We may earn a commission when you buy through links in our articles. Learn more.

Half-Life: Alyx - how Valve brought its most anticipated series to the least understood platform

Valve dishes on the expectations of Half-Life fans, the future of VR, and building a game that feels like an interactive cutscene

Half-Life: Alyx Index

Over a decade after Half-Life 2: Episode Two left us on a maddening cliffhanger, Valve has released a new Half-Life game – though true to Valve’s standards, it’s not the game that many fans were expecting. Half-Life: Alyx is a virtual reality title, and one that’s at risk of getting crushed under the expectations of both Half-Life fans and VR evangelists.

Astoundingly, it lives up to those expectations – but you can read our Half-Life: Alyx review for more on that. We were scheduled to visit Valve earlier this month and speak with the developers in person about their approach to building Alyx, but as you might suspect, the whole ‘global pandemic’ thing threw a proverbial crowbar into those plans.

Instead, we spent a pleasant afternoon Skype call with programmer Kerry Davis – a long-time Valve employee who’s been there since the original Half-Life 2 – and designer Chris Remo – who joined Valve alongside the rest of Firewatch studio Camp Santo in 2018 – about managing those expectations, building a new type of triple-A game, and, of course, the number three.

PCGamesN: What’s the feeling around Valve since Half-Life: Alyx was announced and you’ve seen some of the early public response?

Kerry Davis: For me… relief, I guess? Playing the game internally, you could really feel that it was going to be something special, but you still just never know how it’s going to be received. There’s so many unknowns, especially in VR. Even trying to understand what it is that people are expecting and what they want and what they’re looking forward to.

It was gonna be a big mystery once we announced, to see how people reacted – and the reaction has been great. I’ve been very pleased with it. It just makes me more excited to polish every corner of the game that I can, to make sure it meets their expectations now.

Chris Remo: Yeah, I would echo everything Kerry said. I think relief is definitely a dominant feeling here. In the run up to the announcement, I feel like we were bracing ourselves for the worst possible reaction we could imagine. We’ve been working on this thing for a while, and you never know. It’s like this every time you work on, particularly, a single-player game. You don’t know what it’s gonna look like on the other side.

Then, as Kerry says, particularly in VR, which is less fully-explored than non-VR games. The reaction’s been absolutely amazing. It’s been incredibly gratifying. I’ve been on this project less time than Kerry has – I’m newer to Valve. So I can only imagine for the people who have been working on this longer. But the feeling is pretty great right now.

YouTube Thumbnail

How long has Alyx been in development at this point?

KD: I think it’s been about four years. That’s what I’ve been hearing. I tried to look back through my history to see when I joined the project, and it looked like it was about three-and-a-half years ago. So that sounds right to me.

The announcement of Alyx has opened up speculation about the number three all over again, and some folks at Valve have suggested the possibility of further games in the Half-Life universe. Do you see Half-Life continuing from here?

CR: The sensation, shipping this, is that it’s – maybe shouldn’t be a big surprise – but it’s incredibly fun to work on Half-Life. I think that’s been just as true for the people who have been here a long time and shipped multiple Half-Life games as it is for those of us who are newer to the studio and come to it from the perspective of people who have played all of the Half-Life games and been really influenced by them.

My design sensibility has been enormously shaped as a game developer by the Half-Life series. Along with the relief that we were talking about when it comes to this game finally being out there and almost to the public, there’s also a really strong sense that this is a great world to be working in. I don’t think there are very many people at Valve who wouldn’t want to keep exploring that going forward.

KD: Yes, I agree. I personally want to keep exploring that going forward. [laughter] I started at Valve the summer that Half-Life 2 shipped. I came in at the very end of that process, and I got to see the result of the work, but wasn’t there as a part of the work. Then through the episodes, I got a taste of what it’s like to create a Half-Life game, and it is extremely fun and exciting. But there’s also a lot of responsibility that you feel while you’re doing it, to live up to the franchise.

With Half-Life: Alyx, it was interesting to come into this new project from the beginning and enter back into that space after such a long time away. It gets those creative juices flowing again and it’s been really exciting.

Have the expectations fans have about a new Half-Life game made you nervous at all?

KD: It did. At this point in the project I can say that my nerves are settled, because every time I play through it, I feel like we’ve met that bar. As long as I’m representative of some percentage of our customers, then I know that there’s going to be a lot of happy people who feel like we did it justice. We treated it well.

