John Carmack goes off about online-only games being abandoned
pcgamer.com
pcgamer.com
But modern games that rely on matchmaking through the developer/publisher? Forget it, when they go down, there's no saving them, alas.
So really this is a pretty clear-cut case of functionality deteriorating because the maker of the game has way more power than before to change the game _after_ the customer bought it (the recent brouhaha where BMW or whoever offered a subscription to boost your car's top speed by unlocking a speed lock in software also comes to mind, not to mention "purchased" streaming-only movies disappearing forever once the streaming service dies, etc.).
Really bizarre era of consumer behavior. 'You're gonna buy my crapware games and you're gonna like it! You're gonna read my articles shitting on you and you're gonna like it!'
Cyberpunk (as much as I love it) and Battlefield 2042 were inevitable symbolic trainwrecks. <10k daily players for the latest Battlefield title? Time to wake up.
By comparison Battlefield: Bad Company 2 was clear the best Battlefield game of all time. And the graphics to this day still hold up.
The bugs have been patched but not game ruining (for me). T-posing was infrequent, but there were some issues.
It's the business practice of sending the customer their boots without the sole and toe box sewn together then telling them they get most of the experience earlier this way and theyll stitch it together in a month if they have the time.
I agree it will age well, particularly with the right improvements.
I think the people that bought into the hype got severely disappointed and anyone buying the game on consoles simply got scammed, but as someone who bought the game after it had been out for a bit, I've had nothing but fun with it.
In fact, one thing CP2077 allows that most other games don't, is backing up the DRM-free installer for the patched game. Steam/PS store/Xbox Online will die one day and all their online patches will disappear forerver, but the GOG versions of most games will run in 20 or 30 years as long as you remember make a backup of the installer.
steam usually allows refunds in the first few hours of playtime, which supports this usecase as well.
This is in France, where there's a mandatory 14-day return window for online purchases. But there are restrictions on some items, such as media must be "unopened", and I guess downloading is considered as "opening".
Which is only tangentially related. Greed is the true culprit.
For a publicly traded company, especially one with short term oriented investors, some amount of brand tarnishing is acceptable.
For another similar example see Fallout 76.
It's poor leadership and short sightedness IMO.
You can't run a hacked server that ignores license state or hacking bans if you have no access to the server software. They disguise it under a claim of convenience, a feature that you match against similar-experience players, but you're right, it's ultimately an antifeature.
Gamers put up with a lot of crap, between exploited ring0 DRM, platform exclusives, the loss of consumer rights (like refunds and resale) and crappy ports from consoles. I don't understand how they get away with it. I might just be old, but I've changed my entire purchasing game strategy to ignore games that fall foul of these things.
If a server says: load whatever skin you want... well, why would you pay for it?
This is also why modding got killed off, you can't sell people stuff if someone else will make better assets for free and just puts them up for download. Even if integrity or security was a big issue, that's been solved since quake 3 and ut99 with integrity checks and server-side load controls (ironically, the same ones that are now used in locked-down games to make selling different coloured virtual shirts a thing).
I'm sure that EA would love to kill Battlefield 4. But it will never die, because players can run and moderate dedicated servers.
Battlefield 1 (newer than 4), on the other hand, died years ago. Battlefield 5 is practically the same game, arguably worse, yet was able to steal most of 1's player base because support (and moderation) were effectively dropped by EA during the release of 5.
Anti-cheat is the newest (and deadliest) iteration of this pattern. CoD Warzone is free to play, so long as you run it on Windows, and not in a VM.
Worst of all, this means the burden of moderation has been moved to anti-cheat itself, even though any experienced forum moderator can tell you that moderation itself is based on social interaction, not technical behavior.
Gamers aren't tinkerers anymore.
Times change, gamers might have been tinkerers and hobbyists who accepted needing to mess with their software and perhaps even hardware (memory address allocations, anyone?!), but that was 30 to 40 years ago. We're in the 21st century now, more than 1/5th of the way to the 22nd, and gaming has become mainstream.
Gamers today just want to play games, messing with the computer (what's a computer?) is above their pay grade. Games today route multiplayer through their own matchmaking servers because Player-2-Player matchmaking /fucking sucks/, it's janky under ideal circumstances let alone normal. Gamers today don't want to deal with network shenanigans, they just want to play; so devs and publishers oblige by handling more things on their end simplifying the experience.
Gamers aren't tinkerers anymore.
The gamers of 30 to 40 years ago have moved on to messing with "home labs" and Raspberry Pis and 3D printing and shitposting on HN.
More importantly, their audience isn't the audience anymore.
After all, it was never a hugely significant number of tinkerers that were important, it was the audience each tinkerer brought with them.
That just isn't the business model anymore.
Could it be that said audience depended on the tinkerers to help with settings things up, but since games "just work" now, the middleman has been cut?
With gaming becoming popular I'd say tinkering gamers are the same number as before, just not the same proportion of the overall gamers pool.
Completely with you on matchmaking. The pathetic limits on these things. You've not lived until you've had a dozen friends all try to climb in/on a drop ship in Tribes 2 in a 64v64 match. That was worth carting a CRT to a lan gaming convention alone. But I don't believe players prefer this, they just don't have a choice. I much prefer the old server-list model.
Messing with computers.
