Half-Life 2 Interview

[Ghost [PX]]

In retrospect, would you have imagined that computer gaming would reach the level of complexity and depth that is apparent in the comparison of Half-Life and Half-Life 2?
Well, there's always been this issue posed by people thinking about extraterrestrials. The universe is really, really old. Even if intelligence is super-rare, and even if most intelligent species wipe themselves out, there is still plenty of time for intelligent life to fill up the galaxy even traveling way below sub-light speeds. Using just about any statistical model, we should be ass-deep in ET in our galaxy. So where are all the alien civilizations?

I have my pet theory which I pretentiously call the "Gambeboy Really Really Advanced Theory." About the time that any species reaches the level where it could actually build an interstellar spaceship, they are also building super-powered computers capable of playing super-powered video games. ET then has a choice of going and exploring the real universe (lots of dull travel where the exciting question of the day is "how much space dust will we collide with") or exploring the virtual universe created on their Gambeboy Really Really Advanced. So ETs are out there, but there just too busy playing and exploring virtual universes that are much more immersive and compelling than the real universe.

I guess that means that I have a lot of confidence that we have only scratched the surface of what will be possible in videogames, and that, thanks to the twin forces of Moore's Law and increasing sophistication among game developers, whatever the delta between Half-Life 1 and Half-Life 2, we'll see much greater advances in the near future.
How do you feel Half-Life 2 will contend with such other eagerly awaited releases? i.e. Doom III, Deus Ex 2, etc.
I'm never sure how to answer questions like that. I know people like to see developers get into pissing matches, and I am as guilty of that as any gamer (aw, for the good old days of Paul Steed .plan file updates). However the reality is that I am looking forward to buying and playing Doom III, and I'm looking forward to buying and playing Deus Ex 2. I want Duke Nukem Forever to come out, and Stalker and Max Payne 2 to show us all something new, and so on.

The reality is that the PC games market is not a zero sum game, and the more good games that are out there, the more every developer benefits by growing the number of people who regularly buy PC games.
Do you think that Half-Life 2 will surpass the popularity of its predecessor?
Yes. It's a much better game.
Will Half-Life 2 be catered towards a particular brand of video card? As in NVIDIA or ATI?
It would be great if we could pick one hardware platform with perfectly consistent performance, but we can't.

One interesting problem for game engines right now is scalability. You have to run fast on the DX6 generation of hardware (TNT, Intel 810), and you have to fully exploit the latest and greatest DX9 hardware (R350-based and NV35). There's a far greater difference between old and new ATI cards than there is between the latest and greatest ATI and NVIDIA cards.
Do you think that Half-Life 2 will effect the computer gaming world the same way that Half-Life has?
This is another question I get asked a lot and don't know how to answer. I'm not really sure that Half-Life has had that great an impact on computer gaming. When you look at Doom, you can see that within 12 months that every developer was trying to do the same thing. I think there were over 400 Doom clones shipping or in development at one point. When you look at Command & Conquer and Warcraft, you can see how that impacted developer and publisher plans almost immediately, and the same can be said for Ultima Online and Everquest. Hell, even Deer Hunter had a much bigger impact on the industry than Half-Life.

Right now I can think of less than a handful of games that I can look at and say "I see where they learned something from what we did and are building on that." In a lot of ways we are playing out the string that id software initiated with Doom, whether it's the FPS genre, multiplayer, or supporting the MOD community.

Where I hope Half-Life 2 does have an impact is in solving a set of technological, authoring, and business problems for the MOD community that will create an explosion of really innovative games.
From your standpoint, is there any particular feature in Half-Life 2 that you are especially fond of?
It's a toss-up between the work Ken Birdwell did with acting and emotions, and the work that Jay Stelly did on physics.
Is there anything that you wanted to do with Half-Life 2, but were hindered from doing so due to a particular limitation?
There was some really cool image-based rendering technology that Gary McTaggart prototyped. It would have given us some super-cool capabilities for apparently complex environments. We weren't able to get it sufficiently integrated by our September 2002 milestone for it to make it into this game.
Are there any un-fostered ideas that failed to make their way into Half-Life 2 that you can divulge to us?
I'm still looking to get the team to put moss in.
What was the most difficult challenge to overcome when developing Half-Life 2?
The length of the project. We've been working on Half-Life 2 for five years, and it's been a huge amount of work the whole time. It's just hard to maintain your focus over that length of time. We've had some of our long-time playtesters going from being seniors in high school to being college graduates during Half-Life 2 development.

