How far are we from AI recreating EQ99?

[Dulu]

https://www.youtube.com/watch?v=MWl2qQRyutc

Watched this video, where the guy basically recreates Doom in an hour.

Keep in mind, Doom is a single-player game, and came out in 1993.

But... the video is 3 months old and was probably recorded some time before that.

So the question is, how far are we from making a game like P99 in AI?

[DeathsSilkyMist]

https://www.youtube.com/watch?v=MWl2qQRyutc

Watched this video, where the guy basically recreates Doom in an hour.

Keep in mind, Doom is a single-player game, and came out in 1993.

But... the video is 3 months old and was probably recorded some time before that.

So the question is, how far are we from making a game like P99 in AI?

Even an older MMO like Everquest is significantly more complicated to produce when compared to Doom. You can already download an open source Unreal 5 multiplayer shooter template, so there's no real need to remake the wheel in a probably worse way with AI.

Complex games like Everquest (and beyond) are going to be one of the last frontiers when it comes to AI generated content, especially multiplayer games. Vibe coding networking code is usually the easiest way to open up security flaws and cheating. You still need experienced programmers at the wheel to check AI's work at all times.

[BradZax]

I saw a youtube video about putting people on a moving boat in an MMO style game and it looked like he was building a nuclear bomb so I think yeah we're a ways away still.

You still need experienced programmers at the wheel to check AI's work at all times.

But how long until AI turns each experienced programmer into the equivalent of 50 experienced programmers :eek:

[DeathsSilkyMist]

But how long until AI turns each experienced programmer into the equivalent of 50 experienced programmers :eek:

This is true. AI can act as a force multiplier for experienced programmers.

The main concern I have with that thinking is this will probably lead to hiring less junior programmers.

The question is: Is that a good thing? The only way to create experienced programmers is to train junior programmers. Until AI can do everything for us, we don't want to stop training people.

Ideally you keep hiring junior programmers, and you use AI to increase productivity.

[BradZax]

The main concern I have with that thinking is this will probably lead to hiring less junior programmers.

Maybe, or glass is half full take:

Jr programmers (who have the most up to date knowledge of cutting edge tools because they are young and adaptive) will be able to compete with companies that wont hire them, and decimate them in the market.

Ye old "disruption" technique :cool:

PRBSKaTDrqQ

[Dgc2002]

Watched this video, where the guy basically recreates Doom in an hour.


That's a CRAZY representation of the video.

The 'better' version he made using grok had:
The most basic of WASD movement
Shooting is just point and click
Enemy behavior is

move towards player position at a constant speed if they're not close enough
face player
shoot every n seconds


The shooting every n seconds portion wasn't even finished within the hour time limit but then worked randomly after adding the imp.

I wouldn't be surprised if you can just find a free boilerplate project that covers all of that and more as a starting point.

[Swish]

Until AI can do everything for us, we don't want to stop training people.


You don't want this.

[loramin]

OP the short answer is that we're a longer ways off than you think.

I code with AI daily, and it's incredibly powerful. I can say "go do this task", and it does it. Most of the time it even gets it right.

However there are two major issues. First, the context window. The AI can only "remember" so much at a time, out of everything it reads, everything you tell it etc. The context limits are inherent to how AI works: until there's some new innovation occurs, this isn't going to get much better.

The second problem ... last night I asked it to fix a test, and it tried to change the test to check the wrong thing (so it passed, but the issue wasn't fixed :(). AIs can't (really) understand what's "right", as they just know what their linguistic model predicts.

It's like that old joke about the repair man and the hammer. AI is a powerful tool, but it doesn't fix things, the human intellect still does.

A repairman was hired to repair a large machine in a factory. He showed up, examined the machine, then tapped it once with a hammer. It started up. The factory owner was pleased, but not when he got a bill from the repairman for $100. He thought that was outrageous, and he asked for an itemized bill. So the repairman handed him a bill which said:

Tapping machine with hammer: $1

Knowing where to tap: $99

(Re)creating real/serious games still requires humans because of the above ... just not as many as it used to.

[Kich867]

Having started using it more at work, also programming, it's not bad at certain things but anything that requires any sort of context its often dumb as hell at. Or, to reinforce loramin's point, like I had it help me spin up some AWS infrastructure, it organized it and wrote it well but it included options that don't exist on those resources / tried to use conflicting options / just had general bugs in it. So it requires extensive review.

It helped me because it wrote up like 900 lines of cloudformation template, but the fiddling afterward to correct it was also time consuming. I'm skeptical on whether in the end it actually did save me any time from the amount I have to baby it.

I think it was MIT that recently did a study on this that: despite the project managers claiming AI is improving speed, and despite the devs themselves believing that AI is improving their workflow, when you actually observe and time how long tasks take to be done, tasks with AI are being completed about 20% slower.

(IIRC they did a study with various open source developers, found similar tasks, had them do tasks with AI and tasks without AI, and they were consistently faster without it)

[BradZax]

First, the context window. The AI can only "remember" so much at a time, out of everything it reads, everything you tell it etc. The context limits are inherent to how AI works: until there's some new innovation occurs, this isn't going to get much better.

I think we're just waiting on more power. For example. Google image creation with Gemini using context costs about 3-5$ in wattage to generate a 6-10 second video. Vs their diffusion model for video generation is more like 5 cents or less.

That doesn't mean anything except it just gives you an idea about how much wattage context adds to the equation!

I am hoping corporations using their own nuclear reactors will be give us a big jump in usage.

https://www.latitudemedia.com/news/open-circuit-inside-metas-massive-nuclear-push/

[DeathsSilkyMist]

I think we're just waiting on more power. For example. Google image creation with Gemini using context costs about 3-5$ in wattage to generate a 6-10 second video. Vs their diffusion model for video generation is more like 5 cents or less.

That doesn't mean anything except it just gives you an idea about how much wattage context adds to the equation!

I am hoping corporations using their own nuclear reactors will be give us a big jump in usage.

https://www.latitudemedia.com/news/open-circuit-inside-metas-massive-nuclear-push/

Personally I do not think the current implementation of LLM's will ever have enough of a context window to play a 40 hour game while maintaining the state of the game. Nor will it give us artifical general intelligence.

We need to wait for a company to come out with a different approach to AI.

[BradZax]

Personally, I do not think these guys would be investing trillions in a glorified clippy app.

If they were doing it because the subscription model made sense, nvidia would still be selling personal computers instead of swapping to some kind of Death Star customers only model.

My guess is they have seen what it's actually capable of if they had:

https://i.imgur.com/KfYu0m3.png

https://i.imgur.com/W59PAWv.png

[loramin]

I think we're just waiting on more power. For example. Google image creation with Gemini using context costs about 3-5$ in wattage to generate a 6-10 second video. Vs their diffusion model for video generation is more like 5 cents or less.

That doesn't mean anything except it just gives you an idea about how much wattage context adds to the equation!

I am hoping corporations using their own nuclear reactors will be give us a big jump in usage.

https://www.latitudemedia.com/news/open-circuit-inside-metas-massive-nuclear-push/

Well power is certainly a factor, but there's something more fundamental to the tech.

You have to understand "AI' is just math. Everything you tell the AI (ie. all context) becomes part of its calculation. There are fundamental limits to how much context it can process. Throwing more processing power (which requires electrical power) can increase the "context window" some, but only so far.

It's a diminishing returns thing with how the math works: the thousandth context token takes more power than the first, and the ten thousandth takes way more, and so on ... much like how level 60 requires way more effort than level 1.

You might pay a neighborhood kid to level to 2, but it will cost wayyyy more (more than 60x) to pay him to get to 60. Even if you inherit $100 more, it probably won't get you to 60.

[OriginalContentGuy]

Personally, I do not think these guys would be investing trillions in a glorified clippy app.

I do. It's not that simple but economic flywheels are a concept if you insist on modeling an economy as an electronic circuit.

[BradZax]

Well power is certainly a factor, but there's something more fundamental to the tech.

You have to understand "AI' is just math. Everything you tell the AI (ie. all context) becomes part of its calculation. There are fundamental limits to how much context it can process. Throwing more processing power (which requires electrical power) can increase the "context window" some, but only so far.

It's a diminishing returns thing with how the math works: the thousandth context token takes more power than the first, and the ten thousandth takes way more, and so on ... much like how level 60 requires way more effort than level 1.

You might pay a neighborhood kid to level to 2, but it will cost wayyyy more (more than 60x) to pay him to get to 60. Even if you inherit $100 more, it probably won't get you to 60.

5 years ago you could have said the exact same thing about LLMs just not being mathematically possible to get to the level they were 3 years ago.

We're handicapped right now not with how much power it can use, or how much power it can process etc.

We're handicapped with:

It wants more power, it can use more power, but that power is just too expensive to be worth it commercially right now.

[BradZax]

I do.

The economy isn't just some funny meme that reddit made everyone think it is.

These guys know what they are doing.

The CEO of Nvidia didn't get into component engineering 50 years ago so he could sell snake oil when he was old and rich enough to be retired.

Nor did he start it to be some megalomaniac tech super villian, because we have 50 years of him proving building quality computer components makes him and gaming the world happy.

[OriginalContentGuy]

You're right, I'm not. But name me one large organization whose leaders don't write down their instructions or procedures so that they may be read by the company and executed according to specific directions. I read the notes of the guys to whom you are referring to inform my speculation and not just the public relations releases.

