MCV/DEVELOP News, events, research and jobs from the games industry
So the single and multiplayer elements of DriveClub aren’t totally distinct from one another, are they?
We do try and push some of what’s going on in the wider DriveClub community into the single player as well, just to give you the feeling that there’s something bigger out there.
Does the open access to tracks and vehicles in multiplayer affect how you’ve handled progression in single player?
A little bit. One of the worst things that can happen if you’re in a party session with your mates is that you get to a content lock where you’ve not reached the required level so you can’t play on this track, use this vehicle or whatever. The simplest way around it, and this is the system we ended up choosing for multiplayer, was that to offer you loan cars for it. So you’re playing a bunch of events with your mates, you get to one particular event and it’s a car you haven’t yet unlocked, we’re going to loan it to you for the purpose of that event.
And how does ‘loaning’ differ from ‘having’ in a virtual space?
The only caveat is that you don’t earn fame. Fame is basically our points system in the game. You earn fame by doing all kinds of things in DriveClub. It’s a similar approach to the way the Call of Duty games deal with private matches. To avoid collusion they stop you from earning XP whilst you’re playing a private match. This is kind of like that.
And what does fame get you?
Haha. That’s the $64000 question!
Fame gets you all the rewards in the game – all the unlocks. If you treat fame as more like experience, that’s pretty much it is at its core. You earn fame, you level up, you gain fame ranks, and every time you level up, you get new cars, new customisation assets, and new accolades.
The accolades can be applied to your vehicles and you’ll see them on your profile in-game. If I were to look at your profile, you might have some particularly prominent accolades on it that only you’ve achieved, or maybe you’re in a very small percentage of people that have achieved them. So it’s a way to show off – a sort of badge of honour if you like.
It was hard to get the progression right, to be honest. It took us a long time and a lot of user-testing.
The designer that was responsible for the main meta-game (or meta-progression) simulated a lot of it beforehand in a spreadsheet with a bit of visual basic – he just knocked something up. Now, we go to the master DriveClub simulator, he puts numbers in and tweaks it and can follow and simulate things like level curves, where each unlock appears and also the balance of the vehicle classes and everything else in between.
Using that data he was able to tune the game. You kind of take that data, then from the other side you come in with a bunch of user testing, get actual real players playing the game. You can talk to them directly and get anecdotally what they felt about the experience, but then you also get telemetry data from it as well.
The combination of all of those things – it’s a long, time consuming process, particularly for a game the size of DriveClub. But you have to go to those lengths to make a nice balanced, rewarding progression structure.
Was the first set of data from the telemetry came back, how close was it to those initial projections?
Pretty close. Actually, it was really close.
Chris was the designer behind this stuff – he’s been doing it for a long time and he’s got a pretty good feel for it. It wasn’t like he was coming at this cold. We’ve done three generations of racing games now, and Chris has been with us for a long time.
Even so, this is a different type of game than anything that we’ve made before. In the initial tests, it felt like it was a little too punishing. What we didn’t want to do was drip-feed the content to you as though we were just throwing rewards at the player – the player gets bored of it really quickly, and I think players are quite savvy towards it these days. But at the same time you want to make sure there isn’t this barren patch of no content and no reward.
So first time around it wasn’t so generous. Then we actually flipped it the other way, and we’ve kind of being balancing it since then until we got to where we’re at. The last user test we did a few months back now, the feedback was ‘Please do not change this. This is now perfect.’
Is there much merit in signposting ahead of time what a player is about to unlock?
Yes. It’s really important. You always need to give people a goal to achieve. If they’re just in the dark all the time, there’s no desire to progress. But, you don’t want to lead people by the hand either.
I’m quite cynical about those kind of mechanics. I like some element of exploration, but I also like to know when I’m on the right path. And I think that should be what good game design is about. It’s to give the player multiple paths, but re-enforce when they’re on the right path. So you’re not exactly telling them where to go, but you’re encouraging them to move in a certain direction or behave in a certain way. It feels much more natural and is much more rewarding.
We highlight stuff with the accolades in particular, because each accolade has a very specific goal. You can browse through each and every accolade right at the beginning if you want to, and with each accolade there are multiple levels, so it’ll tell you that to reach level one of this or that accolade, you need to do x number of these things. Really simple, really straightforward.
But if you don’t want to go that far (because not everybody’s going to care that much about their profile – not everybody’s interested in it), we do pull out accolades that you’re close to achieving, or accolades which work if you’ve picked a particular track or a particular car. Some are more suited to your current play style, so we surface those for the player and give you those to latch on to.