PC gaming today is as simple as going out to Best Buy or Costco, buying a "gaming" desktop, come home, unbox it, plug everything in, initialize (note: not install) Windows, install Steam, and Bob's your uncle.
Most of us hailing from the 20th century learned how computers worked because we kind of had to to run the games we wanted to play. Memory address allocations, DLL hell, driver hell, registry tricks, picking out parts and putting it all together; all stuff gamers today simply don't have to deal with. Windows Update will even figure out most if not all the drivers for you.
And before anyone says it: Yes yes, I know buying a "gaming" desktop from Best Buy or Costco isn't ideal. But guess what? We are outdated oldtimers. We definitely know how to get the best stuff, but gamers simply don't need the best anymore. Computers have become household appliances, not tinkerers' toys to be handled by wizards. So what if that desktop from Best Buy craps out? They just go and buy another one like you would buy another Playstation and Bob's your uncle.
Gamers aren't tinkerers anymore, most people don't mess with computers anymore.
My experience ~30 years ago involves a 386, big-box computer, Window 3.1 pre-installed. My options for tinkering (outside the battle for HIMEM) were pressing a turbo button and hex-editing File Manager to File Mangler. There was no hardware modding scene at that level, just a choice of massively expensive storage and RAM. I'm sure there was more that you could describe as maintenance but I don't consider myself any less of a hacker because I don't have to run "hdsit" before I shut down.
Did I do more stuff? Sure. I didn't have a huge number of games so writing your own, playing with tools was pretty much your only option, and I would agree with a characterisation that people fiddle with computers less because they have less contact time with an actual desktop environment outside of the browser. We are spoilt for choice.
But PC gamers still tinker. The have hardware options unimaginable in the 90s and the scale of game modding scenes for things like The Witcher 3, GTA5, Fallout 4, Skyrim, Minecraft, Souls and Rimworld eclipse those efforts in the 90s, many times over. Shader beautifiers weren't even possible. Is it niche? Probably, but PC gaming has always been niche.
Hm? Not quite, there's a solid player base still playing.
How does that make gamers special - aren't most mainstream products optimized for profit at the expense of quality?
This is always the official explanation.
It is just pure greed.
5-10 years ago publishers would just ignore outrage and fix the bare minimum quietly and then leave it alone. Think of all the major bugs just left in games in the 00s and 10s, Fallout New Vegas in particular comes to mind. That might be the single game with the worst long term support ive ever seen. They released it in even worse shape, then made it barely functional, released mod tools and never touched it again because the Obsidian & Bethesda relationship fell apart.
Today both BF2042 and Cyberpunk are going out of their way to improve their games from launch state and make a marketing effort on behalf of it to recover their brand. Gigantic difference.
Although they're both Microsoft now, so maybe it can happen.
Indies who want to do the right thing face a customer base that doesn't really get it and yells at them for spending times on things they don't understand. So they pivot - making games is already risky enough as it is.
Game devs have less courage because too many are funded by publishers rather than self-publishing. Those devs that don't enjoy breakout success are safer but those that make hit games never reap the full benefit of it - publishers take too large a cut - and can't fund themselves into making something bigger, like how Half-Life set the stage for Steam.
"Early access" has become a meaningless label, incentivizing the norm towards good-enough games rather than great games, and muddying the waters for truly great games to be recognized as such, because they often don't start that way, but also used as a defensive mechanism against the entitled creeps who think it's okay to harass the social media void with their complaints, not realizing that the people on the other side of it see it and sometimes emotionally react to it.
I guess I'm shaking my cane too much, but it's not just the companies going down false paths, it's also people following them down. Put another way, the industry is systematically unhealthy.
I see gaming like most forms of entertainment which are enjoyed in the moment and then you move on. When I got to a music concert, I don't sit there and record it for watching again 10 years later. I enjoy it in the moment and move on, there will be newer ones to enjoy later.
There's really a lot of value in retro and mods IMO. Sure it's not for everyone but what is?
HN is always telling us to be outraged when an online game is shut down, but by the time they get shut down, hardly anyone cares about playing them anyway, they moved on to newer stuff.
And in the VR realm there aren't a ton of large story FPSes now. Especially not of this caliber.
Hard disagree here. Some subgenres don't even have any newer games at all.
And even for genres that have been embraced by the mainstream, often they are all polished into the same shape. Yeah, there are plenty of conveniences now that older games might not have but a lot uniqueness is also lost in the process. How many modern games still expect you to navigate an environment yourself instead of feeding you questmarkers and triply-redundant messages and reminders of what you need to be doing next? How many shooters that just focus on the shooting instead of feeling the need to add some skinner-box-inducing RPG-lite mechanics. How many Point & Click adventures that even come close to scope or engrossing art styles of the classics? Why is Planescape: Torment still considered one of the best (C)RPG stories by many when writers had almost 25 years to surpass it?
Yeah, if you pick a random game from the 90s and a random game from today, then the latter will perhaps be more playable. Many old games have plenty of jank that wouln't fly today. Because today's games are designed for just that: so that anyone with half a braincell can complete them. That means ecessive hints. It means refusal to provide challenges that might limit your market. It means limiting the amount of text because everything needs to be voiced which costs a lot. It means repeating mechanics that "work" instead of daring to try something unique.
> I see gaming like most forms of entertainment which are enjoyed in the moment and then you move on. When I got to a music concert, I don't sit there and record it for watching again 10 years later. I enjoy it in the moment and move on, there will be newer ones to enjoy later.
