I need some piece of LLM generated code to make it into some critical TCP library that’s used for the internet everywhere.
Maybe then I can get some peace from the individuals that cry when a company mentions they used LLMs.
Aside: This looks like a proper use of LLMs - we’re not displacing artists with something trained on stolen art. Maybe it impacts some interns that needed to create PowerPoints.
Edit: Fine. I’m wrong. Vote with your wallet and don’t buy the game.
Except for the “developing concept art” bit. That’s not some unimportant intern’s work and does have the potential to displace artists. And while no AI-generated content may make it into the game, it does suggest that there will be art in the game that’s based on the AI-generated content. That’s what concept art is for, after all.
To add to this, concept art is one of the places I would least like AI to be used in. It utterly fails at being creative and actually creating something fundamentally NEW and the more we use it the more our media will just devolve into remixed homogeneity.
At first, Larian had planned to continue working with Hasbro’s Wizards of the Coast division on Dungeons & Dragons, but Vincke said he and his team spent a few months working on a new project before realizing they weren’t feeling the excitement they once did. “Conceptually, all of the ingredients for a really cool game were there except the hearts of the developers,” he said. They abandoned that game last year and pivoted to Divinity, a franchise that Larian also happens to own.
It’s crazy they have the finances to be working on a D&D franchise game and decide “…Nah. Let’s do something else.”
They recently switched to a new engine…
Uh oh.
I know folks like to hate on Unity, and Borderlands 3. Rightfully so. But let me list out some “in house engine” releases:
Cyberpunk 2077, which Nvidia backing
Mass Effect Andromeda, after previously being Unreal
Starfield
Paradox Grand Strategy, like Stellaris
A “smaller studio” example, Distant Worlds 2
All these drug their developers through hell, and we’re still technical messes at release. And after.
Now let’s look at some others:
KCD2: CryEngine
Expedition 33: Unreal
Black Myth Wukong: Unreal
Stray: Unreal
As a “smaller studio” example, Satisfactory: Unreal
…I’m just saying. Making a modern engine from scratch is hard. There are just too many things to worry about. And the record of “RPG studios rolling a new in house engine” is not great.
So what I hope this means is Larian moving to CryEngine or something like that, and not making something from scratch. But if they’re talking about early access so soon, I bet they licensed another engine.
They said very little about what that new engine entails, but much like Starfield, I suspect it’s largely reusing their old engine and only remaking select parts of it. Larian is doing something in the RPG space that, to me, makes nearly all of their competitors feel outdated, and it makes sense to me to make their own engine to do that as efficiently as possible. To make one of their games in an off the shelf engine like Unreal, with all of the bespoke physics objects and the ways every entity interacts with spells, elements, and other effects, could easily result in huge performance costs above and beyond what we saw in Act 3 of BG3.
I’m utterly terrified of them pulling an Andromeda/2077 and getting stuck in dev hell trying to debug the new engine bits instead of actually building the game. This is the advantage of prebuilt engines: someone else has already one all the hardware support/optimization and contemporary architecture stuff for you.
I’m less afraid of them pulling a Starfield, I suppose. The “divinity engine” in BG3 already runs okay. It’s not sleek like CryEngine KCD2, but it doesn’t feel janky or dated either, and even the mildest refresh over BG3 would be fine.
Much less is determined by engine than the average person thinks. Andromeda wasn’t a new engine; it was an engine that was made to make Battlefield games that then had to be used to make action RPGs and racing games after the fact. Capcom made an engine for the games they had in mind 10 years ago, and it’s fantastic at Resident Evil, Devil May Cry, and even serving as an emulation wrapper, but it’s showing cracks under the support for open world games that they added more recently. Larian’s engine is made to support the systems driven RPGs they conceptualized in the early 2010s, and there’s little chance some other engine will do it just as well or better without plenty of custom code anyway. Ask Digital Foundry about all of the “optimization” Unreal 5 has done for developers already.
If they can shoehorn in something akin to KCD2’s or Satisfactory’s Global Illumination, but keep their dev workflows and existing systems in place, that’d be perfect.