CR: I would agree with that. Watching the game come together and seeing how it’s been polished and improved over the last year has been pretty remarkable. There’s always something amazing in game development when you see everything coalesce. Everything starts to turn into the thing that you’re hoping it would become. If you didn’t necessarily even know what that was going to look like, there’s still a hope that it’s going to add up to something more than the sum of its parts.

Seeing that happen over the last year – particularly in this franchise that I have not personally worked on before, but I think everyone feels this way – has been pretty amazing. We’re at the phase now where a lot more people are starting to play the game, and in some cases play it all the way through. We use a lot of external play testers, to make sure that we have people who aren’t deeply familiar with the game playing it all the way through, and getting good feedback and bug reporting and everything else. Then there have been members of the press who have been in here playing through the game for various reasons.

Someone, a week or two ago when they finished the entire game, said something to the effect of “this is what I was hoping it would be, but not what I expected”. They were like, “you guys keep saying this is a full Half-Life game – a full-on, full-scale Half-Life game – and I wanted to believe that but I didn’t, really. But now I get it. Now I see what you’re talking about.” That’s happened a few times in the last couple of weeks and it’s a pretty great thing to hear.

KD: I was so fascinated by VR when we started working with it, even before The Lab, so I volunteered to help with demos for external partners and tech people and stuff. Every time they would come in skeptical, we would put them in the equipment and play theBlu – put them in front of that whale – and it just completely transformed their opinion. That experience was repeated over and over and over and I loved it.

This game is giving that same thing. People come in sceptical and then at the end, they’re like, “wow, that was so much more than I expected,” which is really exciting.

Previous Half-Life games have come alongside major technical transformations. The first Half-Life set new standards for AI, presentation, and set pieces, and Half-Life 2 expanded that with physics. Did it feel like you needed a similar paradigm shift – like VR – for Alyx to justify itself?

KD: My personal thought has always been that this should be to Half-Life 2 what Half-Life 2 was to Half-Life 1. I wasn’t always convinced that we could accomplish that, but I felt like that was a goal that we should be shooting for. VR was a really nice boost to that goal, because it gave us a whole new space to explore with the input, the interactions, the experience, and the atmosphere.

It would have been difficult to make that kind of dramatic leap with another flat screen first-person shooter. We were able to use VR as a way to elevate this and give it that step up from Half-Life 2, that Half-Life 2 was from Half-Life 1.

CD: We released a few gameplay trailers a couple weeks ago. Those were really fun to work on, they’re all direct-feed capture from the game itself – that’s how they were made. When they came out, I was clicking through some web forums and looking at people’s reactions. There was a comment, where I read it and I thought, “that’s sort of an odd thing to say.” The person said, “all these object interactions, it all looks like cutscenes.” I was like, “what do they mean it looks like cutscenes?” I realised I had been taking for granted, because of working on this thing, all of the things you can do with objects in the game. They didn’t mean it pejoratively.

They were taken aback by seeing your character interacting with all this stuff in ways that, yeah, if this were not a VR game, those would all have to be individually animated. But in the context of this game you’re doing all this stuff with your hands, and you’re operating like you would if you were interacting with these things. It ends up looking sort of shocking in the context of a videogame. Working on it, I took that for granted. I read this person’s comment, and I was like, “Oh, yeah, this actually is a totally different kind of interaction than previous Half-Life games have had.”

That was one of a number of things that when you’re just in here working on it, you start to forget how shocking and surprising that is. But then you see people’s reactions to it and you’re reminded, “oh, man, people are sitting down at this whole thing and they’ve got the entire game in front of them.” It’s going to be a really, really, really, really exciting thing to see.

On the opposite end, you’ve got VR fans looking at Alyx as a sort of ‘killer app’ for VR gaming.

KD: I am a huge VR fan, personally, so I really want VR to stay around. Outside of my connection to a company that has invested in VR, I just want VR. So I’m doubly motivated to make this succeed. I really hope that it helps. I mean, it certainly can’t hurt to have something really fun to play in VR if we want VR to stick around. I’m pleased that at the very least we’re not gonna hurt the platform. Hopefully we help it.

CR: We’re in a somewhat unusual position in that, because of the way that Valve is run, and because of the way it’s set up as a company, Valve is in a position to be able to make something like this – for reasons that are maybe not the the traditional, extremely strict type of financial considerations that would usually go into making a game of this scale. I wasn’t here when this game kicked off development, but it’s pretty clear that a big potential effect a lot of people are hoping this game could have is, “look, you actually can support a full-scale, triple-A game campaign in VR.” That’s a thing you can just do.

