will the server engine architecture ever be released/for sale?

[Shaman]

title. since the CEO seems to be so convinced on this metaverse technology the NQ team has created I am partially surprised that there doesn't seem to be a way to buy it off them or something.

[Maxim Kammerer]

JC was even more convinced on the potential of DU as MMO. So what?

[blundertwink]

What do you even mean by "server engine architecture"...? Why would you want a backend that's failed to scale at every turn -- especially one that's likely vendor-locked into AWS? 

 

Regardless, there is a way to buy it off them...clearly NQ is wanting to be acquired.

 

Their timing and strategy sucks...because they are hoping their early trademarks around "metaverse" and whatever patents they hold would make them an easy target for some big entity investing in the metaverse.

 

Of course, Microsoft just dropped their metaverse project and no one believes Meta's platform will amount to jack shit. There's no "big entity" left other than Meta, and they have no need for NQ's silly trademark or poorly implemented tech. 

 

People in this metaverse space are dropping like flies for a good reason.

 

I do believe NQ's leadership genuinely believes in web3 (hence their announcement that their current development focus is on metaverse tooling like "3d blogging")...and this belief absent any evidence, valid use case, or basic understanding of the technology in general is (as time will prove) not a healthy way to run a company. 

[Aaron Cain]

To make a long story very short:

 

No

[GraXXoR]

In short answer:

No / Everything is for sale if the price is right.

[CptLoRes]

Funny thing is JC has a PhD in Artificial Intelligence and Robotics, but didn't want any kind of intelligent behavior in the game unless coming from players.

And big corp is literally throwing money at AI currently, and all the big name game companies are now investing heavily into AI driven game mechanics.

[Atmosph3rik]

1 hour ago, CptLoRes said:

unny thing is JC has a PhD in Artificial Intelligence and Robotics, but didn't want any kind of intelligent behavior in the game unless coming from players.

And big corp is literally throwing money at AI currently, and all the big name game companies are now investing heavily into AI driven game mechanics.

 

Do you have any examples of an AI driven game mechanic?  I have no idea what that means.

 

 

 

[CptLoRes]

 

 

So the obvious stuff like more realistic human like behavior from NPC's, but also things like player characters moving in a realistic way in all situations and reacting to the environment, and AI driven graphics optimizations etc.

[blundertwink]

4 hours ago, CptLoRes said:

So the obvious stuff like more realistic human like behavior from NPC's, but also things like player characters moving in a realistic way in all situations and reacting to the environment, and AI driven graphics optimizations etc.

 

Well, more realistic and complex animations, anyway. That isn't the same as "game mechanics", it's asset production -- which is maybe what you meant, so I don't mean to be nitpicky. 

 

Just saying, there's a big difference between "AI driven game mechanics" and "AI as a tool to author assets". Procedurally generated animations have been a thing for a while now, so this is definitely an evolution of an idea that some studios have been using for a long time. 

 

AI driven game mechanics are far harder to imagine via AI unless you're making very simple clones of existing games.

 

Good mechanics require creativity and a deep understanding of engagement, which even the best designers can't really quantify in objective terms...so machine learning is a long way from being able to drive "mechanics" but it will be fascinating to see if and how it changes asset creation. 

Edited February 1, 2023 by blundertwink

[Atmosph3rik]

On 2/1/2023 at 3:20 AM, CptLoRes said:

So the obvious stuff like more realistic human like behavior from NPC's, but also things like player characters moving in a realistic way in all situations and reacting to the environment, and AI driven graphics optimizations etc.

 

 

Video game characters moving and acting realistically is already AI.  It always has been. 

 

What you're describing is better AI.

 

It doesn't seem like you or the people who are making those videos are really clear on whether AI is the new technological breakthrough that will change everything, or that AI will create the new technological breakthrough that will change everything.

 

It's like saying you've discovered the secret to turning lead into gold, and all you need to do it is lots of gold.

 

[CptLoRes]

One of the major differences is that technology using AI is derived from training data sets and/or goal/reward driven evolution, and the resulting solutions are often able to perform under situations they have never encountered before. That is very different from the handcrafted/scripted list of behaviors that you will find in a typical game character 'AI' in current games. It would not be possible or at least not practical to program the complexity you get from training AI's by hand.

