Return to “Suggestions”

Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#46
Flatfingers wrote:The key advantage of sentience is environmental adaptability. I like the notion that drones can do just one thing well and don't have financial or emotional goals, while NPCs are OK at several things (and possibly really good at one thing) but need some form of upkeep.
That is certainly one of the distinctions I'm personally hoping for between drones and NPCs. Drones should be treated as specialised tools. Whereas an NPC and the player might be able to fly a ship where you can swap different modules and weapons in and out quite easily, everything inside a drone should be hardwired into it. You can have drones that specifically use lasers to deal damage, drones that produce EMP damage, drones that mine, drones that relay intel to the controlling agent, etc.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#47
I think there is a option where "combat drones" makes sense: attacking subsystems on moderate to capital sized ships, though I am picturing them more like a homing missile objects that can make fine adjustments to intercept the target. Once they are on the hull of the target they start firing their lasers and will eventually cut through the hull to cause damage to the subsystem. Picture something similar to the drones in the Matrix where they attach the Nebuchadnezzar. Why subsystems? Well, subsystems are relatively small and quite hard to hit for a small fast moving craft like a fighter. It makes sense that drones would be used for this kind of attack. They are different than a missile in that they continue to cause damage over time if they are not destroyed (great long term damage output), where missiles only damage the ship once (great short-term damage output).

I don't think that combat drones make sense as highly maneuverable objects in a battle where fighters are trying to shoot each other. If they are similar in size to a fighter (which they will have to be as fighters are somewhat difficult to hit) and are about as maneuverable as a fighter, and have one or two guns, what is the purpose of a fighter with a human/NCP as the pilot? If something that is basically the size of a fighter is built and can do fighter-like things then why aren't all fighters AI controlled? And, if they can fit an AI in a drone even smaller than a fighter, then surely the AI can be retrofitted into a into a NPC controlled fighter.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#48
Any 'bandwidth' thingie just doesn't make any sense. Like, at all. Constructions drones are there to work autonomously, completing a given set of orders; I expect no less from my combat drones. If I need to baby-sit my drones, I might as well do everything by hand. Bloody ridiculous. From the gameplay point of view, I have something to say.

What might work without the fabled H-drive:

Specialised combat drones - similar to what pulsar9 said, but maybe even crazier - imagine drones with some sort of drill attached, penetrating capital ship's shields, drilling deep into the hull, finding and destroying or hacking vital systems...or just general-purpose combat drone to shoot down enemies.

Size of the drone may be the best limitation, preventing installation of really powerful shields and guns so that a single barrage of flak from a mid-sized capital ship would be enough to take them down by hundreds. This would stimulate both player and NPCs to develop specialised drones to use for certain situations.

Fabled H-drive solution, the autonomous one:

Since drones are only that big, their CPUs are limited in power, so they are actually slower, less manoeuvrable and less powerful than any ship in the game. Best use: throwing a barrage of combat drones at the enemy to cover your escape.

Fabled H-drive solution, the bandwidth one:

Ok, you absolutely can't live without cloud computing and all that deplorable always-online stuff; in that case, drones should have limited operating radius (less than 1k from the ship, or something) since the delay in calculation would mean bad things happening. That way, drones are basically just your turrets, but de-attached from the ship and flying around to get an optimal firing solution. The big catch is - since you spend your ship's CPU time on your drones, it makes your ship slower, less manoeuvrable and generally less powerful. A nice trade-off.