It’s understandable why there haven’t been a lot of those so far, because it has been kind of an unproven notion for a while. Fortunately, as a company, we’re in a position where we can go out and put the resources into saying, “No, this is actually totally feasible, you can make this. You can do it. It can be really good.” That’s hopefully something that has an impact in that area.

Valve is in a unique position to put resources into a triple-A VR game like this. How does it feel, creatively, to work on a project that just couldn’t happen at a lot of studios?

KD: Having the flexibility that our company situation allows, to forge ahead with this work, and not have to stress every morning that they’re going to shut it down because we’re spending too much money on it – it frees you up. To know, we’re investing a lot in this and there’s a responsibility to really make it great, but I can also focus my time and energy on the work that I’m doing and not worry about the politics of it or whether this project is being viewed as worthy of our time or worthy of our investment. Just knowing that, “hey, you’ve got the reins go out and do this thing and make it amazing – do what you have to do to accomplish that” is great. It’s a great situation to be in.

CR: I think that definitely factors into what I was saying earlier about seeing the game come together. Seeing the level of polish, care, and overall work, effort, and improvement that has gone into the game over the last year is pretty incredible. I think a testament to what Kerry is saying. It’s very clear that people are overwhelmingly focused on putting their best work into the game, really above all else. That feels like what the main – arguably only – consideration is from a development standpoint.

Do you feel like you have to address the cliffhanger from the end of Episode Two in some way, even with this being a prequel?

CR: To that, I’m probably not going to give a very detailed answer. Other than, yeah, It wouldn’t hurt to brush up on how Episode Two ended before playing this game, but beyond that, we’ll let you let you make your own discoveries.

How do you feel Alyx works for players who are new to the series?

CR: I would hope pretty well. I think it’ll be a pretty cool thing for players to experience the dialogue in a Half-Life game. By dialogue, I mean, two ways instead of just going one way. That’s something that is facilitated very much by Alyx being the protagonist rather than Gordon Freeman – which isn’t a permanent ongoing statement about the Half-Life franchise or anything. But for this particular game, it’s a really cool thing to have this back and forth between Alyx and Russell, this other character who’s very important in the game.

This game is a big payoff for people who are deeply familiar with Half-Life. But if you’re not familiar with Half-Life, and maybe you’re coming in more from the VR angle, for example, I think that it’s a pretty easy transition in. I think having that back and forth is just a fun thing that most people are going to enjoy a lot.

KD: When I started as a developer at the end of Half-Life 2, I actually didn’t know anything about the company before that – which I probably shouldn’t admit. So I hadn’t played Half-Life, and my first experience with Valve was playing Half-Life 2 about three months before release. Of course, I realise now that I missed out on a lot of the lore, the callbacks, and the characters’ history, which adds a lot of depth and color to the game. But that didn’t matter because it was so much fun to play. Just as a standalone game, it was amazing for me.

I think Half-Life: Alyx is going to have the same thing. There’s a lot of great callbacks, bits of information, and story in there that Half-Life fans will get to absorb. But even if you’ve never heard of Half-Life before, it’s just going to be a great game experience. I think that’s probably enough for a lot of people.

CR: I totally agree with that. Particularly City 17 – and this definitely was established with the very opening scenes of Half-Life 2 – it’s just such a rich, evocative world. You can drop right into that and so much is communicated just by being in the middle of it.

After so many games with Gordan Freeman as a silent protagonist, here we’re playing as Alyx, who has a lot to say. Were there particular challenges with having that sort of dialogue in a VR game, where the player is embodying the character they’re playing?

CR: In any first person game, there’s always a challenge when your protagonist speaks. It’s a writing challenge, and a design challenge, and, honestly, an audio challenge. Any particular element of a design is going to spiral out into a bunch of different disciplines, and you have to solve them for the game in question. Even things like how many back and forth lines do you have.

Working on Firewatch before this – pretty much the entire game was solving problems like that. It’s a smaller percentage of Half-Life: Alyx, in the totality of what is in this game, because you have combat and traversal and world-building and exploration and object interaction and all these things. But it’s certainly still a design challenge, and you have to be incredibly cognisant of making sure the players understand who they are, and who they’re talking to, and what this means, and make sure it doesn’t get stomped by things, or that it doesn’t stomp over other important things that the player should be cognisant of. There’s a whole raft of things you have to be aware of. It operates in a slightly different design space narratively in that regard than other Half-Life games do.