Yet people still listen to classical music and everything in between then and now. Some even mainting their own collection. Are you really going to say that for every Mozart or The Beetles or Black Sabbath or whatever there is a 2020s group that pulls of the concepts better? Does the warm fuzzy sound from old analog recordings prevent you from enjoying them as much as older graphics do?
Of course the answer to what is better also depends a lot of taste. Just like with music, the predominant styles of games shifts over time.
I'd rather sink time into a shoestrung indie studio than anything EA wants to publish, after that.
That’s hardly true for modern games. One I’m working on currently would be on the order of a few thousand US a month to run the entire infrastructure. Trimming that down does not help me launch, and only helps the very small minority trying to run a particular game after its death.
Most modern games don’t run on a single server piece, it’s distributed.
It's infinitely expensive, because older games compete with new ones(not everyone cares for looks and there are mods that can make this argument obsolete). And new ones don't always win. Especially with introduction of game as a service. There are many tactics used by modern developers to squeeze more money from players, that modern gamers don't like. And if gamers had a choice many would go retro. This is why WoW Classic exists. This is why emulation exist. This is why people develop open engines for old games. And this is why it is in developers best interest to prevent their older games from competing with new ones. Especially if people were to learn that they could get free skins for their games in past, while modern ones demand cash for even basic palette swap.
I kinda like the idea of more open gaming type systems.
I joined in the fun, and for an hour I felt like I was 15 again, it's held up surprisingly well! If epic ever turns off the Fortnite server, it's just gone.
I had so much god damned fun playing that game
EMPRESS is the last line of defence.
plenty of games that have lost their matchmaking by GameSpy shutting down have found a second MP life through third-party matchmaking services like Game Ranger.
But do NOT buy games where MP doesn't work at all without official servers (or only through Steam MP or equivalent).
You make it sound like a good thing. And it indeed is for gamers. But it's the worse nightmare for established dev/pub that people play the old games, on their own servers, instead of your latest title.
It actually happened quite a lot. Starcraft I was (is?) more popular than Starcraft II in Korea, where the biggest Starcraft E-sport scenes are. I recently learned that Red Alert II is still fairly popular today in China, and Red Alert III players don't exist.
"How to out compete our previous title" is a real problem that established studios face. The lack of durability is a "solution".
Not just games, it's really.. everything.
I predict that a 100 years from now the collectors of old things of any kind will have a lot more from prior to 2000-2020 (depending on the item) than anything newer. Whether it is computers, electronics, games, cars, doesn't matter.
All of them are becoming so dependent on third-party cloud services and phone apps that simply won't exist and will be impossible to recreate in a century. All that will survive is the older generation of products that did not depend on anything. Very sad.
In other industries, changes in material usage can be cost saving in nature or more robust.
But in the games industry, new generations of players are literally losing out in features that previously existed, or quality concerns that one never had to think about before.
I generally disagree on that if you stick to Windows based games. Now if you're playing on any of the consoles, well longevity is decided by the platform owner.
In the indie games industry there are likely more games than ever that are better than ever and can be self hosted.
The unaddressed problem here is that 'online only' skinner box style games are massive money makers when they are a hit. Couple that with ever increasing costs in making top of the line games and you can see why the people handing out funding want to go that way.
Indie games are getting better, but if you want AAA graphics, etc. your competition is far more limited.
Cyberpunk 2077 shows they are still tossing out unfinished bug ridden messes. The only difference is so many AAA games like Assassins Creed, GTA, Call of Duty, etc are just incremental releases they have more time to polish the same crap.
So we end up in this situation where to be consumer friendly they should either adopt a rolling release F2P model (like fortnite, dota, cs:go, etc) or at the very least pledge long term support to each release.
They do neither so games like Call of Duty end up as a money pit when they get infested with hackers a couple of years after the series moves on to the next title.
Unfortunately, as this year's CoD release has shown, they can't even do incremental updates correctly anymore. They've broken major parts of the game, removed key features (like leaderboards), and taken months to release content.
Speaking of Assassin's Creed, the reason they're able to polish is not because it's the "same crap". On the contrary, every time they've had to deal with an engine upgrade they've suffered huge bugs (and I'll touch on this later on). But they're able to mitigate this by having large teams operate as cogs, with higher level tools, and this has been battle tested for years. Call of Duty is another, huge teams, working on cycles. CD Projekt Red did not have this in place. Their expectations were utterly delusional.
Speaking about engine upgrades, maintaining your own custom engine is incredibly costly in every aspect. Harder to find talent, harder to maintain, harder to test. All of this compounds, and gets worse when you're trying to stretch your engine away from "the same crap" and into a new genre. This is why more studios now are relying on the likes of Unity or Unreal - including CDPR. Or EA, who once upon a time had both Eclipse (single player RPG engine) and Frostbite (FPS shooter engine), and are now betting on Unreal.
That’s what I mean by the same crap. It’s not that the engine is unchanged and they never add new gameplay elements, but rather the team/company knows more or less exactly the kind of game that they’re making on day one and therefore what kinds of people they need etc.