From the other Larian article in this community, it seems their engine improvements are largely things that they claim will allow them to iterate on ideas faster, like going right from mocap to a usable animation more quickly.
I truly love that Larian leadership frames everything they talk about around devs and their needs/wants. Another D&D game? “Oh, that’s great and all, but our devs hearts weren’t in it so we dropped it like a rock.” New engine? They ramble about improvements to dev workflows. It is so obviously a top priority.
Keep in mind that also comes with Vincke championing AI, and though he says no genAI assets will make it to the final product, there’s still some dissent. Here’s hoping though.
Well that can be reasonable. Obviously don’t vibe code an engine, but LLMs are great for basic code autocomplete, or quick utility scripts, things like that.
Really specialized AI (not LLMs/GenAI) can be great at, say, turning raw mocap into character animations. Or turning artist sketches into 3D models. Cogs in their pipeline, so to speak, which has nothing to do with GenAI slop making it into a final product.
The line is very fine though, and most in the business world skew to the side of pushing slop.
I think very little about AI compared to most people, for or against, but it largely seems to me like a solution in search of a problem, and it’s very cult-like how many CEOs get on board with it so quickly despite its very public lack of actually good results. On paper, the way Vincke describes their use of it sounds fine to me, but hopefully he’s not doing something so idiotic as to mandate its usage, as is happening at workplaces for friends of mine right now.
“Mandate its usage” could mean the motion cap/animation people have to learn some kind of automation tool, that’s now part of the engine.
That’s fine.
And that’s very different from the “you MUST make X hits to Microsoft Copilot” type garbage that’s so common now.
I’m harping on this because I’m afraid Larian will try something reasonable, yet get immense, unwarranted backlash (both internally and publically) because of other workers’ experiences with enshittified ML. And how politicized “AI” is becoming.
Machine learning is not bad. Tech Bro evangelism and the virus they spread among executives, is. And I don’t want the garbage they sell to poison tools studios like Larian could use to get ahead of AAA publishers.
Hasbro could have done nothing and made a bunch of money, but they chose temporary short term gains. Baldur’s Gate 4 will arrive far sooner than you think, and it will be terrible.
WTF. That’s awful, and also totally baffling. “This single game is responsible for a huge chunk of revenue and introducting countless people to D&D; let’s lay off its staff and leadership.”
Baldur’s Gate 4 will arrive far sooner than you think, and it will be terrible.
What do you mean by this? An outsourced spinoff is already in the works? I don’t see that in the linked article.
Nothing has been announced as far as Baldur’s Gate 4 goes yet. It looks like Hasbro is being a little bit smart and are going to try and make (“make”) a handful of other smaller games, like the recent Warlock game announcement.
But at some point Baldur’s Gate 4 will be announced, but Hasbro isn’t going to be willing to invest properly into it in order to make a good game.
…I’m just saying. Making a modern engine from scratch is hard. There are just too many things to worry about. And the record of “RPG studios rolling a new in house engine” is not great.
Larian’s track record is good. They used an in-house engine for Divinity: Original Sin, Divinity: Original Sin 2, and Baldur’s Gate 3. They also made their own game engine for their older Divinity titles (Divine Divinity and Beyond Divinity). And Vincke attributes at least part of their success to using in-house tools instead of “off the shelf” engines.
I can understand this hesitation, but I don’t expect that from Larian, they’ve delivered in the past and I suspect they’ll deliver again.
(So had CD Projekt Red of course, but Cyberpunk’s launch issues were largely stability/performance related, IIRC. Whereas Hello Games over-promised and under-delivered core features on No Man’s Sky.)
Let’s be straight: as amazing as Baldur’s Gate 3 is today, Act 3 launched half baked and half broken. My first playthrough experience was horrible, largely thanks to broken flags and missing content from the Upper City, and I’d be lying if I said I didn’t have comparable experiences with early versions of Original Sin 2. Hell, they rewrote basically the entire final act of that game with the definitive edition, and I’m under the impression Original Sin 1 had a similiar situation, though I didn’t play it enough between the original and the definitive edition to experience it.
