to that point and make the proper adjustments. It was grueling. We didn't have development tools like we do nowadays. Everything was a gruel. Getting assets into the game was done by hand by an engineer, and nothing was easy. Nothing came easy.
I remember I was making the interface for [the skills trees]. There was a stone slab, and the skills would all fit into these sockets. And I thought, "OK, we're basically done, I'll just make this skill tree and this stone slab, and then we will fill it out with skills". At least ten, twenty times, almost per character, I'd have to redo the stone slab with different positions for the skills and different hierarchies between them. And we never had a tool to do this, so I had to render out a whole new stone slab every time we changed a single thing. I made a ton of stone skill tree slabs art over and over and over that all just got thrown away. So, that was a great example, if we had better tools, or if we knew what the process was going to be, I wouldn't have put all that work into it.
I think people were loopy. We worked really hard and we worked late, and people played music, and they would go out and get a dinner and come back. Then sometimes we would do things like, "Oh, it's midnight. Farscape is on, so let's watch that for an hour," or something like that. It was 95 percent work, but there were little breaks here and there. There was a lot of office camaraderie around the late nights. Not everybody stayed late, but there was a good crew of, let's say ten people in particular that would stay late all the time. Some people would still stay late, don't get me wrong, stay until 9pm or 10pm, but that wasn't the 2am crew, or whatever. And then I'd get up at 6am and go to work and we'd do it all again.
We were always a couple months away, so if we just grind it out hard we'll be done with it. But it was mentally and physically exhausting. And people started to break down. People were getting sick all the time, they were breaking up with their girlfriends. It was bad. It was physically hard.
It was physically hard. People would be sweating and you can just see the stress on their faces. We barely had relief. If someone had to take a week off or couldn't stay a few nights, other people would say, "Well, why aren't those people staying? We've got to stay." And we, as managers, as the people in charge, we didn't have good answers for that. We didn't handle it very well.
It wasn't all the worst times. Again, there was a lot of camaraderie, and there was good times to be had amongst this, but it's probably comparable to people on the front lines of a war. They get drunk and have a good time once in a while, but then it's back to the grind in the morning. What made it OK for me, and this is different for a lot of people, but I knew the game was going to be great. I knew we were on the right track. So, it was worth it. But that was not true for everybody. There were a lot of doubters who thought, well, I don't really like how it's going, either because they're stressed out or because they really didn't like it. So for those people, it would be even harder. I had total confidence, and that's what made it OK, that there was going to be an end and it's going to be great. But for people who didn't have that, it made it even worse.
For me, personally, a really weird experience is that I got married in May of 2000, which we thought was a very safe day to set the ceremony. And it turned out, it wasn't a safe date. We were working right through that time. And so, my poor wife had to handle all the arrangements and took on all the stress of the wedding. A lot of people picked up, my parents helped out a lot. But it was very odd that my wedding was coming up in the middle of this horrible crunch. So, at one point, I was just like, "OK, I'm leaving work, and I'm going to get my tuxedo, and the next day I'm getting married." I almost couldn't even think about the wedding. And then we kind of knew we were about done, so we were going to do a honeymoon at the end of June. And, we left on our honeymoon the week before we shipped. Now, all my stuff was really done, and it was just technical support things. But I just felt so weird. I've been here this whole time, and here at the very last days, I'm just taking off and leaving the crew behind.
Launch: Critics, Bugs, and the Stone of Jordan
Diablo II finally launched in the U.S. on June 29th, and in Europe on June 30th. The team wasn't finished yet, though. They still had an expansion to make, and there were bugs and hackers to deal with, too.
I was in Paris when Diablo II released. And on the Champs-Elysees there was big banners, and the big Virgin megastore had a huge Diablo display. There were even TV commercials for Diablo in Paris. It all just felt so cool.
We knew immediately that it was selling really well. And then soon after, we learned it was selling really well in Korea. So, that was all great to hear, even though we were still fighting a lot of these technical issues. That would fall much more on Dave and the Battle.net team staff. Max and I were more on the creative end of things, and pretty much had washed our hands of the whole project. So it was easier for us than a lot of those guys, I'm sure.