As for saving real CPU cycles - unfortunately, I have no bright ideas here. Making drones a sub-optimal solution from the gameplay point of view is a good way of limiting their power and their use (and as a direct result, their numbers in game); any of the aforementioned trade-offs do that to a certain degree. Also, hacking. Hack them. Turn them against their mothership. Hack them all, and re-program them to shoot everything that moves. It would be a good reason not to use them at all :)

Edit: edited for clarity.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#49
outlander4 wrote:Any 'bandwidth' thingie just doesn't make any sense. Like, at all. Constructions drones are there to work autonomously, completing a given set of orders. If I need to baby-sit my drones, I might as well do everything by hand. Bloody ridiculous. From the gameplay point of view, I have something to say.
It makes complete sense - you actually make it make sense in your own post below, and I'll cover that. As for construction drones, a sufficiently powerful processor bank would exist at the construction site, allowing construction drones to operate within its vicinity.
outlander4 wrote:Ok, you absolutely can't live without cloud computing and all that deplorable always-online stuff; in that case, drones should have limited operating radius (less than 1k from the ship, or something) since the delay in calculation would mean bad things happening. That way, drones are basically just your turrets, but de-attached from the ship and flying around to get an optimal firing solution. The big catch is - since you spend your ship's CPU time on your drones, it makes your ship slower, less manoeuvrable and generally less powerful. A nice trade-off.
Yes, this is precisely how I see it. Your ship has a finite amount of processing power and bandwidth. If you deploy drones, you will have less bandwidth to allocate to other tasks (and I can think of a few) and less processing power to allocate to other systems. It's a trade-off that you manage. Drones also have to stay within a certain range of your vessel (like with EVE Online) or another friendly vessel/station that you're communicating with.
outlander4 wrote:As for saving real CPU cycles - unfortunately, I have no bright ideas here. Making drones a sub-optimal solution from the gameplay point of view is a good way of limiting their power and their use; any of the afore-mentioned trade-offs do that to a certain degree.
I'd like to see Josh's input on how computationally expensive drones would be to implement.
outlander4 wrote:Also, hacking. Hack them. Turn them against their mothership. Hack them all, and re-program them to shoot everything that moves. It would be a good reason not to use them at all :)
What makes you assume that drones would be that easy to hack, or can be hacked at all? The aim isn't to design them to be frustrating to use.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#50
ThymineC wrote: It makes complete sense - you actually make it make sense in your own post below, and I'll cover that. As for construction drones, a sufficiently powerful processor bank would exist at the construction site, allowing construction drones to operate within its vicinity.
Yeah, I certainly can see it from your side. I'm just not sure I like H-drive. It's a nice concept and a brilliant idea, don't get me wrong; it's just that I don't really need any scientific or pseudo-scientific explanations for ship behaviour in the game where planets don't move from their bloody places. So, binding drones to H-drive implementation is not something I'd like to see in game.
ThymineC wrote:
outlander4 wrote:Also, hacking. Hack them. Turn them against their mothership. Hack them all, and re-program them to shoot everything that moves. It would be a good reason not to use them at all :)
What makes you assume that drones would be that easy to hack, or can be hacked at all? The aim isn't to design them to be frustrating to use.
In non-H universe - because they have a PC that can be re-programmed. As easy as that. You can counter it with drones where instructions are hard-coded and no re-programming-on-the-fly takes place, but those drones would be much less responsive to changing combat environments.

In H-universe - well, they send their data somewhere - you can intercept it, decrypt it (I'm sure a CPU that calculated where individual particles are can hack encryption), and provide conflicting data causing drone to do something stupid or just become inactive.

As for real CPU cycles - aye, we need Josh's word on that. It's a complex problem though, so he might not know precisely.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#51
outlander4 wrote:
ThymineC wrote: It makes complete sense - you actually make it make sense in your own post below, and I'll cover that. As for construction drones, a sufficiently powerful processor bank would exist at the construction site, allowing construction drones to operate within its vicinity.
Yeah, I certainly can see it from your side. I'm just not sure I like H-drive. It's a nice concept and a brilliant idea, don't get me wrong; it's just that I don't really need any scientific or pseudo-scientific explanations for ship behaviour in the game where planets don't move from their bloody places. So, binding drones to H-drive implementation is not something I'd like to see in game.
They're not bound to the H-drive implementation at all. All you need to assume is that the tech required for drones to operate consumes a lot of processing power, more than the drone's microprocessor can provide. H-tech provides a plausible justification for that but even in its absence you can imagine that drones rely on some form of technology that requires a lot of processing power.
outlander4 wrote:
ThymineC wrote:
outlander4 wrote:Also, hacking. Hack them. Turn them against their mothership. Hack them all, and re-program them to shoot everything that moves. It would be a good reason not to use them at all :)
What makes you assume that drones would be that easy to hack, or can be hacked at all? The aim isn't to design them to be frustrating to use.
In non-H universe - because they have a PC that can be re-programmed. As easy as that. You can counter it with drones where instructions are hard-coded and no re-programming-on-the-fly takes place, but those drones would be much less responsive to changing combat environments.