KD: Early on, there was a moment where you reach a point where Alex is having a very serious conversation about her father. At that particular point in the map, we’d dropped some surprise barnacles sort of hidden around the corner. So it always timed out that you turn the corner and in the middle of this very heartfelt, serious speech, you’d get caught by a barnacle, and you’re getting pulled up and strangled, and pulling out your pistol and shooting over your head wildly trying to save yourself.

That friction, that’s the issue you deal with. But fortunately, that’s not really a VR-specific problem. If the character is expressing a feeling and the player is having a very different feeling, then you have that collision. That’s a tricky thing to solve, but it’s a fun problem to work on.

CR: It’s also a great example of the kind of problem that playtesting will really surface.

The first level of Alyx – where you’ve got some quiet rooms filled with objects to interact with – almost feels as much like a tutorial for VR as it does an introduction to this game. Was that intentional?

KD: Yeah, absolutely. You always want to be able to teach your player as early as possible how the game works, what the rules of the world are, and what kind of agency they have within it. You can just tell them that through an instruction manual, or you can try to lay it out in front of them, and encourage them to play in ways that will answer those sorts of questions. Like, “Hey, here’s some fun stuff you might want to pick up and play with.”

In the process of that, you’ll learn what it means to pick up something, pass it from hand to hand, and drop it on the ground and things like that. That’s very intentional that the game is trying to teach you as you go as early as possible.

CR: Yeah, totally. Even just in the two years I’ve been here, the opening of the game was revised more heavily than almost any other major area of the game. It changed so much. A big part of that is to make sure this is a good place for you both to learn these things, and also to enjoy being in while learning them.

When the player first gets in there, they’re gonna want to use all of these affordances they have in VR, and pick things up, look at them, and put stuff around. So that environment better be an appealing place that makes you totally fine with doing that, and isn’t like some sterile cube or something – not that it would ever be a sterile cube. But that area definitely was redesigned a few times to try and work towards the goal that Kerry is describing.

KD: Or sticking the player into a dark, terrifying cave, expecting them to explore and figure out the interactions when they’re already frightened. You want to make it safe for them.

Did the contiguous, first-person pacing and structure of previous Half-Life games make it easier to adapt the setting to VR?

KD: Pacing-wise, it worked well, although we had to adjust to the fact that the pacing is slower in VR. People take more time in each space – they spend more time lingering over the stuff that they’re exploring and poking into corners. It’s a little bit slower paced than the previous Half-Life games. The difficulty that it added was with one of the hallmarks of Half-Life: the interactivity of the world, where the player feels like they have a real physical presence, and they can affect the world and interact with it. In Half-Life 2, it was all the physics that gave us that.

In Half-Life: Alyx, we felt from the beginning that, to continue that, we needed to have hands that were solid, physical, and collided with the world. When you picked up an object, you gripped it in a reasonably convincing way. Those were all fine goals, but they’re also really difficult things to do in VR, which is why a lot of games make the very reasonable choice of having your hands just be non-solid – they’re just sort of ghost hands. Or when you go to pick something up, your hand disappears, and now you’re just holding the object. Or it’s like a point-and-grabber type thing. Having the hands need to be physical in this way presented a huge number of challenges. But it was necessary to keep the feel of Half-Life continuing in this game.

Was there anything about the Half-Life universe that was especially challenging to adapt to VR?

CR: I didn’t really see a lot of things that felt that way. The genesis of the project, which I wasn’t here for, started by essentially bringing Half-Life 2 assets into VR, and just playing around with the mechanics and principles of a Half-Life experience. People found that it actually worked really well, without even needing to do a complete overhaul. There were definitely challenges throughout, but I don’t know that there were a lot of really core Half-Life things that felt fundamentally at odds with the medium.

KD: Everything seemed to fit pretty well. It was actually an advantage, the cast of characters that we’ve had in Half-Life, because there’s such a range of scales. Scale is fantastic in VR. The fact that we already have these Striders – having them march through the city over your head is wonderful in VR. So we didn’t have to invent a bunch of new characters or mechanics to show that off. We already had all of that at our disposal.

