Author Topic: Superluminal Communication  (Read 2254 times)

0 Members and 1 Guest are viewing this topic.

Offline liveware (OP)

  • Bug Moderators
  • Commodore
  • ***
  • Posts: 742
  • Thanked: 88 times
Superluminal Communication
« on: July 02, 2020, 04:49:27 PM »
Currently, all communication between ships/colonies/etc in Aurora is modelled as an effectively instantaneous process. How difficult would it be to add in a realistic time delay feature to account for the effects of the finite speed of light on radio communications? Below are some of my thoughts on possible implementation methods I've been brainstorming in my idle time. Take it or leave it as you may.

Firstly, it is important to define a standardized communication path between ships, colonies, and other game elements (such as buoys/satellites). Suppose I have a single colony (Earth) and a single ship in Earth orbit. The ship is placed in a fleet, lets assume it's a naval ship so its under a naval admin command which is ultimately answerable to a naval headquarters on Earth. I propose that whenever a player attempts to communicate with a ship, the player must do so via whatever naval headquarters the ship is assigned to. The naval headquarters would then relay the command to the applicable ship (possibly through a network of communication relay nodes). The existing fleet interface window would suffice from graphical point of view, but the important game mechanic change would be that as the ship becomes farther and farther away from it's assigned naval headquarters, it would respond to commands more slowly. Additionally, if the ship detected something interesting, say a hostile contact or a new mineral deposit, there would be a time delay between when the game engine registers this and the player receives notification from the naval headquarters. This communication time delay could be diminished via a researchable technology, say superluminal communications, which reduces (or possibly eliminates) the communication time delay.

As a follow up, it would be interesting if ships (and even colonies) had a limited effective communication range, rather than the effectively unlimited range they have now, and if they stray too far out of range they receive some sort of malus. This would provide the possibility of another interesting game mechanic in the form of communication relay satellites or dedicated command and control ships which could possibly also benefit from the aforementioned superluminal communications research option.

Together, the communication time delay and limited communication range would make naval headquarters facilities much more important than they are currently. With the mechanics described above, there would be a strong incentive to build naval headquarters facilities on many colonies in order to reduce communication time delay. In this system, it might even be sensible to allow a planet/colony/naval HQ to be designated as a higher level command HQ of some sort, with several sub-HQ's under it's prevue with improved communication links between them.

I think this sort of system would add some interesting player options. One would have to consider defending not only your lines of supply, but also your lines of communication, which may or may not be deployed in a similar fashion. Additionally, you could realistically cut off an enemies' lines of communication and then strike while their fleets are in disarray, perhaps experiencing something similar to jump-shock. This might even give a technologically inferior race a fighting chance against a superior foe. I'm thinking along the lines of how the initial Bugger invasion of Sol was stopped by the humans in Ender's Game (book). Additionally, the communication delay would place much more emphasis on conditional automatic fleet commands, as a fleet commander would be able to react immediately to a new contact but would not necessarily be able to report and receive new orders from central command for quite a while if the fleet is located very far from a naval HQ.

What are other peoples thoughts? There's definitely room for refinement on what I've posted here.

This entire comm delay/finite comm range concept could be added as an option item in the main game menu and be player toggleable, so it would not be necessary if not desired. Also, there should probably be an option somewhere to 'assume tactical command' of a fleet that so that the player can RP the fleet commander / ship captain role if so desired, as I suspect that is a major reason many people play Aurora.
« Last Edit: July 02, 2020, 04:56:18 PM by liveware »
Open the pod-bay doors HAL...
 

Offline TheDOC

  • Registered
  • Sub-Lieutenant
  • ******
  • T
  • Posts: 128
  • Thanked: 17 times
Re: Superluminal Communication
« Reply #1 on: July 02, 2020, 05:43:56 PM »
I think that from a practical standpoint, having this kind of penalty would not be good. The idea of some kind of communication technology works well when the ships are abstracted (ie: Stellaris) enough, but in a game this detailed, you play as both the Governor AND the Captain of your ships at the same time. Normally i RP in the delay, so i personally don't feel the need, but it still can skew decision making.

That said, i like the idea for one thing: passive sensors. Have the contacts of DSTS and passive EM sensors be delayed with your formula if they are not in your focus system. Sure, you can bypass via active sensors being constantly on, but it should be an interesting mechanic. Maybe with a game option toggle.
 

Offline liveware (OP)

  • Bug Moderators
  • Commodore
  • ***
  • Posts: 742
  • Thanked: 88 times
Re: Superluminal Communication
« Reply #2 on: July 02, 2020, 05:45:55 PM »
I've tried RP'ing the delay in but it gets... tedious.