In H-universe - well, they send their data somewhere - you can intercept it, decrypt it (I'm sure a CPU that calculated where individual particles are can hack encryption), and provide conflicting data causing drone to do something stupid or just become inactive.
Yeah, I'd personally like to see drones that can get hacked, but as with everything I think it should be balanced.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#52
ThymineC wrote: They're not bound to the H-drive implementation at all. All you need to assume is that the tech required for drones to operate consumes a lot of processing power, more than the drone's microprocessor can provide. H-tech provides a plausible justification for that but even in its absence you can imagine that drones rely on some form of technology that requires a lot of processing power.
That's exactly the problem I see. With reasonable software, a single Raspberry Pi can do all the calculations required for the drone. I mean, people build robots using it, and space is actually easier to navigate than surface of the Earth.

So, what calculations are we talking about?

It makes sense only in the H-drive context. In non-H scenario (I keep inventing those terms) it's absolutely pointless. The rule is: complicated solutions for simple problems break suspension of disbelief faster.
Last edited by outlander on Sun Feb 23, 2014 6:05 pm, edited 1 time in total.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#53
outlander4 wrote:
ThymineC wrote: They're not bound to the H-drive implementation at all. All you need to assume is that the tech required for drones to operate consumes a lot of processing power, more than the drone's microprocessor can provide. H-tech provides a plausible justification for that but even in its absence you can imagine that drones rely on some form of technology that requires a lot of processing power.
That's exactly the problem I see. With reasonable software, a single Raspberry Pi can do all the calculations required for the drone. I mean, people build robots using it, and space is actually easier to navigate than surface of the Earth.

So, what calculations are we talking about?
Will drones not have shields?

If they do, then that's enough. If not, then I'd prefer they are based on H-tech. Especially if drones follow the same flight model as ships and have a limited top speed.

And yeah, I'm building a robot with BrickPi too. :)
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#54
Shields...well..If it's not an H-field shield, than I see no reason why they should require a lot of computing power. And surely people who manage to travel between stars can pack more processing power into the car-sized flying contraption to handle this. I just don't see any reason to make drones depend on the mothership unless it's an H-scenario we are talking about.

edit: I guess I was ahead of your edit or my phone just misplaced a part of your message. So we basically agree on the matter, I suppose.
Last edited by outlander on Sun Feb 23, 2014 6:16 pm, edited 1 time in total.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#55
outlander4 wrote:Shields...well..If it's not an H-field shield, than I see no reason why they should require a lot of computing power. And surely people who manage to travel between stars can pack more processing power into the car-sized flying contraption to handle this. I just don't see any reason to make drones depend on the mothership unless it's an H-scenario we are talking about.
Even if shields didn't rely on H-tech, that doesn't automatically necessitate that the computations involved are any easier. Since shields will have to be based on technology well beyond anything we've developed in real-life, I could just say "Well, due to the technology involved, shields require 10 exaflops of processing power to manage" and there's no reason why that should be disputed. Since we're basing shields on either one imaginary technology or another, we can make any claim we want about how computationally expensive generating shields is. Sure, chips might be 1000 times as powerful in Limit Theory as they are in real-life, but generating shields can be so computationally expensive that even chips as advanced as that might not be able to handle the computations involved.

