Skip to main content

Long read: The beauty and drama of video games and their clouds

"It's a little bit hard to work out without knowing the altitude of that dragon..."

If you click on a link and make a purchase we may receive a small commission. Read our editorial policy.

Overwatch: Blizzard answers the big questions

Competitive play! Server tick rate! Balance! More!

The Overwatch launch appears to have been a phenomenal success. With more than seven million players having registered since launch, across three different platforms, Blizzard's first multiplayer shooter is already one of the largest in the world. But for this game to truly achieve its ambitions, it's going to need to hold on to that playerbase in the months and years to come. Blizzard has reached its payload, but there's a ways to go yet.

The Overwatch community mostly communicates in Play of the Game GIFs at this stage, but there are a few concerns they'd like to see addressed. We're still waiting on the game's competitive mode and it's not yet clear how that will compare to the earlier version we saw in the closed beta. A number of players would like to see Overwatch introduce a server tick rate of 60Hz or higher, similar to Counter-Strike: Global Offensive and Battlefield 4. And balance! We always need to talk about balance. It sounds like the team are looking at McCree and D. VA, but what exactly can we expect there? And when?

With all of this in mind, we spoke to Overwatch's game director, Jeff Kaplan. Join us in the interview below as we hit all of these topics and a few others besides. It's a good one.

The next big update we're waiting for is the return of competitive play. Is that still planned for the end of this month?

Jeff Kaplan: pre-launch, post-beard.

Jeff Kaplan: Yeah, in fact, we were just playtesting it this morning. Competitive play is the big focus for the team right now and our hope is that it will release at the end of this month. That's what we're driving towards and it's looking good. Obviously we'd like to leave ourselves an out, in case something goes sideways in the middle of development, but right now things are looking good.

Do you have any date penciled in at the moment?

Jeff Kaplan: We actually don't have a specific date right now. Each morning we look at it and say, you know, here's where we're thinking and that date has shifted forwards and backwards numerous times already. It'd be very misleading if I gave you a specific date at this point, because I think it would probably shift.

So what exactly can we expect from the competitive play? Will it compare to what we saw in the beta?

Jeff Kaplan: We were pretty pleased with the beta system in a lot of ways, but it was a little bit frustrating to us because it was unfinished. A lot of the feedback we got was just about what was there and it seemed to ignore the parts we mentioned that were also coming.

But we got enough sentiment from the competitive community that we realised that we needed to make some changes. First and foremost, the system that we had implemented was what we like to call 'progression-based'. We thought there was a lot of coolness to doing this kind of system and that's why we had the one month reset, because it feels good to go through the progression more than one time.

What was not implemented, was what we called our 'heroic rank', where it would show a stack ranking of top players who hit that rank. There was a misunderstanding among our playerbase that everyone was going to get to heroic rank, which is not true, just like it's not true that everybody will get to legendary in Hearthstone.

The perception of our system was off. So the new system, when we unveil it, I think you'll see that it's much more skill-based.

We saw an early version of competitive play during the game's closed beta.

Another thing that there was a lot of complaints about was the time for the season. The playerbase at large felt that if they'd gotten to a certain place in a competitive system, they wanted to enjoy time at that place before it got reset. So when we unveil the new system, we're going to have longer seasons - most likely, the seasons will be about three months and they will match the real-world seasons. They'll probably actually last about two and a half months and then we'll do like a week or two off. We're hoping that this speaks to the competitive community and addresses what their concerns were. There's obviously way more details than that, but we're not fully ready to unveil the system as a whole, but I think those were the things that people were most concerned about.

Could you clarify the difference between a progression-based system and a skill-based system? From my perspective, as the player, how does the game convey that to me?

Jeff Kaplan: The way our system worked before, we had different tiers with challenger, advanced, expert, master and heroic. Within the whole system, you could never drop from a tier. So even if you got to master, you would never drop out of that tier, no matter how much you lost. In the player's opinion, that meant that everybody was going to end up, eventually, in the same place.

In our new system, there's no safety net. If you lose, you're going to go down, if you win, you're going to go up.

How will the matchmaking work in this new system? And will there be a solo queue?

Jeff Kaplan: Yeah, we're going to go out the gate with what we call a 'dynamic queue'. This is what we use for quick play right now, meaning you can queue in any size that you want. So if you want to queue by yourself, or if you want to queue with two people, or three, four, five, six, the system will allow you to do that.

The way that our matchmaking works in quick play and will also work in competitive play, is that it always tries to find... let's say you're a group of five, it will try to find a group of five equally skilled players. Only at extremes of waiting time will it expand its search and be willing to match you with a different group size for whatever reason. That's the way it's always sort of worked in quick play and that's how we'll go out of the gate with competitive, and then we'll just monitor it. The thing that's very important to us, and that we keep trying to prove over and over again to the community is that we're listening and open to change. Hopefully they'll know that if something like the queuing system isn't working how they like it, they realise that instead of just being angry at us, they'll know that Blizzard actually listens to its community, they'll probably make changes based on the feedback.