The dual role of governer/fleet commander is an odd balance to strike.
Open the pod-bay doors HAL...
 

Offline Zincat

  • Captain
  • **********
  • Z
  • Posts: 566
  • Thanked: 111 times
Re: Superluminal Communication
« Reply #3 on: July 02, 2020, 06:15:47 PM »
I would not be a fan of such a mechanic, and so I would play with it off. I do not think it would be a good idea as it would be a very large change of paradigm. But that's just my personal opinion on the concept, of course.

That said, I honestly don't see how it could work. So the player is notified of new contacts with a delay... Unless the delay is very very small (and then I don't see why you'd add it), this basically means that a lot of engagement will be over without the player being able to do anything about it.

Imagine fighters/missiles  detected in course to intercept your fleet in 5 minutes. But you have a delay of 10 minutes. You will know of the engagements after your fleet has been destroyed. I don't see how that works. No matter what the player may or may not "know", surely the fleet would defend itself!

And even if the delay is smaller, surely the fleet can react immediately. Even if it's just reversing course to make the distance from the incoming hostiles as large as possible, surely the fleet does not need some officer back on earth to tell them to do that. And also, would you then enforce that for any new command, or just the first time? By logic, any new command should be delayed by 10 minutes or so in the above case.

You have mentioned "conditional defense orders", but I don't see how it could work because you cannot decide in advance how a fleet should react. It depends on too many factors. What are the hostile forces? What are the relative speeds? What is the range of the fleet, the fuel level, its distance from home, its defences? How is the system where the fight has to take place? Shoud the fleet even fight or just flee, depending on the opposing forces? How could you possibly choose what "conditional defense orders" to give without knowing all of that? Every engagement is different.

Sorry, I just don't see how this could possibly work out in a way that makes any rational sense. Because while information back earth should certainly be delayed in such a scenario, a fleet should be able to immediately fight at full capacity anyway. I just don't see how you can automatically resolve this conundrum in game terms in a way that makes sense for the player.
« Last Edit: July 02, 2020, 06:17:33 PM by Zincat »
 

Offline SpikeTheHobbitMage

  • Bug Moderators
  • Commodore
  • ***
  • S
  • Posts: 670
  • Thanked: 159 times
Re: Superluminal Communication
« Reply #4 on: July 02, 2020, 06:51:13 PM »
There is already a reaction delay mechanic that is supposed to simulate communication lag, both ship-to-shore and internally between crew members.
 

Offline Ulzgoroth

  • Captain
  • **********
  • U
  • Posts: 414
  • Thanked: 73 times
Re: Superluminal Communication
« Reply #5 on: July 02, 2020, 07:01:33 PM »
A strategy game with significant communications and signal delay mechanics needs to have a lot of attention given to those, and to enabling units to function acceptably when they're not under 'real-time' player control.

I'm interested in a game that does that, but I don't think Aurora wants to be that game.
 
The following users thanked this post: DIT_grue, skoormit

Offline liveware (OP)

  • Bug Moderators
  • Commodore
  • ***
  • Posts: 742
  • Thanked: 88 times
Re: Superluminal Communication
« Reply #6 on: July 02, 2020, 11:15:42 PM »
Relativistic effects such as I proposed would certainly be a major addition/paradigm shift for Aurora. However, as all ships at present are subluminal (jump points excluded) the game seems to me to be amenable to such a modification.

As stated previously, automatic conditional commands would be much more important in the mechanic I proposed, as fleet commanders would be expected to behave much more autonomously in the face of new situations than at present. The game does currently allow the conditional order of 'if hostile forces detected, etc...' so I do not think that what I am suggesting is necessarily beyond the scope of the game engine. A new or expanded set of conditional commands may be necessary to supplement this change in detail, however. It is possible that this is beyond the scope of what Aurora strives to be, and such a change would not be welcome on such grounds, even if possible in theory.

Perhaps it is time for me to embark more seriously on my 'Relativistic Aurora' project, which at present is naught but some spreadsheets and napkin calculations, but considers in detail what I suggested in my original post. I hate to reinvent the wheel though.
Open the pod-bay doors HAL...
 

Offline liveware (OP)

  • Bug Moderators
  • Commodore
  • ***
  • Posts: 742
  • Thanked: 88 times
Re: Superluminal Communication
« Reply #7 on: July 02, 2020, 11:30:18 PM »
A strategy game with significant communications and signal delay mechanics needs to have a lot of attention given to those, and to enabling units to function acceptably when they're not under 'real-time' player control.