Like curating a feed of what the player could or should be going for?
Yeah, based on the decisions you make. So the player is conscious in their decision (we’re not making decisions for the player – they’re going to make their own decisions), but based on those decisions we’re going to push certain content.
There’s a lot of push content in DriveClub – that was a big, big thing for us.
It feels to us that gaming time is pretty precious. Most of us don’t have as much time to game as we maybe used to, and there are so many games out there competing for your attention right now, that we felt that with DriveClub we needed to be respectful of the player’s time. We needed to make sure that if they only had an hour or two a week to play the game, that that hour or two was great and was really rewarding.
So when you play games where there are half a dozen progress metres at the end of a session, do you look at it as being more of a scattershot approach?
Sure, but you need both I think. Everybody likes a summary – the match summary, the race summary, the round summary – as long as that information is meaningful in some way.
Stats for stats’ sake is something I just don’t necessarily agree with. But stats that are meaningful are useful.
Destiny is a great example of something I’ve been playing recently which isn’t DriveClub. It makes your end of round (or end of mission) summary screen split into three sections, and they just give you that very broad summary on the first page. If you want to, you can skip through the pages and drill down even further. That’s not a bad approach, it’s saying to the player ‘Look, there’s all this information, but we’re just going to give you this window into it, and if you want the rest of it, you can go there and get more data.’
DriveClub is a bit like that. Like I said, you can see all the accolades if you want that information, but we never surface that to you. We only surface the things that are meaningful and relevant.
Is this unusual for the sorts of games you make?
Driving games usually don’t have these rather elaborate design mechanics to them – they tend to be a lot more traditional. They tend to mostly be based in motorsports, actually, where you have a championship and you get points in that to progress to the next one and so on.
I have a kind of OCD when it comes to being a completionist in games, and I love the way the original Forza series just gave you the event grid. It’s gave you all the events in the world and a little tick or a little point when you’ve achieved one of them. That’s great for players like me who need that visual sense of completion.
But I understand that that’s quite daunting for other types of players – being able to see absolutely everything.
I think it was Sid Meier who said that the perfect number of options you can give a player when they’re making a decision about new information is three. Would you agree with that?
Yeah, three really is a magic number isn’t it? A choice of two just feels like it isn’t even a choice.
You know, now that you mention it, magically it has happened that way in DriveClub. We didn’t consciously set out to do that, but actually if you go to the drive menu there are three options, and there are never more than three options on the list in the tour menu either. It just magically happened, that was absolutely subconscious.
But I suppose the rule of threes is prevalent in all design, not just videogames; you’ll see the rule of threes used all over the place.
I think if you’re not careful, you can easily overwhelm players. There’s so much to do in DriveClub, especially considering all the social aspects of it (because all the challenges happen asynchronously), so these can be going on in the background while you’re doing other things.
Trying to manage all of that logistically is challenging, so we had to be really careful with the user interface (because that’s the point at which the player interfaces with the game) to make sure that we’re not overwhelming the player with too much content – we’re only really surfacing the stuff that’s relevant and meaningful, but again there is enough agency for the player to kind of pick the things that they want to do at that point in time.
Was presenting the right information from a myriad of content one of the main challenges of making DriveClub then?
Absolutely, yeah. And that’s why we delayed it.
It wasn’t delayed for technical reasons then?
It was design reasons, yeah.
That’s fairly uncommon.
Very much so, yeah. I mean the game was finished. What we had was perfectly shippable – it would’ve reviewed reasonably well and it would’ve sold well, but the ambition of the vision initially wasn’t being done justice, particularly because we wanted to release this as a service – as a racing platform.
Evolution has said before that DriveClub a year after release was planned from the outset to look very different to DriveClub at launch…
And we don’t know what that’s going to be yet either. We had a vision for the game and we have a vision for how we want to extend that, but we’ve left gaps in the schedule next year because we want to work with the community on this. We want to leave enough space to respond to feedback from the community and to try and deliver features or even remove features that aren’t quite working.
I like DLC in games. I like that it extends the experience for me, but I feel like I’m just getting what I’m expected to be getting – there’s nothing that surprises me. We were quite keen to move beyond that as a developer.
Instead of just giving players more of the game, it was to try and open a dialogue with the community, be much more transparent and much more open and let the players know that it’s a shared experience, just as much for us as it is them. We’re all shaping what it’ll be a year from now together.
Thank you for your time!