On the other end of scale, Alyx is filled with all these tiny objects you can examine to get more information about the world – sort of like Gone Home. Like the newspaper early on that has a bunch of detail about recent events. How do you approach narrative design with all those extra opportunities for environmental storytelling in VR?

CR: Yeah, that was really fun. One of my small contributions to this game was writing all the text on that newspaper, for instance. I spent a couple weeks doing a bunch of in-world little environmental storytelling props and stuff. It’s incredibly fun. You brought up Gone Home, for instance – that’s a game where you can rotate items around you in this sort of disembodied way. Every non-VR game has to solve that problem uniquely.

Every single game that uses a mouse and keyboard or controller, you have to decide: Can you pick up objects? If so, how? Can you inspect them from different angles? If so, how? When you put them back down, how does that work? Every single one of those is sort of a bespoke solution for a game. In VR, you still have to implement things, but the baseline assumptions are kind of screaming in your face about how obvious they are. Yes, people have hands. They want to pick things up and look at things.

That’s both a technical demand and a creative one. Kerry certainly can speak a lot to that. When it comes to filling out the world, you want stuff that ideally is going to hold up when players are putting things close to their face and looking around. And, as Kerry said, spending more time in these different environments and making use of both their mobility and their ability to interact with objects.

On top of all that, when you’re in a world like Half-Life, where there’s 20 years of games with a really rich world and environment. There’s just a ton of opportunity. You don’t want to go overboard and make everything a lore dump left and right, but there’s a lot that you can dig into there, and there’s really rich soil to till when it comes to giving the player stuff to chew on.

KD: Yeah, I didn’t even realise some of the props that were in there. I’ve YouTube videos where people were exploring the stuff that came out in the Steam VR Home environments that we shipped. Going through each little prop that they found and examining them. I was even surprised by some of them. There’s just so much stuff in the game. No single person can be aware of it all. [laughter]

VR still has limitations in terms of audience – whether because of price, or because people get motion sickness or have mobility issues. How do you feel when designing a game that has an inherently limited audience?

KD: I think it would have been a lot harder – I don’t know how many years ago, time compresses as you get older – but it would have been more difficult before the prevalence of Let’s Play videos and people watching people play becoming a common thing. That’s what people do. If you can’t play the game or don’t have time to play the game, you just go watch someone play the game. That’s gonna help a lot. People will be comfortable with being able to watch someone else play if they don’t have access to VR.

My feeling of it is that this game is going to come out and it’s always going to be there. Because I am so hopeful for the future of VR, I feel like some people may have to wait, but the time will come when a friend will have a headset, or it’ll be available publicly somewhere, or it’ll be cheap enough that they can get their own. And then they’ll go back and play this game. It’s gonna be a slow spread. But I’m keeping my spirits up by thinking that more people will get to play it in the future than maybe get to play it on March 23.

CR: The landscape is changing all the time. The entry level VR headsets are at this point getting more affordable than ever. And with respect to accessibility, we’ve tried really hard to support as many different options as we can, to try and close that gap as much as possible. There are the three different core movement modes, the blink, the shift, and continuous movement – and then continuous movement has an additional variant within it. Then we have sitting and standing.

I saw an email from someone internally the other day who has difficulties with VR from a motion standpoint, and they were saying that sitting mode plus continuous movement was the sweet spot for them. Once they found that combination, everything clicked into place, and the game was super playable. So there are a lot of different ways we’re trying to allow people to try different options and see which ones work for them. I think people online have really, really strong opinions about this stuff, particularly people who are already super bought-in to VR. A lot of people feel very passionately that continuous movement is the only valid way.

But I think people should just try all the different options we’re giving them. You might actually be surprised by what you end up preferring, and the game is fully playable with all of those modes. It’s playable with any PC-based headset you have. So hopefully it’s totally true that, over time, that gap will naturally shrink, but we’re trying as hard as we can right out of the gate to really smooth that over as much as possible for as many people as possible.

KD: It’s been fun testing on what we call our min spec, the low end machines, and seeing how well the game runs. It still looks really good on those. It’s been exciting to see that people are going to be able to play this on whatever hardware they have, which is great. Also, single controller mode is kind of a fun option, and the ladder free climb. That used to be connected to continuous movement.

I have to teleport because I get motion sick very easily. But I really like climbing the ladders. For some reason, that doesn’t bother me, so we made that a separate option. So you can still teleport from place to place but if you want to turn on the continuous ladder climbing, the physical climb, you can do that. You can build your own experience that way.