Edit: I replied to an edited draft by Exlax suggesting I can't got perspectives based on the same models any CEO uses and perhaps then some

[BradZax]

You're right, I'm not. But name me one large organization whose leaders don't write down their instructions or procedures so that they may be read by the company and executed according to specific directions. I read the notes of the guys to whom you are referring to inform my speculation and not just the public relations releases.

Are you asking me to name one company that doesn't publically send out all of its business strategies, emails, and meeting recordings to social media readers?

Ok:

All of them.

I just named at least one there.

[OriginalContentGuy]

No and based on your history I don't believe your misunderstanding wasn't intentional.

Sorry exlax I'm busy later maybe.

I think I can dumb down what I said so it would make more sense to you specifically, I expect. And to everyone, but I'm working I was just on break.

[BradZax]

No and based on your history I don't believe your misunderstanding wasn't intentional.

Sorry exlax I'm busy later maybe.

I think I can dumb down what I said so it would make more sense to you specifically, I expect. And to everyone, but I'm working I was just on break.

I think what you are saying is:

Everyone who is rich is dumb and everything they build is fake.

Im simply saying that I don't believe they would spend trillions of dollars on something that produces no revenue.

That the people who literally built the current economy shifted to decide, "you know what that clippy app was actually really on to something"

[OriginalContentGuy]

I still don't believe you are arguing in good faith. Not everyone does.

If we are going to discuss revenue then we must discuss money. Is this correct? If so, then by the same reasoning, If we are to discuss money, then we must discuss energy.

Do you accept this as accurate?

If not, then neither can I assist you presently, nor do I know or care if thar's what you want to happen.

[loramin]

5 years ago you could have said the exact same thing about LLMs just not being mathematically possible to get to the level they were 3 years ago.

We're handicapped right now not with how much power it can use, or how much power it can process etc.

We're handicapped with:

It wants more power, it can use more power, but that power is just too expensive to be worth it commercially right now.

Right, before the LLMs we have today were impossible (with a reasonable amount of processing power). Innovation changed that and we got LLMs.

It's like with airplanes, you didn't go from leaving the ground to circumnavigating the globe: it took many separate inventions.

But, there were periods, often many years, between those inventions. When the Wright brothers first flew, some people were like "I'll be flying to China tomorrow!". Those people were very wrong.

AI will get better, no doubt, but no one can predict innovation timelines. And without some new invention, AI will never make EQ.

[BradZax]

Right, before the LLMs we have today were impossible (with a reasonable amount of processing power). Innovation changed that and we got LLMs.

It's like with airplanes, you didn't go from leaving the ground to circumnavigating the globe: it took many separate inventions.

But, there were periods, often many years, between those inventions. When the Wright brothers first flew, some people were like "I'll be flying to China tomorrow!". Those people were very wrong.

AI will get better, no doubt, but no one can predict innovation timelines.

Sure, but I am not predicting that timeline.

Im arguing that trillions of dollars being invested into a timeline that has no predictability wouldn't make sense unless everyone in the world went mad at the same time.

It took the airplane industry 50 years to go from paper and wood to nuclear intercontinental space traveling rockets.

People worked on flight for melinia, and then after someone figures out lift, suddenly we're riding on busses in the sky.

It's hard to predict where we'll be in the near future, but one thing is for sure the same people that told me that AI was going to crush human creativity and I was evil for supporting it, are now the same ones saying that we're stupid for thinking AI has a lot of untapped potential.

And without some new invention, AI will never make EQ

This is objectively false though. More power would allow modern technology to do this.

Unless you're going to make arguments like, a new invention is a new way to put transistors onto a computer chip, or building a space fairing data center.

[DeathsSilkyMist]

Sure, but I am not predicting that timeline.

Im arguing that trillions of dollars being invested into a timeline that has no predictability wouldn't make sense unless everyone in the world went mad at the same time.

Predicting its around the corner, or a long way off is both a wild guess.

It took the airplane industry 50 years to go from paper and wood to nuclear intercontinental space traveling rockets.

People worked on flight for melinia, and then after someone figures out lift, suddenly we're riding on busses in the sky.

It's hard to predict where we'll be in the near future, but one thing is for sure the same people that told me that AI was going to crush human creativity, are now the same ones saying that we're stupid for thinking AI is a valuable tool.

Think of the metaverse. Facebook bet everything on VR becoming ubiquitous in a decade, to the point where they changed their company name. I don't see anyone using the metaverse in 2026.

Technology can plateau, and the companies who heavily invested in AI are not going to simply tell everyone AI may have plateaued. It would crater their investments.

I don't forsee current LLM's being the future of AI. Hopefully the amount of money put into AI will yield a better AI model, but AI could end up being the next plateaued technology like VR.

[BradZax]

Think of the metaverse. Facebook bet everything on VR becoming ubiquitous in a decade, to the point where they changed their company name. I don't see anyone using the metaverse in 2026.

Technology can plateau, and the companies who heavily invested in AI are not going to simply tell everyone AI may have plateaued. It would crater their investments.

I don't forsee current LLM's being the future of AI. Hopefully the amount of money put into AI will yield a better AI model, but AI could end up being the next plateaued technology like VR.

Yeah i agree, but that was FACEBOOKs bet, not "every company in the world"

Not: every hardware manufacturer, and every technology investor.

WAY more investors did not invest in metaverse, than did.

More people have left the VR space than entered it in the time that meta was investing heavily in their vision for the future of their own company.

While the opposite is true about AI, year after year.

And remember, right now we're severely limited on power!

[DeathsSilkyMist]

Yeah i agree, but that was FACEBOOKs bet, not "every company in the world"

Not: every hardware manufacturer, and every technology investor.

WAY more investors did not invest in metaverse, than did.

More people have left the VR space than entered it in the time that meta was investing heavily in their vision for the future of their own company.

More investment does not equate to more smart people knowing the future. I brought up VR because multiple technologies in recent times have had this pattern:

1. A leap in the technology
2. A mass investment into the technology to be the first company to corner the market
3. Pleateau

Cryptocurrency is another technology that did this. Bitcoin came out in 2009, and it is still just a speculative market. You don't see a mass adoption of cryptocurrency as a primary method of payment.

[BradZax]

More investment does not equate to more smart people knowing the future. I brought up VR because multiple technologies in recent times have had this pattern:

1. A leap in the technology
2. A mass investment into the technology to be the first company to corner the market
3. Pleateau

Cryptocurrency is another technology that did this. Bitcoin came out in 2009, and it is still just a speculative market. You don't see a mass adoption of cryptocurrency as a primary method of payment.

No, Crypto didnt. It took over a decade for any of these companies to invest in crypto.

Also crypto flattens your argument: Because there was an actual return on crypto investments.

And in many high profile cases 10-100x their investments.

That is FULL STOP not the case for AI investment.

There is no revenue being made, yet year after year, there are 10x the investment.

No product to "invest" in.

The argument that is being made is across the internet is, former anti AI people who thought it would crush the human spirit, now think that all the people building it are stupid idiots and the AI can't do anything.

Crypto crashed, when it became a non interesting product to sell to people.

AI could crash if it provides a non interesting product, but up until now, it hasn't even offered a product!

And its going full steam ahead!

[DeathsSilkyMist]

No, Crypto didnt. It took over a decade for any of these companies to invest in crypto.

Also crypto flattens your argument: Because there was an actual return on crypto investments.

And in many high profile cases 10-100x their investments.

That is FULL STOP not the case for AI investment.

There is no revenue being made, yet year after year, there are 10x the investment.

No product to "invest" in.

The argument that is being made is across the internet is, former anti AI people who thought it would crush the human spirit, now think that all the people building it are stupid idiots and the AI can't do anything.

Crypto crashed, when it became a non interesting product to sell to people.

AI could crash if it provides a non interesting product, but up until now, it hasn't even offered a product!

And its going full steam ahead!

Return on investment and investment itself is irrelevant to the state of the technology. The technology around crypto hasn't made large amounts of progress. That was my point. The technology has largely plateaued, and people are just coasting on the speculative returns.

[loramin]

This is objectively false though. More power would allow modern technology to do this.

Unless you're going to make arguments like, a new invention is a new way to put transistors onto a computer chip, or building a space fairing data center.

No, you're still not understanding. Let's say you pay someone to level your toon. Each new level takes more time, so it costs more money. Up to 20 you can pay them for a level a day, but after that they take multiple days to level.

You can get a new job that pays more, and then you can afford to pay for faster leveling. This is current AI.

But now you want a live toon instead, max level, with all AAs. It doesn't matter if you pay your player more, they still can't get you a level 100 with all AAs. It doesn't matter how much money your player gets, he can only play so many hours a day (and at some point he gets sick and quits).

That's getting an AI to make something like EQ. Throw the very best models at it, give them their own personal nuclear reactor even ... they still won't be able to do it. New tech is required, not just better-powered existing tech.

[Kich867]

These guys know what they are doing.

Why would you ever assume that.

[loramin]

P.S. check out Steve Yaegee's Gas Town. It's basically what you want (throw gas on the fire of current tech) ... and it's barely a thought experiment, let alone something that can produce working software.

[BradZax]

No, you're still not understanding. Let's say you pay someone to level your toon. Each new level takes more time, so it costs more money. Up to 20 you can pay them for a level a day, but after that they take multiple days to level.

You can get a new job that pays more, and then you can afford to pay for faster leveling. This is current AI.