 

For example a AI driven walking solution if trained, would start to move completely different in a heavy armor without having to be told do so. And that movement would be dynamic. Lose a piece of the armor and the movement changes. Add a slippery floor and hilarity ensues etc.

While current solutions on the other hand would just move exactly as programmed regardless.

 

And I predict that games are going to start looking and feeling much more dynamic and realistic in years to come. Much more so then the current progression in games.

[Atmosph3rik]

3 hours ago, CptLoRes said:

One of the major differences is that technology using AI is derived from training data sets and/or goal/reward driven evolution, and the resulting solutions are often able to perform under situations they have never encountered before. That is very different from the handcrafted/scripted list of behaviors that you will find in a typical game character 'AI' in current games. It would not be possible or at least not practical to program the complexity you get from training AI's by hand.

 

For example a AI driven walking solution if trained, would start to move completely different in a heavy armor without having to be told do so. And that movement would be dynamic. Lose a piece of the armor and the movement changes. Add a slippery floor and hilarity ensues etc.

While current solutions on the other hand would just move exactly as programmed regardless.

 

And I predict that games are going to start looking and feeling much more dynamic and realistic in years to come. Much more so then the current progression in games.

 

How do you reward an AI?  Are there AI biscuits?  Do you let it kill just one human as a treat?

 

You're still just describing better NPC AI.   Your example of with the heavy armor is exactly how it works in Elden Ring and all the previous Dark Souls games.  That's not anything new.  The only reason some games don't have it is that it wouldn't improve the game, or they didn't want to pay someone to code it.

 

The only change that your describing is that a game company might hypothetically not need to pay an actual human to code anymore.  

 

You still haven't given an example of how AI specifically would improve a video game.

 

What does it bring to the table that isn't already there?

[CptLoRes]

9 hours ago, Atmosph3rik said:

 

How do you reward an AI?  Are there AI biscuits?  Do you let it kill just one human as a treat?

 

You're still just describing better NPC AI.   Your example of with the heavy armor is exactly how it works in Elden Ring and all the previous Dark Souls games.  That's not anything new.  The only reason some games don't have it is that it wouldn't improve the game, or they didn't want to pay someone to code it.

 

The only change that your describing is that a game company might hypothetically not need to pay an actual human to code anymore.  

 

You still haven't given an example of how AI specifically would improve a video game.

 

What does it bring to the table that isn't already there?

 

Not sure if you are joking or not with the AI biscuits comment, and it is not my job to educate you.

But you should at least make an effort to get to know about basic principles like Reinforcement learning etc. before making derogatory comments about things you appear to know little about.

 

Some more examples below, and remember these videos are early prototypes and some of them are more then a year old.

But you also have to ask yourself that if this AI stuff in games is nonsense, then why are all the big names like Intel, Nvidia and game dev studios spending big money and lots of time and energy on developing this technology?

 

 

 

 

 

 

 

[le_souriceau]

AI with learing obvious will be future of gaming, when they figure out how to make it fast/cheap (effieicent).

 

Because hard-coded scripted behaivors of "enemies" are at soft cap pretty much sinde 90s. Or even before. In still often have same dummy bots in many games we had as far as in 80s. Sure devs can make them more intricate, wasting exponantionally more resources (in fighting games etc, where its critical for gameplay), but still no jumping above head.

[Atmosph3rik]

4 hours ago, CptLoRes said:

Some more examples below

 

 

You're claiming that AI will improve video games.  But all of your examples are just general improvements to video games.  Better NPC behavior, better graphics etc.

 

But you still haven't explained how AI is essential to any of that.

 

All of those improvements were happening anyway, they always have been.  AI could speed up the process, and lower costs.  But those aren't improvements to the games, that just makes them easier and cheaper to make.

 

Currently making a video game requires filthy capitalists, management, and people who code and play video games.

 

Do you really think removing the need for people who code and play video games from the process is going to result in better games?

 

 

 

 

 

 

[Tordan]

1 hour ago, Atmosph3rik said:

...removing the need for people who ...  play video games...

 

I'm pretty sure that removing that jerk Tyim would make any game better.

[Atmosph3rik]

I'm probably not explaining myself very well here.  So let me try this a different way.  Just to be clear i have no issue with "AI" in general.  It's just a computer program.  They're great.  I'm a fan.