Now, part of all this is because Larian opts to make decisions to cut content and reduce scope rather than abuse their staff or delay a project. In Baldur’s Gate specificslly, I won’t say I am perfectly happy with the outcome, but they are a good studio that practices reasonable employee ethics, and ultimately puts in the work to get there with the product as well. I’d have no issue buying Divinity day one or even pre-ordering, but I do not expect a perfectly complete and polished experience on release.
It still has bugs to this day. I played through the whole game two months ago. The printing press mission was extremely broken. It’s a mission where you are supposed to swap out the headline in a printing press so a magazine doesn’t shit-talk your party. The mission progressed as intended, the press even praised me for swapping the article out and on the next day I still got shit-talked.
I had to do the whole mission again and talk to the printing press twice (for no reason) to fix it. Yenna in my camp also never cooked for me.
Larian announcing their next game to be even bigger than before makes me a bit cautious. I hope they don’t bite off more than they can chew.
I think with CDPR people had conveniently forgotten how much of a broken, buggy mess Witcher 3 was at release tbh. It wasn’t as broken as Cyberpunk but I think that it was also easily forgotten because people weren’t remotely as hyped for the game when it came out. CDPR actually has always had a track record of putting out really buggy games that get patched into great ones later.
Cyberpunk’s launch issues were largely stability/performance related
I played the first release when it came out. There were a LOT of mission-breaking bugs, missing content, much less customization options, entirely missing features, a really messy perk system etc. It feels like a very different game now, since they patched in more content that was initially missing.
They should’ve pushed back the game at least for another year. 1.0 mostly focused on the cutscenes and Johnny Silverhand/Keanu Reeves since that’s what sold the game initially and left a lot to be desired in other areas.
If I remember correctly, you couldn’t even skip the first training mission. You also had to combine clothes that looked like ass because the armor rating was tied to them and there was no cosmetics system (that came with 2.0).
Stirring hype has failed too many times. I’m sure they have big plans for the game and it’s going to be very good. However, promising so much so early just screams incoming drama to me.
I don’t think it was a slog, but I do find DND 5e an unsatisfying system. You spend a long time waiting to get to the cool parts of your character, and unlimited resting breaks dnd’s already dubious balance.
I tried a Pathfinder mod, but it wasn’t quite doing it for me. I’m not sure what would do it for me exactly. I’m being a stereotypical customer where I don’t know what I want, but I can tell you what I don’t like, heh.
There’s a single mod that purports to convert the game to pathfinder 2e. I’m not sure if the balance isn’t to my taste or if I was doing something wrong, but my characters all had very low hit chances on their first attack. Missing isn’t fun for me
The story was excellent, the combat was a slog. Still finished it and overall enjoyed it, but it felt like they were being limited by the DnD system a lot, ultimately worsening the experience.
The first Divinity was called Divine Divinity, and it was closer to Diablo than Baldur’s Gate. As per this interview, this game is going to be the same style as BG3 and the Original Sin games.
Learned about this for the first time earlier today watching Mortismal’s Divinity recap videos lol
I did know about Divine Divinity (which, fun fact, canonically features Lucian the Divine as your player character) but Dragon Commander completely missed my radar.
With the money they made from BG3 they have the means to make Divinity the best it can possibly be and I couldn’t be happier for them. The first two Divinity games were great, don’t get me wrong, but they were basically low budget when compared to the money they got from Hasbro and Wizards of the Coast to make BG3. I imagine they basically have “fuck you” money now and can do whatever the hell they want. They definitely deserve it and I can’t wait.
Edit: Turns out Larian is going to use gen AI for concept art. I guess fuck all those concept artists trying to get entry level jobs. Very disappointed.
Their concept artists are allowed to use some generative AI tools to explore ideas and speed up their workflow. They’re currently hiring a bunch more concept artists (both juniors and a senior character artist) so if you’re trying to get a job: https://larian.com/careers/4fd694b3-ece7-4307-9949-15cac512a815
Great place to go if you’re looking for a concept artist job.
Edit: Turns out Larian is going to use gen AI for concept art. I guess fuck all those concept artists trying to get entry level jobs. Very disappointed.