But now you want a live toon instead, max level, with all AAs. It doesn't matter if you pay your player more, they still can't get you a level 100 with all AAs. It doesn't matter how much money your player gets, he can only play so many hours a day.

That's getting an AI to make something like EQ. Throw the very best models at it, give them their own personal nuclear reactor even ... they still won't be able to do it. New tech is required, not just better-powered existing tech.

Right now you can generate an image on google gemni, it uses recursive learning and LLM models.

You can also generate an image on one of their diffusion models, these do not use recursive learning or LLM models.

The recursive memory generation costs 10$ a second of wattage to generate.

The problem isn't "this is as much wattage as technology allows us to pipe into it"

It's that, "currently it costs 10$ to generate that much wattage."

So, they don't allow users to use the full version of their LLMs or they would go bankrupt in processing costs.

Every question, causes the LLM to turn on compactors that need to be cooled.

It could turn on more compactors now, but they don't have the power to cool them.

If overnight we built enough pwerplants to generate enough power so that it costs 10 cents to generate 1 second of video, you would see massive improvements in LLM results that you thought were impossible, overnight.

The new version of chat GPT and shit, thats trying to do MORE with less. Optimizing their limited model to function as efficiently as it can with current power limitations.

But the thing they are investing in, is capable doing more, with MORE!

[DeathsSilkyMist]

Right now you can generate an image on google gemni, it uses recursive learning and LLM models.

You can also generate an image on one of their diffusion models, these do not use recursive learning or LLM models.

The recursive memory generation costs 10$ a second of wattage to generate.

The problem isn't "this is as much wattage as technology allows us to pipe into it"

It's that, "currently it costs 10$ to generate that much wattage."

So, they don't allow users to use the full version of their LLMs or they would go bankrupt in processing costs.

Every question, causes the LLM to turn on compactors that need to be cooled.

It could turn on more compactors now, but they don't have the power to cool them.

If overnight we built enough pwerplants to generate enough power so that it costs 10 cents to generate 1 second of video, you would see massive improvements in LLM results that you thought were impossible, overnight.

The new version of chat GPT and shit, thats trying to do MORE with less. Optimizing their limited model to function as efficiently as it can with current power limitations.

But the thing they are investing in, is capable doing more, with MORE!

Let's run with your idea for a moment. If we built 100x the datacenters overnight, genie 3 still wouldn't be able to make Everquest. The tech just doesn't work in a way that is condusive to persistent online worlds with thousands of players. Maybe genie 3 with 100x datacenters would allow you to play a basic singleplayer game for a few hours.

[loramin]

Right now you can generate an image on google gemni, it uses recursive learning and LLM models.

You can also generate an image on one of their diffusion models, these do not use recursive learning or LLM models.

The recursive memory generation costs 10$ a second of wattage to generate.

The problem isn't "this is as much wattage as technology allows us to pipe into it"

It's that, "currently it costs 10$ to generate that much wattage."

So, they don't allow users to use the full version of their LLMs or they would go bankrupt in processing costs.

Every question, causes the LLM to turn on compactors that need to be cooled.

It could turn on more compactors now, but they don't have the power to cool them.

If overnight we built enough pwerplants to generate enough power so that it costs 10 cents to generate 1 second of video, you would see massive improvements in LLM results that you thought were impossible, overnight.

The new version of chat GPT and shit, thats trying to do MORE with less. Optimizing their limited model to function as efficiently as it can with current power limitations.

But the thing they are investing in, is capable doing more, with MORE!

Did you read what I wrote about EQ? You think that because you can pay some kid to get you from level 3 to 4 that money = leveling ... and it does ... to a point. Even if level 40 takes a lot longer than 4 you can (probably, depending on the kid) pay them more to make them spend the hours needed to get you to 40.

But if you try to make that kid farm you a live toon with max AAs, the kid will go off to college before he finishes. You can throw ten times as much money per hour at him and it won't change a thing.

It's an analogy, but that's the gist of it for AI too. If an AI can (extending the analogy) get you to 60 with today's tech, then yes, throwing more power at it might get you to 60 with 1 AA. Even more power (a lot: remember, diminishing returns) might get you 2 AAs.

But if you want all AAs (ie a working EQ), even with all the power in the world, current LLMs could never do it. But don't try to argue with me: go learn more about how this amazing tech works yourself!

[Goregasmic]

It is all fun and games in theory but right now it costs like 10B$ to build a powerplant over 10 years and pretty much all of these companies have zero path to profitability. Funding will dry up before they get online. AI doesn't have much social acceptability right now, not sure how much funding they'd get from public money either. There's already huge clashes over data center water usage for what gain exactly?

[BradZax]

It is all fun and games in theory but right now it costs like 10B$ to build a powerplant over 10 years and pretty much all of these companies have zero path to profitability. Funding will dry up before they get online. AI doesn't have much social acceptability right now, not sure how much funding they'd get from public money either. There's already huge clashes over data center water usage for what gain exactly?

Well in foundation the anti ai people killed all the pro ai people and all the ai.

The anti ai crowd is getting pretty far left behind IMO but the reactionary karens that want to protect children from deep fakes are still clinging to their paranoia.

I genuinely think we're at the 70s and some of us like computers and some of us hate them.

[OriginalContentGuy]

This is like Nexii's Erotic Revelations thread except with Brad. 'Nuff said.

[OriginalContentGuy]

Respect to OP though for sharing

[BradZax]

But don't try to argue with me: go learn more about how this amazing tech works yourself!

Just curious.

Why do you resort to like this "im smarter than you, your argument comes form a lack of knowledge" approach?

[Goregasmic]

Well in foundation the anti ai people killed all the pro ai people and all the ai.

The anti ai crowd is getting pretty far left behind IMO but the reactionary karens that want to protect children from deep fakes are still clinging to their paranoia.

I genuinely think we're at the 70s and some of us like computers and some of us hate them.

I think the average person looks at this and all they see is shitty image generation, hallucinations, broken google overview, more sophisticated scams, increased workers/citizen monitoring, deep fakes, disinformation and we'd have to trade our local power/water for this? And we aren't even talking about the more nefarious uses like defense departments using it to dodge accountability but that's probably beyond what the average person notices I guess.

[loramin]

Just curious.

Why do you resort to like this "im smarter than you, your argument comes form a lack of knowledge" approach?

Because in life some things require knowledge and effort to understand?

Look, you are claiming the only thing stopping AI from writing EQ is power. That's like claiming disease is caused by spirits not viruses.

I just spent half a day (was at jury duty the other half) writing software with Claude's latest model, on a $100+/month plan. I use this technology as part of my job: I know how it works like a carpenter knows a hammer.

I'm telling you: more power (alone) will not get AI to make EQ. If that's not enough to even make you curious (go ask an AI!), nothing I say will change your mind.

[BradZax]

Look, you are claiming the only thing stopping AI from writing EQ is power.

No, that is what you are arguing with me about.

I am arguing that the trillions invested in Ai indicates that the Ai has not plateaued and right now and i know for a fact it is severely limited by power costs and the Ai tools that we're using are heavily capped so the examples of mistakes and stuff most people experience are because they are using a very limited version of their Ais.

The OP question is "how far are we from making a game like EQ in AI" like the examples.

Right now, right now someone could do that.

Someone already did with unity (https://www.youtube.com/watch?v=Ny8F8va9Zi8), without AI.

So with AI, right now. Someone can make an everquest game with AI.

We just need someone to want to do it!

[Kich867]

No, that is what you are arguing with me about.

I am arguing that the trillions invested in Ai indicates that the Ai has not plateaued and right now and i know for a fact it is severely limited by power costs and the Ai tools that we're using are heavily capped so the examples of mistakes and stuff most people experience are because they are using a very limited version of their Ais.

The OP question is "how far are we from making a game like EQ in AI" like the examples.

Right now, right now someone could do that.

Someone already did with unity (https://www.youtube.com/watch?v=Ny8F8va9Zi8), without AI.

So with AI, right now. Someone can make an everquest game with AI.

We just need someone to want to do it!

Why?

Also, no, someone didn't already do it in unity, that's not very real. It looks like they just spat some zone files into something and called it a day, it's single player, the site doesn't appear to exist anymore, inevitably they ran into the thing that most people run into when trying to make something: it's hard.

It's also a bizarre take to be like "we just need to make power cheaper!" oh word? Is that all? Ok cool gl bro.

[loramin]

The OP question is "how far are we from making a game like EQ in AI" like the examples.

Right now, right now someone could do that.

Someone already did with unity (https://www.youtube.com/watch?v=Ny8F8va9Zi8), without AI.

AI absolutely could help a seasoned programmer do work on a project like that ... but you are quite wrong if you think you can just ask Claude to port the EQ client to Unity and get that video. I will bet my life savings on this fact, I'm that certain of it.

Also, even if AI could make complex tweaks to the existing client ... that's still modifying existing software, which is different from making new software.

Plus ... we haven't even gotten to the server (which AI also can't make).

I'm not trying to be condescending, but when you keep making wildly false claims while refusing to educate yourself about reality, it's hard not to be.

[BradZax]

but you are quite wrong if you think you can just ask Claude to port the EQ client to Unity and get that video. I will bet my life savings on this fact, I'm that certain of it.

Keep your life savings you retard, not one single person in this thread even wondered if that was possible.

The question was: Look someone built doom like this.

Can someone build eq like this?