Specialization isn’t a bad thing, but it does mean you can get away with less generally competent team, which very much plays a role in staffing etc.
the game part is mostly ok, just boring
they rebooted Modern Warfare 1 and 2 for fuck's sake. how more unoriginal one can get!?
that's absolutely on business and not on the coders IMHO
In fact there are striking similarities between Hollywood VFX artists and AAA game developers in how shitty the work conditions are.
Even media with all streaming going on it's sometimes just gone.
Having been playing computer games since the mid 90's I don't know if that claim is actually true. Once upon a time there were games that would delete your entire hard drive when uninstalled and wouldn't last a 15 minute game session before blue screening the whole system. Honestly, given the complexity that modern games require I don't think it's bad at all.
I think the problems we're seeing with modern games actually being worse (the new Arkham game compared to the old) are strictly dollar and cents choices. The ability is there, but the industry has become far too pragmatic.
I honestly would challenge you to find games from the mid 90's+ that won't work on a modern machine. Many times if there's a feature that straight up doesn't work it's due to online services being discontinued. There's been some work done to rebuild services like GameSpy or Westwood Online, but once more forms of encryption and anti-cheat or DRM were implemented in these sorts of services, the harder it's going to be to preserve their experience.
But yeah, we're in a new phase where publishers kill games and then they are gone forever and that's a real shame. I think there needs to be some sort of legislation around abandonware so that people can host live services and MMOs after the publishers give up on them.
> before blue screening the whole system
TBF at the time the number of possible configurations a game could run on was close to infinity
Nowadays platform are much more stable, drivers are much more polished and you have 2/3 max manufacturers that make all the hardware gamers use which mostly is top of the gamma components specifically built to run games.
I remember assembling my PCs in the 90s by putting together components cannibalized from everywhere I could, using esotic hardware that took weeks of debugging to make it work, CPU, RAM and BUS were constantly abundantly overclocked, air flow was a joke at best and software configurations were highly customized
The fact that they even booted it's a miracle
Random number generators, premium stores, pay to win, etc... Seeing anything appurtenant to these features take top priority is frustrating.
What about recipes and such?
But, with longevity, what other industries are there? Furniture is one that frustrates me, as everyone is sold on the idea that some furniture is for a lifetime, at least. Which is borderline bullshit. There can be some things that will last with maintenance, but a lot of really pretty things that last are also done using materials that we flat don't allow the harvesting of nowadays. Such that things can only last if they aren't getting used. :(
(Upholstered items, though, I'm with you on.)
Many of the problems with games are to do with game engine architectural decisions made 20 years ago - anything based on Unreal Engine has code that was last touched in the mid 2000's for example, and some of that is _not good_. Lots of issues around sloppy drivers, bad memory management stem from an overuse of the "old way" of doing things, which is driven by the programmers you're glorifying in your post. Some of the smartest people I've worked with are still working on their first game, and some have been in the industry long enough to be veterans.
The quality isn't deteriorating, the market has expaned hugely. Companies that are now considered "small" would have been considered enormous productions 25 years ago, and those companies are pushing out incredible quality experiences. Companies that are absolute behemoths now are putting out iterative works that people _clearly_ enjoy because there are alternatives to them and yet they buy Call of Duty because it's fresh, fun, exciting, nostalgic.
> I don’t personally know any other field whose products have deteriorated over time like the game industry's
Ignoring the fact that I disagree about the quality having fallen (stray, tunic, and vampire survivors are some of the best experiences I've had gaming, and they're all relatively new), people lament the same way about everything. Cars aren't built like they used to be, household appliances used to be more robust, movies were better in <X> time, children's toys are just ads for <y> show, there's no good music these days. Look on any thread here and you'll see the same thing (and if you speak to your parents they'll say the same about things that were better when _they_ were younger)
Movies were better, they made less movies which increased quality, had better actors who had more talent (singing, dancing, acting). Kids TV shows are created to sell toys. Products are cheaper and made to be thrown out vs repaired.
Profit keeps going up, quality goes down while prices try to stay the same.
I don't think the decisions John is talking about are in the hand of individual devs.
They're usually really good at getting the job done and balancing quality + productivity. However, most things are written with a one-off mindset because chances are your next game will need things to work completely differently. Everywhere else in tech you find "scalable" and "backwards compatible" solutions; not in gaming.
Part of this is the job description. Unlike the rest of tech, you're usually building a game that will be popular for 2-3 years then disappear. It's not common to find that model anywhere else in tech.
Just my 2 cents.
Either way, I would never attribute this to ICs, they're just doing the best or fastest job they can with the requirements they're given. A lot of the time anyway.
Watching a friend playing Ghost of Tsushima and it's the same. So good. Horizon : Zero Dawn. Crazy stuff.
I don't know what these guys are doing, but the scenes are gorgeous. The gameplay is loads of fun (MW2 is sooo much fun). The choreography is great. Love it all.
Household appliances?
People won't grind on small details for low pay, so simplified games are the way forward.
> "Most of game development is a panicky rush to make things stop falling apart long enough to ship, so it can be hard to dedicate time to fundamental software engineering, but there is a satisfaction to it, and it can pay off with less problematic late stage development."
When you're at id Software and don't have to answer to anyone with regards to deadlines or funding, it's fine to think like this. Unfortunately a lot of developers have a lot more constraints than Carmack did when he made his best games.
Carmack's company didn't have this problem when they made the games they eventually open-sourced. When they did get bought by a publisher, they no longer open sourced their games as far as I know, so who knows what the engineering practices were like at that point.
