All right, time for another weekly blog!
This week we’ll talk about how we approach designing menus, and how complicated it can get. This is topical, as we’re currently designing brand new menus for Wreckfest, and you’ll get to see them soon enough
The first step of the process is to recall this mantra: “Form follows function.” This means we want to have menus that look good, yes, but most of all are easy to use.
To get the ball rolling, we talk. A lot. What do we want is not as important as what we need? To map out what these needs are, the artists and game designers gather up – the designers explain what kind of flow they have in mind, and the artists pitch their ideas to the designers. This means a lot of meetings, and often quite a bit of arguing, critique and pondering. It sounds rougher than it really is, for most of it is really good natured! Plus, if you can’t back your ideas up with proof and/or solid points, it’s probably not that good an idea
This process is repeated over and over again, as it’s important to give everyone a chance to mull things over, and to spot mistakes and fallacies in their proposals. After all, this is just the very first stage, so even if the “core loop”, meaning the very basic game flow, and all the major menu items are already known, it’s easy to miss some crucial detail.
Once the basic frames are in place, the artists create a grey box prototype. No bells and whistles, no graphics, no nothing – just the basic functionality, so we can spot if this design really works.
When the result is good to go, we move on to the next stage: finding that right feel for the menus. For this, references are the key. What games got that right atmosphere we’re after? What newspapers have the right way of displaying information? What movies work that light and shadow magic just the way we like it? This can go as detailed as need be, e.g. we spot a game that has that 0.5 second fadeout with each menu click that looks real cool!
This is all references, mind you. It doesn’t mean copying others’ work 1:1! It means we gather inspiration from things we like, and build from that. Imitation is, after all, the sincerest form of flattery
We have a huge in-house reference library for just this purpose, so our artists don’t have to hit Google all the time, but can add items to the library that look sweet whenever they come across them, and then just check the library when the need arises. It makes the whole process a lot smoother, and allows for much faster prototype construction.
We also create tapestries of different reference materials to give us a sort of “mood board” that is not about particular details, but rather about the overall feel you get when you glance at it.
The third step is creating a more finalized version of our menu prototype. At this point we already know what the menu flow will be, and what the style will be. This stage is all about detail. We hit Photoshop like nobody’s business and delve on minor changes of font size, line thickness and color/light gradient. We create as “final” version as we can, and check that it all looks good.
Fourth stage is motion design. Because menus are rarely just static, we do a mock-up so people get the idea of what our new menus will look like in action. This requires some 3D sorcery as well as placing lights and objects in the mock-up environment, and often this means improvising solutions as all the necessary bits and pieces are not in the game yet.
The fifth and final stage of the design process is proof of concept. So, we have menu design that’s all the way up to how it will look, and we have a motion mock-up to give an idea what it’ll look like when you’re actually using the menus. If both look successful, we create a playable prototype: a collage of grey boxes that link to each other, but even if they look really rough, you can use them like you’d use the final thing. This stage shows if the whole design really, truly works or if (or, rather, where) it needs some more tuning.
Once we have the proof of concept prototype in place, too, we can start the hard part: actually creating the menus. We’ll go over that stage at a later date
Thanks for reading our weekly blog!
As always, stay safe, people.
- Team Bugbear
This week we’ll talk about creating cars – where do we start, where do we go from there, what to look out for, and how long it all takes? For Wreckfest, existing cars took about two months to complete – per car. Yes, it really takes that long!
It all begins by, obviously, figuring out what type of car do we want? A sedan? A truck? After we settle on e.g. a muscle car, we dig deeper: what type and style of this particular car type do we want? An olden goldie, or a more modern one? European, American or what? After this part of the process we have an idea of what kind of car we’re looking for, so we can start looking for reference material from their real world counterparts, up to and including blueprints of real cars.
Next up is the basic modeling of the car: modeling the car body without paying too much detail to panel curvatures, specific details and so on. These very basic models have all the obvious parts in place: four tires, bumpers, headlights and all that jazz, and about in the right places, too. It’ll look clunky as heck, but that’s okay, as we’re mostly interested about the overall feel of the car, and not so much of any specific details.