Not, can someone wish EQ into existence because someone built Doom using AI?

[BradZax]

It's also a bizarre take to be like "we just need to make power cheaper!" oh word? Is that all? Ok cool gl bro.

Thats not my take but if you think it is then that would explain why you're confused.

In response to someone saying that AI is not good at managing multiple tasks because it's not great with recursive memory. I mentioned that Power consumption is the reason.

I did not say with more power alexa could build whatever your heart could imagine.

I said recursive memory is EXPENSIVE right now, and soon it wont be because we're building private power plants.

Maybe try reading instead of misunderstanding and getting mad.

[Kich867]

Thats not my take but if you think it is then that would explain why you're confused.

Maybe try reading instead of misunderstanding and getting mad.

ok cool gl bro

[BradZax]

ok cool gl bro

thanks bro gl.

[DeathsSilkyMist]

The question was: Look someone built doom like this.

Can someone build eq like this?

Not, can someone wish EQ into existence because someone built Doom using AI?

With current AI tools (and the upgrades to said tools coming in the near future), you cannot build Everquest using AI to do the majority of the work. The techniques in the video OP posted would be woefully insufficient.

Can a team of developers build Everquest with the assistance of AI? Sure. The majority of the work would still be done by human programmers and artists.

This is not going to change in the next few years, unless we get a new breakthrough in AI.

[BradZax]

With current AI tools (and the upgrades to said tools coming in the near future), you cannot build Everquest using AI to do the majority of the work. The techniques in the video OP posted would be woefully insufficient.

But to answer OPs question:

When can we make EQ with AI?

Right now. The answer is right now.

Can a team of developers build Everquest with the assistance of AI? Sure. The majority of the work would still be done by human programmers and artists.

A team of developers could build it without AI in 1999.

One developer could build it without AI right now.

One person who doesnt know how to make games can make it with AI right now too.

Making it doesn't mean wishing it into existence.

It means making it. And using AI tools and youtube (LIKE THE GUY IN THE OP VIDEOS), they can figure it out.

This is not going to change in the next few years, unless we get a new breakthrough in AI.

I wouldn't expect much change in anything over a few years, but currently we're not hobbled by "breakthroughs in ai" right now, we're hobbled in "power consumption for AI"

Will more breakthroughs in ai make ai better? Yes.

Would more power right now (without breakthroughs in ai) make ai better? YES.

Right now the tools you're using, the highest paied tools, offer you like 20% of what their tools are capable of.

Because costs.

This is not speculative. It's a fact. Let models think longer, branch more, run more internal passes, use bigger context windows, or spawn helper agents performance goes up.
Reasoning improves.
Fewer mistakes.
Better planning.
That’s pure compute.
Throttled & limited because of power consumption.

[DeathsSilkyMist]

Cool.

But to answer OPs question:

When can 1 developer make EQ with AI?

Right now. The answer is right now.

A team of developers could build it without AI.

One developer could build it without AI.

Can a single newbie game developer use AI to make Everquest? No.

Can a single veteran game developer make Everquest by themselves without AI? Yes.

My guess is OP is talking about AI because they are more interested in the first question "Can a single newbie game developer use AI to make Everquest?"

The answer is no. It will be no for at least the next few years, unless we get a new breakthrough in AI.

[BradZax]

Can a single newbie game developer use AI to make Everquest? No.

I would disagree.

Can a single newbie developer who is capable of learning new things and is smart, and has passion and drive, build EQ with AI?

YES. (They could do that right now as a hobby, without any use of AI. but AI is going to reduce about 90% of the time it would take to learn things)

Can a single retard with no passion and no previous success in any aspects of their lives, is a quitter, never finishes anything, never tried to learn a new skill before make EQ with AI?

No.

The answer is no. It will be no for at least the next few years, unless we get a new breakthrough in AI.

This is ironic, because I would argue that it would be mor elike 20 years before someone could build EQ the way YOU ARE ASKING. Which is not what OP asked.

Op gave 2 examples of how he expects the game to be made with AI, using traditional means aided with AI, and you're answering a different question.

You're answering the question: How long until Alexa can make EQ.

[DeathsSilkyMist]

Can a single newbie developer who is capable of learning new things and is smart, and has passion and drive, build EQ with AI?

The answer to this question is no.The reason for this is AI is not perfect at debugging it's own code. A newbie would not be able to debug the code either, because they are not experienced enough to understand the code AI is writing.

[Belambic]

It would take the newbie developer decades to make an EQ style game with AI.

[Kich867]

The question was: Look someone built doom like this.

Can someone build eq like this?

Not, can someone wish EQ into existence because someone built Doom using AI?

ok so I finally actually watched that vid. Lol, lmao, kek even.

That guy didn't build doom, not even close. Plus, doom is open source, which makes it surprising how hood that was considering how it works top to bottom is already inside AI.

And I'll be honest I don't see a difference between your two questions. They're both trying to wish it into existence because someone did it with AI and then adamantly pushing back after being told "no" by people who actually use this shit and vaguely know how it works.

But to answer your question directly, no not really. You could use it to assist you in developing an mmo in unity or whatever, if you wanted, but no, it couldn't actually build the whole game for you. You'd come into too many problems with its inability to be consistent.

The bottom line is that AI can never, ever be used for anything important because it isn't idempotent.

There would have to be a substantial jump in its capabilities in order to do so.

[BradZax]

Where in the video is the AI building the whole game?

It isnt because that isnt what OP asked.

He asked when can he build EQ using AI like the guy was using AI to learn how to build doom (with 1 hour of work).

You could do that now.

[DeathsSilkyMist]

You're answering the question: How long until Alexa can make EQ.

By definition, a newbie game developer is incapable of building a game like Everquest, due to their lack of knowledge and skills. If building Everquest was so easy a single newbie could do it, there would be more MMO's on the market. AI would have to do most of the work in this scenario.

Can a newbie become a veteran? Of course. But then the question is no longer "Can a newbie with AI make Everquest?", because the person is no longer a newbie.

[Kich867]

The fact that he didnt buil ddoom is another argument that you guys are being retarded.

He didnt even give an example of someone asking AI to build a game, he was literally using AI to build a game after never having done it before.

Showing that anyone can learn how to build EQ with AI.

I mean its your argument, so, I'm not really sure where you are anymore. The question has been asked and answered.

It sounds like you want to do this, so, don't let us stop you. Go learn how to program with AI big dawg glgl. You don't need a bunch of strangers permissions on a forum for a 30 year old emulated game.

[BradZax]

The bottom line is that AI can never, ever be used for anything important because it isn't idempotent.

There would have to be a substantial jump in its capabilities in order to do so.

THIS is where we're arguing about power consumption.

Going back to the first time I brought it up, it was in response to an absolutist statement like this, that is based off people using AI that is hard capped at about 10-20% of its actual capability.

The reason we see so much investment into AI is because at 100% usage, you don't see the mistakes we all see, you see better agents, you see worker replacement.

I'm just talking about the EXCITING and COOL technological advancements that are soon to come to blow our minds, because recently companies like META have been given the permission to build and operate their own power.

I mean its your argument, so, I'm not really sure where you are anymore. The question has been asked and answered.

It sounds like you want to do this, so, don't let us stop you. Go learn how to program with AI big dawg glgl. You don't need a bunch of strangers permissions on a forum for a 30 year old emulated game.

I didn't comment if EQ would be build-able by just asking AI to make it.

I don't know if that will ever happen.

Can someone make a film like jurrasic park in their bedroom with AI video tools right now? I actually don't think so.

Could someone make everquest in their bedroom right now with AI tools? Yes absolutly.

[DeathsSilkyMist]

Could someone make everquest in their bedroom right now with AI tools? Yes absolutly.

You aren't providing the proper context when it comes to what the AI is doing to help create Everquest. You are keeping it ambiguous.

As I said in my previous post:


By definition, a newbie game developer is incapable of building a game like Everquest, due to their lack of knowledge and skills. If building Everquest was so easy a single newbie could do it, there would be more MMO's on the market. AI would have to do most of the work in this scenario.

Can a newbie become a veteran? Of course. But then the question is no longer "Can a newbie with AI make Everquest?", because the person is no longer a newbie.

[BradZax]

By definition, a newbie game developer is incapable of building a game like Everquest, due to their lack of knowledge and skills. If building Everquest was so easy a single newbie could do it, there would be more MMO's on the market. AI would have to do most of the work in this scenario.

Can a newbie become a veteran? Of course. But then the question is no longer "Can a newbie with AI make Everquest?", because the person is no longer a newbie.

Yes. A newbie, with no experience, can watch videos on youtube, ask questions and code with AI, to build everquest from the ground up right now.

[DeathsSilkyMist]

Yes. A newbie, with no experience, can watch videos on youtube, ask questions and code with AI, to build everquest from the ground up right now.

Of course. Nobody said people are incapable of learning how to make video games. But in this scenario the human is learning how to become a veteran game developer. That is what allows them to build Everquest. You don't need AI to learn how to make video games.

It would take years for a determined newbie with current AI tools to properly make Everquest. Most of that time would be spent learning how to program, after realizing vibe coding with AI isn't producing the results they want. Current AI tools would not allow a total newbie to make Everquest in a few months.

[BradZax]

Of course. Nobody said people are incapable of learning how to make video games. But in this scenario the human is learning how to become a veteran game developer. That is what allows them to build Everquest. You don't need AI to learn how to make video games.

