Solo Play / Co-Op mechanics

ManicMan

ManicMan

Lord
Okay, while there is a bit about this every so often, I decided I wanna bit of talk on this ^_^ Basically, how to solo-play games.

I think some Solo-play rules are basically no different from 'Co-Op' rules, which is when you kinda play as the 'good guys' with the 'bad guys' being controlled by all players often via a chain rule set.

now, For dungeon crawlers (like HeroQuest) this is pretty easy. You already have alot of the chain set up in that you have squares for movement, you have stat cards which say how many squares you move etc. All you really need is a basic extra rule set which basically control who they go after (do they move towards the closest player, the weakest player, etc). If you go a bit more advanced, you can factor in when best to attack and when to retreat, or even factor in the enemy Mind points (or whatever the game system has) to decide how 'smart' there actions are.

For Skirmish games, it becomes a bit harder. By Skirmish I'm saying that a player controls a small group of figures, more then just a single miniature. In theory you get the same kinda simple method used for Dungeon Crawlers but Skirmish often doesn't use a square based system but swap to a measurement based system, mostly using inches or millimetres. things like Single page rules (which are... to be printed on pretty large pages these days) use a decision tree method which CAN work okay, but of course, you end up with alot of strategy being lost as the rules as basically 'if enemies between you and your objective, Attack them if you can, else run towards your objective but that often feels.. kinda hollow at times.

Then you get Army combat. This is when you have a bunch of units under command. It follows on from Skirmish but gets even more complex cause in theory, you have the interplay of unit vs unit to deal with.

Really, I wanna work on a Solo-play system which can kinda be 'plug and played' into whatever system (though I'm not aiming to make one any time soon). My approach is from a video game coding point of view. I might not have any 'hits' under my belt but I have released a few games over my time and I very much have knowledge of various systems which include enemy control. If you look at some of the earliest games like Space Invaders, you have pretty much VERY dumb Enemy Logic.
Code: 
10 Set Direction = right;
20 Move direction;
30 if (hit path end right) then Set Direction = left else set direction right;
40 move down one;
50 Move direction;
60 Speed +=1;
70 go to 20;

This system was used for MANY MANY years and is still used to a degree these days. Most platform games basically use it. While not the first, we have Pac-man adding in a major overhaul. Each ghost (given a name which is just flavour, and a nickname which is based on there own logic) was given slightly different logic.
Red Blinky (nicknamed Shadow) is designed to directly follow Pac-man through the maze in normal 'chase' mode, 'head to the upper right in 'scatter' mode (when ghosts turn flashing) and will speed up based on the number of dots left in the maze
Pink Pinky (nicknamed Speedy) targets a dot which is 2 spots infront of Pac-man, thus heading him off. in Scatter mode heads to the upper left.
Blue Inky (nicknamed Bashful) targets using a bit of maths based on Blinky and Pac-mans relative distance from each other in Chase mode, heads to the bottom Right in Scatter mode.
Orange Clyde (nicknamed Pokey) heads toward Pac-man in Chase mode UNLESS he is within 8 dots of Pac-man, then he heads to his scatter target, which is the lower left.

well.. that's about right.. there are variations of course, but that givens each 'unit' a different characteristic and they play differently (you can always tell a poorly made Pac-man clone cause they don't have different characteristics.

ALOT of games still use that logic where different units have a base line code but then have tweaks based on characteristics, so you can have them function differently.

Going back a bit, I think the Computer versions of HeroQuest, used the first code. each enemy acts in the same way. There is no real characteristics based on unit type. Which I think it's pretty fine for Dungeon Crawlers, it's just when you get to more advanced games like Skirmish and Army, you really need something different. You need the pac-man style to have logic tweaks.

Oakbound's Thunder under mount Silence uses 3 types of Chain logic, with different unit types being one of them three. This gives some variation without being overly complex for a dungeon crawler. The role are 'Marauders' which do what they can to rush into combat, 'Guards' who will guard over a target and mostly stay close to that, and 'Sentries' who patrol the board. There is also a 4th 'Agent' which is a lot more loose and is a catch all for any special case.