Once everything is modeled and in their right places, we create several variations of the car with different grills, lights and light configurations. The body of the car itself remains untouched, but it’s the small details that give the car that look you come to love, so it pays to create a few variants to be sure that this is what we want. Again, it’s not as much about creating finalized models as it is about giving the car several discernible feels, as all this helps us figure out what would work for this car and what doesn’t.
Once the prototype is ready, we start working on it in earnest. We start tuning curves, adding panels, modeling specific parts like axles and upholstery. Basically every detail will be added at this point – yes, it’ll often be just a rough semblance of what it’ll end up looking, but all parts will be put in place. For example, detailing the drivetrain means that if you flip the car and focus your view on the axles, you will see the brake shoes, springs, suspension parts, bolts, nuts and so on.
Despite the detail, we don’t actually animate that many of these parts, simply because it wouldn’t be cost efficient for the drain it would have on the engine. We totally could, though
The last stage is finalizing the car. At this point we know exactly what we want. The rest is polish, more polish, and then some more polish. We put all those nice curves in place. Get that geometry in place to the tee. Make sure that everything is in its place, and it’s all smooth and sweet. This will also help in creating smoother shading, which gives the car that golden touch. This final stage takes agonizingly long, as it’s all about the detail, and there’s always something that you could do some more work on.
After this stage is done, after everything is neat and tidy, we can start working on the textures… but that’s a whole other blog entry for a later date
Also, lately we’ve turned our demands on every detail to eleven, but we’d rather let the results speak for themselves than talk about what we will do. Let’s just say that, for example, everything is now modeled to the tenth of an inch. No, we’re not kidding. We figured that if we’re going to ramp up our physics and tracks, our cars should live up to that level of detail, too. Sweet, huh?
That’s it for this week!
As always, stay safe, all y’all
(Note: while you can spot that one of these cars is indeed American Muscle 2, the blue car is a model that was abandoned, so don’t take its appearance as anything else as serving as an example )
With the new physics model coming up, we want our tracks to live up to the standards that kind of detail brings. That’s why this week’s blog will be about designing race tracks – what’s our philosophy there, and our golden goals?
In general, we are not aiming for easy tracks. However, we don’t want to create unfair tracks, far from it. We want the players to be able to learn the optimal paths to drive, and then hone their skills to shave precious seconds off from their lap times. We want the tracks to have this kind of detail – that you can enjoy thoroughly the tracks when you’re cruising for bruising, but if you want to push your limits and break your speed records, there’s absolutely room for that.
The essential philosophy in creating a track like that is that it should not fight or punish the player, that the tracks won’t let you go easy but also won’t kick you when you’re down. This means our tracks should have room for errors. You might floor it too heavily on a straight and brake too late, which will result in the car swinging off the road. But, having that room for error, that wide patch of grass before the tree line, ensures that you can get back on track and up to speed without losing your momentum entirely. (At this moment, not all our tracks are this kind.)
In addition to these safety zones, we aim to pace our tracks. You have your curves and your chicanes, but you’ll also have straight patches of road where you can put the pedal to the metal. This is not only because we appreciate the need for speed, but also so that you can have a moment to breathe and collect yourself after a not-that-optimal swerve and slide at that latest curve that got your heart racing. Take a few seconds to just keep on accelerating, and use that moment to wipe your brow and sigh from relief. That’s what we’re after.
With these two basic goals in mind, we can make tracks that are challenging, but not punishing, and can be enjoyed by Sunday drivers and gung-ho speed freaks alike. Especially the delicately designed optimal routes should provide a lot of substance to people who aim to be in the top 1% fastest drivers in each given track.
To make tracks like this, we do a lot of research. We examine thoroughly how classic real life tracks look like from the ground and from the air, so we can learn exactly what makes them legendary? We have obtained a bucketload of new tools that allow us to create a track by adding and moving control points and then creating a procedural geometry and terrain for a specific area, as well as tweaking the inclination of the road and angles of banks with one degree accuracy.
The best bit? We are making a library of road parts so that we can quickly mix and match, so we can create new track prototypes faster and faster. This means more speedier development times for tracks, which means we’ll be able to push out tracks on a faster pace.
