Flanking and Shield Facings

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
Ran Taro
Space Squid
Posts: 58
Joined: Wed Oct 06, 2004 2:51 am

#31 Post by Ran Taro »

I think daveybaby's idea is good, but as suggested it adds too many variables for the AI to run. Requiring an AI to run complex algorithms to manage something nominally under player control results in a high risk of unintuitive or exploit prone gameplay (see MOO3).

Consider the example he gave (Total War). In this game, certain sides (flank/rear) are always weak and this produces the desired result. Why do we need to add complexity on to this? The Solution Dreamer suggests is more similar to Total War than the more complex ones suggested, and therefore seems more likely to work, since it has worked before in a completed game.

Why not just invent some fluff to say shield technology by it's very nature focusses more energy in the opposite direction to the main thrusters?(hence shields are always strongest at the front). Then you would end up with a similar system to Total War, where flanking is very important - especially if ships are grouped into 'units'.

Perhaps technology advances would allow more even (or differently skewed) shield distributions. However, if the distribution is always fixed at the design stage, and the number of designs per empire has some limiting factor, then this could make for some interesting strategy too. For example, you could use espionage to discover what the shield distributions of an empires main ship designs are, and design counter tactics based on them.

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#32 Post by Daveybaby »

noelte wrote:My main concern goes with having a shield which only covers one sector and leave another completely unprotected.
Thats not what i was suggesting. I'm saying that the shield always expands automatically to protect the ship from attack from any and all directions at once. But the shield's 'leakage' increases the more it has to expand, so the shield's effectiveness is reduced.

Its a simple formula: Shield leakage is proportional to the angular distribution/spread of incoming attacks.
noelte wrote:So i would suggest something like that:
facing : front => 75% front + 10% left + 10% right + 5% back
facing : default => 50% front + 20% left + 20% right +10% back
facing : equal => 25% front + 25% left + 25% right +25% back.
By making it a number of discrete steps rather than an analogue system, youre forcing intelligent decision making into the loop - either from the player (in which case, micromanagement hell) or from the AI (in which case, obfuscation and player frustration, and possibly Ran Taro's argument below might then start to hold some water). At what point is it optimal to switch shield facings? (If) space combat was realtime, then this becomes a serious problem. Fancy losing a battle because you didnt click the button at the right time, as opposed to because you didnt manoeuvre for position as well as your opponent?

Ran Taro wrote:I think daveybaby's idea is good, but as suggested it adds too many variables for the AI to run. Requiring an AI to run complex algorithms to manage something nominally under player control results in a high risk of unintuitive or exploit prone gameplay (see MOO3).
This isnt an AI issue, its a purely mechanistic, automatic adjustment of shield facings (important note: automatic does not involve AI). No decision making, just a few simple mathematical calculations per ship. A trivial amount of processing even for a 20 year old PC. If you think that this is anything that is going to tax a processor at all for anything less than tens of thousands of ships (in which case you would never be working things out on a ship by ship basis anyway, but by groups of ships) then you dont understand how computers work.
The COW Project : You have a spy in your midst.

noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#33 Post by noelte »

Hmm, from your first post
e.g.
0 to 90 degrees coverage = 100% efficiency
90 to 180 degrees coverage = 75% efficiency
180 to 270 degrees coverage = 50% efficiency
270 to 360 degrees coverage = 25% efficiency
What happend, when the facing is between 0-90 degrees coverage with the uncoveraged area??
By making it a number of discrete steps rather than an analogue system, youre forcing intelligent decision making into the loop
I can't understand this point. And btw, i wouldn't allow to change facing within a battle, facing is a design decision. (was that the misunderstanding?) I would rather say, having a discrete number of variations reduces micro and decision making.
Press any key to continue or any other key to cancel.
Can COWs fly?

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#34 Post by Daveybaby »

noelte wrote:Hmm, from your first post
What happend, when the facing is between 0-90 degrees coverage with the uncoveraged area??
It isnt covered. But that doesnt matter, because you arent being attacked from those angles. If you ARE attacked from those angles, coverage is instantly expanded, and your efficiency drops.

