Return to “Suggestions”

Post

Mind Acceleration

#1
This isn't intended as game design for LT. Maybe LT2, but really this is just game design for the heck of it. I'm just expanding on earlier ideas of mine.

OUTLINE
In Thoughts on Worker and Executive NPCs, I originally imagined workers (low-LOD NPCs) and executives (high-LOD NPCs and the player) as being conscious entities whose minds are hosted on an artificial substrate.

Part of my proposal was that workers could be allocated to controlling different systems on a vessel/station or even set in charge of a vessel or station, and that CPU could be allocated to them to improve their performance, as allocating more processing power to a worker allows their consciousness to be simulated at a higher rate, effectively allowing them to "think faster".

Since executives are also digitally mind-uploaded entities, the question arose as to why executives' minded couldn't be accelerated through allocating CPU in the same way as for workers:
Cornflakes wrote:when ships have x times the amount of computation, why do i have to use this computation for x employees instead of making myself x times faster?
I did some handwaving to claim it wasn't possible - but why shouldn't it be? This proposal considers the implementation of a "mind acceleration" mechanic and explains what benefits and other implications this idea has.

THEORY
All workers and executives (including the player) are mind-uploaded agents. Their consciousness is hosted within a physical hardware unit that I guess we could call a "brain module" or "mindcore" or something along those lines. mindcores encapsulate all the data associated with a particular agent's consciousness, but require CPU processing power allocated to them by an external computer system in order to simulate the consciousness. The more CPU that is allocated to a mindcore, the faster it simulates the agent's consciousness, but the more heat it generates. There is an upper limit to the heat generation rate of any given mindcore that prohibits faster simulation of an agent even if additional processing power is available - exceeding this would risk damaging the mindcore and thus effectively cause "brain damage" or death to the agent being simulated.

Agents can also allocate additional processing power to their mindcores if they're the commander of the vessel, but this generates excessive heat. When agents "board" a particular vessel and take control of it, their mindcores are kept within a commander support module, which acts as the interface between the agent and the ship's systems and main computer facilities. The CSM of a vessel also provides a certain capacity for excessive heat storage generated by the mindcore, allowing the mindcore to accelerate itself for certain amounts of time with (literal) cooldown periods in between.

As I write in A Reconstruction on Death Mechanic, agents already have their consciousnesses running by default at an accelerated rate of a few hundred times faster relative to us. Let's say that this speedup factor is 500x. What that means is that 1 second to the real-life player would (in terms of lore) be perceived as around 8 minutes for the player character and other agents in Limit Theory. I posit 500x speedup to be the maximum rate at which mindcores can be operated with stable heat levels. "Mind acceleration" is really then temporarily bumping this up to, say, 550x speedup, for a perceived slowdown in the passage of time of 10%.

PROPOSAL
Here are the mechanics I propose for mind acceleration:
  • All vessels need to have a commander support module installed. Alternatively, instead of being a separate module, the functions of the CSM could be incorporated into the core module of the vessel, if such a thing exists. The core module would then not only form the core component of a vessel or station, but host the mindcore of the agent.
  • Commander support modules come in different sizes and offer different heat storage capacities. Larger commander support modules typically offer higher heat storage capacities. Larger vessels can fit larger CSM's.
  • The higher the heat storage capacity of a vessel, the greater the acceleration factor that an agent can achieve for a given amount of time, or the greater the length of time that an agent can sustain a given acceleration fator.
  • When an agent accelerates their mind, heat will build up in the CSM over time. When an agent falls back to the default level of 500x speedup, any excess heat stored in the CSM's heatbanks are dissipated over time.
  • If a CSM's heatbanks become full, safety measures drop the agent's simulation rate back to the 500x default.
  • It may be possible to allow agents to override these safety measures and continue mind acceleration past the maximum safe threshold at the risk that of causing malfunctions such as blackouts or damage to the mindcore.
The implications here is that the larger the vessel an agent acquires, the greater their capability to perform mind acceleration. Agents in fighters and small vessels will only be able to accelerate their minds by limited amounts for brief durations, while those in command of battleships or capitals could achieve greater mind acceleration levels for longer periods of time.

What benefits does this have to gameplay?
As Cornflakes suggested, flying a battleship should feel a lot different from flying a fighter and the difference between both should be reinforced not just by quantitative differences in the scale of their attributes, but in the actual gameplay involved with handling both. Smaller ships should play like in Freelancer, while larger ships should play like in EVE Online.