How challenging is it to develop around all the different VR headsets and their differing capabilities?

KD: That’s a tricky problem. We have many long debates around the office about how this control should be mapped to a particular controller. It’s not even just finding enough inputs to connect the actions to, but having them make sense within the context of a single controller. You can just stick stuff on the buttons and call it a day, but if it doesn’t feel right when you’re interacting… There are similar actions, like releasing the clip from the pistol or opening the shotgun, those should probably be done in a similar way, because they’re a class of action. You’re trying to categorise things and group them.

Since I’m working on interactions, I have four different controller types at my desk, and I have to switch between them. That is really straining on the brain, to recalculate all of my inputs every time I switch controllers. I don’t recommend switching while you’re playing, but we’ve got them all working really nicely. So I’m happy about that.

How has the current state of VR influenced Alyx’s development? It feels like a lot of the movement options in particular are building from what others have done in the space.

KD: That’s a big part, since VR is growing so quickly. People are exploring what it means to play a game in VR. The options and the ideas are changing constantly. You want to be aware of what’s happening, what people are liking, and how they want to experience their VR, so that you can make sure you’re delivering the sort of experience that they’re looking for.

CR: I think the important thing is to take ideas from anywhere. Take good ideas from anywhere. If there are other games that are implementing things that seem like they would allow people options that are going to make the game better for them, definitely look at those. Don’t worry about the sort of big, ideological, ‘the game must play this way versus that way.’ Take the good ideas from wherever they’re coming from.

Are there any concepts you tried that you found just did not work in VR?

CR: I think undue movement, like moving the player a lot without the player’s input, is certainly something to be used very sparingly and carefully.

KD: Yeah, definitely that. There are certain behaviors that can be tricky to replicate in VR, because we really only have three points of data. We know where your head is, and we know where your hands are. Things like when you’re standing on a catwalk and you want to look over the edge to see what’s below you, it’s difficult to determine from the input whether you’re just looking over an edge or you’re actually walking off the cliff. Some of those types of movements and interactions are difficult in VR, because you can’t just push a button and say I’m leaning – oh, actually, maybe you could. Hm. I’ll have to go back and look at that.

But yeah, things like that. In general, most concepts just work. They just translate. You just have to figure out how to enable the player to perform them.

What’s the weirdest issue you’ve seen come up in playtesting?

KD: As the interaction person, that’s always where my focus is. It’s usually just watching people fail to interact with something that seems so obvious in real life. I talked before about doors, and how long it took to encourage people to actually open a door like you would physically. With gamers it was almost worse because they’re so ingrained with this ‘press E to interact.’ You just walk up to the door, press the key, and it opens. You go into VR, and you’re in a game, so you assume things are going to work the same way. So we were trying to free people of those expectations that they’ve learned over the years, and encourage them to just treat it like a world, and explore it physically and see what you can and can’t do that way.

The timing with Black Mesa’s release has sort of made March into an unofficial Half-Life month. How do you feel about these releases coinciding with each other like that?

CR: Yeah, I didn’t even realise that until pretty recently. It certainly was a coincidence. I mean, it’s cool, good for them. They shipped this very ambitious thing that’s been in development for a long time, like ten years. It’s been in the works for a while. I watched a trailer the other day and a lot of the stuff in there looks incredibly beautiful. I hadn’t really thought about it from the standpoint you’re saying, like “Half-Life month,” but yeah, that’s cool, if Half-Life is even more in people’s consciousness.

KD: Just have everybody thinking Half-Life all at the same time. I mean, give them something else to think about right now.

CR: We can all use that.

KD: Give them a reason to stay home.

CR: Yeah, there you go.

What’s going on with In the Valley of Gods?

CR: In the Valley of Gods is, I guess, on hold for the time being. We were working on it when those of us from Campo Santo came up to Valve, and something we found over time was, in contrast to working on that game, Valve was a place with a lot of teams, a lot of projects, a lot going on.

There wasn’t really ever like a particular moment or decision or anything, but people have been working on other things here at the company. I worked on Underlords for a bit. I’ve been working on Half-Life: Alyx now for a while. We just eventually came to a conclusion as a team like, ‘Oh, it looks like we’re all kind of working on other stuff right now. Let’s put this thing on hold for a bit and figure out over time what that means.’ Really, that was about it.