It might be easier if you dont think of it as 'coverage', but more along the lines of: if the shield is being *hit* from more than one direction, its efficiency drops.

I can't understand this point. And btw, i wouldn't allow to change facing within a battle, facing is a design decision. (was that the misunderstanding?) I would rather say, having a discrete number of variations reduces micro and decision making.
Ah, yes, that was the misunderstanding.
But having fixed facings doesnt help with flanking. In fact, it could actively discourage it (especially with the 25/25/25/25 setting), encouraging the enemy to concentrate on one facing to wear it down.

What i want is something the exact opposite of what the Moo2 shield system produced.
The COW Project : You have a spy in your midst.

noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#35 Post by noelte »

Aha, now i get your idea, nice.

but how can an attacker tell the shield efficiency from his point of view. Isn't it hard to tell how effective a shield will block an shot? It surely depends on from which directions the shield is hit at the same time.
Press any key to continue or any other key to cancel.
Can COWs fly?

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#36 Post by Daveybaby »

Well, i guess to a large extent that depends on the scale of space combat and how much detail in general you want to present to the player. e.g. are you going to show separate status for shields, armour, hull, crew, weapons - or just one general 'health' bar? To a large extent this depends on whether space combat is going to involve dozens of ships or dozens of task forces (hundreds of ships).

It could be represented graphically either by highlighting the active part of the shield, or using different colours or brightnesses to represent shield efficiency. Or it could even be shown numerically for a selected ship or task force.

But if youre talking a scale of hundreds of ships then you might not even need to be shown explicitly. It might be enough that you are aware that flanking a ship reduces its shield efficiency by (say) 50% - as long as you can notice some difference in play (i.e. you can tell that enemy ships die faster). e.g. in total war, there is no indication of the effects of flanking other than the fact that you tend to annihilate the enemy very quickly if you manage to do so, compared to a head-on attack.

Note : 'hit at the same time' is a key issue. There should be a minimum interval after a flanker is eliminated before the affected part of the shield can 'de-activate' and thus before your efficiency goes back up, because its actually very unlikely that 2 shots will hit at exactly the same time. Maybe for higher tech shields this interval would be smaller thus quicker recovery of shield efficiency. Regardless of this, a shield segment should always turn on instantly.
The COW Project : You have a spy in your midst.

noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#37 Post by noelte »

hmm, if those shield facing have some effect on how the battle is going and we allow player to control the battle in detail, we need to display shield stats on which players can made their decision. If shield facing is done under the hat, we shouldn't implement it.

On the other hand, we can simply calculate ship-to-ship weapon efficiency. Say if a ship is quicker than another, it's weapons hit that ship harder. This way the ship doesn't have to flank out the other ship explizit. That makes it less tediouse for players to control the battle themself.

I would say, that issue goes along with the question of how many ships (ship - stacks) we will have at a single battle. If we have fewer shield facing would add tactical possibilities. if we have thousands of ships, i would leave it out (no real value).
Press any key to continue or any other key to cancel.
Can COWs fly?

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#38 Post by Daveybaby »

I would tend to agree. Shield facings would tend to be something which would apply to single ship control level combat.

For task force level control i would still like to see something similar though - possibly through the use of task force different formations (offensive/defensive/evasive etc) which would confer different strengths and vulnerabilities.
The COW Project : You have a spy in your midst.

Impaler
Creative Contributor
Posts: 1060
Joined: Sun Jun 29, 2003 12:40 am
Location: Tucson, Arizona USA

#39 Post by Impaler »

"RED SQUADRON TO GOLD SQUADRON SET REAR DELFECTORS TO MAXIMUM, THAT SHOULD KEEP THEM OFF OUR TAIL!!"

I think squadron/taskforce level redirection of shield energy would be quite apropriate and usefull (and cool).