I don't know. I think it's silly for decades-old-games to run into licensing issues when trying to go open source. The tech is effectively obsolete except in the instance of that specific game.
I'm also not sure why you think remakes are that different from cover bands - neither are a 100% identical copy.
Those games are far simpler (by at least an order of magnitude, maybe two or three) than most of the modern AAA titles.
What was the last AAA game Carmack shipped? Rage? Hardly a smashing success, even if it was technically impressive.
I have a huge amount of respect for Carmack’s skills and work ethic and track record, but I wonder if he really has a good handle on the sheer amount of work that goes into a modern game. With often dozens of programmers, and hundreds or thousands of contributors, the coordination overhead is just massive. I think that’s why overall quality control is harder to maintain.
actually Id Software people worked in much harder conditions than game devs of today.
https://preview.redd.it/iehd38ucuuz71.jpg?width=450&format=p...
He's the one who made the whole thing viable in the first place! I would hate to stand in his shoes, to watch the fruits of my labor rot on the ground below, begging someone to simply reach down and pick them up: someone whose entire business model forbids them.
Was it worth it? Sure, he got to work on something really cool, and get paid lots of money... Yet it seems obvious from where I'm standing that this story could have been written with much more satisfying goals, and still played out with most of the same benefits reached.
Just look at the valve index! Huge success without any of this walled-garden bullshit or compromise, and it's still king 4 years in. The only advantage Oculus has had is affordability, but I'm still not convinced walled-garden paths are the only ones that lead to cheap.
Huh? Oculus is the best selling VR headset available, it sold way more than the Index.
> The only advantage Oculus has had is affordability
...yes, this is how consumer sales works. People buy what is affordable.
> Carmack goes on to say that Bosworth greenlit the release of the Oculus Go root build (an unlocked OS allowing full access (opens in new tab)) that he had long pushed-for, but "after seeing how much internal effort was involved to make it happen, I almost felt bad about it," said Carmack. "The constraints are just different in a company the size of Meta."
The "constraints" are their business model: underprice the hardware and overprice the software. It's not a new model, this is what consoles have done since Carmack made Doom!
Release date of Valve Index: 2019-06-28
Release date of first Oculus¹: 2016-03-28
There were about three years of tech adoption by consumers between these releases, Oculus was, before their first commercial HMD release, was acquired by Facebook for US$2 Billion, FB which could also use its near monopoly in social networks to place ads and thus influence billions of people to buy the product(s) of one of its subsidiaries.
In contrast:
> It sold an estimated 149,000 sets in 2019, 103,000 of which were in the fourth quarter due to the announcement of Half-Life: Alyx, which buyers received for free. The sudden demand caused the unit to be sold out in all 31 countries except Japan in January 2020. As of December 2019, 6.67% of the VR units connected to Steam are Valve Index sets. While Valve had anticipated supply for many of those that had ordered the Index in time for the March 2020 release of Half-Life: Alyx, the COVID-19 pandemic impacted production of the Index which left Valve with a reduced number of units available on the release date.
- https://en.wikipedia.org/wiki/Valve_Index#Release (emphasis mine)
¹: since you didn't specify which of the range of Oculus HMDs you're referring to
He’s done it already with id software, he didn’t end broke. Feels like you don’t know who this guy is
When I grab my Quest, because I finally got some time and want to play, now, I press the power button, grab the controllers, select a game, and play. When I consider grabbing any other headset, I realize I'll have to boot my gaming rig, connect cables, and then likely spend at least 10 minutes getting it to work properly and will encounter random software issues.
That alone makes the Quest amazing.
Now, slowly, competitors are coming out (whose main advantage is "not owned by Facebook"), but most of them either aren't released yet, don't have a broad selection of games, or both. Unfortunately, that means that for casual gamers, the Quest is still king IMO, and has created quite a bit of lock-in.
(Edit: The Quest also had the advantage of being available - while Index is still not officially available from Valve in most countries, so resellers easily drive the price up to $2k for the kit with controllers and base stations, on top of the gaming rig you need).
When I grab my Quest I quickly remember how nauseating any games are that are actually good games. The only kind of decemt content that doesn't make me think "this game would be more enjoyable on my regular display" are arcade style like beat saber or pistol whip.
Personally I still don't think the current approach to VR will ever get mass appeal for gaming.
I'm told state-of-the-art quest on PC is running SteamVR through windows remote desktop. It's bad. Incredibly bad.
Once the Index is set up and calibrated, all you have to do is turn on a controller, and put on the headset. It's really not as inconvenient as you have made it out to be.
But my point here is not Index vs Quest. My point is closed-platform vs open-platform. The very issues that Carmack laments here are inherent in the closed-platform business model that Oculus (and later Meta) is built around. The very business that he was a core member of was explicitly designed to fail him in these respects. There's no such thing as a free lunch.
Carmack got to do most of what he wanted: create VR tech and get paid big money; but he did not get to have the result he wanted: an open VR platform that works with (instead of against) its users. It's a serious compromise that he made. He didn't need the money. This is the man who famously gave away a Porsche at a Quake tournament! He could have put more value in his legacy than in his financial stability, but he didn't. I can only speculate on how that would feel, but I would be surprised to learn it feels good.
