Archives by Day

December 2024
SuMTuWThFSa
1234567
891011121314
15161718192021
22232425262728
293031

MX vs. ATV Reflex

Platform(s): Nintendo DS, PC, PSP, PlayStation 3, Xbox 360
Genre: Racing
Publisher: THQ
Developer: Rainbow Studios
Release Date: Dec. 1, 2009

Advertising

As an Amazon Associate, we earn commission from qualifying purchases.





'MX vs. ATV Reflex' (ALL) Developer Interview #2

by Adam Pavlacka on Nov. 6, 2009 @ 8:45 a.m. PST

MX vs. ATV Reflex introduces revolutionary physics that allow for real-world terrain deformation and an intuitive, dual-stick control scheme that separates man from machine, as well as a wide variety of race modes, vehicles and worlds to explore while utilizing the all-new Rhythm Racing 2.0 physics engine.

MX vs. ATV Reflex introduces revolutionary physics that allow for real-world terrain deformation and an intuitive, dual-stick control scheme that separates man from machine. MX vs. ATV Reflex will deliver a wide variety of race modes, vehicles and worlds to explore while utilizing the all-new Rhythm Racing 2.0 physics engine.

WP: Who has the honor to speak with us? State your name, rank and occupation!

I'm Nick Wlodyka, and I'm general manager at Rainbow.

WP: Rainbow has been producing MX vs. ATV games for a while, with the first one starting back on the PlayStation 2. How has the game changed over the years with each iteration? What were the key things that you've focused on and learned as the franchise has progressed?

NW: Over the years, Rainbow has probably been most renowned for the physics that we put in our game and the feeling that those created that probably best represented off-road racing, especially as it pertained to quads and bikes. We've really learned that what we're able to create with that is what gets people really, really excited, but there's a lot more to the game than that. Expectations have changed significantly as a result of direct competition and as a result of what technology has provided for us in terms of opportunities to really push the experience forward. As we look back, I think a lot of focus was initially put on a physics system that dates back to the initial MX vs. ATV, and that physics was adapted, changes were made, every subsequent iteration, to a point where we felt that now, with what we have at our disposal, we really need to bring a significant leap forward in the overall driving experience, the racing experience.

That's what we did this year with Reflex. It wasn't just an incremental improvement over our old physics system. Now that we've got this power available to us and we really know how to harness it, let's create a system that's going to provide for an unparalleled experience for players. So by breaking apart the player from the vehicle, it now affords them the luxury of having a significant additional element of control. For a new player, they can continue to play it like they have been, with just the left stick. For somebody who wants to take it that step further, wants to make their turns tighter, wants to stay on their bike in a moment when they would otherwise fall off, that's where the right stick really comes into play.

We're really excited because it's not just a little bit better this year. We've gone right back to the grassroots, the foundation, and created a system where beyond just what I talked about with respect to separating the rider from the vehicle, we now have everything that's physics-based. Where we used to cheat values in terms of how much slip there was on the tire going around a corner, how much the suspension gave, all that is now done with real-world data, which is pretty powerful.

WP: As game journalists, we're always hearing that the graphics were redone from the ground up. In this case, you've redone the physics engine, but what does that mean? Can you be a little more descriptive?

NW: Sure. Absolutely, a lot of times people talk about, "Oh, it's brand-new." We're not touting what we've done from a visual standpoint with our product this year, although we've taken a lot of steps back and really redone a lot of work to give it a much more high-fidelity, real feel to it. It's not something where we feel we really need to write home about it because a lot of games look fantastic out there. That was one of our issues coming from Untamed: We maybe don't compete on a visual level with some of the other titles out that are there, and what fans expect from the product, so we put a lot of hard work into that. Again, it's not something that we're going to say is built from the ground up, but we feel really good about where we've gone.

The physics system is a little bit different in that we really threw out the old system and spent the last 2.5 years on it. Physics is obviously the underlying foundation of the game, and it was a very risky thing for us in removing the old physics system and putting in a brand-new physics system. Over 2.5 years, it started from no code, no technology and was built from the ground up to work with the game, with what you seeing today. It's not just a claim that we've redone this. We literally took out the old one, worked on a new one and put it in, and this is what we've got as a result.

WP: When you're putting together something like that, how do you balance that with the power of the system? Obviously the current-gen systems are more powerful than the previous generation, but do you just start out, throw everything and the kitchen sink in there, fire it up, watch it crawl and then decide what to pull out? Do you try to make educated guesses as you're building it? Do you build it up slowly? What is the developer's process when developing a physics engine?

NW: With respect to how we approach it, there are several key steps. It's not one where you can just build it out to complete, drop it in and see what happens. There are key checks and balances along the way, and it was the same thing for us. With that said, there is a certain amount of investment from a time standpoint that you have to put into getting something working at a basic level. That's what we did.

We had a first key milestone after probably a year, where we took what we created, dropped it in the game and looked at it. Does it feel right? Is it delivering on what our ultimate goals were? Equally importantly, what's the performance? Are we dropping down significantly from where we thought we'd be? From that process forward, we had another check about six months later. We started with just the bike, so we worked on two wheels and then we extended it to four wheels. It was really, really tough for the team because all of our track tunings are based on this new physics system, so the track designers couldn't build tracks for the bikes or the quads until that physics system was done. With all those dependencies, it made it a really, really stressful time.

Ultimately, we still had to do a lot of performance things. We had to spread performance across multiple processors, which we didn't think we'd initially need to do. What you get is a process that involves seeing something as you initially designed it, testing it out, seeing what you didn't expect, the positive consequences, the negative consequences, and working through those, along with everyone else who's dependent on it when it's such a key piece of the game.

