Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

SC's Subsumption #nobullshit

13»

Comments

  • MaxBaconMaxBacon Member LegendaryPosts: 7,760
    RingLdr77 said:
    I am worried about an AI controlled economy as nobody has gotten a game economy right yet, so somehow scripting AI into is doesn't seem like a good idea. It's like the blind leading the blind.
    Their intent shows to be AI driven economy, player-influenced, I personally dislike when players have so much control to the point we see in several MMO's the literal control of markets and manipulating prices to accumulate unbelievable amounts of wealth.

    Play by the AI rules can work but it can't be a static system where players feel they can't make any relevant impact.
    Excession
  • MinscMinsc Member UncommonPosts: 1,353
    Minsc said:
    MaxBacon said:
    Most companies that try making massive advancements in games end up abandoning them before launch.
    If it was easy everyone would be doing it now wouldn't they? xD

    See The Sims 3, one great tech take at it yet they gave up mid-way on it (as it has issues, the hugely known optimization matter and other bits), then they decide to do the 4 and cut it back so much in TS4 it literally ends up as a "meh".

    It's that nature, when you go through "uncharted waters" in development it's by trial and error until it works, some have more maneuverability than others to risk such (publishers obviously hate those risks so it's uncommon to see such happen in large-scale/budget projects).
    I just don't think crowdfunded games are the place to try to test all of these things.
    Well who better? A publisher is never going to take the risk. An indie company is not going to have the resources without a publisher backing, which again is never going to take the risk.
    It really isn't a case of who better, because you could easily say that a brand new studio headed by someone who hasn't released a successful title in (I think) the past decade isn't the best case scenario for a studio trying to do something that's never been done.

    Maybe they'll pull it off and it will be a tight, smooth experience integrating a head-spinning number of interdependent, robust systems.  But it leaves a lot of room for skepticism until we see that finished product.  So far, I've seen very little that would make the title truly revolutionary.  There are some revolutionary ideas, but if the execution falls apart, it would be nothing more than a cautionary tale.

    The gamer nerd in me wants to see it come into fruition just to experience such a game.  The realist in me thinks that one of two things will happen: a whole lot of scope trimming, or a mess of convoluted systems that don't complement one another well.
    Right but CR is the founder of the company. His brother Erin is the lead dev on the project and he has been in the industry all this time. It's also pretty foolish to think that even though CR wasn't actively in the industry that he wasn't still keeping up with what was going on.
    ErillionExcessionConstantineMerus
  • Octagon7711Octagon7711 Member LegendaryPosts: 9,000
    edited October 2017
    Excession said:
    MaxBacon said:
    Specific to subsumption though it's already in for 3.0 so what they talk about is way more fleshed out than what was just one idea back then. This obviously by iterations and the first release is the basic version of what the AI meta is.
    You "think" it is in 3.0, since nobody has played/seen 3.0 yet, nobody knows if it is in or not.
    I would agree with this.  If it was in 3.0 they would have put in the live demo.  After all they had AI in the exploration video with the sand worm.  They had problems with it.  I'd really be surprised if it made it into 3.0.  Even the last mission they showcased had NPCs in the background but not a mention of improvements to AI.  

    I didn't check but it should be on the current schedule if it's a part of 3.0.

    Exactly what are the new upgrades in 3.0 suppose to allow the NPCs to do, that the current NPCs aren't doing?  

    "We all do the best we can based on life experience, point of view, and our ability to believe in ourselves." - Naropa      "We don't see things as they are, we see them as we are."  SR Covey

  • MadFrenchieMadFrenchie Member LegendaryPosts: 8,505
    Minsc said:
    Minsc said:
    MaxBacon said:
    Most companies that try making massive advancements in games end up abandoning them before launch.
    If it was easy everyone would be doing it now wouldn't they? xD

    See The Sims 3, one great tech take at it yet they gave up mid-way on it (as it has issues, the hugely known optimization matter and other bits), then they decide to do the 4 and cut it back so much in TS4 it literally ends up as a "meh".

    It's that nature, when you go through "uncharted waters" in development it's by trial and error until it works, some have more maneuverability than others to risk such (publishers obviously hate those risks so it's uncommon to see such happen in large-scale/budget projects).
    I just don't think crowdfunded games are the place to try to test all of these things.
    Well who better? A publisher is never going to take the risk. An indie company is not going to have the resources without a publisher backing, which again is never going to take the risk.
    It really isn't a case of who better, because you could easily say that a brand new studio headed by someone who hasn't released a successful title in (I think) the past decade isn't the best case scenario for a studio trying to do something that's never been done.

    Maybe they'll pull it off and it will be a tight, smooth experience integrating a head-spinning number of interdependent, robust systems.  But it leaves a lot of room for skepticism until we see that finished product.  So far, I've seen very little that would make the title truly revolutionary.  There are some revolutionary ideas, but if the execution falls apart, it would be nothing more than a cautionary tale.