The cherry on top? This library of road parts will be a terrific help for modders, who can either use these bits of track as they are, or they can alter them as they see fit!
All this is in line with our usual design philosophy: more depth than you’d ever guess, but with low enough a learning curve that you can thoroughly enjoy taking it casual.
That’s it for this week!
Enjoy the weekend, and, as always, stay safe.
This week we’ve made some really great progress on two fronts: physics and tracks. The original plan was to write this blog about the latter, but after those really intensely focused meetings we felt that we should process it a bit and then talk about it.
So, this week’s blog will be not be about tracks, nor will it be about physics. It will be about two new game modes.
Elimination will pit players on a racing track to compete against each other, but not in the traditional sense. This game’s winner won’t be determined by who finishes first, but who can outlast everyone else. Every lap (or, say, 45 seconds, there will be variations) the game checks who is the last driver in the race, and weeds the poor racer out of the competition. In addition, all those unfortunate enough to get wrecked are also out of the race.
This means that all players need to drive fast and furious so they won’t be the unlucky last player that gets whacked after each lap. This, in turn, means that people will resort to dirty tricks to stay alive, and that, in turn, will result in a lot of people getting banged up beyond recognition during the race itself.
Elimination should prove to be a fun game mode, especially since Wreckfest is so delightfully cruel when it comes to damaging your rides. Elimination will be a mixture of reckless driving, insane risks, pure skill and manic mayhem. We hope you’ll enjoy it as much as we are.
Our second new game mode is Team Race. This mode will allow up to four teams to race for victory. But, like always, we’ve added a few twists into how this whole thing works.
Yes, the first to cross the finish line gets the most points. Yes, second and third places net their drivers a hefty score, too. But so does wrecking other players, and driving the fastest lap times. In short, Team Race will reward your team in several ways: by being a menace to others, and by being a fast driver. If you get left behind and have no chance of being in the top bracket, take it out on your fellow drivers, and watch as your team’s score goes up.
Team Race aims to be a really interesting mixture of aggressive driving, team spirit, and gung ho attitude. Risks are rewarded, but the real sweet spot here is tactics: when driving with an experienced team who works well together, you can designate different roles to each other for maximal efficiency.
No holds barred, folks! Track shots are allowed, even encouraged
For those more interested in clean racing, do not fret! Even if in their current state both Team Race and Elimination reward aggressive driving, we are working on alternative clean racing versions of both. We’re still figuring out how to reward players in both modes, but as soon as we got some stuff figured out, we’ll let you know.
Mind you that both Team Race and Elimination are still under development. They need to be refined into true gems, and that takes many iterations before we’re done. This is where your help comes in – when the next build is launched, test these new game modes out in earnest, and tell us what works and what doesn’t. That way you’re doing your part in making Wreckfest be all it can be.
That’s it for this week!
Stay tuned for the track status report in our next weekly blog.
And, as always,… Stay safe, all y’all.
- Team Bugbear
This week’s blog opens our design philosophy and the lengths we go to so we can deliver that vision to our players.
The focus is on physics, and what we’re doing with all that jazz.
As a company, Bugbear is committed to making delightful car games. Not only do we strive towards excellent entertainment factor, we want our games to last for years. With Wreckfest, we really let our ambition run rampant, because we don’t want to create just another run of the mill action racing.
Often cars in games have only a few discernable factors, like acceleration, top speed and turning speed. That’s it. They’re really simple creatures, and this usually results in each new car being just plain superior to what you previously had, because of those overly simplified characteristics.
We don’t want to do that. We want our cars to be the main characters of Wreckfest. We want each and every car to be their own person, with their own quirks, strengths and hiccups. To accomplish this, our cars can’t only just feel different, but actually be different.
This feat requires some true magic under the hood. In order to give the cars enough personality, we need to have enough variables in handling and behavior so you could really spot those differences. This, of course, means amping up the physics engine, and to be honest, we’re going out of our way to do this right.