I understand this to mean that controlling fighters in LT should be very twitch-based, with little need for complex puzzle-solving thinking and really more just quick reflexes and a good ability to fly and shoot straight. On the other hand, large vessels like battleships should involve less twitch-based reaction skills and more managerial play, where the player focuses more on the following kind of things:
  • Locking onto targets and assigning turrets and other weapons to track and fire at them.
  • Managing shield frequencies and orienting the vessel to minimise damage being sustained.
  • Establishing communication pipelines to other vessels in order to share valuable information with them to boost their combat performance as I've discussed here.
  • Controlling the allocation of ship resources to different systems including power, CPU and possibly others such as "repair nanite streams" or coolant. I discuss the need for and implementation of bullet time for players when accessing menus in that thread, and this provides a more concretely developed design for bullet time.
  • Importantly, RTS-like control of other vessels within the fleet, including fleet formations and tactics, without the need for a non-diegetic pause mechanic as others (such as Flatfingers) have proposed.
The mind acceleration mechanic supports this - it allows larger vessels to enter what is effectively Bullet Time to handle complex and demanding tasks while in the heat of battle.

In addition, this mechanic also naturally leads to making the battlefield seem epic, as well as making fighters and other nimble craft appear to fly by slower to an accelerated agent within a battleship, as Josh was concerned about in A Request For Fighters.

How do we display relevant information to the player?
In Limit Theory, I have proposed two allocable ship resources: power and CPU. Josh has previously implemented a boost capacitor mechanic which allows for agents to sustain brief periods of higher power outputs, and the relevant information for this is presented to the player as a blue bar in the lower-right corner of the HUD. I prefer to imagine the mind acceleration mechanic as the CPU analogue of this, and for that reason I imagine the UI representation of the mechanic to be pretty similar. Along with the blue boost capacitor bar, the player will be able to see a red CSM heat level bar:
Image (Direct link)

The speedup rate relative to the default level can be shown below this bar as well.

Unlike the boost capacitor bar which starts off full and diminishes with use, the CSM heat bar starts off empty and builds up towards its maximum as the player keeps his mind accelerated beyond the default 500x level. When the player relaxes back to the 500x level, the heat bar diminishes back to zero again.

While the player is keeping his mind boosted, special filters and other effects could be used to give the impression of Bullet Time, such as making all sound effects become longer and lower-pitched.

How does the player control mind acceleration?
I imagine that the player could perform configuration of mind acceleration (such as the speedup factor) in a menu before entering into combat. Alteratively, the speedup factor might be a fixed property of the CSM.

I'd imagine that the boost capacitor could be triggered by holding down Shift, and mind acceleration could be triggered by holding down Ctrl.

Is this mechanic compatible with other proposed mechanics?
Yes. This mechanic is compatible with other mechanics that have been proposed. Hyperion's reconstruction idea - that when an executive dies, a certain amount of time (such as subjective weeks or months) elapses while they're being reconstructed - can be made to work with this. At first I didn't see how it could plausibly take weeks or months to recompile what is essentially a computer program (if we assumed that executives are mind-uploaded agents), but this mechanic makes complete sense when we consider that everything in LT is happening at an accelerated factor of 500x anyway. If, say, it took 12 "base" hours to recompile an executive, where a base hour is an hour as it would be experienced by one of us, then that would be the equivalent of 250 subjective days passing within LT for all agents operating at the default 500x speedup rate.

This mechanic is compatible with my ideas in Thoughts on Workers and Executives. In addition, it resolves the issue of worker->executive promotioon as discussed in What Do Workers Do When You Kill Their Queen?, since we are imagining workers and executives to both be mind-uploaded agents and so there is no need to imagine an implausible change in the nature of a worker from a computer program to a flesh-and-blood entity when they are promoted.

This is also compatible with my thoughts in Eugenics (and Evolution).

What are the limitations of this proposal?
The main problems associated with this proposal are:
  • People don't generally like to imagine themselves as being computer programs for some odd reason. They'd be happier imagining themselves as being flesh-and-blood.
  • A game that utilises such a a mechanic necessarily cannot be multiplayer.
  • Ship windows and glass cockpits, as aesthetically pleasing as they might be, would be rendered even more illogical than they'd otherwise already be.
  • Some people may not like the idea that utilising bullet time becomes necessary for handling the systems of larger vessels in the heat of battle.
Post

Re: Mind Acceleration

#2
As always, an excellent presentation of an idea. This really helps me understand what a suggestion is for and how it might play out in a game.

To summarize it back to you to see if I understand it correctly, you're proposing two mechanics (beyond the "everyone's a program" assumption):

1. Speeding up one's mind has the in-game effect of slowing down all action.
2. Speeding up one's mind is on a timer.

If I have this right, then my reaction is something like this:

1. Cool. Slowing down time sufficiently is functionally equivalent to being able to pause the game in order to perceive patterns and formulate plans.
2. Ack.