You weren't told that. You don't know what "Virtual Desktop" is (an app that makes using SteamVR wirelessly on a Quest a one click process) and apparently assumed it's the same as Remote Desktop, the two are unrelated.
If you're willing to do wired you don't even need Virtual Desktop, AirLink and a normal Link cable both work just fine.
First of all, You've never worked at a corporation? Sounds like it. Corporate politics can easily eat-up an individual influence on the direction a product goes. Especially when you're judging long periods of time, like entire history of Oculus from the looks of it.
Second of all, go read his public rant after resigning. you will find answers there.
I'm sure the rant is interesting. I haven't really been following his life story.
I suspect that when Carmack was making PC games all the way from Commander Keen to Rage, he understood the value of PC as a free/liberated platform. Do I have proof? Well, here he is lamenting the consequences of working on a proprietary platform. I'm content with that.
This isn't some kind of trial. My point is not to condemn the man. My point is to use his character to articulate the story of my own perspective.
-----------------------
"Be disciplined about your build processes and what you put in your source tree, so there is at least the possibility of making the project open source," said Carmack. "Think twice before adding dependencies that you can’t redistribute, and consider testing with stubbed out versions of the things you do use. Don’t do things in your code that wouldn’t be acceptable for the whole world to see.
"Most of game development is a panicky rush to make things stop falling apart long enough to ship, so it can be hard to dedicate time to fundamental software engineering, but there is a satisfaction to it, and it can pay off with less problematic late stage development."
The last phrase is quite the euphemism for "the game disappearing entirely, forever" and through Carmack's repeated exhortations and examples you get a sense of someone who's very frustrated at seeing things built on foundations that can ultimately prove self-defeating. We've entered an era where even many singleplayer games require some sort of server ping, while other games don't work at all if you're offline (which outside of MMOs has sometimes felt like industry overreach).
Obviously I don't know for sure, and there will be other factors like money / time spent on remastering, but I'd still wager that e.g. the fact that the Metroid Prime remaster was able to add in a (seemingly) whole new lighting system and control scheme whereas the the GTA trilogy remaster couldn't even get half of its weather effects working probably speaks volumes to the respective code quality of their originals.
Either way, I'm certain that I'd much rather work on porting / patching / remastering a game from a studio that's following the advice Carmack gives here.
> Another alternative would be spinning off the project: Meta letting it go and allowing team members to leave, take over the rights for a nominal fee of $10,000, and maintain it.
In my experience, studios have a contradictory FOMO mindset of letting other people run the game:
* If someone else makes money off it, they want to be part of it
* At the same time, they don't want to run it themself
* It's not worth the effort to open source it, also see argument #1
What usually ends up happening is the community reverse engineers the protocol and writes their own OSS backend.
Shameless plug: I founded Rivet (W23, https://rivet.gg) to simplify multiplayer game server ops. We talk to a lot of studios with hard-to-manage old games (i.e. 3+ years old can be ancient in gaming). We often help studios move these to our standardized ops so they don't have to worry about it anymore. We also help lower the server cost for old games by auto-scaling their servers based on player demand.
"Hey, there was this multiplayer game my friends and I loved to play called ___ and it had the following mechanics. Here are some maps and the instruction manual. You can even check out some examples from the following YouTube links." Etc.
Reverse engineering. Please wait...
Meanwhile my kids are literally playing Super Mario Bros right now while I'm typing this comment. That's a 38-year-old game. Every time Mario falls down a hole or gets nailed by a ricocheting turtle shell, I hear squeals of delight and frustration, and then they hit the "continue" option to try the level again for the 15th time. It's not really "angry mad," but "happy frustrated."
There's far more quality interaction going on with Super Mario Bros than with whatever random "modern" FPS they sometimes play, where they are mostly zoned out and are getting more "angry mad" rather than "happy frustrated" when they are killed in those games.
I heard my girlfriend's then 10 year old son complaining 5 years ago now about fortnite players being garbage because they had no skins and were called 'noskins'.
I didn't play it so I asked if they gave stat increases or anything and he said no, just cosmetic. I asked what a skin had to do with playing ability and was told 'Tch, you just don't get it' -replete with eyeroll.
A few months ago heard the same thing from a friend's kid (~same age) who'd just gotten his first xbox. 'Eww noskins', with the added cherry of him complaining about using a controller as it was so much harder than using a phone screen.
YouTube was the cause of it back 5 years ago, idk about now, but the companies get their messages in no matter what.
It all works far too often and worries me about where we are headed.
I often wonder how I'd have turned out if watching GI Joe and Transformers as a kid had beaten me upside the head with more than 'dont steal, drugs bad, police good, buy toys' messages of the 80s.
Given the throughput increase in computational power and broadband this should be a decreasing cost even if the use stays static in perpetuity.
I guarantee that you can run them in perpetuity. I worked for a game hosting company in my early career.
Running a process to simulate an entire game world is very expensive, far more than storing a bit of data and responding to the occasional HTTP request.
You aren’t getting away from needing to run many game servers ready for a connection. The most cost effective way to do that is bare metal for base load, which means you can’t scale down to zero. Worse still, in a game where players choose their server, most servers will have many empty slots.
Of course the self hosted versus company hosted equation is much different here. If the game publisher eventually just offered the server software for self hosting and freed themselves from the management burden it in theory would be much cheaper. But likely they've boxed themselves in with license deals that make this difficult.
