The Development Of A Continuum: Andrey Iones On TimeShift

TimeShift is the first major game from Saber, a studio run out of New Jersey with the bulk of its development staff in Saint Petersburg, Russia. Here, VP of production, Andrey Iones, talks about the unique structure of the developer, the strengths of working with in-house tech, and working on a title that switched publishers during development as well as being sent back for a year of re-development after originally being completed in 2006.

First I want to ask -- you guys are based in Saint Petersburg?

Andrey Iones: We actually have two offices -- one in New York/New Jersey, and another one is in Saint Petersburg. The bulk of development was done in Saint Petersburg. We have over 80 people over there, and we have the management team and production team in New York.

So these offices... they're both owned by Saber? You don't outsource to other developers?

AI: We run both offices. Both offices are a part of Saber and we run as-is -- integrated team. But we do some small outsourcing to some other teams in other parts of Russia and the Ukraine, to do small things like individual assets, characters, and things like this.

What's the background on how this company got set up and how you developed having that satellite office? I guess that satellite office is the one in America? It's smaller?

AI: I wouldn't really call either office a satellite office. The company was set up seven years ago by three guys: myself -- Andrey Iones -- Matthew Karch, my partner, and Anton Krupkin. I grew up in Saint Petersburg, Russia, and moved to the U.S., and so did Anton. We met this man who used to be a lawyer, who was doing some immigration case for me. We got [to be] friends, and we decided to do something on our own.

I had a background in computer science in general, as well as in gaming. He was passionate about games. He owned every console since he was a kid. So we decided to do our own thing, and that's how it all started in New York, but because I have connections and I grew up in Saint Petersburg, Russia, it was natural for us to do an expansion over there. We just kept growing and growing the team.

We started with a small title called Will Rock done for Ubisoft, which was done PC-only, but it was a small, decent title just to get our feet wet, and to form the team and to establish technology. And then, seven years later, here we are done with TimeShift on three SKUs.

First of all, do you have your own technology, or do you use an engine that's provided from another company?

AI: No. We use our own technology. We do work with some middleware providers such as Havok or GameSpy... but the core technology was done in Saber, by Saber, so we have full control over it. We have to, because in a game like TimeShift, you need to have full control over the tech if you want to do something as challenging as time control.

Right. How does that play into it when you're developing the engine? Because this game has that time control mechanic, which is different than most of the FPSes on the market. Is that something that you find is really important -- catering the engine's technology to the gameplay design?

AI: It was really important for us to have a gameplay hook which would put Saber aside from the pack of other games similar to ours, so we'd have something unique to us. Having our own technology was really instrumental in pushing the game out on all three SKUs, and at the same time it allowed us to implement all the complexities related to time control in general, and time reversal in particular.

So you guys, did you internally develop all three SKUs?

AI: Yes, we do.

Did you have any challenges, moving your technology across three different platforms? They all have their own quirks.

AI: Working on three SKUs at the same time is really hard. We had to do a significant redesign of the engine to take full advantage of next-gen capabilities of high-end PCs, Xbox 360, and PlayStation 3 hardware, doing things like native multithreading in the engine, and doing things like supporting all these multicore, multi-CPU hardware. It's challenging, but I believe we did a good job of harnessing the power of all three SKUs. I believe that it is what it is.

This game has an interesting development history, in that you started out with a different publisher, and the title was picked up and moved over. The development cycle was greatly extended from the original plan, wasn't it?

AI: It was extended over and over again. A brief history of the title that was published: it was picked up in 2004 by Atari, and by that time, we already had a fully functional prototype with time-control mechanics fully working. It was like a complete vertical slice of the game, and it wasn't bad. It had two or three good levels with really good visual quality and fully gameplay mechanics in there. It was picked up opportunistically by Atari.

Was that just on PC at that time?

AI: It was on PC only. The original plan was to put it out on PC and Xbox -- the first Xbox. Saber was a relatively small company coming from somewhere in Eastern Europe, and nobody really knew where Russia was.