Each car has a huge number characteristics that define how it functions. The values given for the car’s frame affect how it turns, lulls, nods and bends in any given circumstance. In Wreckfest, the suspension portion of the car, meaning springs, sway bars and so on, is just a small part of the whole, but that alone contains over sixty different values. Each of those values affect directly how the car behaves, and each value needs a bucketload of tuning so you can get that feel of the car just right. The suspension is both a treat for sore eyes, as the cars bounce on their springs realistically, but also an important part of modeling car handling to the tee.
The tires of the car get a lot of love as well. They’re actually a perfect example of our attention to detail. For example, the surface of the tire is divided into small segments. Each segment has several layers: our physics modeling takes into account the rigidity of the tire’s surface, the elasticity of the rubber mixture, and shock absorbance of the layers beneath. Not only that, we monitor things like friction and how the tire heats up – and how that heat changes the various characteristics! In short, our tires behave differently when they’re warm, and they have enough values to handle simulating any surface conditions we want.
Despite all this hyping up of how detailed we are with our physics modeling, the main goal, however, is not because we want to be a simulator. No. We need our physics modeling to have this much detail because we want the cars to behave like they’re supposed to when you’re sliding. We want that feel when the car remains in your precise control even when you’re drifting like crazy. That feel when your steel horse bucks and whines under you, but you, you remain in control. You can’t do that without digging deep into physics. So, we dug deep.
This is why it’s taking so long to get the next build ready for launch.
This is also why we think it’s worth the wait.
This week at Bugbear!
*cue drum roll*
Aaaand no, the new build is not ready yet. We’re sorry that it’s taking so long, but the physics are a huge chunk of stuff you need to get just right. Once that’s done we should be able to focus on other stuff and push out new builds on a more regular basis, so there’s that, at least. We’re still doing our utmost best to launch the new build for everyone to marvel, but also working on asynchronous gameplay and new game modes.
We envy how well Forza does their asynchronous gameplay with Drivatar. It is absolutely fantastic in how it learns from the way you play, then mimics your style in someone else’s game, effectively letting you race against people who you have never, ever played with! We’d love to have our own Drivatar in Wreckfest, but we can’t. Our studio is just too small to pull it off. Instead, we’ll approach asynchronous gameplay from a less ambitious angle.
In effect, the way we’ll aim to create an avatar of you in the game will be through keeping record of your statistics. What’s your favorite car, your favorite track, your best lap time, your average position upon finishing, how much damage you do, and so on. Mostly they’re basic statistics, ones you see in many a game. These will be useful in themselves, helping you see how well you race with any car on any track, and also fun, when you want to e.g. check on how much time you’ve spent flying through the air!
With the statistics, we will try to build a guesstimate of you in the game. If your record shows you like crashing more than clean driving, your stats show that. If you tend to speed too much and take a corner way too fast, your stats will show that. And based on those stats, we’ll basically assign an AI behavior to your avatar. It won’t be 100% accurate, since it won’t mimic your driving to the tee, but it should be close enough to guess your preferred style of driving.
That’s the plan, anyway. Creating an asynchronous AI model is very difficult, but we’ll try nonetheless. In any case, the career statistics should be a cool thing to have.
This past two days we’ve also started developing a new game mode to Wreckfest. It’ll be a cooperative race, dubbed cleverly Team Race. Players will be split into two teams, Red and Blue, and face each other off on the track. Here’s what we got so far:
The teams gain points in three ways. The first is, obviously, your position when the race ends. The second is from best lap time: the player driving the fastest lap time nets their team a certain number of points. The third way of scoring is wrecking opponent cars. When the race is over, the points are added up, and the team with the most points wins. So, even if a Red car crosses the finish line first, it doesn’t mean their team will necessarily win, if Blue has managed to rack up points by other means.
Team Race is still very much under development, and we’re yet to get to internal testing. Still, plans are made, and they’re being executed as you read this. Once we’re ready, it will be included in the next build, and we eagerly wait for your feedback!
Also, regarding physics, here’s a small video from this Monday, showing how bad a broken physics engine can get
Don’t worry, it won’t happen for you – this is a result of internal testing and tweaking the settings a bit too much