Half-Life 2: 20th Anniversary Update
half-life.com
half-life.com
I could never get used to Ravenholm though. Thinking about it even now years later stresses me out a tad.
To be fair, I did enjoy it and could feel the intent of it being the horror section. I could see it usettling me if I played it at the time, but I see a lot of people saying it holds up really well so I still feel I've missed out on something.
If you want a similar feeling from a more modern game I recommend trying Prey (the new one). Terrible name that doesn't fit the title imo, but playing it felt similar to when I first played half life years ago, more than any other game I've played since then at least (except Half Life: Alyx).
I definitely think the atmosphere of Ravenholm contributed to my feelings about it, but I suspect if it was just the slow zombies or was all puzzles or something I would actually really enjoy it. There was something about the atmosphere + claustrophobic spaces + fast zombies in places I didn't like. Or rather, I did like it but not as much as the rest of it.
Also, at 0:30, turning off that living place hazard is very kind of you. And you also saved Bob and Jim from being cut in half by doing that. Father Grigori might not like it, but he's crazy too.
e.g. the bit where I hide behind behind a headstone in the graveyard is by no means reliable (maybe 80%?)
Thanks to HN the view count on YouTube has increased dramatically! If it gets past 1000 I'm going to have to record and upload the final section with the combine troops.
EDIT: assuming that is still possible: Valve have made some changes for the twentieth birthday which I noticed today.
e.g. there's more grass and bizarrely the lift button has switched to the other side of the elevator
It highlights some of the architectural decisions (and how players exploit them), and has gems such as one of the commentators having their entire section skipped by the speedrunner.
The day that dev looked at player time-location heatmaps, investigated outliers, and then cried...
(disclaimer: I work for CodeWeavers, we sell CrossOver which should be a great and easy way to play)
edit: tested out the 20th anniversary update on M2 Pro, it works great!
Better translation layers are what made the Steam Deck possible.
One needs to realise graphics APIs have historically not been terribly efficient.
There was a reason Mantle/Vulkan came to be.
If Apple doesn't care about backwards compatibility on their platform, why should Valve or other developers be burdened with the added cost, especially on a platform with small market share that won't drive many sales?
* I first installed Whisky, a Wine/Game Porting toolkit/app thingy (https://getwhisky.app). Then I installed Steam for Windows inside Whisky, then downloaded HL2. The download was a little bit hiccup-y, but it worked. Testplay of HL2 was great, apart from the mouse.
- and then debugging / figuring out how to hack together a version that would compile in Visual Studio (back when you had to have a license… in theory…) as a preteen -
had a way bigger impact on me than I gave it credit for at the time. I think seeing the sheer scope of that codebase and realizing how much I didn’t understand and still had to learn is part of what put me in my path into both the game industry and software startups. I found both Valve’s and Gabe’s stories that much more compelling after gaining an appreciation for how much there was under the hood.
Cheers to everyone who has worked on this engine, this project, and this company over the years. I suspect your influence runs deeper than you know.
I suppose this is the very definition of developer integrity.
However it’s also worth noting that they were spending that time on other projects making them likely even more money. HL3 isn’t going to have live service or esports applications that make money like their other games do. It probably won’t be a big earner compared to those other games.
Valve are also a privately held company and the rumour is the top people there are crazy rich, and even those further down the ladder earn a lot more than they would elsewhere. They don’t need HL3.
We all want HL3, but other than the story it would probably be fairly stagnant in terms of gameplay - nothing new has really happened in the genre.
They dont have to do anything new, just make it good enough, like Titanfall 2.
> I believe they also said they don't want to do more of the same
I've heard this too, with Half Life they want each game to be groundbreaking. Arkane have done great games, but they're not all groundbreaking, and arguably none of them are to the level that HL1/2 were.
But Gabe has said Half-Life specifically is what they consider their ground breaking series now. So there won't be a 3 until they can push a boundary again like the first 2 did.