That's most likely what we'll do: keep an open dialogue and make changes based on feedback. We always try to educate the community as to the trade-offs. Nothing comes for free. There's usually a reason why we made a decision that we made and it wasn't because we were looking to upset people or anything like that. There's usually some very difficult trade-offs to be made that led to the decision in the first place.

Watch on YouTube

Whilst we're on that topic, something that has cropped up within the Overwatch community a few times now, especially when we're talking about things like competitive play, is the server tick rate. Players would like it to be higher. Will that happen?

Jeff Kaplan: I'm really glad you asked about that, because there's so much confusion. First of all, most people don't even really understand the way that the network code works. For example, the server does tick at 60Hz, it's the client update rate that is lower. That just shows a general misunderstanding.

I think players have latched onto server tick rate as being the reason that certain things happen. One of the things that players are upset about is that if they get shot, where they perceive they were behind a wall, that this is a problem with server tick rate. Certainly there are contributions that could happen with both the server and the client update rates that could cause something like that to happen, but usually, in most cases, you're talking about latency.

You haven't seen a lot of shooters that move as fast as Overwatch, with abilities like Tracer's blink or Genji's dash. There's also not a lot of games using a killcam like Overwatch does, so it all sort of makes the problem more evident than I think it would be in other shooters. We're addressing it in a number of ways.

We put out a 17-minute video with our lead gameplay engineer and our network engineer, who wrote the netcode for the game, explaining exactly how the game works, to try to educate people on which parts of gameplay issues were latency-related vs. netcode-related. Trying to give everybody more transparency and understanding of how the game works so they can help comment intelligently on what is and isn't a problem. That video is widely available and I would encourage people to watch it.

The other thing that we did, after that video, is we actually added what is called a 'high bandwidth option' to our custom games, that allows players to play at a 60Hz update rate. We wanted to explore what the game would be like at 60Hz, so that way, we could explore if it's possible for us to try other versions of the game with this update rate. What we've been trying to do is encourage players to play with it. It's kind of ironic that they're demanding it, it's in the game and playable, and right now we're seeing about 0.08 per cent of all matches that take place in Overwatch are actually using the feature.

I've heard a lot of players say "god dammit, Blizzard, just add it to quick play". It would be wildly irresponsible for us to add that to something like quick play or competitive without getting more testing on the feature. This is something that you care about and you want to see more of: well, please use more of the feature that's in the game. Give us feedback, so we can iron out all of the kinks with it and if things are looking good, we can absolutely add it to other parts of the game.

That's something that's been very near and dear to us. There's nothing more important to us than the game feeling fast and responsive. We've said, since day one, we wanted it to be the high bar of what a shooter should feel like and if any part of it doesn't feel that way, we're concerned about it and want to fix it. Kind of like what I was talking about with competitive play and players immediately getting angry about stuff, I'd just like to remind people that it's an active issue. Not only have we been communicating about in that 17-minute video, but we've also added the feature to custom games to try to get testing on it, so we can explore if we can do this in other parts of the game.

Right now, for example, what we've seen is that it wouldn't work for about 20 per cent of our users. That's a problem we can solve if we get more testing on it, but we couldn't just put it live. As you know, we announced that number of seven million players last week. The last thing we want to do is take 20 per cent of those people and tell them you can't play Overwatch anymore because people were angry on the forums and we flipped the switch and added this thing.

The more people that test the custom game feature, the more likely we are to able to proliferate that to other parts of the game.

Watch on YouTube

Right now, what else is the team looking at? We've heard mention of balance changes for heroes like McCree and D. Va. Can you give us any specifics there?

Jeff Kaplan: Yeah, the McCree nerf - I hate using that word, but I just said it - the McCree balance changes will be coming sooner than the D. Va buff.

McCree is pretty straightforward. What we're looking at right now is his fan the hammer damage. We're going to reduce it. The goal there is to make it so that McCree can still use his combo that we love, which is the flashbang and fan the hammer on somebody like Tracer. McCree should absolutely kill that Tracer. We want McCree to be a counter to people like Tracer, Genji and Reaper. What we're not crazy about, right now, is the way in which McCree can absolutely shred tanks. It's a little too easy, so we want to bring the fan the hammer damage down, so he's still killing the squishies and the medium strength heroes, but he's less effective against the tanks. If he times everything perfectly and gets every single shot off, he's got a shot against a tank, but it's not the instant 'I win' button that it is right now.

So that's what's coming for McCree. We've been testing it for a while and it feels good. You still want him on your team.

With D. Va, we're still in exploration mode. There's a couple of directions we could take her and it's still unclear which direction we'll go. Obviously her damage has come into question from a lot of people. You have to be really close to do effective damage with D. Va and we might look at that. We might also look at her survivability. A lot of time, D. Va can get herself into situations where she's a tank and it feels like she should be there, but then she gets knocked out of that mech so quickly. We probably won't do a bunch of buffs to her damage and her survivability; we'll probably pick one direction or the other, but right now, we've been exploring both.