Like I said before, One page rules gives a system for Skirmish games but I think it's a bit too simple but also defines units under one of three types (this one based on weapons (Melee, Ranged, Mixed).

As I play it a fair bit, Zombicide (a more modern system) uses a very VERY basic system. Enemies always hit, always move towards line of sight or 'loudest' target (when you fire a bullet based weapon, you make a noise, each character is a noise, etc). the odd special rule for some special characters but that is about it. very automatic with little variation as the game is more about skilful players vs brainless hordes.

So.. lets discuss various solo-play options, the good, the bad, everything in between ^_^ maybe what you think a solo-play system should have or want it to have. Problem with any solo-play system (though sometimes, maybe not a problem) is you mostly have the option of deciding of the enemy goes easy on you or not.
 
yep. very different. If you wanna really score high on the game, you need to know that ^_^
of course, a game like Pac-man was never... finished (there is no winning state) and for high score-ers, it's not so much the score you get, but level you get to. Due to the variables for data storage, there is 256 levels in 'normal' play, before it loops around, but when it loops, due to the data corruption with the variable resetting, you only have half the screen visible, the other half is very corrupt.. But it's still playable and how many levels you can clear like that is a BIG thing. Anyone that just boosts at their score with pac-man as an achievement isn't that great.

Personally? I'm not too good at it ^_^ depends on my mood, maybe a good number of levels in but with each level being the same just getting harder, it gets a bit dull unless you are in the right atmosphere. I kinda perfect more story based games.
 
In my WarpStrike skirmish game (see my sig), random tyranids or necrons pop up on the board for every twelve '6's rolled during the game (and several blips start out one the board at the beginning of each game). I gave each 'blip' unit a behavior, which basically comes down to "target the closest unengaged model". Hormagaunts will try to double up on opponents, while genestealers go solo. Melee-oriented units will try to melee, while ranged units (like gargoyles or ranged-weapon-equipped Tyranid Warriors) will try to get within short range and blast away. (A fun trick is to get a blip to flip when it's closer to your opponent's troops than to your own. :grin: Scanners can help with this.)

Horde Mode has slightly more involved rules, with Horde units moving toward the nearest visible player unit, then the nearest visible Objective, and then the player's table edge. The Horde deployment zone is divided into a number of spawning zones equal to the player army size divided by 500. Random Horde units spawn at the beginning of each Horde turn, and then follow the 'behavior algorithm.' It's been designed as a co-op game, but they have rules for single-player vs the Horde as well (double the supply point gains and discard/redraw and objectives that require interacting with your ally).

Here's the 11e Horde Mode AI algorithm:
5.1 The Horde utilizes all datasheet abilities it can.
All datasheet abilities should be utilized at the appropriate time if possible.

Certain datasheet abilities do not function due to the lack of Horde stratagem usage, army rules, and detachment rules. (Khal, Sisters of Battle, and Farseer have example abilities.) Ignore these.

Abilities that require you to choose a mode (Belisarius Cawl, etc.) are up to group discretion. You can either choose the option your group believes is the best for the Horde, or choose randomly.

If they spawned this turn and did not make an ingress move, Horde Units with Scouts may add their Scouts distance to their move characteristic.



5.2 During the movement phase each Horde unit attempts to move directly toward the following, in order of importance:

The closest visible enemy.

The closest visible objective not controlled by another Horde unit, ignoring any objectives in the Horde Spawn Zones.

The Defender’s board edge.

Designer’s Note: This means some Horde units will camp objectives if not baited away by visible units.



5.3 Each unit moves if appropriate. Exceptions are as follows:

FLYING units take to the skies (See Core Rules 21.03) if doing so would allow the unit to get closer to their movement target (See 5.2).

Units where the majority of models are equipped with [HEAVY] weapons and an enemy unit is visible and within range of the weapons move a maximum of 3" to use the [HEAVY] ability.

Single-model units whose primary ranged weapons have the [HEAVY] ability and have an enemy unit visible and within range of those weapons instead move a maximum of 3” to keep the [HEAVY] hit roll bonus active.