This is all speculation based on code leaks, and it's possible this is all true and they still don't release anything because the project stalls. Supposedly the game has moved into the later stages of development, but I'm not fully convinced by that part of the leaks.
https://luxurylaunches.com/transport/gabe-newell-luxury-yach...
Not excusing this behaviour, but there is a reason for it.
HL1 and HL2 each changed the industry. Alyx is mostly forgotten.
[citation needed]
Valve has been mostly silent on the issue (after hyping up EP 3 themselves).
Portal 3 seems like the easy money grab. That had more lasting pop culture relevance, and "it's more Portal" would be enough for a bunch of people to buy it.
For what it's worth, I never even wanted Half-Life 3. I just wanted Episode 3, which was what they originally promised. It didn't need to be some gigantic groundbreaking thing, just another episode to finish off the story.
In short, they started working on episode 3, and made it about the point where the game development progresses from "here's demos of what could be set pieces in the episode" to putting together the sketch of the game narrative [1] when the developers were pulled to crunch for Left 4 Dead. After Left 4 Dead released, Gabe felt the window for an HL2e3 had closed, so it would be HL3 instead... and that because HL1 and HL2 had both heralded genre shifts, HL3 needed something to push that shift as well. But there weren't any such ideas for that shift (except maybe VR, which hasn't exactly panned out), and Gabe also felt that releasing a HL3 just to complete the story wouldn't cut it, so it never got made.
Which pretty much matches everybody's speculation for the past decade.
[1] About 18-24 months away from being releasable at that point, to give a sense of how far from complete it was.
I guess they could, if not flat structure and billion-dollar yachts and what else is there.
Basically, they wanted more than just simply finishing the story, they wanted to make some cool new features, similarly to what HL2 was compared to HL1.
https://store.steampowered.com/app/220/HalfLife_2/
Also try clicking on the gravity gun at the end of the anniversary page.
https://projectborealis.com/prologue-release/
Looks great, and feels pretty close to HL2 mechanics! Definitely has to be optimized more, but a very promising start (albeit short, ~10-15 minutes).
But the graphics looked much nicer in UE and not to mention the tooling is nicer than what one could use with mods. Surely, the upfront cost of porting aspects of HL2 over to UE is not investment lost.
Recently, it was revealed that even the S.T.A.L.K.E.R. 2 benchmarks used upscaling and the problems with visually stunning but horribly performing games like The Forever Winter were also pretty obvious.
I did very much appreciate what Black Mesa did, though. Maybe Project Borealis will also be a success story in due time.
However, it should be noted that this isn't an issue for consoles (where precompiled shaders come with the game) or Steam Deck (where shaders are compiled via Fossilize before the game is even launched). It's most notable on DirectX programs during the first run, which hits benchmarks quite badly but also becomes less stuttery as more shaders are cached.
On console, precompiling makes sense because you know exactly what hardware a user will have and you can optimize for one or two sets of hardware. The effort required to automatically package and download these shaders for users on their first load is worth it, so it's a viable fix to shader stutter in games where it crops up.
On Linux, precompiling makes sense because shaders take an extra long time to process due to the DirectX -> Vulkan translation. Since this causes stutter in every game, a precompilation step is pretty much mandatory for everything but conveniently also solves the UE5 stutter issue at it's roots.
On PC, it's basically a maelstrom of worst case scenarios. You don't know what hardware a user will have, so you can't package precompiled shaders. You're not translating shader calls so you have to rely on each version of DirectX's specific DXIL features instead of the unified Vulkan 1.2+ SPIR-V that you get from DXVK. And of course, even if you did get a magic "Compile the Shaders!" program working most users wouldn't bother since they're impatient. Some games try adding optional precompilation screens, but I wager most people just skip them when given the opportunity.
edit: on iPhone
I tried to play it much more recently, maybe six or so years ago on an Intel MacBook before they broke 32-bit, and once again even on modern hardware the game took FOREVER to load.