But as the project matured, it was rising up in the ranks at Atari, and at some point, they realized they had something bigger than they originally thought, so they refocused it from PC and Xbox to high-end PCs and Xbox 360.

Unfortunately, in 2006, Atari ran out of money, so they decided to sell the title to somebody. There were a number of publishers bidding for the title and Vivendi eventually acquired the title. They liked it. It was an opportunistic pick-up for Vivendi, because something that was big for Atari wasn't so big for Vivendi anymore.

But just like with Atari, the title kept rising in the ranks, and eventually, the guys in power realized that they had something bigger than this. They gave us another year, and they expanded the title from PC and Xbox 360 to PlayStation 3 as well. A year later, we have a title we're releasing on three SKUs. It's completely redesigned in terms of graphics, in terms of gameplay, and now it's fully a next-gen title.

The title's changed a lot over the course of its development, from its original plan until where it's arrived at now. Can you talk about some of the processes you went through and what you learned when you were dealing with that situation?

AI: The first, big-ticket item that changed was the visual style, from steampunk to a much more mass-market gritty "destroyed beauty" type of thing. I think we really succeeded in building this universe, which fits the title. The main gameplay mechanics were tweaked over and over again until we arrived at what we currently have, which is really feasible, which is really easy to get into.

Which is really important on consoles, because the title started on PC and Xbox, and now it's Xbox 360, PlayStation 3, and PC. To nail down the time-control mechanics to make it challenging, but it's an easy learning curve. The switch to the console mentality was really important to us.

Do you feel the game is better than it would have been if it had followed that original plan in its development, coming out back when Atari had it?

AI: Yeah, definitely. The quality of the game was significantly improved over the course of the years. We were fortunate enough to work with some of the more experienced people in the industry -- people like Chris Miller, who produced F.E.A.R. and N.O.L.F. and other games. Working with those people, they brought a lot to the mix of an already talented team, and helped us to bring the quality of the game really to where it belongs.

It's interesting. The one thing I would say is that you've got a lot of time to improve the title, but by the time it's come to market, the playing field is a little bit different, on the Xbox 360 particularly. There's a lot of shooters. How do you feel about that aspect of it?

AI: I believe that even last year, we didn't have a bad game. It was a good game. At the time, it was a good game, and people were considering it to be one of the better-looking Xbox 360 games a year ago.

Clearly, when the decision was made to extend the title, the first question we asked ourselves, as well as Vivendi, was how to stay competitive a year from now. The decision was to to do a complete overhaul on a lot of engine technologies, such as rendering. We really upped the ante in terms of the things we were doing.

Things like shadow maps -- it's something that nobody else was doing. Maybe Crysis was doing it, but that's only one SKU right now. Things like a host of rendering technologies, like detailed normal maps, and next generation physics, and post-processing effects and per-pixel effects. These are things we are doing but nobody else is doing. Graphics-wise, we are definitely where everybody else is, and maybe a little bit... we're definitely keeping up with the pace of competition, and in many respects, we are better than other games.

I'm sure those things are true, and they do make the game look really good, but how do you think that gamers, when they make a decision about what game to buy -- psychology-wise -- do you think technology matters, or do you think it's the theme? How do you take that into account when you're working on the game?

AI: Technology is important, but we're not selling a tech demo. Definitely what TimeShift is all about is the universe, the experience, and the look and feel of this universe. The goal was to bring the player into this universe, and to leave him there and to experience all the twists of the story and let him interact with the environment and with the universe. It's definitely much more than a tech thing. That's why we spent so much time on tweaking the gameplay, balancing the gameplay, making sure the shooting mechanics are what they should be, as well as the time-control mechanics what they should be.

Did all the creative direction come out of your office in Saint Petersburg, or was it your office in New Jersey, or also from the publisher?