Mrs. Flatfingers and I play the iPhonePad game Scramble, in which you try to find as many words as you can in a 4x4 grid of letters within two minutes. It is incredibly, infuriatingly annoying because the &^#$ timer interrupts me just when I'm getting in the zone. I have to hurry through every choice because I am not given enough time to grok the structure of the board. The ability to perceive high-level patterns (long, high-value words) is penalized in favor of quickly finding many baby words.

I think, just speaking for myself, that I would find a mind acceleration timer, regardless of any diegetic presentation or explanation, equally irritating. Regardless of whether the expiration of the time means Bullet Time is over, or if the player can extend the timer at some cost/penalty, it's still a &^#$ timer, and it still discourages thoughtful play in favor of "hurry, hurry, do something, even if it's sub-optimal, time's running out!"

(Side node: The equivalent of Bullet Time was actually a key mechanic of F.E.A.R., which I generally enjoyed. It's not a timer per se to which I object -- it worked fine in F.E.A.R.'s purely tactical design. It's being forced to hurry operational or strategic thinking, as in perceiving word patterns or managing a fleet, that feels counterproductive to fun.)

Although it's not my primary objection, I'm also curious how the mind acceleration effect would be employed by executive NPCs in a way that maintains player/NPC parity.

One way to try to salvage the cost aspect of the mind acceleration idea is by replacing the "heat" mechanic (which is just a timer) with a reduction/penalty to something else of the player's. Switching to some other consumable resource doesn't help as that's just a different form of a timer.

Some kind of passive effect, that persists as long as the player chooses to continue the Bullet Time effect, might be palatable, though. Maybe the price of being able to increase your local perceptiveness is a loss of global perception: for example, as long as you're accelerated, your sensors go blank beyond a few kilometers of your current position. (This would also work if you wanted to support variable-rate mind acceleration. Global awareness would be inversely proportional to one's mind acceleration factor.)

Would something like this be tolerable versus a straight timer? Or is there a significant benefit to requiring quick decision-making (despite enabling Bullet Time in the first place) that I'm not seeing?
Post

Re: Mind Acceleration

#3
I just don't see the need to cap the player's ability to issue orders by limiting orders while paused / bullet time.

It's a fine mechanic for a FPS where you control a single actor and don't have to make any strategical or organisational decisions but in a game where you can control a galactic empire, the UI will become your enemy eventually.
Delegation is fine and dandy but you still have to set up and manage the delegation.

Bullet time / orders while paused are not a new suggestion.
Limiting the ability to pause the game is something I'm strongly opposed to. =)
There is no "I" in Tea. That would be gross.
Post

Re: Mind Acceleration

#4
Flatfingers wrote:To summarize it back to you to see if I understand it correctly, you're proposing two mechanics (beyond the "everyone's a program" assumption):

1. Speeding up one's mind has the in-game effect of slowing down all action.
2. Speeding up one's mind is on a timer.
Correct.
Flatfingers wrote:I think, just speaking for myself, that I would find a mind acceleration timer, regardless of any diegetic presentation or explanation, equally irritating. Regardless of whether the expiration of the time means Bullet Time is over, or if the player can extend the timer at some cost/penalty, it's still a &^#$ timer, and it still discourages thoughtful play in favor of "hurry, hurry, do something, even if it's sub-optimal, time's running out!"

(Side node: The equivalent of Bullet Time was actually a key mechanic of F.E.A.R., which I generally enjoyed. It's not a timer per se to which I object -- it worked fine in F.E.A.R.'s purely tactical design. It's being forced to hurry operational or strategic thinking, as in perceiving word patterns or managing a fleet, that feels counterproductive to fun.)
Well, there are two reasons why I have to disagree with you here:
  • If the player is in the heat of a battle, I don't think they should be able to indefinitely pause the game to assign orders and survey data and such. I believe this kind of defeats the point of battles being intense and exciting events, because it keeps breaking up the fast-paced action. I think that LT should reward deep strategic thinking that's done before battle in preparation for it.
  • A pause mechanic is not only non-diegetic but also unfair to the NPC and breaks player-NPC parity; the player is able to pause indefinitely and take however long to assess tactical information and make decisions as they like, but the NPCs can't.
Flatfingers wrote:Although it's not my primary objection, I'm also curious how the mind acceleration effect would be employed by executive NPCs in a way that maintains player/NPC parity.
The commanders of other vessels (which may be executives or workers) will be able to accelerate their minds and the game will keep track of the heat buildup as it does everything else. While the NPC is mind-accelerating itself, more (real-life) processing power will be allocated to it. Mind acceleration generally becomes more powerful the larger in ship size you go, and there are far fewer larger vessels than smaller ones, so this shouldn't be too much of a burden on the CPU. In addition, when the player mind accelerates himself, the entire simulation slows down, making the whole game less of a burden on their CPU.
Flatfingers wrote: One way to try to salvage the cost aspect of the mind acceleration idea is by replacing the "heat" mechanic (which is just a timer) with a reduction/penalty to something else of the player's. Switching to some other consumable resource doesn't help as that's just a different form of a timer.
Well, the heat mechanic aside, in order to accelerate your mind you'd need to allocate more CPU to your mindcore, reducing the amount of CPU available to distribute to other systems. So you could just assume that there's a small window between 500x speedup and, say, 600x speedup or whatever in which the player can set their mind acceleration rate, and assume that the excessive heat build up is all handled automatically. That way players can accelerate their minds if they wish, and can sustain that accelerated state, indefinitely, but may have to reduce CPU allocation to other systems.