We've tried high-damage D. Va with lower survivability. She's almost more of a skirmisher. We've tried other directions, like not buffing her damage, but buffing her survivability. I think the D. Va buffs will take a little bit longer for us to get to, but the McCree balance changes should come sooner. D. Va's not in a horrible place, we just feel like she could be in a slightly better place, whereas McCree is causing a lot of concern in the community and we want to make sure they know we're responsive.

When those balance changes do come, can you hit the PC and the consoles at roughly the same time?

Jeff Kaplan: Yeah, I think when the McCree balance changes happen, they should happen at the same time. We've scheduled to pass console certification at the same time for those. With that said, if we end up in a situation where we've passed on one of the console platforms and PC, but for some reason, the other console platform is lagging behind, we'll probably make a game-time decision of saying 'hey, we might need these changes out for where we can get them'. We're going to do our best though. We feel like we've scheduled in such a way that we should be able to clear any hurdles, but we'll be smarter about that in the next week or two.

Watch on YouTube

How about the Play of the Game? It sounds like you'll be tweaking that for some time to come. Is there anything new you can discuss there?

Jeff Kaplan: I can tell you about what changes we would like to make. The next step for us with Play of the Game - and this one's not on the immediate horizon, I'd say it's something we'd like to get to this summer - we would like to show play of the games in a more cinematic way. We feel like for your standard play of the game, like a Reaper getting a death blossom kill on four people, they look great and they're really fun to watch. But some of the other ones like the Torbjörn who's dead and his turret's killing people, or a Widowmaker getting a really impressive shot from across the map: those aren't showing great right now. They're actually really cool moments but we feel like there's a way to show them more cinematically.

We feel like if we can refine how we're showing play of the games, we'll actually open the door for us to do more things like our saviour play of the games, where you're about to be killed and somebody intervenes. It's actually really cool, but it doesn't show very well because the camera might not be tracking from the best angle. Right now it's all from first-person. So that's on our list right now, but I'd say that even though play of the games aren't perfect, and there's a lot of funny ones or accidental ones that happen, overall the feature has been great for the community discussion it's spurred. I would say that at least 70 per cent of the time, they are actually really cool moments that people feel pretty happy about.

Now that the game has launched and you're starting to look ahead at future content, how regularly do you expect Overwatch to see things like new heroes and new maps? Is there any kind of timeframe there?

Jeff Kaplan: It's really funny because launch happened and it was kind of this cool moment where the team is actually working on patches for the end of the year and early next year. We have parts of the team working on so much of the game right now that a lot of us we're like 'oh yeah, that whole launch thing just happened'. We have plans for some hero and map updates, which you'll be hearing more about as we get a little bit further into the summer.

But we don't know what the right cadence should be just yet. We have a gut feeling from when we added new maps and heroes to the beta, but what we want to do is release our first new maps and our first new heroes and then get a sense from the community as to like, was that too much? Was that too little? Is it better to release them this way or that way?

We feel like we're going to learn a lot from those first few updates that happen. An example from the beta is that we felt like our release rate was pretty good for the most part, but we widely on the team felt like releasing Mei, D. Va and Genji all at the same time was way too disruptive for the game. That was a lesson that we learnt. When we come out with our first new heroes post-launch, we're going to do them one at a time for a while, until we feel the game really stabilises.

If you go back and read some of that beta feedback when we unleashed those three heroes after Blizzcon, they were demanding that all those heroes be removed. Like, they're all overpowered and we should get rid of them. If you think about it today, Genji, D. Va and well, maybe not Mei, are some of the most loved heroes. I think Mei is probably the most loved hero in terms of her cute, adorable personality, but a lot of players claim she's annoying. But at the time, they were demanding that we remove those heroes. We want to be careful about overdoing it with a patch.

Watch on YouTube

Whilst we're talking new heroes, how did you find it when people started speculating that a new hero, called Sombra, was about to be revealed around the time of your last animated short? Was that strange, from your perspective?

Jeff Kaplan: No, it was not strange. We have put a lot of hints, all over the game and out of the game, so I would say it's about time that people are concerned with who Sombra is. I hope they keep researching more and find more hints that are out there.

So they haven't found everything yet?

Jeff Kaplan: No. No sir.

Last time we spoke, I asked you about the possibility of crossplay between the PlayStation 4 and Xbox One. You said that it was something you'd be interested in, but it wasn't something you could look at just yet. Is that still the case and has there been any progress there?

Jeff Kaplan: There hasn't been any progress, but we're still very open-minded about it. I know that Rocket League was coming out with cross-platform play between PlayStation and Xbox. I'm a huge Rocket League fan . I primarily played on PC, but I want to check it out and see how it feels. I also want to see what the fan reaction is and if they're liking it. I have friends at Psyonix and I want to get their reaction too. We are very open-minded, but we're going to wait and gather feedback from developers and from fans.

For all the latest Overwatch hero, map and gameplay guides, visit our sister site MetaBomb