A possible compromise idea. Durring ship design a player can put shields in the ship with one of the 4 facings or pay a little extra fee to create "Omni-directional" Shields. Much like the relationship between turreted weapons and spinal mounts, the 4 cardinal shields cant move to protect other sides. Only the Omni's can change their direction and bolster the defence of a particual side (or be the only defences if you so wish). Durring combat you can change that side at taskforce level and do neat things like "over-shielding" (dose for shields what over-drive and over-load dose for engines and weapons). The Omni shields take some time to move and can point in one or many directions at the same time (the power drains from one side and shifts to the new face rather then switching instantaniously). If your forces attack form more then one direction you have an advantage, also if your literaly running circles around them faster then shields can move you will tear them up badly (this will help small fast fighter ships and sneak attacks do more damage). It punishes continualy attacking from a single direction as desired but dose not completly eliminate the more traditional moo2 style of shielding.

Along a similar vein, we can focus on the idea of power consumption being the primary limiting factor in a ships combat effectivness. The ship most likly WILL have the physical equipment to protect every side but lacks the power to run them all simultaniusly (or perhaps to run them all AND do anything usefull like move or shoot). Again we give the player some redirecting options in combat but here their blended with overall power distributions that direct power to all parts of the ship. Again we require a delay to raise and drop the shields so out manuvering and flanking take their tole.
Fear is the Mind Killer - Frank Herbert -Dune

Dreamer
Dyson Forest
Posts: 228
Joined: Sun Mar 13, 2005 6:44 am
Location: Santiago, Chile

#40 Post by Dreamer »

noelte wrote:I wouldn't allow any distribution, only a set. Maybe the shielding ability differ from shield type to shield type. Say basic shields can only have the facing equal (or only front). Advanced shield might be adjustable to different facing models (front/default/equal).
I would in fact allow distribution, maybe restricting a minimum of 1 point in any side to mantain shield integrity. This way I can arrange my capital ships like roman legions, combining the use of shields from several ships to defend. Turttle formations! Each character a different ship, note how to create a spear shock formation with 8 ships. (Im using "." becuause html eliminates spaces)

. . . . . . . . . . . . . . . _ _ _
enemy this way . . /
. . . . . . . . . . . . . . .\ _ _ _

And same stuff for blocking screens, carrier defense, etc.

. . . . . . . \ . . . . enemy here
. . . . . . . . \
carriers . . . \ . . . . . . . . . . . . . . . . . .planet here
here . . . . . . .\ _ _

my bombers using
the cover --->

The other advantage over automatically adjusting shields is that you can choose to actually ignore some sources of minor damage to focus your shields on actual threads, kind of like "forget about the Tie figthers, all shields facing the death-star" (for obvious reasons). If automated the player would not have the choice.
Daveybaby wrote:Note : 'hit at the same time' is a key issue.
And you are still ignoring the actual damage you will receive when diverting shields. I would put a single laser on my planet bombers to divert your shields and blast you from the front with reduced shield efficiency. And belive me, my bombers can hold quite a punch, they need to get to the planet alive to drop bombs after all.

Diverting shields automatically will only call for "distractor ships". I make some very thought ships with lesser weapons and some weak ships with a lot of weapons (all space spended in weapons). Then I move my distractors into your fleet, if you choose to blow them before you get flanked... good, you spend all your firepower onto ships prepared for that. If you choose the contrary... also good you get flanked and get blasted anyway.
Daveybaby wrote: e.g. in total war, there is no indication of the effects of flanking other than the fact that you tend to annihilate the enemy very quickly if you manage to do so, compared to a head-on attack.
Also in total war you can see wich one is the actual weak point of your troops (they are mostly rectangles with a front). Here can not be that evident. I could design my ships to have all weapons on the back and shoot you while I run.

Anyway all this discussion is meaningless unless we discuss if combat will be in ships or task forces. Then we can adapt it to have better strategy options for the player. Myself, I like the mixed aproach. Task forces but moving capital ships individually. See Jupiter: Nexus Incident for a good idea of space combat.

Ran Taro
Space Squid
Posts: 58
Joined: Wed Oct 06, 2004 2:51 am

#41 Post by Ran Taro »

Daveybaby wrote: This isn’t an AI issue, it’s a purely mechanistic, automatic adjustment of shield facings (important note: automatic does not involve AI). No decision making, just a few simple mathematical calculations per ship. A trivial amount of processing even for a 20 year old PC.
If it's a simple mechanistic adjustment, it's too unintuitive and exploitable. If involves complex calculations, it's likely to be either too exploitable, or unintuitive, or both.