WP: We played through some of the game, and we noticed that the AI seems to be very aggressive, almost to the point where it doesn't care if it crashes. How do you go about programming an AI versus a physics system, which is based on reality and how materials interact?

NW: Fantastic question. AI has been a really, really challenging one for us this year. Part of it was because we had to get the physics system working before we could start working on the AI. Another part of it was that we wanted to create a system that wasn't rubberbanding, where if you drop far behind, you come back to the pack after a while just to keep the races intense. We wanted to reward people for racing great races, and on the flipside, we wanted it to be challenging so if you weren't performing well, you weren't artificially rewarded. We spent a lot of time building in tools to our AI system to allow the designers to give the different AI riders different characteristics on how they ride, what types of corners they're most likely to blow through, what kind they're not based on human behavior. By examining human behavior, we're able to create characteristics for the AI that allow them to race a lot closer to what a human would be.

From an aggressiveness standpoint, that's part of the sport, that it is aggressive. We didn't want a situation where riders would back away from you as a player because we didn't want them feeling aggressive. Hopefully what happens is you'll feel more aggressiveness in the harder difficulty levels and less so in the beginner — or Rookie, as we call it — difficulty level.

WP: What about the different vehicles? Focusing on one vehicle is easy, but when you're designing a track that can be taken by a bike, ATV and truck, and you have the different track types — supercross, waypoint — how do you balance that?

NW: Another great question. That's one of our greatest challenges: track design around multi vehicles. What we always talk about at the start of every project is, "Do we need to have these multi vehicles? Why do we need to have these mufti vehicles? What are we going to need to explore to make sure all the tracks work for the different vehicles?"

We put all the different vehicles on a track if we're doing a multi-vehicle race to make sure that from a baseline standpoint, they fit. If we build the track too narrow, suddenly the truck will be flying off the track left, right and center, and that's why some people might ask, "Why can't I take a quad into a supercross race?" That's part of it. That track is specially tuned for just bikes.

What we've tried to do this year is that for tracks that have multiple vehicles on it is to create some splinter paths. If you're in a truck, you can go down one path because there is a lot of different debris on that path, and you're able to easily plow through that, whereas if you're on a bike, you're not going to go over that path with a lot of debris because it's going to be too difficult to traverse. In turn, we create more narrow paths off these main tracks so a bike can go down that narrow path whereas a truck wouldn't. There are a lot of checks and balances, there are certain concessions we need to make in terms of overall track width because we don't want to make these tracks super, super wide because the bikes will feel lost on them. Frankly, it's a really tough challenge for the designers, and I think they've done a fantastic job.

WP: After you've put all this work into the game focusing on the realism and getting all the different vehicles in there, how do you ease new players into the game? What tweaks are happening behind the scenes to adjust that difficulty level?

NW: That's why we have our beginner difficulty level. This year, we also introduced a tutorial system to hopefully ease people into that. From a difficulty standpoint, it's primarily the AI riders that are going to make the biggest difficulty adjustment. They don't carry as much speed, they're more apt to make mistakes in the corners, and they're less aggressive. The same mistakes that a human player would make, we put those in the AI players as well. Ultimately, we hope that the races feel fair without feeling unrealistic. In a Rookie race, you'll see the AI crash more often, take corners a little bit more like a beginner user player than once you start to step up into the greater difficulty levels.

You bring up a great question, which is, "How does a game like this, which is very much rooted in the true essence of the sport, appeal to a more casual audience?" It's one of the key questions that we ask ourselves all the time, and what we hope is that by providing a variety, we're really giving players the opportunity to try and experiment with different things and see what feels right for them. Somebody who has a bike and feels good with it is different from a person who chooses a truck because he's used to driving more vehicular platform. Where we're trying to move with it ultimately is not be a jack-of-all-trades and master of none, where we've got all this stuff and some of it works and some of it doesn't. We're conscious throughout development to make sure that everything we put in there is fun and has a purpose. If something's not working or something doesn't feel right, we're not going to release it just for the sake of having tons of stuff in the game.

WP: If you had to sum it up in two to three sentences, what really makes MX vs. ATV Reflex a game that's worth playing?

NW: From a sheer variety standpoint, we have a number of different experiences that hopefully appeal to a broad range of players. We've got new features in the game this year, and what's really important about them is they're not just copy on a box cover; they have fundamentally changed the racing experience. No other racing game offers the level of control that we do over bikes and quads. In addition to that, you've got the terrain deformation, which is a really neat technology feature because not only does it let you leave your mark wherever you go, from free ride worlds to races, but it actually changes the dynamics of the races themselves. Depending on how you use that deformation, as ruts form, you can go through ruts around the corner and take those turns faster. On the flipside, if you take a rut incorrectly, it's going to throw you off a little bit. The game looks really neat, but there's also more to it than a matter of learning how to steer left and right. There's a great amount of depth that's included.

WP: Is there anything about the game that we haven't talked about that you wanted to add?

NW: One of the things that we also worked on a lot this year that bears notice is the audio system, and I hope people pick up on it when they play it. We've had a lot of commentary in the past about the bikes and quads not sounding very realistic, so we really make a conscious effort to bring in an audio soundscape that truly represents the intensity of the sport. We've had some great feedback so far from people who are hardcore into that sport and know exactly what those machines sound like, to people who are just a little more casual and are looking for an intense experience. The other thing we did is rather than licensing a bunch of tracks for the game, we did a lot of individual competition that was intended to play up the experience and really hit the beats of what's going on in the race. We're happy with where that's come, and hopefully for players, it completes the experience.


More articles about MX vs. ATV Reflex
blog comments powered by Disqus