Ai fits into this scenario because in the videos OP posted, someone with ZERO experience making games, managed to build a functioning doom executable.

Ai helped him write code he didnt know how to write. Ai helped tell him where to put stuff he didn't know where to put it. Ai tells you create accounts, and register to download all the programs you need to build something.

I am not sure where AI comes into this scenario.

This is where AI comes into the equation:

You could sit down with Chat GPT right now and say:

Hi, Im a total newbie, and have never made a game before. i want to make everquest, like from 1999!

Can you help me, step by step, starting from TOTAL scratch, build everquest? Create an outline that we will follow, step by step to build a functional everquest clone.

I will need help to learn how to do every step.

And as you work, through each step, asking AI what to do, and how to do it. You will, by the end of a year or 2, have a fully functional everquest style game, as a newbie (presuming you were doing this in your downtime).

[DeathsSilkyMist]

This is where AI comes into the equation:

You could sit down with Chat GPT right now and say:

Hi, Im a total newbie, and have never made a game before. i want to make everquest, like from 1999!

Can you help me, step by step, starting from TOTAL scratch, build everquest? Create an outline that we will follow, step by step to build a functional everquest clone.

I will need help to learn how to do every step.

And as you work, through each step, asking AI what to do, and how to do it. You will, by the end of a year or 2, have a fully functional everquest style game, as a newbie (presuming you were doing this in your downtime).

Using the methods you describe, the only scenario in which that might happen is if the newbie or the AI just takes the entire EQEMU codebase as a starting point.

But then it wasn't AI that did the work. It was the newbie/AI taking someone else's work.

If we are talking about the scenario in which a newbie uses AI to build Everquest from scratch, it would take many years with the current AI tools. If building MMO's was that easy, we'd have a lot more indie MMO's on the market.

[BradZax]

The only scenario in which that might happen is if the newbie or the AI just takes the entire EQEMU codebase as a starting point.

But then it wasn't AI that did the work. It was the newbie/AI taking someone else's work.

If we are talking about the scenario in which a newbie uses AI to build Everquest from scratch, it would take many years with the current AI tools. If building MMO's was that easy, we'd have a lot more indie MMO's on the market.

You could like download a tutorial on youtube (https://www.youtube.com/watch?v=tCclPNYcJ_w&list=PLOZ8jGta4Ld-0jOEtFjDQtvwNsSPpasoU) that if you follow step by step make your own clone of eq in like 100 hours without even using AI.

Most of the stuff is hard to do, but with AI anyone could follow the tutorial, because you can ask questions, it can tell you how to do stuff you don't understand bla bla bla, its awesome.

[DeathsSilkyMist]

No i disagree, you're like hungup too specifically on weird everquest minutia.

If you wanted to make "your own" mmo, like everquest, like the OP asked.

You could like download a tutorial on youtube from 12 years ago, that if you follow step by step make your own clone of eq in like 100 hours.

Most of the stuff is hard to do, but with AI anyone could follow the tutorial, because you can ask questions, it can tell you how to do stuff you don't understand bla bla bla, its awesome.

If there are so many amazing youtube tutorials from 12 years ago that can hold your hand enough to make a MMO on the scale and quality of Everquest in 100 hours, where are all of the indie MMO's over the past decade?

To me it just sounds like you are vastly underestimating the complexity of video games, while vastly overestimating an AI's ability to teach someone.

[OriginalContentGuy]

You could like download a tutorial on youtube (https://www.youtube.com/watch?v=tCclPNYcJ_w&list=PLOZ8jGta4Ld-0jOEtFjDQtvwNsSPpasoU) that if you follow step by step make your own clone of eq in like 100 hours without even using AI.

Most of the stuff is hard to do, but with AI anyone could follow the tutorial, because you can ask questions, it can tell you how to do stuff you don't understand bla bla bla, its awesome.

LoL do it then. Go on

[BradZax]

If there are so many amazing youtube tutorials from 12 years ago that can hold your hand enough to make a MMO on the scale and quality of Everquest in 100 hours, where are all of the indie MMO's over the past decade?

To me it just sounds like you are vastly underestimating the complexity of video games, while vastly overestimating an AI's ability to teach someone.

https://store.steampowered.com/app/2006120/GameZero/

https://store.steampowered.com/app/2176920/Flempire/

https://store.steampowered.com/app/1456420/Manikineko_Online/

https://store.steampowered.com/app/2890220/The_Universe_of_Red_Hope/

These are MMOs that have been released/early access/in development by 1 person.

Also, that guy was making VR request. There’s the Unity request project. There’s like 1 million fucking crazy MMO projects on YouTube from Rando’s.

Again, I’m not saying that any retard can do this. I’m saying that someone who’s smart and capable and passionate and driven and really wants to can, even if they know nothing about computer programming today

[DeathsSilkyMist]

https://store.steampowered.com/app/2006120/GameZero/

https://store.steampowered.com/app/2176920/Flempire/

https://store.steampowered.com/app/1456420/Manikineko_Online/

https://store.steampowered.com/app/2890220/The_Universe_of_Red_Hope/

These are MMOs that have been released/early access/in development by 1 person.

I don't get your point. We both agreed individual developers can create video games, even MMOs, without AI.

That has no bearing on the subject of AI. None of these games prove the idea that AI was a significant factor in their creation, whether it be code generation or teaching.

[BradZax]

LoL do it then. Go on

I’m smart and passionate. I just don’t have the drive to.

I don't get your point. We both agreed individual developers can create video games, even MMOs, without AI.

My point would be that you don’t disagree with me and yet here you are seven pages later still arguing with me.

That has no bearing on the subject of AI. None of these games prove the idea that AI was a significant factor in their creation, whether it be code generation or teaching.

You asked where are the indi mmos made with YouTube tutorials, I gave them to you.

[DeathsSilkyMist]

My point would be that you don’t disagree with me and yet here you are seven pages later still arguing with me.


Let's go back to OP's question:

https://www.youtube.com/watch?v=MWl2qQRyutc

Watched this video, where the guy basically recreates Doom in an hour.

Keep in mind, Doom is a single-player game, and came out in 1993.

But... the video is 3 months old and was probably recorded some time before that.

So the question is, how far are we from making a game like P99 in AI?

The question OP asked is: How far away are we from AI being a significant developement partner when making a complex game like Everquest? The context comes from OP's amazement at how far the youtuber got in one hour using AI.

The answer is we are years away from AI being a significant development partner in the creation of complex video games like Everquest. AI will be coding assistants watched over by experienced programmers at best in the next few years, unless we get a new breakthrough in AI.

AI will not be able to level up newbies to the point where they can make a game at the quality of Everquest in 100 hours.


You asked where are the indi mmos made with YouTube tutorials, I gave them to you.

To be honest, those four games look terrible. Three of them are not released yet, so we don't know if they will collapse before they release, or collapse soon after release.

The one game that is released has 1 review. These are not good examples.

[Kich867]

My point would be that you don’t disagree with me and yet here you are seven pages later still arguing with me.

I'll be real dawg, its because you're a layman and you aren't being specific enough with what you're talking about, so we have to draw inferences on what we think you're trying to ask, to which you repeatedly just said "that's not what I mean" to an extent that I can barely wrap my head around what you're even doing here.

You made up your mind zero seconds in and are engaging with social media at this point just to talk to people. Which is fine sort of, just don't get all haughty over it.

If your point was "can someone with no programming experience talk to AI a bunch to learn how to do this?" I'd say probably, yeah. The debatable part beyond that would be scrutinizing the project and dealing with the hard parts of game development, especially multi-player persistent world.

It would be really challenging, particularly due to the lack of idempotency--which basically means it doesn't produce a consistent result. If you ask it A, it does not always respond with B. Because fundamentally that's just not really how they work, right.

And so the biggest hurdles will be anytime you need to revisit things, or append things, or expand on a thing you've already built, it's not necessarily going to "know" that and give you back what it thinks you want and how that not really being specifically what you're trying to do, and then trying to figure out how to get it to do what you actually wanted.

Honestly what I would expect to see happen, what I think is far more likely, is that AI would get someone like that far enough through the project to have that person begin to understand what it is they need to actually do and then rely less and less on the AI as they just simply aren't that helpful.

But yeah, at the end of the day, building an MMO isn't a secret, it's a pretty well known thing how to do it's just a huge pain in the ass. So like if you asked AI about it it should be able to tell you these concepts and what is generally considered best practice (although even that is suspect, remember, AI can't think, it isn't intelligent, it doesn't really understand things, so it's just going to say whatever it sees enough people saying is a best practice).

[BradZax]

You are engaging with social media at this point just to talk to people.

This is an internet forum.

But yeah, at the end of the day, building an MMO isn't a secret, it's a pretty well known thing how to do it's just a huge pain in the ass. So like if you asked AI about it it should be able to tell you these concepts and what is generally considered best practice.

Yep! And it will hold your hand as you accomplish each step with zero prior experience!

So now, it's possible now.

Honestly what I would expect to see happen, what I think is far more likely, is that AI would get someone like that far enough through the project to have that person begin to understand what it is they need to actually do.

Yes!

[DeathsSilkyMist]

This is an internet forum.

Yep! And it will hold your hand as you accomplish each step with zero prior experience!

So now, it's possible now.

Something being possible does not mean it is probable.

Is it possible for monkeys banging on typewriters to generate Shakespeare? Yes. Is it probable? No.

