The Wayback Machine - https://web.archive.org/web/20000620093601/http://www.voodooextreme.com:80/games/interviews/timsweeney.html




Talkin' nasty with Epic's Tim Sweeney
Posted on: May 25th, 2000
by Billy "Wicked" Wilson, Chris "Outlaw" O'Brien, Manveer "Eidolon" Heir, and Robert "Apache" Howarth
     
 

Once again, after a bit of bribery, we managed to get Epic's code mantrain, Tim Sweeney into the hot seat for a little Q&A, this time a bit of post-E3 Q&A, to discuss some of the things that were discovered/announced/etc. at this year's little convention down in LA, we also managed to get Tim's thoughts on Brandon's departure, on writing the next-gen Unreal engine, and just about everything else we could think of (one of the better interviews that we've had with The Man, and we've had quite a few!).

 

Voodoo Extreme: What is Brandon "GreenMarine" Reinhart's departure from Epic Games going to mean for Epic in general, your upcoming engine, next title, and the gaming community?

Tim Sweeney: It means you're going to be missing his k00l .plan updates! :-) He was a really big contributor all around (gameplay code, porting, ideas, and lots of other miscellaneous stuff).

The big question now is, can Epic find another really good programmer to join the team? I have a feeling we can. :) But, to be clear, our goal isn't to find a replacement for Brandon, but to bring on a new programmer with his own unique set of talents. To a larger extent than most developers, we tend to mold projects according to peoples' talents and interests rather than arrange the team according to a master plan. Things change; every time someone joins or leaves Epic, we become a slightly different company.

VE: Why, in your own words, would you say that Brandon Reinhart left?

TS: You should ask him for the definitive answer, but in general Brandon's real passion has been working on gameplay: gameplay programming, design, and so on. At Epic he did a lot of that, but probably not as much as he would have liked. This is the downside of a lean-and-mean development team in general, that you often have to do what needs to be done, rather than what you'd really like to be doing. (Ask me how I enjoyed fixing Unreal 1's network code sometime!) I gather that the 3D Realms decision gives Brandon a chance to spend a larger percentage of his time on the gameplay stuff he loves, and we respect that.

VE: Is he leaving on good terms? How is your relationship with 3DRealms?

TS: Brandon is leaving on good terms. We're good friends with the 3D Realms guys, and have lots of good discussions with the programmers there...Chris Hargrove and I love to get into tech debates! Now if they'd just let us see Duke Forever...they're super-secretive; we haven't even seen the game playing. Maybe Brandon can help get us in? :-)

VE: What, so far, do you think of Duke Nukem Forever?

TS: Really excited and looking forward to it!

VE: The X-Box is looking pretty dope, what has Epic's involvement been with Microsoft to put the Unreal engine on this exciting new set-top box?

TS: We're there, baby. The Unreal engine is already on it!

VE: I'm sure you've seen Volition's Red Faction press release, do you anticipate adding any type of "Geo-Mod" technology (basically arbitrary geometry modification) to the next-gen Unreal engine? Do you think this is more of a buzz word right now, or do you think that we have the hardware to pull off the type of system requirements this type of enhancement would require? Do you think we're still a few years off until we really see this type of feature _really_ working?

TS: Hey, UnrealEd had support for realtime constructive solid geometry operations as early as 1996. The engine is actually set up so that a licensee, with a couple weeks work, could use UnrealEd's CSG code in order to do that same kind of realtime geometry modification during gameplay, such as blowing arbitrary holes in walls anywhere in a level. That's easy; the hard part is implementing the FEM, physics, and collision checking that makes complex objects collapse on their own, without any pre-scripting, when supports are shot out.

VE: On a scale of 1-10, what are the odds of Linux/Mac eventually supporting Direct 3D?

TS: That's an interesting thought. I have no idea.

VE: On a Voodoo5, Unreal Tournament runs really, really well in OpenGL mode, even though FSAA does not work (3dfx said something about UT's OpenGL not running in true full screen mode as the reason why) are there any plans to have unofficial/unsupported OpenGL drivers in Epic's next engine?

TS: Yea. On Windows, Direct3D will be the API we optimize for, be we'll also support OpenGL for portability (even on Windows).

VE: Deus Ex is coming to the Dreamcast, how about Unreal Tournament?