The added costs are studios running MMOs, and those aren't insane either. People hosted their own servers since the beginning of the net.
The odd runaway hit MMO shells out more but also makes a ton more. Even then we see companies cheap out and make us sit through server queues.
Live services these days amount to little more than checking license to play, loot-boxes, battlepass, etc.
And yet because of those evil cash-grabs I can only play the game a few years even for single player games.
Live services have become a worse dlc, as we knew it would in tge beginning.
> No,they are not. Especially since op mentioned most popular games having ~10k concurrent users.
> The added costs are studios running MMOs, and those aren't insane either. People hosted their own servers since the beginning of the net.
Do you know how much it costs to run an online game?
Assuming we're not talking about MMOs (because you're the one who said the added costs are studios for MMOs), a game with 10k peak concurrent users, and 10 players per session will need 1k concurrent sessions. Assuming the server needs ~ 2 cores and 4GB ram, you'll need 2000 cores and 4000GB/4TB RAM. We're talking "old games" here, so I think it's reasonable to assume that we're not running autoscaling containers, and frankly it's probably a bit of a stretch to assume that there's anything other than a fixed set of servers if we're talking games from the 2005-2012 era.
Hiring a 32 core/128GB Server is ~$1000/month from OVH, so to keep the peak capacity for the game available, you're talking ~$60k/month, plus various online services (login, auth, party, matchmaking, etc), licenses for things like SQL server, windows (yes, windows is commonly used in game development even up until recently). That's assuming no maintenance, or development costs happen. You have to budget for handling things like log4j, keeping CI/CD pipelines up to date to keep SSL certs valid and rotated, handling player support for those games, etc.
That's not nothing, and while it might be manageable for EA/Ubisoft for your game, what about the other 10 games they have with 10k players each?
Google Reader.
It seems like a lot of huge corporations will blindly chase profits with zero consideration of what their fans want.
I've been building a company (https://hyperspacelogistics.com/) where we believe that doing what's best for the fans is actually best for the company. In fact, we state "We strive to support actively used products for as long as possible" as one of our core values.
Our upcoming game GunZone will even include a dedicated server so players can still play even if the company dies out. Funny that this standard of old games is now considered a big feature.
It's infuriating how many times I've run into games where the potential commands/switches have been documented online across 30 websites, but every site is incomplete. Tons of games don't even document how to set the listen address or port, or even what runtimes might be needed in order to run.
I've always loved the SimCity series and at various times have spent a large number of hours tinkering with a large city. It's relazing. SimCity 4 was a great entry to the series. it had some oddities (eg regions allowed you to just send trash into the void) but it's a single player game so who cares? Much like No Man's Sky and "balance", as an aside.
But EA decided SimCity 5 had to be played online. Why? Because reasons. Often there's hand-waving about "piracy" but really it's just about control and "encouraging" micro-transactions as a future revenue source. Being able to flaunt your city to other players is a prime driver for people buying digital cosmetics.
But this meant SimCity 5 was vastly more complicated as software. The city size was a lot smaller. They wasted time on multiplayer features nobody cared about. Basically they had zero understanding of who their target market was and what theye liked. As long as some VP can put up a graph in PowerPoint showing projections on micro-transaction revenues and get a fat bonus, who really cares about what the players want.
So what happened/ SimCity 5 was a flop nobody even talks about anymore and it caused Cities Skylines to come into existence.
Fallout 76 [1] is another big example of turning something people loved (ie Fallout 3 and New Vegas) into something despised just because you, the publisher, can't get out of your mind that "online = more revenue".
Old games are loved, so much so that people build software emulators of hardware that is no longer made just to play them.
As for two of John's points:
Why not allocate a single developer at Meta to maintaining the game? Because that's career suicide for that developer. You'll never get promoted for that. You'll probably be viewed as non-essential and gently shown your way to the door with a subpar performance review when some director later needs a better rating for someone they care about and the ratings have to fit a bell curve.
Why not sell it off? Because then it's gone. if you hold ownership you might be able to sell it for a lot in the future. Sell it for $10,000 now and you've gained money that is immaterial and lost any future potential. It just doesn't make sense. You'll spend more than that on lawyers just reviewing the contract.
It was going to be an instant buy for me but then as I read the developer statements before release ... it started getting really weird. It was clear it wasn't going to be Sim City, or at least a lot of design decisions had nothing to do with what I thought of as Sim City.
Such a sad waste.
Really strange too because if there was an obvious formula to make a ton of money there it was and they still screwed it up.
I'm glad a game developer is saying this to other game developers. Maybe they'll listen.
I know that as a player, I think twice before buying games from publishers like Ubisoft that have a history of shutting down games - even games that I don't play.
Just one example is that he petitioned Samsung to remove the requirement for triple-buffering on their screens, to allow for better latency and in-game experience.
Samsung declined, so he hacked apart his phone to make a proof-of-concept prototype of the screen without the triple-buffering. He demoed this to Samsung, and they agreed to make the change because the experience was so much better.
It's tough to say "what game has Carmack had an impact on lately", because the scope of his role has was to impact the ecosystem of games and VR as a whole.
You don't see pushback on Joe Biden along the lines of "But what has he done for Scranton lately?", because the role of president encompasses a larger scope.
Similarly, the role of John Carmack has been larger in scope than improving one game at a time (and has been for a while).