The games examples you provided all look terrible, and are a farcry from the quality and scope of Everquest.

If your point is simply that AI can generate slop that nobody wants to play, then I agree with you.

[loramin]

It's possible because Brad declares it to be so, based on no programming background and (poor) inferences from a few YouTube videos.

Nevermind that reality disagrees ... Brad has declared it to be true!

Do I have the substance of your argument more or less correct?

[BradZax]

Something being possible does not mean it is probable.

Is it possible for monkeys banging on typewriters to generate Shakespeare? Yes. Is it probable? No.

Yeah but these like, philisophical arguments though compelling, are not quite based on actual contextual data of today.

Like I could say, its possible the world is flat because I haven't seen the world, and that would not be probable because of the amount of other people who have seen or navigated the world.

Just like learning how to make 1990s video games there are many examples of indi projects and developers doing it successfully, countless hobby projects, open source unity versions of everquest, downloadable VR mods, emulation servers etc, that it would indicate that what I am saying, IS more probable than not.

And the bet isn't: "what are the odds that someone does this in 3 years?" it's, "how long until someone CAN do it."

The answer is right now.

[DeathsSilkyMist]

Yeah but these like, philisophical arguments though compelling, are not quite based on actual contextual data of today.

Like I could say, its possible the world is flat because I haven't seen the world, and that would not be probable because of the amount of other people who have seen or navigated the world.

Just like learning how to make 1990s video games there are many examples of indi projects and developers doing it successfully, countless hobby projects, open source unity versions of everquest, downloadable VR mods, emulation servers etc, that it would indicate that what I am saying, IS more probable than not.

And the bet isn't: "what are the odds that someone does this in 3 years?" it's, "how long until someone CAN do it."

We keep going in circles. Let me ask you directly.

If a single newbie developer were to create Everquest from scratch using the current AI tools, what percentage of the work would be done by AI, and what percentage of the work would be done by the developer?

Is it 80% AI and 20% 1 newbie developer?
Is it 50% AI and 50% 1 newbie developer?
Some other percentage?

[BradZax]

We keep going in circles. Let me ask you directly.

If a single newbie developer were to create Everquest from scratch using the current AI tools, what percentage of the work would be done by AI, and what percentage of the work would be done by the developer?

Is it 80% AI and 20% 1 newbie developer?
Is it 50% AI and 50% 1 newbie developer?
Some other percentage?

That's a tough question to answer because it's like, "If AI taught you how to write and mail a letter, how much of the work is done by AI and how much of it is done by you?"

Idk, that is a weird question. So i asked chat gpt your question and it said:

The honest answer is: AI can probably do 40–70% of the raw production output, but you (the developer) will still do 60–30% of the total work, because the “total work” is mostly decision-making, iteration, debugging, and taste.

For further clarification I asked, "of the work I have to do, how much of it can you teach me how to do, step by step?"

It said:

All of it.

[DeathsSilkyMist]

According to chat GPT (because there is no way I would give you an accurate answer to such a strange question, because like we're talking, "if AI taught you how to write and mail a letter, how much of the work is done by AI and how much of it is done by you?"

Idk, that is a weird question. But i asked chat gpt and it said:

Chat GPT is wrong. There is no scenario in which current AI tools could even do 30% of the work on an Everquest-style MMO. If you did that, the game would be buggy and easy to cheat in. It wouldn't be a game people would want to play. This is not going to change in the next few years, unless we get a new breakthrough in AI.

[BradZax]

Chat GPT is wrong. There is no scenario in which current AI tools could even do 30% of the work on an Everquest-style MMO. If you did that, the game would be buggy and easy to cheat in. It wouldn't he a game people would want to play. This is not going to change in the next few years, unless we get a new breakthrough in AI.

I don't think you can make that claim.

[DeathsSilkyMist]

I don't think you can make that claim.

I can. Secure networking code is not easy code to write. AI code generators often make mistakes, and are not always able to figure out that they made a mistake. There are plenty of occasions where AI has declared it was correct, while the code it generated was broken.

You can test this yourself with AI code generators. You don't need to take my word for it.

The issue is an inexperienced programmer will not he able to debug a lot of code generated by AI, because they are too inexperienced to understand the code being written. So you are at the mercy of the AI claiming it is correct, even when it isn't.

[BradZax]

Ok heres the answer to op:

Right now anyone (except for DSM or Loramin) can.

[DeathsSilkyMist]

Ok heres the answer to op:

Right now anyone (except for DSM or Loramin) can.

You didn't include yourself in that list:)


I’m smart and passionate. I just don’t have the drive to.


I appreciate your enthusiasm for AI. But the reality is AI just isn't advanced enough yet. You should try to play around with something like Claude Code, so you can see for yourself that the AI makes mistakes when it comes to code generation.

You still haven't addressed the issue where an inexperienced programmer is unable to debug code generated by AI.

[BradZax]

I could do it because I'm a mexiCAN not a mexiCANT.

You still haven't addressed the issue where an inexperienced programmer is unable to debug code generated by AI.

You copy and paste the code and tell the LLM you're working with: "it didnt work"

and it tries again.

And you say "it didnt work again is there a forum I can post on to ask for help?"

And it tells you where and how to ask for help.

[Danth]

If we limit ourselves to a human directing some sort of AI-driven code generator, if it can produce a Wolf3D or DooM-type game at present, we'll see an EQ-style game within ten years, possibly before this decade is out,. That's especially likely if you allow for an experienced human, not just a total programming newbie. MMOG's are complex, but broken down into their component parts--host, client, network code, database, graphic engine, textures, etc--it's manageable. You're just replacing the junior coders with automation. We'll get there.

If you want to simply provide "OmniAI" a prompt of "produce me a circa 1999-style multiplayer virtual world with a medieval fantasy setting" then it spits out a fully completed game a couple hours later, I think we're farther away for the reasons discussed. AI's an oxymoron term, it has no genuine intelligence, and as such it can handle math or produce functional code--known things with a logical basis--has no concept of things like enjoyment or fun. Your AI-generated game would be, to use the current buzzword, so much "AI slop," perhaps functional but likely soulless and unenjoyable.

Same as automation in a lot of other industrial fields, really: The more reptitive and lower-skill positions end up replaced by machines, but you still need skilled operators keeping the overall system functional.

Thinking about it, if AI gets to the point where it can also replicate the "human factor," I don't think a 1990's style video game will matter because it'll by extention be able to do so much more.

[DeathsSilkyMist]

I could do it because I'm a mexiCAN not a mexiCANT.

I look forward to playing the Everquest-like MMO you are going to make in the next 100 hours with AI. Put a link in this thread to the steam page when it's ready.


You copy and paste the code and tell the LLM you're working with: "it didnt work"

and it tries again.

And you say "it didnt work again is there a forum I can post on to ask for help?"

And it tells you where and how to ask for help.

In this scenario most of the code ends up getting written and debug by humans across various forums. So again, you aren't even going to get 30% of the game coded by AI. This also assumes people will answer your questions every time, and answer them correctly. Forum posts asking for coding help are not always fruitful.

[loramin]

I don't think you can make that claim.

No one but Brad gets to claim anything, but everything he claims is true: reality be damned!

[BradZax]

No one but Brad gets to claim anything, but everything he claims is true: reality be damned!

What did I claim?

[loramin]

That modern AI could make EQ (with more power).

[Goregasmic]

You guys are getting trolled hard.

[BradZax]

That modern AI could make EQ (with more power).

Never claimed that.

[BradZax]

You guys are getting trolled hard.

I know these two retards enough to know that this is serious to them, and it really matters.

And it does to me too.

[OriginalContentGuy]

Ok heres the answer to op:

Right now anyone (except for DSM or Loramin) can.

Bet your life.

[BradZax]

No one but Brad gets to claim anything, but everything he claims is true: reality be damned!

Im going to use this moment to be teachable.

You know how someone made the loramin is an asshole thread?

Here's the thing, this is what you do:

I am making a general claim about technology.

Your claim is that I am wrong, and uneducated.

You offer no counter point, no shared progress-able goal. No interesting shared information.

Just a new topic: That I am an idiot, and should shut up.

That is why you are an asshole, Laura.

[DeathsSilkyMist]

You guys are getting trolled hard.

Probably. It has crossed my mind.

For BradZax and anyone else interested in AI coding, go and try something like Claude Code. You will see for yourself that it makes mistakes. BradZax's suggestion:


You copy and paste the code and tell the LLM you're working with: "it didnt work"

and it tries again.

And you say "it didnt work again is there a forum I can post on to ask for help?"

And it tells you where and how to ask for help.

Admits that he will outsource his problems to other humans, rather than the AI. So I am not sure why he believes ChatGPT when it says AI could do 40% of the production work in an Everquest-style MMO.

Basically the pipeline seems to be:

1. Generate AI code.
2. Get stuck debugging it since you do not understand it.
3. Ask other humans for help and hope they answer. Many programming forums are not keen to simply write your program for you.
4. Repeat until finished with the game.

I don't see this pipeline working too well.

[BradZax]

For BradZax and anyone else interested in AI coding, go and try something like Claude Code. You will see for yourself that it makes mistakes. BradZax's suggestion:.

Your ASSumption that only you has done this is what makes you a donkey.

[OriginalContentGuy]

_4s1pPI12FQ
Sure thing exlax.

[BradZax]

Is that what reading this thread feels like for non dsm loramin or brad posters?