AI: We got a lot of creative direction from our publisher. We're really fortunate enough to work with one of the best creative people in the industry. But Saber as a team matured a lot through the course of the years, and I believe we were able to add a few significant improvements to the overall look and feel and the gameplay experience of the game.

I believe it was a great collaboration effort across all three locales -- Los Angeles, New York, and Russia. The people traveling across all three places... we have a lot of guys coming from Los Angeles to our Russian office, and we would bring the best from Russia into the U.S. to New York, so that the people would get familiar with U.S. sensibilities, and with U.S. perception of things in general. It was really important.

There are some big Russian and other big Eastern European developers. Sort of up-and-coming, I guess, to an extent, at least in terms of the American market and the broader international market. What do you think that culturally, from a game development perspective... what do Russian developers bring, and is there something you say is quintessentially Russian, or interesting about the way games are developed or conceived of, in Eastern Europe?

AI: (laughs) That's a good question. I believe Russia is great when it comes to finding top-notch developers in terms of programming skills and artistic skills. I believe it takes a lot to make a game produced and developed in Russia to comport to Western sensibilities, so it was really important that we had the U.S. designers and minds behind the game. Matthew Karch, my partner, is the lead designer on the game, and the vision that he is bringing to the mix as well as the vibe they're getting from Vivendi was really instrumental in getting the game where it needed to be, in terms of look and feel and design, and just how the game plays.

What's challenging about developing the game in Russia is that you want to try all the small companies in Russia. We're trying to do something, but it's not really working out, at least not yet. Let's say that you have money and a great game idea in the U.S. All you have to do is just put out an ad and eventually people with experience will come to you.

Not in Russia. In Russia, you can't find people like designers and producers. They simply don't exist, because there are no triple-A titles coming out of Russia. It's simply a challenge. You have to bring these people in, and you have to train them so that they become triple-A developers. It's much more than programming, because you have to have people with production skills and with designer skills. It's really hard to come by those in Russia.

Is that because the gaming culture in Russia is different? Obviously a lot of people come into the game industry because they love games. What do you think affects that?

AI: There are a lot of people who love games in Russia. But there are not a lot of console games done in Russia, as well as there are no consoles which are sold in Russia. If you want to find people to develop a game for consoles, obviously you want to find people who understand consoles, at least from gameplay perspective, as users. You simply can't find those, because you can't go out and buy a PlayStation 3. Well, that has changed as of about six months ago, but it's obviously nothing compared to what you can get in the U.S. It's really hard to find those people.

There's no established game industry, in the U.S. sense. In other words, there are some local publishers and some local developers, but they're working on really small PC titles, and it's PC-only development. It's basically teams of 10 to 20 people, which is nothing compared to what you need to have if you want to produce a triple-A game on three SKUs. You need to have a team of at least 80 people, all of whom are experienced and have at least some titles under their belts. You simply can't form this team in Russia, unless you stay in business for a long time so that you can bring in people, train people within the team so that they grow mature and fill those leadership roles.

Are there people coming from other disciplines into game development in Russia because there isn't an industry? Is it just people who are interested in games but didn't have an outlet for it before -- that they were brought into development like that?

AI: We have programmers who used to be mainstream web developers or database programmers who come in and bring a unique skillset, but obviously you can't form a team around those people. You need to have people who understand games and who develop games. Luckily for us, there are a few smaller developers in Russia from where we could cherry-pick the best people -- the best designers, and the best programmers. They formed the core of Saber's team a long time ago, because they're working so closely with Western publishers -- first with Ubisoft, and now with Vivendi. We're able to train those people so that they become true leaders in a company that develops a triple-A franchise.

Obviously the quality of the game is good and it does have creative ideas, but it is, essentially, a conventional game. Are there any ideas that you think would come out of your staff that aren't as steeped in the past ten or fifteen years of game development, and what's been out on consoles already? Do you hear unusual ideas out of any of them? Sometimes people who come from other backgrounds don't really know what they're up against, and they'll suggest atypical ideas. Have you gotten any of that?