Alternatively, if you really don't want there to be any kind of timer, you can effectively eliminate it by entering the data editor and changing the constant that governs the ratio of speedup factor to additional heat generation. If an excessive heat generation rate of 10 units.s^-1 occurred when you used a speedup factor of 200% w.r.t. default, you would only perceive that as a buildup rate of 5 units.s^-1. If an excessive heat generation rate of 10 units.s^-1 occurred for a speedup factor of 1000% w.r.t. default, that would be perceived as a buildup rate of 1 units.s^-1. If you set the ratio of speedup factor to heat generation rate at something like infinity, you effectively get a pause mechanic.
Gazz wrote:Limiting the ability to pause the game is something I'm strongly opposed to. =)
I'm totally in favour of letting the player pause the game, it's absolutely necessary, but during this time the player shouldn't be able to interact with the actual game e.g. by assigning orders. I'm not sure if they should be able to pan around either.
Post

Re: Mind Acceleration

#5
ThymineC wrote: If the player is in the heat of a battle, I don't think they should be able to indefinitely pause the game to assign orders and survey data and such.
And that's what I'll never agree with. =)
Unless and until the game AI can provide a staff of human-like intelligence and the UI lets me order combat maneuvers of great complexity as effortlessly as ordering another human to do it, I need to be able to observe and correct what all my minions do.
There is no "I" in Tea. That would be gross.
Post

Re: Mind Acceleration

#6
ThymineC wrote: Well, there are two reasons why I have to disagree with you here:
  • A pause mechanic is not only non-diegetic but also unfair to the NPC and breaks player-NPC parity; the player is able to pause indefinitely and take however long to assess tactical information and make decisions as they like, but the NPCs can't.
i disagree with your disagreement.
assesments the player needs 10s of seconds at least the AI does in single digit amounts of frames.
so a pause mechanic would actually re-introduce NPC-player parity.

and limiting pause for players would give NPC's an edge
Post

Re: Mind Acceleration

#7
Cornflakes_91 wrote:
ThymineC wrote: Well, there are two reasons why I have to disagree with you here:
  • A pause mechanic is not only non-diegetic but also unfair to the NPC and breaks player-NPC parity; the player is able to pause indefinitely and take however long to assess tactical information and make decisions as they like, but the NPCs can't.
i disagree with your disagreement.
assesments the player needs 10s of seconds at least the AI does in single digit amounts of frames.
so a pause mechanic would actually re-introduce NPC-player parity.

and limiting pause for players would give NPC's an edge
The AI should be programmed to act like humans, though. That means not having 100% accuracy when shooting, and making assessments/coming up with strategies in around the same time as the average player.

It's better to make the player and AI equal in as many respects as possible rather than have the AI have a vastly superior advantage in one respect, the player in another, and hope the two balance out.
Post

Re: Mind Acceleration

#8
ThymineC wrote:The AI should be programmed to act like humans, though. That means not having 100% accuracy when shooting, and making assessments/coming up with strategies in around the same time as the average player.
Unlimited orders / information while paused does not help the player with the fighting, though - only with the decision-making which is just about impossible to "correctly" scale for the AI.

The AI can parse information and give orders while paused all the time... except it happens too fast for you to notice.
All I want is parity.