I'm interested in a game that does that, but I don't think Aurora wants to be that game.

I have considered this problem in detail. I have a couple of potential solutions:

1. Utilize an existing (preferably open source or friendly source) implementation of an astronomical star map or stellar evolution program as a base engine. Add sci-fi elements on top of the existing engine.

2. Utilize an existing implementation of a software suite capable of simulating electromagnetic particles. In such a system one could specify the behavior of sensor arrays, and from that, a more generalized sci-fi universe with stars and planets and ships and everything in between.

Obviously both approaches would require significant development effort. The first approach is advantageous as it builds on existing purpose-built tools. The second method also builds on existing tools, but in a very different way and would require much effort to implement even the simplest of Aurora style sensor satellites.
Open the pod-bay doors HAL...
 

Offline SpikeTheHobbitMage

  • Bug Moderators
  • Commodore
  • ***
  • S
  • Posts: 670
  • Thanked: 159 times
Re: Superluminal Communication
« Reply #8 on: July 02, 2020, 11:57:35 PM »
A strategy game with significant communications and signal delay mechanics needs to have a lot of attention given to those, and to enabling units to function acceptably when they're not under 'real-time' player control.

I'm interested in a game that does that, but I don't think Aurora wants to be that game.

I have considered this problem in detail. I have a couple of potential solutions:

1. Utilize an existing (preferably open source or friendly source) implementation of an astronomical star map or stellar evolution program as a base engine. Add sci-fi elements on top of the existing engine.

2. Utilize an existing implementation of a software suite capable of simulating electromagnetic particles. In such a system one could specify the behavior of sensor arrays, and from that, a more generalized sci-fi universe with stars and planets and ships and everything in between.

Obviously both approaches would require significant development effort. The first approach is advantageous as it builds on existing purpose-built tools. The second method also builds on existing tools, but in a very different way and would require much effort to implement even the simplest of Aurora style sensor satellites.
Significant development work in integrating the new libraries, and in designing, implementing, and testing the new rules, including rule changes, database changes, UI changes, and AI changes.  And it would eliminate random stars games.  And that is before we get to licensing issues, since Aurora is closed source and those existing tools may require license fees for closed use, if they allow it at all.

There are simpler ways to do it by using existing in-game distances to determine lag, but it would still involve structural changes and rebalancing to make it all work.
 

Offline Zincat

  • Captain
  • **********
  • Z
  • Posts: 566
  • Thanked: 111 times
Re: Superluminal Communication
« Reply #9 on: July 03, 2020, 05:13:10 AM »
The problem is not how to determine the lag. I will say this again, the problem is that this would be a major paradigm shift where fleets have to be autonomous once they're out of your borders. Unless the lag is so small that it does not even matter, but then I don't see the point of creating all this complex system.

You send them out and have to account for the possibility that they do their thing based on conditional order, you can intervene from time to time but there are also cases where the combat is completely autonomous.

That is the exact opposite of what combat has been in aurora up to now. I'm not saying it's a bad thing in principle, mind you, just that I don't see how it would fit with how combat works in Aurora.
Try to think of the conditional orders nightmare, you'd have to either set literally a hundred different conditional orders for each fleet, or to accept a significant loss of possibilities in combat.

Just for sensors, you'd need like 10 different conditional orders
Should they be on all the time or not? Should you turn them on once you have a passive contact? Should you turn them off after seeing what  is present? Should you turn them off mid combat in an attempt to flee if the combat is going bad? How should the fleet decide if the combat is going bad?
Should the fleet engage the enemy? Should it try to retreat? Should it engage from afar and then retreat? Should it engage from afar and then close in? Should it close in and do hit and run tactics?
What is the loss tolerance the fleet has? Should it try to flee after sustaining some damage? Should it stay and fight to the end? Should it try to split and flee? How should it try to split? In how many different groups should it try to split? What routes should the  split fleet want to use to retreat?
Regarding missiles, should the fleet fire them? Should it fire them all at the same target or split them? How many volleys should it fire? How many missiles should be kept in reserve? Should it stop firing if the enemy seems to have strong PD? Should the fleet retreat once out of missiles?
Regarding beam weapons, should the fleet try to kite enemy at max distance? Should it close at point blank? Should it target one ship down until destroyed? Should it target small or large enemies first? How should it behave in case the enemy is faster?
Regarding AMM, should the fleet use them? When should it use them? How many of them should be used depending on incoming volley size?
Regarding fighters, should the fleet launch them? Should they close directly or take a roundabout route? Should they close to point blank range? Should they stay at max range? Should they launch missiles and get back? Should they not engage if enemy fighters are detected?