In the former case, as has been pointed out, it isn't likely to produce the kind of tactical manoeuvring that is your stated aim. What it is likely to produce is players specifically designing the smallest, fastest, sneakiest, least damaging, most damage evading ship possible and getting very small numbers of these ships around an enemies flanks, just to spread their shields thin. At that point massed firepower from the front is just as effective as from anywhere else. So a core fleet doesn't have to manoeuver, or be very manoeuvrable. It just has to have maximum concentratable firepower, and a few specialist flankers. Doesn't sound like tactical mane overing to me. Sounds more like scout + heavy artillery is the single most viable tactic.

So the latter case is where you'll likely end up with this idea because people start recognising the problem inherent in your approach as stated above. So then you start adding other factors to the simple mechanism like the relative damage from each facing, shield recharge rates, balanced over shields, energy delivery systems, the square root of the inverse proportion of Pythagoras big toenail etc to the simple mechanism and it becomes complex. Then you realise that in different situations the premises of this mechanism need to be different to reach the correct solution. So then you need a branching decision tree... and now I hope you see where I think this is going.

It all seems pretty unnecessary when KISS tells you that you have a perfectly acceptable answer right in front of your face. You want to reproduce an effect that works well in Total War, why wouldn't you simply take the model that makes it work in Total War and adapt it to your use? It's fun to imagine a more complex delivery system, but if the complexity is extraneous, it can only cause problems.
Daveybaby wrote:If you think that this is anything that is going to tax a processor at all for anything less than tens of thousands of ships (in which case you would never be working things out on a ship by ship basis anyway, but by groups of ships) then you don’t understand how computers work.
I don't think I said anything about taxing the processor, and I'd appreciate it if you didn't patronise me. The argument 'If you think X then you are ignorant' is worse than useless, it's flame bait.

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#42 Post by Daveybaby »

What it is likely to produce is players specifically designing the smallest, fastest, sneakiest, least damaging, most damage evading ship possible and getting very small numbers of these ships around an enemies flanks, just to spread their shields thin.
Youre basically saying "i can do this to exploit the mechanism, therefore your idea wont work". But i'm saying - thats NOT an exploit because it can be countered.

Specifically, it forces me to counter it by, for example, building my own ships specifically designed to shoot down small, fast, sneaky ships (i.e. pack them with sensors and PD weapons) and using them to intercept your ships. At that point, your ships are dead to no effect, so youve wasted all that money. So maybe you take some heavy cruisers with them to take out my PD ships so that your small ships can do their job. And then maybe i send some cruisers with my PD ships. And so on.

Thats not an exploit, its EXACTLY the sort of effect i am trying to achieve. Its the whole point of my idea.
It all seems pretty unnecessary when KISS tells you that you have a perfectly acceptable answer right in front of your face. You want to reproduce an effect that works well in Total War, why wouldn't you simply take the model that makes it work in Total War and adapt it to your use? It's fun to imagine a more complex delivery system, but if the complexity is extraneous, it can only cause problems.
What i have done IS take the total war model and adapt it to my use. Its really no more complex than the total war system, i honestly dont see why you think it is.

And for the record, the total war system *is* vulnerable to attacks from the rear by fast, manouevrable, lightly armed units. Which is the whole POINT. The game actually specifically includes units to do this - light cavalry.
I don't think I said anything about taxing the processor
it adds too many variables for the AI to run
This to me sounded like somebody worried about either AI (development) complexity or processing capability - wasnt sure which so I addressed both possibilities (and just to reiterate, neither possibility is relevant). Note i said "if" as in "IF this is what you meant".
...and I'd appreciate it if you didn't patronise me. The argument 'If you think X then you are ignorant' is worse than useless, it's flame bait.
I never said ignorant, so i'd appreciate it if you didnt put words into my mouth - and if you choose to feel patronised by someone countering something you've said (or indeed not said) then thats up to you. Its only flame bait if you over-react.
The COW Project : You have a spy in your midst.