TL;DR: I can make the claim that shields require a lot of CPU whether or not they're based on H-tech.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#56
Oh, so we don't agree after all.
ThymineC wrote:I can make the claim that shields require a lot of CPU whether or not they're based on H-tech.
And a directly opposite claim can also be made, I suppose. If we go all realism, fields can be electromagnetic and require little in the way of calculations, with crazy contraptions like magnetically stabilised cloud of plasma around the ship to absorb/deflect lasers; LT is obviously not about that...

All-in-all, in non-H world (hah) it'd be better to search other ways to balance drones. For example, limiting their functionality because highly programmable drones are easily hackable or something.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#57
outlander4 wrote:Oh, so we don't agree after all.
ThymineC wrote:I can make the claim that shields require a lot of CPU whether or not they're based on H-tech.
And a directly opposite claim can also be made, I suppose. If we go all realism, fields can be electromagnetic and require little in the way of calculations, with crazy contraptions like magnetically stabilised cloud of plasma around the ship to absorb/deflect lasers; LT is obviously not about that...
And what about bullets/missiles? Neutrally-charged objects wouldn't be affected by plasma, will they?

Edit: Conceivably they might be vaporised on contact.
outlander4 wrote:All-in-all, in non-H world (hah) it'd be better to search other ways to balance drones. For example, limiting their functionality because highly programmable drones are easily hackable or something.
What's wrong with balancing based on bandwidth?
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#58
Quoting is a nightmare in Opera Mini, so I'll just respond:

Neutraly charged objects will penetrate shields. Of course, plasma should be pretty dense to reflect lasers and it will probably explode when neutral object enters it; a kind of dynamic protection we have on tanks. Would throw the spacecraft away from its course rather violently. But again - it's not LT; it's more fitting to E:D where planets do orbit their suns.

To me, the whole reason of making drones is to let them do stuff without my control; fire-and-forget them, so to say. So this bandwidth limitation goes against my idea of having drones in the first place. What if it's an exploration drone? What can it explore if it's limited to communication range? At that distance, I can go and see stuff for myself. And so on.
Image
Survivor of the Josh Parnell Blackout of 2015.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#59
outlander4 wrote:To me, the whole reason of making drones is to let them do stuff without my control; fire-and-forget them, so to say. So this bandwidth limitation goes against my idea of having drones in the first place. What if it's an exploration drone? What can it explore if it's limited to communication range? At that distance, I can go and see stuff for myself. And so on.
That's not how I'd personally like drones to be. I'd like drones to be fewer in number, individually more important and controllable by the player, like in EVE Online. To differentiate them from fighters, they would be smaller, mass-producible, more specialised, "dumber", not require hiring NPCs or paying wages, and require a constant connection with the vessel.

As for exploration drones, I'm not entirely sure if I want them. Exploration vessels that can serve under your command, sure; but I want drones to be entities that stick close to their parent vessel, in general.

On the other hand, they could be quite cool. If so, then the lore can be adapted to fit that. Gameplay is primary, lore secondary.
Post

Re: Combat Drones & Why I Oppose them for Limit Theory

#60
Aye, so here's the key difference. Well, each one is entitled to his own thoughts, I guess. Me thinks I should go to bed. Need to make another presentation about yet another way to mess up our planet tomorrow...

The last thing - I had a random thought: H-technology is so good at explaining everything because it's just a generic 'dedicate processing time to make stuff happen' thing.

For example, you don't need a diaphragm, a ribcage and a set of muscles to breathe; with H-breather you cat just put air in there if you have enough CPU processing time. If not, well, bad luck!

So of course H-thingie makes perfect sense in explaining pretty much everything; you don't even need to say it because it's a given. It's the nature of things.
Image
Survivor of the Josh Parnell Blackout of 2015.

Online Now

Users browsing this forum: No registered users and 21 guests

cron