You would need to set ALL of the above for every single fleet you send out there. And I'm being generous as I can think of many, many more conditions you'd need to set. A nightmare.

The alternative would be to have a significant loss of control in that scenario, where you can choose just 4-5 conditional orders for each fleet. This however would mean there's only a number of different ways in which the fleet would behave, and it would have SIGNIFICANT repercussions on fleet design.
After all, if every fleet you send outside your borders is limited to a pre-determined number of tactics, a lot of ship designs become sub-optimal, and perhaps not something you should invest your limited resources into.
Why even bother going for a stealth fleet design or a hit and run harassment fleet design if the conditional orders do not allow for the control required to utilize those fleet designs well?

I'll say it again, it's not the idea in principle that I'm against. I've played plenty of sci-fi games where the combat was barebone and engagement rules were simple, and the game was still fun and engaging. But combat and engagement in Aurora is extremely complex and diversified, and as such I cannot see how such a huge loss of control in fleet orders would be acceptable.
« Last Edit: July 03, 2020, 07:45:39 AM by Zincat »
 
The following users thanked this post: SpikeTheHobbitMage

Offline Cobaia

  • Warrant Officer, Class 1
  • *****
  • C
  • Posts: 88
  • Thanked: 16 times
Re: Superluminal Communication
« Reply #10 on: July 03, 2020, 05:29:05 AM »
I don't get the need for a delay.

Fleets of ships are given a general order followed by the ROE instructions when they depart: Protect sector X and engage when engaged on. After that all the decision making is up to the commander of each ship or the fleet commander. This is the system that is already implemented in Aurora (Governor and Captain roles that the player has). This delay you propose is only appliable to the Strategic orders given by the Navy Headquarters, all tatical orders are made by the commanders.

And since you referenced Orson Scott's Ender's Game you can also give the player the need to research the 'Ansible' so that communications are instant between the equipment.

I think a simulation of the communications delay is out of scope, it should be kept abstracted or at a conventional start game a research would be needed to achieve it.
 

Offline Jorgen_CAB

  • Admiral of the Fleet
  • ***********
  • J
  • Posts: 2822
  • Thanked: 673 times
Re: Superluminal Communication
« Reply #11 on: July 03, 2020, 11:56:36 AM »
We have to understand that if ships (or fleets) need to store individual data in contacts things will become very complex from a programing perspective and there are other issues with this as well. But I do think there are some merits to communication being somewhat important, I also could see small change to sensor mechanics eventually to be a bit less black and white, some introductions in C# have already been heading in that way.

What I see as realistic is a few things to spice up the contact and the way we can do things in the game.

Communications
Every ship will need a communications module in order to communicate with other ships. This module while active can then be connected with any other ship using FTL type communication and sharing data flawlessly up to a certain range. This module also emit a certain EM rating. When the module is on it will share all data with the ships connected to the "network". Only ships with a command module can act as a network. The command module could a new command bridge type with a new type of officer, the bonus of the highest ranking network officer add to the range that all the ships connected can communicate with each other.

There could in practice be more than one network in the same system if they are separated by too big of a distance.

The game now only have to store information for each network in each system and for any fleet not connected to the network. A fleet would only be able to use target information that it can access.

There would also be a function where you can click on a fleet and the map only show what they fleet can see which is mainly for role-play but can also be useful to know what contacts it can actually see.


Sensors and electronic warfare
Sensors should not automatically detect things within a certain distance automatically but there should be a low chance at the outer distance and then higher and higher the closer something gets. Once you have a sensor lock the chance to loose it should be really low.

In order to make things more interesting if you have ships at different locations the more the distant apart they are the more the chance increase that you will get a sensor lock on something. This would then encourage spreading out your scouting elements even more which in my opinion would give more of a tactical feel to the sensor mechanic.

On top of this you also add electronic warfare, both passive and active. Active system you will easily see with passive EM but you will have trouble locking on with actives, passive simply make it harder to lock on with actives much like the cloaking technology.


I also think that communication module need to go into missiles to put the nail in the coffin for very effective size 1 ASM missile spam or MIRVs (unless super close distance that is).
 

Offline Ulzgoroth

  • Captain
  • **********
  • U
  • Posts: 414
  • Thanked: 73 times
Re: Superluminal Communication
« Reply #12 on: July 03, 2020, 12:33:05 PM »
A strategy game with significant communications and signal delay mechanics needs to have a lot of attention given to those, and to enabling units to function acceptably when they're not under 'real-time' player control.