[DeathsSilkyMist]

Your ASSumption that only you has done this is what makes you a donkey.

This is literally how everything is currently made you donkey.

You think people just born out of the womb programing video games?

They learn how to do it.

Youtube could teach anyone how to do it.

Ai makes it vastly easier.

IDK how to tell you this but you put your pants on one leg at a time.

At no point have I argued that ai will do magic. No matter how many times you and this Laura chick keep trying to say AI cant.

But I did say that in 1 to 2 years you could have a working game like everquest with the help of AI and youtube.

You're a donkey.

We have both agreed multiple times in this thread that people can learn how to make video games via youtube, chatGPT, etc.

I don't know why we keep going in circles on that point. There is no reason to act as if we disagree.

I disagree with the idea that AI will significantly speed up the production of an Everquest-style MMO, whether that is generating training material for the programmer, code generation, art generation, etc.

Right now AI makes too many mistakes, and lacks the consistency to accurately produce thousands of lines of code, hundreds of textures that match the art style, 3d models that are game ready, etc.

This is why you need experienced developers watching the AI output at all times. There are times where the amount of work it takes to fix what the AI did is longer than simply doing it yourself.

[BradZax]

We have both agreed multiple times in this thread that people can learn how to make video games via youtube, chatGPT, etc.

I don't know why we keep going in circles on that point. There is no reason to act as if we disagree.

Because you wont say "oh, i thought you meant something else, my bad!"

[DeathsSilkyMist]

Because you wont say "oh, i thought you meant something else, my bad!"

You are making other claims besides the single claim: "People can learn to make games with youtube, chatGPT, etc.". That is why the conversation hasn't ended. For example:


But I did say that in 1 to 2 years you could have a working game like everquest with the help of AI and youtube


I don't understand how you came to this conclusion.

[BradZax]

Yeah but i wouldnt have a problem with you disagreeing with me about those claims, but I would appreciate you met me halfway after I go through the trouble of clarifying.

I don't understand how you came to this conclusion.

Well I am assuming this is a late night kids are asleep hobby project. And I also am imagining building a smaller but comparable version of ever quest, you could have that working in a year or 2 with trial, error, learning how to use AI, what kind of AI tools are available etc, and building—and doing all with AI being your CORE educator and co-worker. (would it be high quality? idk depends on how cool the person making it is).

I am not arguing that you will ever be able to say, "alexa, make me everquest" and it would just do it. (though I am personally confident that is an inevitable future, just dont know how long it will take at all, could be 10 years, could be 10,000 because we nuke ourselves back to the stone age).

[loramin]

Never claimed that.

Page 1

I think we're just waiting on more power. For example. Google image creation with Gemini using context costs about 3-5$ in wattage to generate a 6-10 second video. Vs their diffusion model for video generation is more like 5 cents or less.

That doesn't mean anything except it just gives you an idea about how much wattage context adds to the equation!

I am hoping corporations using their own nuclear reactors will be give us a big jump in usage.

https://www.latitudemedia.com/news/open-circuit-inside-metas-massive-nuclear-push/

Page 5:


The OP question is "how far are we from making a game like EQ in AI" like the examples.

Right now, right now someone could do that.

Someone already did with unity (https://www.youtube.com/watch?v=Ny8F8va9Zi8), without AI.

So with AI, right now. Someone can make an everquest game with AI.

We just need someone to want to do it!

Last page:
Ok heres the answer to op:

Right now anyone (except for DSM or Loramin) can.

You claim it a bunch more throughout the thread, but I have better things to do than quote troll posts back at trolls.

[DeathsSilkyMist]

Yeah but i wouldnt have a problem with you disagreeing with me about those claims, but I would appreciate you met me halfway and showed me that in the area we agree that you don't also disagree.


We both agree that people can learn to make video games via ChatGPT, youtube, etc. I apologize if you thought I was saying something else. Is there anything else you think we should meet halfway on?


You keep doubling down that im wrong about someone apt being able to build EQ right now by themselves with AI with no prior experience.


When you say things like this:

Ai fits into this scenario because in the videos OP posted, someone with ZERO experience making games, managed to build a functioning doom executable.

Ai helped him write code he didnt know how to write. Ai helped tell him where to put stuff he didn't know where to put it. Ai tells you create accounts, and register to download all the programs you need to build something.


It sounds like you are saying a newbie can use AI to generate a significant portion of the code needed to create an Everquest-style MMO. Maybe that is not your intent, but that is how I read it. I disagree with this idea.


Well for one I am assuming this is a late night kids are asleep hobby project. And I also am imaginging building a smaller but comperable version of everquest, you could have that working in a year or 2 with trial, error, learning, and building, all with AI being your CORE educator and co-worker.

I and AI would suggest, hiring people on fiver, reaching out to reddit communities, etc to accomplish this goal, because that would all be a part of any hobby project like this.

I am not arguing that you will ever be able to say, "alexa, make me everquest" and it would just do it. (though I am personally confident that is an inevitable future, just dont know how long it will take at all, could be 10 years, could be 10,000 because we nuke ourselves back to the stone age).

It would take years for an experienced programmer to solo write an Everquest-Style MMO from scratch if they only worked on it like 2-3 hours a day as a hobby project, even with AI and some part time assistants from Reddit or where-ever else.

Now I don't know what you mean by a "smaller version" of Everquest, so maybe you can elaborate. If you just mean less zones, that doesn't reduce the programming workload, as all of the game mechanics would still need to be in place. If you cut game mechanics, there is a point at which you can say it isn't Everquest anymore.

[BradZax]

By not quoting the post that I was replying to you are misrepresenting what I am saying

Page 1

Quote I was replying to:

First, the context window. The AI can only "remember" so much at a time, out of everything it reads, everything you tell it etc. The context limits are inherent to how AI works: until there's some new innovation occurs, this isn't going to get much better.

I was saying more power would solve context limits you claimed were inherent to how AI works, not some new innovation

NOT that AI will make EQ with more power.


Page 5:

Where does that quote even imply that AI can just make Everquest by itself, or that more power would make AI make evertquest by itself, that quote literally doesn't even support your argument, it breaks down and clarifies why someone can use AI RIGHT NOW to make Everquest if they wanted.


Last page:

Yes: Right now, me and DSM agree, someone can use AI to make their own Everquest game. I stand by that.

That is NOT what you are accusing me of: saying that with moree power AI could generate EQ on its own.

You claim it a bunch more throughout the thread, but I have better things to do than quote troll posts back at trolls.

I never once said what you are accusing me of saying, and you couldn't find one example of me saying it.

[BradZax]

It sounds like you are saying a newbie can use AI to generate a significant portion of the code needed to create an Everquest-style MMO. Maybe that is not your intent, but that is how I read it. I disagree with this idea.


Here's what I am saying about what is possible right now:

A newbie can sit down infront of a FREE version of chat GPT and say:

"Ok here we go, I want to make everquest like the 1999 version, i need you to set me up with a walkthrough and teach me how to build this every step of the way, starting from scratch, with nothing but windows 11.

IM ready to go, i have no idea, what is the first step, lets get going!"

It will first tell you to subscribe to the pro version so you can, and from there on it will walk you through everyone of the next steps.

And if you did that... and you were a SMART and CAPABLE person.

And you were dedicated to building your own everquest in your free time from scratch.

That you would be able to accomplish that, right now.

Will you be essentially, "learning how to do it yourself?"

Yes.

[OriginalContentGuy]

I never once said what you are accusing me of saying, and you couldn't find one example of me saying it.

_4s1pPI12FQ

[DeathsSilkyMist]

Here's what I am saying about what is possible right now:

A newbie can sit down infront of a FREE version of chat GPT and say:

"Ok here we go, I want to make everquest like the 1999 version, i need you to set me up with a walkthrough and teach me how to build this every step of the way, starting from scratch, with nothing but windows 11.

IM ready to go, i have no idea, what is the first step, lets get going!"

It will first tell you to subscribe to the pro version so you can, and from there on it will walk you through everyone of the next steps.

And if you did that... and you were a SMART and CAPABLE person.

And you were dedicated to building your own everquest in your free time from scratch.

That you would be able to accomplish that, right now.

Will you be essentially, "learning how to do it yourself?"

Yes.

I agree. I think where my disagreement came from is some of the timelines you have previously posted.

In this specific scenario, it would probably take 5-10 years for a newbie solo developer to make an Everquest-style MMO from scratch in their free time using AI and paying people for some part time work. This assumes the newbie is a very good learner, is dedicated to the cause, doesn't take long breaks, etc. This also assumes the scope and quality of the game is similar to Original Everquest without expansions.

[BradZax]

excellent post

:o

In this specific scenario, it would probably take 5-10 years for a newbie solo developer to make an Everquest-style MMO from scratch in their free time using AI and paying people for some part time work. This assumes the newbie is a very good learner, is dedicated to the cause, doesn't take long breaks, etc. This also assumes the scope and quality of the game is similar to Original Everquest without expansions.

Im comfortable with 5 years, and you get the kids on weekends.

[DeathsSilkyMist]

Im comfortable with 5 years, and you get the kids on weekends.

Lol sounds good!

My biggest worry with a newbie creating a multiplayer game is the networking code. Multiplayer games often die due to rampant cheating. Even AAA games have had bad cheating issues due to code created by seasoned programmers. AAA games at least have the money and time to mass ban cheaters, but indie studios don't always have that luxury.