It’s misinformation. They have almost 30 concept artists employed. They use GenAI for quick ideation, not for concept art.
Also, corpos are alllowed and in some cases are required to lie, even the “good ones” like Larian. And now, when they have more money then ever, they become less trustworthy than ever. This slope is very slippery. Nothing stops them from overextending their ideas, and when a lot of money involved, I can forsee “well, we need to finish quickly, and we’re already use llm anyway, let it help with the script, and since it’s already in the script no reason not to let it generate some art, and well, since it’s already everywhere, why don’t we generate the code with it”
My point is pretty simple: they said they only use LLM “for good”, but the more they get, the more insensitive they get to lie, so your “but they said [bla bla]” argument can’t hold. If they started using it for something, the only thing stopping them from using it for everything is their reputation and the desire to make a good game, and the more money is on the line, the less value that desire holds in the face of immediate profits.
I love everything Larian did before, I’m a huge fan of the Divinity series, BG3 is still my top 5 favourite games of all time, but this doesn’t mean all that can’t go to shit, wouldn’t be the first.
I don’t agree on a fundamental level, though. Anything and everything could go to shit at any time. You could get killed on a bus stop, your favourite grocery brand might be outed to be using slave labour, Larian could start using AI instead of human work for everything…
With that approach, might as well hide in the woods, disconnect from civilisation and wait for the world to end.
I personally reject that attitude. I think we should support what’s good while it’s good and stop supporting it when it goes bad. And, to me, the way Larian uses AI is not “it went bad already”. Like it or not, the tools exist. They’re everywhere. A single (small-ish) company rejecting its use out of principle is not going to make a dent on that, it won’t even be registered within the margin of error among the billions, if not trillions, of monthly impressions AI companies get from teenagers talking to chat-bots.
And even if it did, it doesn’t matter - AI companies are not making profit anyway. Actually, fewer users is better for them, because they’re actively losing money every time someone uses their product. The whole AI bubble is propped up on the largest circlejerk in history and users are the least important, if not flat out insignificant, aspect of it.
And don’t get me wrong, the fact that it’s originally trained on stolen data is important, but… Kind of irrelevant in this case - and that’s for two reasons:
Companies like Corridor Crew mostly use AI that they self-train, which means that no stealing happens. We don’t know what Larian does.
Even if Larian uses publicly available models that are trained on stolen data… fuck me, we should be going after the people who stole the data, not the end users!
Would I prefer if they didn’t use AI at all? Sure! But am I going to start shitting on the entire company just because they do? Hell no! Their products are still made with care and love and humanity at their core. Just listen to interviews from Jennifer English or Neil Newbon - they praise both the VO company and Larian for their amazing approach.
And finally - the “war on Larian” would make much more sense if we also learned that they’re firing concept artists - but they’re doing the opposite - they have open positions for concept artists, character artists, environment artists, etc., etc. They’re currently hiring these people!
So, yeah, it’s just a lot of noise over practically nothing, in my opinion.
Shovel Knight was responsible for kicking off the second renaissance wave of indie games, as Cave Story was the first wave. It's quite a bit saddening that they've put themselves in this corner.
Shovel Knight is one of the most successful indie games ever released. We’re not talking of a “moderately” successful game that sold a few hundred thousand copies, like Hyper Lighr Drifter or CrossCode: SK sold over 2.5m copies back in 2019, and I’d wager at least as much since then. How do you go from there to almost bankruptcy?
They did a lot of extra work on it without charging for it, and it’s been a long time since they put out a hit. California salaries and real estate are expensive.
Paying a bunch of salaries when your revenue streams are Shovel Knight (good but old game that kept getting free DLC and made a lot of its money before release) and… Shovel Knight Dig. They had to go back to Kickstarter for Mina, after all.
If it was one guy or a tiny team, SK’s success would be enough for them to be “set for life”, but a business is more expensive to run than a team is. They probably don’t expect Mina to be a phenomenal income stream either, since (like SK) it’s already mostly done making them money.
bloomberg.com
Najnowsze