The only thing I remember… there may have been some problems on launch day. We finally got those solved, and we brought everything back up, and got things running, and then there was this gold duping exploit, like, with splitting piles of gold. And I just remember it, day two, I felt like the whole thing that we had worked for, where we were doing this client server thing where everybody could be protected, and then we had blown our economy, or whatever, on like, Day 2. We were like, "Oh, my God, what did we do, we've blown it."
And so, it was just this emergency fix, getting those repaired and the exploit fixed, and stuff like that. Those are sort of the first memories I have of it. It goes straight into the frying pan of, you're in a live environment, and things are broken and you've got to fix them as fast as possible. Which was a new kind of concept for myself and the team, when you're doing these live products that are always on, that you have to respond to things in a much different way than you did with single player games, right? It's like, there's an exploit and we'll patch that at some point. We'll work on it over the next couple weeks, and maybe we'll put out a patch by the end of the month that fixes it. But if people want to destroy their game, I guess they can go ahead and do that. But most people won't.
That was kind of the mentality, and then there was even less of a mentality of that if there was a bug or an exploit like that on a cartridge game or on a physical CD console game, you know, they weren't fixed ever. So, now, things are put out and they're patched regularly, but that was very uncommon practice back then. It was really unusual. One of the things that really separated Blizzard from a lot of the pack was the fact that we continued to support and patch our products and get rid of bugs and problems and include enhanced features and stuff like that even after the game launched.
Yeah, I think the first week was kind of a disaster, wasn't it? I was sort of luckily gone on this honeymoon.
Yeah, it was a disaster, but it was a good disaster, because it was over capacity, and that's why we were having the problems.
I think by now, everyone has their cheating Diablo 2 story. They don't say it as if Diablo's a crap game because there's cheating, now they kind of talk about the fun and the weird experiences with the cheating. So, it doesn't bug me anymore.
At some point the Stone of Jordan became a kind of unit of currency. We we made a bunch of items, and sometimes when you're making items, you're making ones that are better than others, and [the Stone of Jordan] was deemed the most valuable item in the game. Even though it wasn't necessarily true, it was deemed as a rare item to find, as well as extremely valuable. So people started trading things for, and it would cost Stones of Jordan. But then people started duping Stones of Jordan, so Stones of Jordan were everywhere, and there were lots of them in the game, because there were lots of duping bugs and things like that.
"We finally got the [initial problems] solved, and we brought everything back up, and got things running, and then there was this gold duping exploit, like, with splitting piles of gold. And I just remember it, day two, I felt like the whole thing that we had worked for, where we were doing this client server thing where everybody could be protected, and then we had blown our economy, or whatever, on like, Day 2. We were like, "Oh, my God, what did we do, we've blown it." - David Brevik
Then we fixed the duping bugs, but there was still a lot of them out there. So we came up with a way to remove Stones of Jordan out of the economy, a sort of Stone of Jordan sink to get rid of them, so that they weren't all over the place any more. And that was this donation thing, where you could donate the Stone of Jordan and summon a Super Diablo or something like that. People today still, when they think of items in Diablo 2, it's the number one item that comes to mind, just because it was so valuable, and then used mainly as a currency for trading, and then just became the most popular duped item, and then there were just gobs of them in the economy.
We were surprised when the SoJ became a unit of currency, but we were happy to see that dynamic occur. I always thought the gold cap was a bit too blunt an instrument, and it was neat to see players figure out this work-around.
It was a learning experience. It was the first time we had done client-server stuff, and so some of the mistakes that we made with the way that we were doing left too many holes, and items were being generated, and we figured out a way to generate unique IDs for them, but that wasn't necessarily in the code at the beginning. And, trying to improve on those systems to make duping more difficult and fixing all the flaws and the places you could cancel, and actually transferring the items instead of destroying them and recreating them.