    The gamer nerd in me wants to see it come into fruition just to experience such a game.  The realist in me thinks that one of two things will happen: a whole lot of scope trimming, or a mess of convoluted systems that don't complement one another well.
    Right but CR is the founder of the company. His brother Erin is the lead dev on the project and he has been in the industry all this time. It's also pretty foolish to think that even though CR wasn't actively in the industry that he wasn't still keeping up with what was going on.
    "Keeping up with the industry" is not the same as actively participating in project management in the industry.
    ExcessionKefo

    image
  • MaxBaconMaxBacon Member LegendaryPosts: 7,760
    edited October 2017
    @Octagon7711 https://robertsspaceindustries.com/schedule-report

    3.0 is meant to actually have NPCs, SC currently does not have NPC AI, it has Ship AI and some character models in shops that ain't AI (as in Arccorp since 2015).

    You seem to have missed in both the demos since the first one last year, the mission giver that runs under subsumption AI and the NPCs roaming (were originally static in the first one) the landing zone area were already there.

    As you can see by the schedule, even on bug-fixing changelog "Fixed an issue where an AI ship in escort missions would become stuck while scanning." shows the type of AI upgrades in subsumption that allow for missions as escorts, missions part of the new mission system, meant as procedurally generated type of content.

    It's a matter of being informed, not of what you or I believe or think, at this moment it stands as part of the update in-schedule.
    Excession
  • Octagon7711Octagon7711 Member LegendaryPosts: 9,000
    MaxBacon said:
    @Octagon7711 https://robertsspaceindustries.com/schedule-report

    3.0 is meant to actually have NPCs, SC currently does not have NPC AI, it has Ship AI and some character models in shops (as in Arccorp since 2015).

    You seem to have missed in both the demos since the first one last year, the mission giver that runs under subsumption AI and the NPCs roaming (were originally static in the first one) the landing zone area were already there.

    As you can see by the schedule, even on bug-fixing changelog "Fixed an issue where an AI ship in escort missions would become stuck while scanning." shows the type of AI upgrades in subsumption that allow for missions as escorts, missions part of the new mission system, meant to be procedurally created.

    AI

    MISSION SYSTEM

    • The overall system that will be used to create mission flows when we start building missions like Patrol, Assassination, Smuggling, etc.
    • We have now moved into the bugfixing stage of this feature

    ALREADY COMPLETE IN THE 3.0.0 BRANCH:

    • AI Turrets

    "We all do the best we can based on life experience, point of view, and our ability to believe in ourselves." - Naropa      "We don't see things as they are, we see them as we are."  SR Covey

  • MaxBaconMaxBacon Member LegendaryPosts: 7,760

    AI

    MISSION SYSTEM

    • The overall system that will be used to create mission flows when we start building missions like Patrol, Assassination, Smuggling, etc.
    • We have now moved into the bugfixing stage of this feature

    ALREADY COMPLETE IN THE 3.0.0 BRANCH:

    • AI Turrets
    ALREADY COMPLETE IN THE 3.0.0 BRANCH:
    Subsumption Base Functionality
    • This is the initial release of the foundation technology that drives all of the AI, mission, dynamic content, and conversation logic. The mission content previously created with FlowGraph has been replaced and the implementation time to achieve a desired effect has been cut dramatically.

    Subsumption is one underlying tech, it's not a "feature" per self, it's not one of those superficial features they can just cut, that'd be like cutting IS 2.0 from 3.0 and breaking all the game systems converted and created on the new tech.
    Excession
  • MaxBaconMaxBacon Member LegendaryPosts: 7,760
    edited October 2017
    Sim Citizen! \o/
    Mind you, I say that as someone who's played the hell out of The Sims, so that's not a bad thing for me.

    The only concern I have with more advanced NPC AI that's just there for flavor is... how will it affect the overhead? The main reason other MMOs don't do stuff like this isn't because the devs are incompetent, but to cut down on the number of scripts running simultaneously. Sometimes a nifty feature just isn't worth the load it adds to the game.

    I remember The Sims getting laggy a few times when there was way too much stuff going on, and that's just a single player game. At least it won't be a concern during combat, since we'll be away from hubs where they're all moving around. BDO also does this to a smaller degree - the NPCs go about their daily lives, and go to different locations in the mornings and in the evenings, and it's probably part of what makes people complain about lag in the big cities so much. That, plus a gazillion players doing their own thing of course. xD
    Server services by layers, the game is meant to have a server mesh, if they don't localize the services they may just have central services that handle AI and other things globally.

    This would not be a big deal because you can prioritize, say the Sim dogfighting AI obviously needs to be very responsive, but NPCs walking around to places and doing sims and simulating their own stuff is the type of stuff that can be optimized as something like NPC it won't need to be that responsive or reactive, that must be how they have been handling it.

    The case of The Sims, The Sims ends up laggy because it's poorly coded to handle it, just see the expansion packs and the colossal loadings showing how they didn't really prepare their codebase to handle what they ended up doing with it.

    The case of BDO, that's not Sims simulation, the BDO is a script in a loop that never changes, the NPCs always take the same path, always say the same thing, always appear/disappear at the same time, it's all scripted. In this case on SC; the NPC would be relying upon in its behavior trees instead of on a loop script.

    BDO lag I don't think it has to do with their AI (it's really simple, even on combat you notice how their AI is so unresponsive in agro and so), we can easily notice the huge asset loading pop problems and the cities are the max representation of that "chunk loading" problem with the added large number of players in the area, the large majority of the NPCs you load in the cities is you rendering the character and its animation.
Sign In or Register to comment.