So under the assumption that they do get it done in 5 years, they could still shoot themselves in the foot. That is why the time range I gave was large. Ideally you would spend a lot of time testing and refining after the game is largely done. That would take a long time as a single dev without a full time QA team. You could try and do early access, but this assumes the players would actually tell you about the cheats. They may keep it to themselves so they have an advantage when the game comes out.

[OriginalContentGuy]

This is why Red server has no players frankly.

[BradZax]

That, and most of them are all in jail IRL.

[BradZax]

Lol sounds good!

My biggest worry with a newbie creating a multiplayer game is the networking code. Multiplayer games often die due to rampant cheating. Even AAA games have had bad cheating issues due to code created by seasoned programmers. AAA games at least have the money and time to mass ban cheaters, but indie studios don't always have that luxury.

So under the assumption that they do get it done in 5 years, they could still shoot themselves in the foot. That is why the time range I gave was large. Ideally you would spend a lot of time testing and refining after the game is largely done. That would take a long time as a single dev without a full time QA team. You could try and do early access, but this assumes the players would actually tell you about the cheats. They may keep it to themselves so they have an advantage when the game comes out.

Yeah, I'm sure that security would be a lot for someone to really understand, good or bad security really depends on the knowledge of the person managing something I reckon.

I wonder how long until AI can actually act as cheat security in games as a totally different discussion.

Like i wonder how long until the first AI anti cheat feature is added to a major video game.

who knows, 3 years from now you might be able to purchase that for your AI everquest game right off the shelf! :eek:

[Danth]

My biggest worry with a newbie creating a multiplayer game is the networking code. Multiplayer games often die due to rampant cheating. Even AAA games have had bad cheating issues due to code created by seasoned programmers. AAA games at least have the money and time to mass ban cheaters, but indie studios don't always have that luxury.

EQ is case-in-point here because it trusts the client with *way* too much and has always been hugely vulnerable to various exploits and cheats. As you said, EQ had the financial backing behind it to hire a fairly large GM staff and work on bans/etc directly...and even then the effort was always imperfect. But then in order for your game to have rampant cheating issues, you must first have a game, which satisfies the basic requirement of the thread.

Some games are already trying AI-assisted anti-cheat or login verification. In the cases I'm personally aware of, it works only poorly and is almost universally loathed.

[OriginalContentGuy]

That, and most of them are all in jail IRL.
Yeah but with you that could mean they're crashing on your couch.

Prisons are for burning good night.

[Goregasmic]

1. Generate AI code.
2. Get stuck debugging it since you do not understand it.
3. Ask other humans for help and hope they answer. Many programming forums are not keen to simply write your program for you.
4. Repeat until finished with the game.

I don't see this pipeline working too well.

Story time:

When I came back to EQ I revived my old live account and made a 3 box. I was annoyed at logging all 3 toons every time so asked a tech savvy friend if I could automate a couple mouse clicks and keystrokes. He told me to use autohotkey and asked chatgpt to make a template then sent it to me.

Nothing worked. After digging around I found out Chatgpt couldnt differentiate ahk1.0 from 2.0 which used a different syntax. I had to rewrite all the code using the documentation, googling forum posts and trial and error. I'm not a coder but I'm good at troubleshooting stuff so it wasn't overly hard but still took me a couple hours to automate STUPID KEYSTROKES.

It seems pretty clear to me that anyone who ever coded for 5 minutes knows this process would take several lifetimes for a non coder to get a working p99 which probably has several hundred thousand codelines?

[BradZax]

Some games are already trying AI-assisted anti-cheat or login verification.

"Run in circles."

https://i.imgur.com/eRIZf5f.png

[loramin]

Here's what I am saying about what is possible right now:

A newbie can sit down infront of a FREE version of chat GPT and say:

"Ok here we go, I want to make everquest like the 1999 version, i need you to set me up with a walkthrough and teach me how to build this every step of the way, starting from scratch, with nothing but windows 11.

IM ready to go, i have no idea, what is the first step, lets get going!"

It will first tell you to subscribe to the pro version so you can, and from there on it will walk you through everyone of the next steps.

And if you did that... and you were a SMART and CAPABLE person.

And you were dedicated to building your own everquest in your free time from scratch.

That you would be able to accomplish that, right now.

Will you be essentially, "learning how to do it yourself?"

Yes.

Still not possible. It's mind-boggling that you just keep doubling down on the fantasies in your head ... instead of just engaging with reality and trying to do what you claim is possible.

Also, the idea that ChatGPT could do it is laughable. Claude and Gemini can't either, but they'll get you a whole lot closer than ChatGPT will.

[BradZax]

"Run in circles."

https://i.imgur.com/eRIZf5f.png

"You have 10 seconds to comply."

[Belambic]

Story time:

Nothing worked. After digging around I found out Chatgpt couldnt differentiate ahk1.0 from 2.0 which used a different syntax. I had to rewrite all the code using the documentation, googling forum posts and trial and error.

Asking an AI to make something like EQ would end up with COBOL code, not C++, or any other high level language. It might eventually look and feel like EQ, but it would probably run slow as hell, have numerous glaring exploits exposed, not be able to align textures correctly, everyone would have 4 arms and 15 fingers.

[Baugi]

So the question is, how far are we from making a game like P99 in AI?

If we accept the dubious premise that LLMs can spit out a doom-sized game now with minimal intervention(human intervention cost probably scales more like exponential with project size) and oversimplify a lot -

Compute cost as a function of code-base size scales between O(n^2) and O(n), with traditional attention vs. various linearity hacks/Chinchilla

Doom was ~50k LOC

Everquest through Velious probably had between 500k and 1m LOC (EQMacEMU's server has ~2m for example)

Compute scales at something like 2x per 2-3 years according to Gelsinger vs. Huang's updating of Moore's law.

So maybe log2(500k/50k)*2 to 2*log2(1m/50k)*3, or ~7 to 26 years for the compute power.

You will almost certainly run out of training data long before that. We're basically out now.

I hope that helps. I need to get back to debugging the COBOL Cursor keeps inserting into my secure networking code.

Edit: for an interesting albeit bad comparison, Anthropic claims their C-compiler experiment was basically autonomous(still required expert developer designed test-cases and a CI pipeline that would be impossible for a game) and ran in 2 weeks for $20k, producing 100k LOC. It sometimes works.

[Dulu]

You could sit down with Chat GPT right now and say:

Hi, Im a total newbie, and have never made a game before. i want to make everquest, like from 1999!

Can you help me, step by step, starting from TOTAL scratch, build everquest? Create an outline that we will follow, step by step to build a functional everquest clone.

I will need help to learn how to do every step.


I wasn't expecting this thread to take off like this, and I'm reading everything you guys wrote, but I haven't finished it all yet.

Here's what comes up when you type this into Gemini:

The "Project Norrath" Roadmap
Phase 1: The Foundation (The Engine)
Before you build a world, you need a place for it to live.

Step 1: Choose and install a Game Engine (I recommend Godot or Unity for beginners).

Step 2: Learn the basics of your engine's coding language (GDScript or C#).

Step 3: Create a "Player Controller" (making a character move and rotate).

Phase 2: The "Old School" Aesthetics
Step 4: Import/Create low-poly 3D models (that chunky, nostalgic 1999 look).

Step 5: Set up a "Third-Person Camera" that follows the player.

Step 6: Design a basic environment (a flat plane with a few trees or walls).

Phase 3: The Mechanics (The RPG Guts)
Step 7: Create a Stat System (Health, Mana, Strength, etc.).

Step 8: Build a basic "Targeting System" (clicking an NPC to select them).

Step 9: Implement "Auto-Attack" and basic spell casting.

Phase 4: The "Multiplayer" Magic
Step 10: Set up a basic server (making two players appear on the same screen).

Step 11: Syncing data (if Player A moves, Player B sees it).

Step 12: Database integration (saving your character so you don't lose progress).



Then it goes into deep breakdowns of each step.

[BradZax]

"ok lets start with step one, lets install godot and unity how do I do that idk what that is."

and just keep doing every step like that and in 2 years youll have eq :o

[loramin]

"ok lets start with step one, lets install godot and unity how do I do that idk what that is."

and just keep doing every step like that and in 2 years youll have eq :o

Nope.

Building software with AI is like building a house: it can absolutely tell you how to hammer nails in, install electrical outlets, and so on, because it's good at construction.

But what AI won't do is tell you not to hammer a nail into a wall with an electrical outlet behind it (or not to run that electrical line there in the first place) ... because you never specifically asked it about that (and neither you nor the AI is an architect).

Now, if you asked Claude to come up with a course on becoming a (software) architect, it would probably do a pretty good job by stealing from a lot of existing online courses. Follow it for a couple years and you might become a programmer, and then you could properly use it to build software ...

... although you likely won't be that much fater than Nilbog ... who started building just the server way back in 2008 ... not the client, and even the server isn't from scratch (he leveraged the work of other devs who built EQEmu) ... and he still hasn't finished in 2026.

But keep believing that you and ChatGPT are going to make the next EQ tomorrow without a software background :rolleyes:

[BradZax]

I wasn't talking to you, loserman.

[loramin]

I wasn't talking to you, loserman.

Perhaps you were unaware, but you're talking in a public forum.

[BradZax]

Perhaps you were unaware, but you're talking in a public forum.

Glad to know we agree that your name is a loserman.