Another twist would be that the AI uses "your" bullet time for more extensive planning, too.
There is no "I" in Tea. That would be gross.
Post

Re: Mind Acceleration

#9
Gazz wrote:
ThymineC wrote:The AI should be programmed to act like humans, though. That means not having 100% accuracy when shooting, and making assessments/coming up with strategies in around the same time as the average player.
Unlimited orders / information while paused does not help the player with the fighting, though - only with the decision-making which is just about impossible to "correctly" scale for the AI.
Well that's what I want to avoid - allowing the player to make decisions through the actual pause menu. I'm also hoping/assuming that the AI decision-making can be scaled correctly - this is an implementation-side issue. AI should make decisions at a comparable rate and quality to the player.
Gazz wrote:The AI can parse information and give orders while paused all the time... except it happens too fast for you to notice.
I'm not sure what you mean. When the game is paused, the AI shouldn't be being simulated at all.
Gazz wrote:All I want is parity.
Same.
  • Players and AI should make decisions about as well as each other.
  • Both should be able to mind accelerate themselves. For the player, this means a slow-down in the simulation. For the NPC, this means more processing power allocated to them. This should have the effect of making either kind of agent make better decisions during that time.
  • If the game is paused, neither the player or the AI should be able to make decisions or gather intel.
Gazz wrote:Another twist would be that the AI uses "your" bullet time for more extensive planning, too.
What would that achieve?
Post

Re: Mind Acceleration

#10
ThymineC wrote:I'm not sure what you mean. When the game is paused, the AI shouldn't be being simulated at all.
If "paused" means "slowed by a factor of 1000", then it is simulated.

ThymineC wrote:What would that achieve?
Just like the player the AI could "look around".
3D space means that the number of objects that you include in your evaluation increases with a power of 3.
In pause / bullet time, the AI would have the CPU cycles to evaluate a lot more space/objects around it.

While the regular AI may "take no time", VeryLittleTime^3 is another issue entirely. =)

ThymineC wrote:
Gazz wrote:All I want is parity.
Same.
No, I think you want symmetric parity, assuming that the AI perfectly simulates other human beings.
I'm looking for asymmetric parity because I consider a completely human AI unlikely.
There is no "I" in Tea. That would be gross.
Post

Re: Mind Acceleration

#11
Gazz wrote:
ThymineC wrote:I'm not sure what you mean. When the game is paused, the AI shouldn't be being simulated at all.
If "paused" means "slowed by a factor of 1000", then it is simulated.
And if the AI makes decisions at a comparable rate to the human player, then they should be able to achieve nothing in that time.
Gazz wrote:
ThymineC wrote:What would that achieve?
Just like the player the AI could "look around".
3D space means that the number of objects that you include in your evaluation increases with a power of 3.
In pause / bullet time, the AI would have the CPU cycles to evaluate a lot more space/objects around it.

While the regular AI may "take no time", VeryLittleTime^3 is another issue entirely. =)
You're describing a bug, I think. The AI should not be able to evaluate more information during the player's bullet time. When the player's bullet time is activated, the whole LT simulation slows down, the AI along with it.
Gazz wrote:No, I think you want symmetric parity, assuming that the AI perfectly simulates other human beings.
I'm looking for asymmetric parity because I consider a completely human AI unlikely.
Yeah. Both are forms of parity. But you're right, I want symmetric parity.
Post

Re: Mind Acceleration

#13
Gazz wrote:It's only a bug if it wasn't designed to do that.
I'm getting a bit confused. What I meant as a bug was:
Gazz wrote:Another twist would be that the AI uses "your" bullet time for more extensive planning, too.
The AI shouldn't be able to benefit from the player's mind acceleration. That's not part of the design.
Post

Re: Mind Acceleration

#14
It's part of my amendment to your design. =P
It gives the AI time for far more Big Picture evaluation and maybe decide that instead of evading left or right it should be running like hell.

If you don't use the feature and play a more direct battle, the AI doesn't get this bonus thinking time, either.
There is no "I" in Tea. That would be gross.
Post

Re: Mind Acceleration

#15
Gazz wrote:It's part of my amendment to your design. =P
It gives the AI time for far more Big Picture evaluation and maybe decide that instead of evading left or right it should be running like hell.

If you don't use the feature and play a more direct battle, the AI doesn't get this bonus thinking time, either.
But I don't get the point of this.

In my design, the other NPC commanders already have their own mind acceleration capability. They carry out decision-making at a comparable rate to the player. If the player accelerates their mind, they can get more decision-making done per unit of in-game time, since the game slows down. If the NPC accelerates their mind, they can get more decision-making done per unit of in-game (and real-life) time, since they get more CPU allocated to them.

My design is essentially: everyone gets a speed potion. They can use this potion whenever they want, and it benefits only themselves.

Your design seems to be: the player gets a speed potion, and maybe the other people too, but when the player drinks his speed potion this benefits everybody else as well.

Online Now

Users browsing this forum: No registered users and 21 guests

cron