When you're doing things like that, when you end up in a situation where one thing can get destroyed and another thing can get made, then the thing destroyed doesn't get destroyed because the player disconnected before that gets saved. There's all sorts of tricks that occur when you destroy the item and assume that it gets destroyed before you create the other version of it. So, it becomes a situation that's technically very challenging, but there's a lot that we learned there to make it more secure on future games that I worked on.
But again, I really feel like when people mention Diablo 2, the number one item that comes to mind is the Stone of Jordan. And, even though it became this big duped thing, it was a staple, it was the item to find, the item to seek. You were always happy to get one, and it became super-popular. So, in a lot of ways, it legitimized your account once you had them. And so I still feel like today that people still seek that item, even though maybe it's not the best item in the world.
We were so in the infancy of this type of game at that point that no one could have anticipated the level of effort people put into cheating. I mean, we kind of did because we went to client-server as a model, but it's so complicated to do what we were doing that there's all kinds of little vulnerabilities all over the place, and, for the most part, you could play and avoid cheating problems. And, when people did figure out some horrible way to cheat, eventually it would get fixed by the Battle.net team. And certainly, we've learned a lot from there that we apply these days to help cheating issues. But at the time, it just kind of came with the territory of making some of the first really giant global internet games where there's thousands of people trying to break it at any moment. That was a completely new and foreign thing, and the fact that it worked at all meant that we did a decent job of it.
I do remember one story. I went to a GDC conference, after Diablo II, and one of the topics at one of the forums or whatever was, "Cheating in Games." And I said, "Oh, I should go check out cheating in games, because that's right up our alley right now." And then in the roundtable they were having, everybody just was trashing Diablo the whole time. They were saying, "Here's what Diablo does and why they were stupid, and here's all these things." And I was just getting really mad sitting there, just starting to think in my head. I didn't say anything there, I think I was just in the audience. But I was just thinking, Why don't you guys make a decent game first, and then work on cheating? Which is not fair, and I'm sure a lot of them did make decent games. But I do remember being really mad that everyone was focused on the cheating, when, in my mind, the game was so fun, who cares about the cheating?
A lot of people thought Diablo II was worse [than its predecessor], which wasn't very fun. But I think that in a lot of ways you can't really make games for the critics, you got to make the game that you want. You have to be satisfied with the effort that you put in and the job that you feel that you and your team did. It's like, in a lot of ways, I hate to say it, but it's a little bit like golf. You have to be happy with your golf score, not compare it to everybody else, or really making sure that you're pleasing everybody. Because you're not going to please everybody, and people are going to say stuff for whatever reason, and they may even change their mind. Yeah, they didn't like it at first, but now they really love it, because you made this one little change. And they were ranting and raving about 55 different things, but you make one little change, and they're like, "Oh, I love it now."
So, I think that you're going to get criticism no matter what you do. And this happens, not just in video games, but in any kind of entertainment, movies or music or books or whatever. I liked that one, or I didn't like this one, and everybody's entitled to their opinion, and opinions are free, so people are going to go ahead and say them. It's, how do you let it affect you. You need to focus on, how happy are you with the effort that you put in and the product that you made, and are you proud of what you've done, and are a majority of the people happy with what you created. I think that's the best way to look at it.
There was always people who didn't like Diablo, and that was true of Diablo 1 as well. Some people were absolutely convinced that we were killing the RPG genre by making it too arcadey, and that it wasn't even an RPG, and screw this game. But you could always tell that that was a distinct minority opinion, because people were playing it in droves, and buying it in droves.
Most of my time, I think, after release, in game, in the chat rooms, seeing what people were saying, and by and large those people were very excited, so, you could find the critics. And, we also were learning at this point that when you have a forum, a bulletin board service, on your website, that that tends to aggregate negative opinion. So you could kind of get a false impression of what people were thinking by looking in our forums versus actually going in-game and seeing what people were doing in game and how much fun people were having. But, like Erich, I think that we were so exhausted. We knew the game was good, and we almost didn't care what people were saying, because God, we just needed to get some sleep.