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#43 Post by Daveybaby »

Dreamer wrote:And you are still ignoring the actual damage you will receive when diverting shields. I would put a single laser on my planet bombers to divert your shields and blast you from the front with reduced shield efficiency.
I'm not ignoring the damage. As i've said above, its the whole point of the system. If your bombers are going in harms way just to try to weaken my shields then good. BUT - all that time theyre not bombing the planet, which is kind of a waste.
And belive me, my bombers can hold quite a punch, they need to get to the planet alive to drop bombs after all.
But i assume theyre not indestructible.

The whole point is, youve sent your (expensive and mission critical) bombers out away from the protection of your capital ships to try to flank me and weaken my shields. Thats good, it may work. But maybe i've got some ships specially designed to take out your bombers with minimum fuss (unless of course you have some special magical fantasy design which is indestructible :wink: ) at which point theyre all dead. Or maybe they'll get through after all and weaken me enough for you to win. And maybe you will still have enough left over to successfully bomb the planet. And maybe you wont.

Maybe you will design a different ship for flanking, packing lots of stealth and heavy shields. If i dont have enough sensors on my ships to see them coming i will be dead meat. But *if* i detect them then your (again, expensive) ships might be easily picked off because they will be away from your main fleet.

And on top of all of the above, dont forget that i will be busy trying to flank you. And you will be trying to counter my flankers.

Choices and decisions. Risk and reward. Sounds like fun to me.

Diverting shields automatically will only call for "distractor ships". I make some very thought ships with lesser weapons and some weak ships with a lot of weapons (all space spended in weapons). Then I move my distractors into your fleet, if you choose to blow them before you get flanked... good, you spend all your firepower onto ships prepared for that. If you choose the contrary... also good you get flanked and get blasted anyway.
Theres an implicit assumption here that you are the only one who can design ships for a specific purpose, and that another player will be powerless against them. The whole point is that i can then either design ships to specifically to interccept and destroy your distractors (which, incidentally, still cost money/production to build that hasnt been spent on the rest of your fleet), and/or use tactics to render them less effective, and you will then have to evolve your own designs and/or tactics to counter my counter and so on. This, to me, will lead to interesting gameplay.

Anyway all this discussion is meaningless unless we discuss if combat will be in ships or task forces. Then we can adapt it to have better strategy options for the player.
I disagree - IMO this sort of discussion can be used to drive the scale and design of space combat. We can say "what gameplay effects to we want to achieve?" and then decide on a scale that best fits it.

In reality of course its a much more messy process than that, but getting ideas like this out for discussion will help drive things to an interesting resolution. To me, its important to get interesting tactics into space combat - flanking is one way of starting to achieve that. It doesnt HAVE to use the shield facings method, but its a starting point for discussion.
The COW Project : You have a spy in your midst.

guiguibaah
Creative Contributor
Posts: 441
Joined: Tue Oct 07, 2003 1:00 am

Bombers and Flankings

#44 Post by guiguibaah »

- Of course, you could have ships that are excellent bomber-killers, like the multi-gun corvettes in homeworld (or if you need to skimp and are on a budget, any light corvette would do).

- Or, if we are going with the less is more approach to ships, you design your ships like the assault gunships in Freespace 2 - 6 flak cannons on each side and two missile systems - ensures that no bomb will ever touch that ship - or other ships nearby for that matter. Of course, it would be weak against a long-ranged beam destroyer or something.
There are three kinds of people in this world - those who can count, and those who can't.

Dreamer
Dyson Forest
Posts: 228
Joined: Sun Mar 13, 2005 6:44 am
Location: Santiago, Chile

#45 Post by Dreamer »

Daveybaby wrote:To me, its important to get interesting tactics into space combat - flanking is one way of starting to achieve that. It doesnt HAVE to use the shield facings method, but its a starting point for discussion.
Completely agree. In fact that is why I didn`t only complain about your idea, but instead proposed a variation to archieve the same or similar combat tactics. I`m all for discussion :D

Still I think that this particular system to arcive strategy could be a lot more messier than necesary (for the same final result witch is flanking and other strategy).

How about ships capable of proyecting shields? More than an individual defence you could use it to create a sort of "terrain" in space combat. I`m thinking about a more complex shield use for frigates to doomstars and a less complicated schema for lesser ships (and group this small ones in task forces).

Post Reply