Caveat emptor.
No, it shouldn't be, and isn't.
And claiming it is sounds like an argument by megacorp lawyers, when there's finally a huge class-action suit led by someone who is out for precedent-setting blood:
"It should've been self-evident that we rigged this in such a way that it would suddenly stop working, at a time of our choosing."
When that wasn't what they said in the marketing when they sold it, nor when customers invested many hours creating artifacts within it.
Like it's not self-evident that when you buy a bicycle, the shop will later come steal it back.
Or, when you buy canvas and oil paint, the art supply store won't break into your house and destroy your paintings.
Titanfall: Assault was a great 5 minute mobile auto-battle-arena game, shut down in 2018. I've never had a game that built & escalated into such a fun little conflict before or since.
Dreadnought is a lovely paced spaceship brawler, a great slugest, and is shutting down end of the month.
Neither of these games had anywhere near 10,000 active users, I'd guess. I still there was a chance for the really excellent niche things to do more than fizzle over time & disappear.
EA did them no favors by launching Titanfall 2 at the same time as a CoD and Battlefield title (EA was competing with itself which makes no sense). Had they launched in the spring, away from those two behemoth titles, the game would have had a much better life.
I suspect, however, that this time people got lucky: the Titanfall 1 and 2 games contained most, almost all of the server-side requirements/assets, so hosting a server was mostly a matter of breaking in to one DLL file and a few game script modifications to get access to the entire source engine and game features. As far as I understand, the main missing content were the navigation meshes for the AI, a few scenario scripts, and some party/lobby features. Most, if not all of those have been reverse-engineered or created by the community, so that's not even an issue anymore.
But the more games get individually specialised engines, less and less server features included in the client-side game releases, the harder it becomes to do this type of stuff. In the case of idTech, unreal engine and Source, most things outside of the main menu and settings panel are 'maps' and those are either 'hosted' by your own game client, or by a remote server. That means that due to the architecture, the client package has to have at least some server features. If games have training modes, or explore/tryout/singleplayer modes, there has got to be more server-logic/code/assets in the client package to enable that.
Give people either a defined time period that their purchase will be supported for or open source the server side so it can be run after the company decides to discontinue support.
IMO, if you charge a one time fee, and stop supporting something for economic reasons, users should be entitled to a refund. Or, more specifically, don't sell an infinite commitment for a fixed price.
I could see how EchoVR might not make much of a dent in Meta's earnings, but that'd be a significant customer base to start off with.
This applies to other FAANG companies too - like Google. Why not take a handful of Fuschia engineers, group them together and then have them partner with say, Framework.
Alas - definite agreement - likely structuring acquisitions to account for higher interest rates and future layoffs a decade into the future is probably asking too much of management.
https://web.archive.org/web/20230206155610/https://uploadvr....
They also afforded options blizzard had been too lazy to bother with, such as playing on a favorite patch, etc.
I'd like to hear more of his thoughts on this. The distributed and service orientated the stuff I work in gets, the less we ship. It's glacial.
It's a model I'd like to see included down the road when we can abstract the compute costs
1. Success is not guaranteed. At all
2. Things enjoyed by the masses don't need additional support to stay alive
For games like Quake 3 it was quite open before it was old. The game was community mod'able during its prime. You had mods like OSP being "the" mod for competitive game play (1v1 DM, Team DM and CTF). It being fully open sourced later on was icing on the cake.
Wikipedia said it sold 319,970 copies for about $15 million[0]. It was a 1 time purchase game for like $50 at release. It took 1.5 years to make with a team of 9. Not a bad turn out for a single purchase and open game with a self hostable server component. Keep in mind that was 20 years ago when gaming was way less popular.
There are pros and cons to both. But is this a forced choice, or can we have both?
Why should a matchmaking/game browser use proprietary code? Create standards and use off the shelf components for common problems. Whenever I go into the settings menu of a game I wonder how much time has been spent reinventing it.
I'd like to see DLC die tomorrow please and we get back to passionate people making maps and mods again.. maybe have a community where people build things, become game developers because they enjoyed the hobby.
I completely agree. Game devs have stripped so much away from the user. Matchmaking, lack of server hosting, no mods. It feels like much of the community has gone.
I play BF5 regularly and there's still a large number of 64 player servers full all day, every day. However, there's very few community servers and cheaters are so common. EA/Dice have abandoned it and there's no way for the community to keep it going.
Another game I used to play, Wolfenstein: Enemy Territory, is free to download [1] and still has plenty of community hosted servers/mods. That's a game coming up to its 20 year anniversary.
[1] https://www.splashdamage.com/games/wolfenstein-enemy-territo...
We are people in the world. The reality that our public square--which includes the ways in which we play--is being controlled by businesses who do not answer to us and do not care about that public square except in the way that value can be extracted from it is well within our rights to criticize. And, frankly, legislate.
Carmack's strong views on politics and certain other topics are indeed silly and uninformed, but his commentary in this case is interesting and valuable. He may be wrong in the end, but he's also not speculating like other business pundits.
I don't get it. What qualifications or grounds does anyone need to criticize a large corporations anti consumer actions? None needed, so not having special permission or be pre-qualified for some random person's satisfaction literally doesn't matter.
Anyone can criticize meta and no one needs a grounds aside from their criticism existing.