I'm interested in a game that does that, but I don't think Aurora wants to be that game.

I have considered this problem in detail. I have a couple of potential solutions:

1. Utilize an existing (preferably open source or friendly source) implementation of an astronomical star map or stellar evolution program as a base engine. Add sci-fi elements on top of the existing engine.

2. Utilize an existing implementation of a software suite capable of simulating electromagnetic particles. In such a system one could specify the behavior of sensor arrays, and from that, a more generalized sci-fi universe with stars and planets and ships and everything in between.

Obviously both approaches would require significant development effort. The first approach is advantageous as it builds on existing purpose-built tools. The second method also builds on existing tools, but in a very different way and would require much effort to implement even the simplest of Aurora style sensor satellites.
Neither of those even relates to the problems that would need to be solved.

Almost nobody cares if your game map has unrealistic astrography. Also, you can't have realistic astrography without a 3d map, which is an added game development headache. And for the other thing, there is probably somebody somewhere who really wants a game involving quantum modeling of sensor signals, but I'm not sure there are two people. You certainly don't need that to have a game-suitable model of sensors and communication with signal lag.


The problem is AI. If you have tactically significant control lag, it will rapidly become intolerable for units to to be just following orders with no judgement. If you have strategically significant control lag, you may abstract combat to the point that individual unit judgement doesn't factor into the calculation, but then it becomes intolerable for your strategic maneuver units to lack judgement. Regardless, you wind up with a lot of individual AI agents, each one operating under different incomplete information and outdated orders.

That's some very serious development work. Also possibly a computational burden when the game is running.
 
The following users thanked this post: SpikeTheHobbitMage

Offline liveware (OP)

  • Bug Moderators
  • Commodore
  • ***
  • Posts: 742
  • Thanked: 88 times
Re: Superluminal Communication
« Reply #13 on: July 14, 2020, 12:59:24 AM »
Many of the responses here identify the multitude of challenges inherit in a simulation approach which is more consistent with modern physical theories. As this is supposed to be a science fiction game I am not going to lose any sleep over any diversions from reality I may encounter. There are certainly significant challenges to make things 'realistic' and as previously stated, Aurora may not be the proper venue. Yet Aurora offers much.

A major obstacle, from my perspective, for any 4x game, is reconciling the roles of ship commander, fleet admiral, colony governor, and imperial leader. Each of these roles is inheritly separated spatially, yet each has independent agency. To simultaneously control all four in different spatial locations is fundamentally unrealistic, as any realistic method of communication can propagate at most at light speed. If fiction is permissable, then FTL communication may have a place.

So, if one desires a quasi-realistic strategic situation, one must consider the limitations imposed by light speed communication. With this limitation, standing orders/ROE/automatic engagement orders are a requirement, as waiting several years for a response from central command is not going to be a tenable option for a frontier outpost under siege.
Open the pod-bay doors HAL...
 

Offline Jorgen_CAB

  • Admiral of the Fleet
  • ***********
  • J
  • Posts: 2822
  • Thanked: 673 times
Re: Superluminal Communication
« Reply #14 on: July 14, 2020, 02:50:38 AM »
Many of the responses here identify the multitude of challenges inherit in a simulation approach which is more consistent with modern physical theories. As this is supposed to be a science fiction game I am not going to lose any sleep over any diversions from reality I may encounter. There are certainly significant challenges to make things 'realistic' and as previously stated, Aurora may not be the proper venue. Yet Aurora offers much.

A major obstacle, from my perspective, for any 4x game, is reconciling the roles of ship commander, fleet admiral, colony governor, and imperial leader. Each of these roles is inheritly separated spatially, yet each has independent agency. To simultaneously control all four in different spatial locations is fundamentally unrealistic, as any realistic method of communication can propagate at most at light speed. If fiction is permissable, then FTL communication may have a place.

So, if one desires a quasi-realistic strategic situation, one must consider the limitations imposed by light speed communication. With this limitation, standing orders/ROE/automatic engagement orders are a requirement, as waiting several years for a response from central command is not going to be a tenable option for a frontier outpost under siege.

I think that it simply is easier to role-play most communication problem and leave the AI where it is, the AI are not too efficient anyway no matter what so it does not really matter.

I usually role-play that FTL communication between system really is not possible and sensor buoys on both sides can transmit limited amount of data and that it will take some time as no one can really talk but have to communicate through messaging and it will take time.