AI: Obviously, TimeShift is a shooter, but it's not a standard shooter. It's a shooter with a twist. Obviously, time control is the biggest thing. Not only does it work in single-player, but also in multiplayer. That's where you have the bulk of emergent gameplay situations. You're able to use time control, and just the fact that you can manipulate time in multiplayer -- and how we were able to figure out the time control mechanics in multiplayer -- leads to a lot of interesting gameplay implications.

Obviously single-player as well, it's not just a standard thing, because you have time control. You have to build AI, for example, which is responsive and cognizant of this time control. Not only can you run around opponents in slow time, but you can also do reversal. What happens if you run into a room and you have a bunch of guys -- you activate them and they start shooting, you initiate reversal and then run behind them? They come into this reactive state. You have to program all these reactions to figure which work with time control.

The multiplayer component works with time control. The AI reactions with time control. It's two big things -- big-ticket items -- which make the gameplay experience in TimeShift unique compared to other games. Not to mention other things you can do with physics, for example.

You have to answer more questions about quantum physics than I ever imagined, working on the game, because we always have to control these time streams and alternate time streams, and what happens if you do a reversal and come to a place where -- imagine a scenario where there's a bunch of barrels. You explode those barrels, and go into the place all those barrels were, and use reversal. Essentially you'll end up with a barrel on your head, and what do you do? You'll get stuck there. So we had to come up with a special solution, which is a time continuum breach. If you end up in the middle of something which you reversed on top of your head, that's a time continuum break. You can't continue. That's something you can never experience in any other game.

Something I wanted to ask you about was, your producer Kyle Peschel came over with the title, basically. He was at Atari, and now he moved to Vivendi, and he's risen in the ranks as the producer of the title. Can you talk a little about that, and how that worked out on the project?

AI: Definitely. When TimeShift was acquired by Atari, Kyle was assigned as an assistant producer to the title at Atari, working out of the Santa Monica office. About six or eight months into development, Atari decided to close their Santa Monica office, and he was offered to relocate to New York so he could keep working on the product, which he did. So he moved to New York, and he spent a lot of time in Russia, it was frickin' cold, and he was traveling back and forth.

At one point in time, Atari decided to sell the title to Vivendi, so he was all pissed and frustrated, but he applied for a job with Vivendi, and he was hired. He stayed with us, he stayed on TimeShift, and he actually moved back to Los Angeles to work on TimeShift at Vivendi. He's saw the title through completion. He's been a great partner, and he was really helping us to push the title and to change the title to the way it shaped up.

Do you think that it would have gone as smoothly with changing publishers, if you didn't have someone on your side throughout the whole process?

AI: Definitely. Definitely. He was the one who was with the product from day one, so he knew all the technical details, design details, and the style of the project overall, so he helped to ensure the smooth transition from publisher to publisher. And he has a great understanding of what it takes to make a great first-person shooter, so he was quite instrumental in helping to polish the game on all fronts.

One thing I want to ask about the PS3 SKU is that it's coming out a little after the Xbox 360 and PC SKUs. Can you talk about how it's been to develop the PlayStation 3 version of the game?

AI: We started to work on PlayStation 3 about a year ago. I believe the PlayStation 3 implementation went pretty smoothly. Obviously, we had to spend some time trying to understand the hardware, and figuring out all the details -- things like multithreading and SPUs, and the rendering features of the hardware. But ultimately, the PlayStation 3 SKU is going to be released only three weeks after the Xbox 360 SKU, which is, I believe, a pretty significant achievement on Saber's and Vivendi's part.

It's not easy to master the hardware. You definitely have to put a lot of resources and people on it. But it's a really, really great piece of hardware. You can do a lot with it. Right now, if you run the PlayStation 3 and the 360 version side-by-side, they will look pretty much the same.