Maybe this is just my experience? I haven't heard anyone else complain about the terrible load times?
+map_background none
Lost all of it obviously. Not a single company has my loyalty anymore.
Except if valve were to release a mystery black box with faint lambda symbol on it. I’d pay whatever they asked for it.
FromSoft
Game developers and publishers start shitting the bed when they IPO and need to juggle the conflicting interests of managing investor relations as well as customer demands; that or when they're acquired and turned into a subsidiary.
I couldn't deal with my one of our party members, but have no issues with COD PUG's. Go figure.
Which I imagine doesn't lend itself to doing hard things like making Half Life 3...
Why would any game dev choose to go through a death march to perfection, if they had other project choices?
Does it lend itself to do other hard things like half life alyx?
Which also shows yet another one of Valve's problems with making games, they treat their games like they're "tech demos", so unfortunately they're not as interested in actually moving the stories in their games forward or bringing them to a conclusion. They do a "tech demo", they move on from that tech, leaving the game and it's world and community behind. Plot? What plot? Perhaps they're also stalling on making continuations or even new releases in search of some "gimmick technology" to pair a game with, instead of just telling a story through their games. For those people that do like the narratives and the worlds in their games, it sure is tough luck. There's more to a game than just 'tech', but alas.
They might have had enough original ideas for HL2 (and then Alyx, driven by a new medium), but still not HL3...
The entire media industry on almost every format is chasing nostalgia because they refuse to recreate the environment that made endearing stories and experiences in the first place.
which means they have no obligation to ship. And so it is with the valve-time, they never shipped.
Some pressure (monetary usually) is required. Not to mention that "strive to make art" is not a commercially viable objective - the owners of steam will basically be operating a charity for these artists.
If that's what it takes to make something worth playing, then so be it.
Was Bungie in its day a charity? Or did they just get it? 20 years later the magic is gone and Microsoft is desperately trying to figure out how to make the goose lay an egg. As long as they're optimizing quarterly reports they'll never get there.
Right constraints make things happen.
Roberts was the lead on Digital Anvil's "Freelancer", until the publisher (Microsoft), frustrated at the scope creep and protracted dev cycle, bought out the studio, demoted Roberts, and cut features so they could ship the thing.
Sure single person self-funded passion projects exist. They always have and they always will. And sure what one person can do is more than they ever could in the past. It's still not the same as something that's forged by a team of visionaries each with unique backgrounds and skillsets.
Frost makes the point more well spoken and stylish than me often.
At Cannes and your local 'art' cinema ? To be fair, I don't watch movies much, but I do still go to these sometimes.
> Where are the games publishers on low budget games?
Who said anything about publishers ? (And Valve dumped theirs as soon as they could.)
> Sure single person self-funded passion projects exist. They always have and they always will. And sure what one person can do is more than they ever could in the past. It's still not the same as something that's forged by a team of visionaries each with unique backgrounds and skillsets.
Ok, I have no idea what you're talking about, are you "no-true-scotsmanning" here ?
We have a great recent example : "Factorio (: Space Age)", which started as a one-person idea, took form as a 3-person company, got after release a 20k€ Indiegogo funding, then blazed a trail of success over the next 12 years, now with something like 5 million sales for the base game and a 30 person company.
How is that not "a team of visionaries each with unique backgrounds and skillset" ?!
Or the amateurs at Spring-Recoil / Zero-K / BAR, which show how you can do that even better than the professional, commercial RTS.
Or indeed one person projects like Shadow Empire (with some publisher support), which show how you can make a brilliant 4X/Wargame on what I assume is a tiny budget...
And there are probably many other examples here...
I get that you're trying to discredit the argument by claiming fallacies, but these aren't just my views. Industry insiders (Frost in games and RedLetterMedia in movies) have been talking about this for nearly a decade.