TS: Epic isn't developing for Dreamcast, but Infogrames has the worldwide console rights to Unreal and UT, so Dreamcast is something they could theoretically do.

VE: We haven't heard much from the Unreal PlayStation-1 port guys, are they still working on a port for the PS, or have they moved on the a different system, or was it outright cancelled?

TS: Haven't heard anything for a while. If GT shows us Unreal running on PlayStation 1 and it's cool, we'll approve it for release, otherwise it won't happen.

VE: Lots of super cool games about to be released using the Unreal engine, which ones excite you the most?

TS: Like my mom used to say when asked which one was her favorite, "I like all my babies equally". :-)

VE: When writing the next-gen engine, do you worry that more time will be spent getting the engine to be the best it can and less time spent on actual content? Quake III seemed to suffer from this, IMO. 

TS: This comes down to having a really balanced team, enough great people to do a cool job with the tech, and enough great people to do a cool job with the content, and good communication so everyone is working towards the same goals.

VE: Do you think the Justice Department's proposed splitting of Microsoft will lead to new APIs emerging in the field, leading to more work for the programmer who need to support multiple APIs, instead of just OpenGL or DirectX?

TS: The graphics API's are in pretty good shape with OpenGL and DirectX, but in the periphery areas like audio, input, networking, and file systems. And also, the combinatorial explosion in the number of different configurations that need testing and debugging.

VE: Before Unreal, Epic was a company focused on more arcade style games. Now Epic is moving towards a FPS direction. Do you see Epic expanding their genre's further, and making more games like Age of Wonders, or do you see Epic becoming a FPS company, like id has become.

TS: We'll always be developing one big game at a time, so we won't be "dabbling in lots of markets". I don't see Epic as being a traditional FPS developer forever, though. We just make the kinds of games we ourselves love to play. The only genres that really qualify universally among the team now are first-person shooters and persistent multiplayer games, but we don't at the moment see an angle from which Epic can get into that genre without growing into a huge and financially complex company, something we don't want to do.

But our design possibilities continually change due to technology. 3D action games weren't technologically possible 7 years ago. I'd venture to guess that the kind of game we'll be developing 7 years from now isn't a first-person shooter, but is some genre of game which isn't technologically possible now. If you think about it, when we look back on games like Quake 3 and Unreal Tournament, 7 years in the future, they'll look incredibly low-detail and shallow. By the time NVidia has figured out how to render ten billion triangles per second, I have a feeling we'll have found a better use than building more detailed deathmatch levels.

The thing is, once a genre of game grows old (like realtime strategy games for example), developers have mapped out all the possibilities in such detail that team size, budget, and brand become the things that make games successful. This will be the case with single-player 3D action games in a few years. The further-off games like Duke Forever, Unreal 2, and assuming Valve develops a sequel to Half Life, will up the ante considerably.

We always want to be pushing into new territory, where the development roadmap isn't so well defined, because that's where we really have a chance to innovate with technology, artwork, and design and set ourselves apart from the other games. We don't want to grow into the kind of huge company that one would need to be to compete in a heavily entrenched genre.

VE: Is it true that Mark Rein was actually involved in a midget-porn flick during his college years?

TS: The company policy is not to comment on all midget porn flick allegations.

VE: A certain key member of the industry (*cough* Peter Molyneux *cough*) has announced that his next game will be his LAST for the PC, and that all future games will be developed for consoles. What impact do you think this will have on the industry?

TS: That's nuts! :)

We have to be realistic where the PC falls in the big picture of gaming. PC's won't have the reliability of consoles anytime soon, and consoles won't have the open-ended ness of PC's. Consoles have always been about two-thirds of the game market, because of their mass market appeal and ease of use. PC's have always appealed to the hardcore guys, the people who don't just play games, but compete online with deathmatch, follow the communities and websites, build levels, create mods, run servers. Our feeling is that PC's and consoles will be more symbiotic than ever in the future, when the Internet lets both groups of gamers hook up together.

As always, a big, warm, manly hug (courtesy of Lowtax) goes out to Tim Sweeney for letting us pester him, yet again (Tim really is a stud, regardless of what past girlfriends say).

Voodoo Extreme

Return to Voodoo Extreme's main page