If you look at the multiplatform games that are coming out this fall, there are a great deal of significant games that are running late on the PlayStation 3. Stranglehold was about two months, and The Orange Box isn't out yet, even though it's out on the other platforms. There are other examples. Why do you think that is? What do you think is holding it back?

AI: That's a really good question. I can only explain why Saber managed to put the PlayStation 3 version of the game out within this year, and all those other companies having trouble publishing the SKU, I believe stems from the fact that the Xbox 360 was around for a longer time, so people have a much better understanding of the pipelines and the architecture.

Granted, Xbox 360 architecture is a little more straightforward than the PlayStation 3's, so you have to allocate people to PlayStation 3 SKUs, and let them figure out how to take advantage of it, and make sure that it has a good framerate, and that you fit all that you have into the PlayStation 3's memory. TimeShift proves that it can be done with great quality. Like I said, if you compare the two versions, they run at similar framerates, and the quality of the picture is the same.

But you need to allocate really significant resources. People are probably a little bit concerned about not having enough PlayStation 3 retail units out there, and they're focused on getting PC and Xbox 360 out first and foremost, and they just didn't have enough resources on the PlayStation 3. That's my explanation. But clearly it's a hardware that can run really well, but it's pretty demanding.

People do talk about the difficulty of working with the Cell processor, and they talk about the RAM situation. But there's also little things, like the OS takes up more memory on the PlayStation 3. BlackSite recently announced that they're dropping voice chat from the PlayStation 3 version. Are these things... do they all contribute? Is any one significant, or do you think it's a large number of small problems?

AI: I believe it's a large number of small problems. I don't think there is one thing. There's nothing in particular that can't be overcome. Obviously, things like voice, like you mentioned, do require a significant amount of memory and CPU power, which is even more important, but the Xbox 360 was out for two years, and the PlayStation 3 was out one year. They're still working on tools, and they're working out the kinks of the technology.

But our experience proves that if you put enough experienced engineers on those types of challenging technical tasks, you can definitely resolve them all. Make sure you do feed into memory, you do have good framerates, and you don't downres your textures so that it looks like shit. It just doesn't happen. But it does require a significant investment on the tech part.

If you have your own technology that allows you to fully leverage the complexities of the hardware, you can put it how you want. Whereas if you're running on somebody else's, you might have limited access to how you can configure it, and how you can optimize various aspects of the game, like loading times, feeding into memory, performance. If you have specific technologies such as [time] reversal, for example, you don't have all the control you need. Having our own tech really allows us to fully embrace this hardware.

You're right when you say it came out later, so there's less familiarity and the tools aren't as robust. Do you think that there actually are distinct advantages to the PlayStation 3's hardware, where, as time moves forward, as people come to grips with it, they'll be able to do things that actually take advantage of it? Because right now most games that are multiplatform are leading on 360 and/or PC, and essentially, you're going to buy that version, if you're the consumer and you have the choice, because maybe there's more graphical effects, faster load times, or what have you. Do you think that something is going to bring the PlayStation 3 to the forefront? Like technological advantages to that platform?

AI: I believe both systems are pretty similar in terms of what they can achieve. Things like having, for example, a [guaranteed] hard drive on the PlayStation 3 allows you to do certain things which you can't do if you don't have it on the Xbox 360, but even that is kind of obsolete. The Xbox 360 is coming out with a built-in hard drive, and they're calling those systems that don't have a hard drive obsolete.

Technically speaking, both systems are similar, and what you can do on one system you can do on another system. Granted, Microsoft had more time to work out certain things that you really don't have to worry about on the PlayStation 3. For example, voice is a major component of their Live support, as well as things like ranked and unranked matches, which you have to do because it's a TCR -- it's a technical requirement. All the things which users do care about are mandatory on the Xbox 360, and you simply can't go without them.

On the PlayStation 3, not so much so. You can have much more flexibility with what you do and don't want to put in the game, so you basically can cut corners somewhere if you really want to put the game on the market. But obviously you don't want to do it too much, because you don't want to ruin the users' experience.