"Industry" insiders' opinions are irrelevant, they are just too bogged in the day to day details, they tend to forget that 99% of everything is crap and that's fine (and they do that because they have to make a living there, their incentives are different).
And you cannot predict greatness (you are the one that talked about 'visionaries', remember ?) - specifically of new teams you've never heard about before (of course once they did something great it's another thing, even with reversion to the mean they can have a lot of other successes).
It could be a decent environment for making great games if Gabe wantet it to be. It isn't one right now, Gabe is still a businessman first.
I had the same experience with the Steam Deck: just very well done, including side things like the case that came with the device. I've grown used to accessories bundled with electronics ranging from basically garbage to okay (but not great), while Valve's case was as good as I'd expect from a high-end third-party product.
I threw my CV1 that I bought secondhand in the trash when facebook bought oculus then forced login. Maybe I'll return to the market when it supplies something I want.
https://x.com/MuchRockness/status/1849543449906942094
Even the minority who do buy VR games on Steam are mostly playing them on a cheap Meta headset, so without Meta those sales might not have happened either. The most recent Steam hardware survey shows that of the users who have a VR headset, nearly two thirds of them are using an Oculus/Meta model.
It's a pyrrhic victory, they may have cornered the market but it's still losing them $4-5 billion every quarter with no end in sight.
https://www.cnbc.com/2024/10/30/metas-reality-labs-posts-4po...
Valve tried to make it with Alyx, and while it is amazing, it did not inspire the industry to follow up on.
I do not blame Valve for moving on when nobody followed them.
If Valve wanted more Alyx'es to happen they needed to spread their wealth around until the VR market gained more momentum and became self-sustaining.
VR is a small market to begin with, and most VR people can't play Alyx without buying a whole new computer.
https://store.steampowered.com/hwsurvey/Steam-Hardware-Softw...
When Alyx first came out I had a PC that was the minimum recommended specs for VR from the day the Vive launched (4790K and Geforce 970). The game ran fine.
It sure as hell got better when I upgraded to a 3900X and 3070, but it plays just fine on the original minimum requirements VR PC which was a $1500 PC in 2015.
The idea that PC VR requires a massive rig is just nonsense. Computers that run VR perfectly fine are literally being forced in to retirement, they're officially obsolete.
Also worth noting - you don't need Windows to play Alyx either. SteamVR supports Linux perfectly well, and other games that don't ship native Linux-native builds can still run through Proton. If you own VR in any capacity whatsoever, you should be capable of playing Half Life Alyx; that was Valve's selling point for anyone that had Steam and a headset.
Would be a huge selling point for the steam deck if it could manage it on min specs
It ran, but barely. I probably spent half my playtime restarting the game, trying to find the happy coincidence of playability (because the other sessions were too rough). Being able to play Alyx was one of the reasons I chose the Go over the Deck.
Nope, no need for an expensive gaming PC, just an actual gaming PC.
As with cars, phones, etc. if your budget is tight you can always get so much more value by going for a used model from a generation or two back than you would get by spending the same money on something new.
> Would be a huge selling point for the steam deck if it could manage it on min specs
Steam Deck can technically run a few VR titles but it doesn't do it well. There is a lot of evidence that Valve has prototype standalone headsets running on Steam Deck derived hardware platforms (often referred to as "Deckard") but the hardware just isn't there for full quality PC VR.
Creating a working nuclear fusion device could be cheaper than that.
As neat as it is to see how rare it is these days (w.r.t. the asking prices I currently see online), I've always wished other fans could enjoy and appreciate it as much as I have.
I also highly recommend Divinity OS 1 and 2 for the same level of dedication to every single detail and free post-launch support, even if they didn’t have such an enormous blockbuster budget behind them.
Whoa, what? I only hear people complain more about Windows over time.
Well sure, but windows has never exactly been highly praised outside the enterprise world. Much of the praise that is held is mostly nostalgic.
I haven't heard love or hate for post-acquisition GitHub changes.