 

But the way the term is being used recently in the industry just sounds like hype.

 

Let's say you went to eat at a restaurant that specializes in hamburgers and french fries.  And they have a huge sign on the wall bragging about how they are switching from fresh cut fries to mass produced frozen fries.  As if i'm supposed to be excited about that...

 

As a consumer i don't see anything to be excited about.

 

And when i hear a game company talking about AI as if it's a direction for the company to go in, that means nothing to me.

[CptLoRes]

1 hour ago, Atmosph3rik said:

But you still haven't explained how AI is essential to any of that.

Because the improvement are of such a complexity that it would not be economically or timely to do by hand, like being able to ask a NPC more or less anything and get reasonable answers. Or having the NPC react realistically to any action you can think of and manage to do while in the game, regardless of if the developers planned for it or not.

And some of the graphical improvement like close to photorealistic graphics are only possible using AI for now and a long time away with traditional methods.

 

And the bragging about AI just to generate hype thing was very real 5-6 years ago. But in the last couple of years has some real breakthroughs that bring AI very much into the useful category both in general and for games.

[Atmosph3rik]

1 hour ago, CptLoRes said:

Because the improvement are of such a complexity that it would not be economically or timely to do by hand, like being able to ask a NPC more or less anything and get reasonable answers. Or having the NPC react realistically to any action you can think of and manage to do while in the game, regardless of if the developers planned for it or not.

And some of the graphical improvement like close to photorealistic graphics are only possible using AI for now and a long time away with traditional methods.

 

And the bragging about AI just to generate hype thing was very real 5-6 years ago. But in the last couple of years has some real breakthroughs that bring AI very much into the useful category both in general and for games.

 

So procedural generation would be a pretty good example of what you're talking about already being implemented in game development.

 

I think you could argue that procedural generation has improved game development.  But i don't think it can be credited for improving the games themselves.

 

For a while, game companies were boasting about using it.

 

But now when a company like NQ uses the aid of procedural generation to build planets.  The focus is on the fact that the planets are hand crafted by devs. Not the procedural generation.  Because consumers learned that on its own procedural generation equals quantity over quality.  Without being curated by a human, it's just a bunch of noise.

 

When you describe an NPC reacting to any possible action, i picture how well the trees in DU react to being procedurally placed on any possible terrain.  It works, but it's not great.  Definitely not something to brag about.

 

Circling back around to what you originally said, that i disagree with so strongly.  The words "AI driven game mechanic"

 

I haven't seen any examples of what those words actually mean.  It sounds like hollow marketing speak.  And i would be wary of any game company that was investing heavily in it.

 

 

 

 

 

[CptLoRes]

When developing a AI driven game, the developers define the scope of the game, backstories for characters and key moments to set the frame for what happens and what is allowed inside the game. The AI will then fill in the blanks and bring every part of the game to life in much more detail then what would be possible to script.

 

Here is me spending literally 10 seconds defining a "game" story in chatgpt, and telling the ai about an even that happens while playing.

Imagine what a proper dev studio would be able to do with an AI model trained specifically for the game they are making.

 

 

[Atmosph3rik]

15 hours ago, CptLoRes said:

When developing a AI driven game, the developers define the scope of the game, backstories for characters and key moments to set the frame for what happens and what is allowed inside the game. The AI will then fill in the blanks and bring every part of the game to life in much more detail then what would be possible to script.

 

Here is me spending literally 10 seconds defining a "game" story in chatgpt, and telling the ai about an even that happens while playing.

Imagine what a proper dev studio would be able to do with an AI model trained specifically for the game they are making.

 

 

 

You're still talking about frozen french fries here.  The only improvement that you've brought to the table is cost and time savings.  And the entire premise that there would be an improvement to the product is based on the hypothetical idea that the developers will do something really amazing with the extra time or money that they'll have.  Which could be true, but still isn't anything that would be credited to the AI.

 

The story you posted above is boring and derivative.  A child could write that.  You could blame that on your prompts, i guess.  But either way it's not great.

 

Salesman is also one word.  It's interesting that the AI just went with Sales Man as the name of the character, but still seemed to know that he should be selling something.  

 

I guess that's one point for the AI lol