We're going to recharge a bit on short fun projects once HL-2 goes out the door.
Can we expect to see the same phenomenal, if not more improved, netcode in Half-Life 2?
Yahn Bernier and Mike Dussault have completely re-architected the net code. There are two benefits - it performs much better, and it is much easier for MOD teams to use.
Will players that are new to Half-Life 2 be able to follow the storyline without concise knowledge of the original?
Yes, that's one of the design challenges for any sequel.
Was it a difficult transition for the Valve team to move from the original Half-Life engine, to the newer one? That is to say, did individuals such as the level designers and modelers have difficulty adapting to a much more capable engine?
Not really. The biggest challenge was to raise the level of the art direction. If you only have 500 lit and texture-mapped polygons for a person, there are a bunch of art issues you are constrained from having to think about. In Source, an artist has to think "I can write a program for each pixel on the screen" and then do something really compelling with that power.
Do you think the community will still continue to be as adamant about modding Half-Life once Half-Life 2 debuts, or perhaps it's encouraged that the community slowly digress towards Half-Life 2?
I expect that once the MOD teams get the SDK, that wild horses couldn't stop them from doing a Half-Life 2 MOD. It will be a much more fun environment to work in.
In general, have there been any "close calls" with the Half-Life 2 project?
Not really. The team is pretty experienced, and tends to self-correct by itself.
We assume Half-Life 2 has always been an idea, considering the cliffhanger ending of Half-Life, but has Half-Life 2 developed into the idea that was thought up years ago?
Yep. Given how we work, we need to understand a lot about the characters and the world extending into the past and future.
Will it be possible for levels designed in the current Valve Hammer Editor to be converted into Half-Life 2 levels?
It should be pretty simple. We have taken a lot of existing content and made sure it migrates easily to Source.
In terms of editing, will Half-life 2 use .wads for the implication of textures?
No. In Half-Life 2, we use a file system stored inside a ZIP file. You can embed these ZIP files directly into BSPs if you want to.
As far as textures go, will the limit on their resolution still be 256x256, or is it something larger?
We support 2K by 2K.
Could you inform us of what the maximum "acceptable" polygons is for both levels and models?
That's really an art and design issue more than a technical issue.
Will Half-Life 2's multiplayer be operated from the newly adopted Steam software?
Half-Life 2 will be available via Steam.
Will the multiplayer side of things still be under a thirty-two player limit?
That's really a design decision at this point more than a technical limitation.
With the bar set so high in Half-Life 2, and the now justified wait after the first one, can you even fathom what Half-Life 3 will encompass?
Sure. We don't really have much trouble figuring out what we want to push next, it's usually more of a problem for us to postpone things.
Is it feasible that a Half-Life 3 is soon to follow from the proverbial backburner?
No.
The multiplayer aspect of Half-Life 2 has been kept tightly under wraps, and left most people to guessing. Some people think it may be the classic Half-Life Deathmatch style of gameplay, others think it could even be Team Fortress 2, and others say it will be something we haven't seen or even heard of before now. You might not want to give out any information on the multiplayer aspect of Half-Life 2, but I can tell you that we really want to receive information on it. Could you give us a few details about it?
Not yet.
Doug Lombardi recently gave the confirmation of Half-Life 2 being developed on for the Xbox. Are there any features that have to be scrapped due to the Xbox hardware, or has everything seem to have come threw fine?
The Xbox fits into the middle of our scalability goals for the PC product. It's more of a question of how much we want to optimize for the console than anything.
In the HomeLan Fed interview, you mentioned that the MOD Expo this year will be more about introducing the MOD teams to the new technology of Half-Life 2. Will the SDK only be released to the people at this year's MOD Expo, or will the general public get their hands on the SDK prior to the Half-Life 2 release, and how soon will this be?
I don't know yet. It depends upon the reaction of the first people to get the SDK.
Should we expect to see a playable Half-Life 2 demo prior to its release?
No.
--------------------------------------------------------------------------------------
So what Im getting from this is that HL2's Multiplayer mode is probably gonna be TF2, since they refuse to talk about it.

[THE JEW (RaVeN)]

This is the shack article, yes? I remember reading it; just not where it came from.

[Ghost [PX]]

Yes I believe that it it is the shack article.