As for Windows, I could be in a bubble. But I use Windows, and I hate the UX more every release. Ads, "suggestions", automatically reenabling features, UI complexity, hard to read text, unintuitive UI, performance issues, audio device issues, useless background processes, new layers on top of configuration UI rather than replacing/updating old layers. I think those have all gotten worse since Win7, some since XP. And I thought I saw that opinion corroborated generally. Maybe there's not literally more complaining, but that doesn't necessarily mean people don't agree it's getting worse. What are they going to do, type the complaint in increasingly larger font each year?
It was never about the quality of Windows, but their attacks against FOSS. And I think Nadella largely repaired the damage from that.
It also sounds like the fixed a lot of art/ambience regressions introduced by newer source engines over the years.
Anyway, it's exciting to see an update like this, and a testament to their prior engineering that they can turn up all these dials without weird repercussions.
Regressions do happen which is why they opened a (public) bug tracker at the time, which is still in use: https://github.com/ValveSoftware/Source-1-Games/issues
If you have any HL2 issues, now is probably a good time to get them fixed before the developers responsible for the update get bored and move to another project.
I was a Mac kid. There were a whole bunch of games that I wanted to play but didn't have access to. (At least we had Marathon!) American McGee's Alice was at the top of that list, with the Arkham Trilogy close behind.
The Steam Deck is giving me a chance to catch up on all the computer games that I didn't have the hardware for when they were popular. I've now caught up on Arkham, but I just can't get into Alice. The graphics and the controls are so bad from a modern perspective that it distracts from any desire I have to explore that world. I had to download a mod to even unlock the original Alice, because they don't sell the complete collection on Steam any longer.
Seeing that the Half Life games have been recently remastered, maybe it's time for me to give them a play through. I'm going to be spending lots of time on airplanes over the holidays!
> Reduced chances of birds getting stuck in the world.
I really think I give it a replay. It was such a great game and truly started a new area of gaming.
8 keys, I guess?
As Gabe says, they didn't fulfill their obligation towards their customer and fan base to complete the story. Alyx is cool, but niche.
Oh well, Half Life 3 not confirmed, yet again.
(DOOM and Quake are also good targets and both have a large community)
Sadly slinks away
And yet these days we get "remasters" of games that are not even 5 years old. Most of the AAA industry is just putting out lazy cash grabs and predatory live service garbage. Valve itself has profited immensely from the live service business model. I just wish they would get back to development, or sell someone else the rights to their single-player franchises. Spinoffs like HL: Alyx and Aperture Desk Job don't really cut it for me.
Anyways, this update looks alright. I'm looking forward to the new commentary.
I am so getting the reissued book!
It's hard to put a finger on it, but it was a combination of a truly imaginative setting on-par with a decent science fiction movie or novel and the "show, don't tell" style of proper world-building. It presented a uniquely interesting dystopia, discoverable through a game instead of a movie or book.
I think it was one of the first games I played that was dark. For example, the "police" are cybernetically enhanced, but in a disturbing way ("transhuman"). The low-ranked troopers "just" have permanently worn facemasks a bit like Darth Vader, with an implication that they may not be functional without them. The higher-ranked shock troops have increasing levels of augmentations until some have only a single central "eye" in their visor. You shudder to think what's hidden underneath.
> In a new two-hour documentary from Valve, current and former members of the company talk openly about the creation of Half-Life 2 as well as finally spilling the beans on what happened to Episode 3, and even showing gameplay of early prototypes of the canceled game.
The two hour documentary was just uploaded like an hour ago, so presumably nobody has watched it completely yet.
Still holding out for HL 3. ¯\_(ツ)_/¯
Props to the webmaster (do people still say that?), I love seeing easter eggs like this.
Didn't play Half Life 2 I gather?
I hope it’s not a legend like ken or something, but it’s not obvious from the front page and #1 is a video game.
Tell me we’re not handing out black bars for anniversaries of software now?
Legend.