Units that are within shooting range of a visible enemy unit and will lose some significant bonus from abilities if they move remain stationary.

Units set up within a Spawn Zone from strategic reserves always move. (Even though they entered from strategic reserves. The Horde always wants to clear the Spawn Zone.)

Units within a Spawn Zone always attempt to move out of all Spawn Zones, regardless of other rules.

Designer’s Note: Move unless a unit has no reason to ruin their shooting bonus. Never purposefully stay in a Spawn zone.


5.4 A unit makes an advance move if doing so is advantageous. An advance move is considered advantageous if it meets any of the following criteria:

All models are equipped with [ASSAULT] weapons.

The unit has the ability to shoot or declare a charge after making an advance move.

The unit could not shoot at anything after making a normal move and has no reason to declare a charge. (See 5.7)

The unit could not attempt a charge anyway and the only ranged weapons that it loses access to are Side-arms.

Side-arms are defined as: [CLOSE-QUARTERS] weapons on units with melee loadouts better than close combat weapons. Ranged weapons on attached characters are also considered Side-arms.

(E.G. Infractors should not slow down to shoot pistols. Lychguard should not slow down to let an attached character shoot a Staff of Light. etc.)



5.5 Units inside transports make disembark moves when appropriate:

If there is a visible enemy unit within range of the embarked unit’s ranged weapons and there is not enough Firing Deck X on the transport to fire all weapons (except Side-arms) (See 5.4)

If a disembark move would allow a unit that would attempt to charge this turn to declare a charge. (See 5.7)

The unit will not make a disembark move that ends inside of a Spawn Zone even if the above are true.



5.6 In the shooting phase each model shoots the closest legal target (select randomly if more than one are equally close), with the following exceptions.

If a model is equipped with an [ANTI-X] or [X HUNTER] weapon it attempts to shoot the closest visible X keyword unit. If none are visible within range, default back to the closest legal target.

If a model is equipped with a [PRECISION] weapon it attempts to shoot the closest visible unit with a Leader or Support model. If none are visible within range, default back to the closest legal target.

If a model is equipped with a [MELTA] weapon it attempts to shoot the closest visible non-INFANTRY, non-SWARM unit. If none are visible within range, default back to the closest legal target.



5.7 In the charge phase, each Horde unit charges if the majority of the unit has a melee weapon with a profile better than their faction’s Close Combat Weapon or Armoured Hull equivalent, or is only equipped with Side-arms (See 5.4).

Designer’s Note: This should stop things like tanks, guardsmen and T’au from charging.



5.8 When resolving charge rolls for the Horde, the horde attempts to charge the closest legal target. The Horde models move as far as possible while making charge moves to maximize the amount of Horde models in engagement range at the end of the move.



5.9 Horde models always attempt to pile-in to the closest Player model’s unit, as far as possible.



5.10 Engaged Horde units remain engaged unless they are eligible to shoot or charge after a fall-back move. If they would not charge again (See 5.7) they fall-back move the maximum distance. Otherwise, they fall-back move 2”.



5.11 If possible, Horde models attack the closest Player model’s unit in the Fight Phase.

Designer’s Note: This means sometimes Horde units will split attacks on a model-by-model basis to get the most attacks in as possible.



5.12 Horde units always make consolidation moves if possible.
Some of these directions clearly don't work with older rulesets, but it's not too hard to adapt them.
 
In the 2E codex, genestealers spread ahead of the hive fleets to weaken defenses and guide the fleets to easy pickings. They still create the cults and hybrids and such. By the time you get 4th-generation stealers, the cult is usually about ready to rise up in rebellion. The Magus is born from a 3rd gen hybrid, and when it matures, the Patriarch and/or Magus generate a psychic call that attracts the tyranids.

When the hive fleet arrives, if the cult is still around, it gets absorbed into the hivefleet along with the rest of the biomatter on the planet. The genestealers themselves are probably incorporated into the tyranid forces, but all the cultists and hybrids get recycled.

Back on topic, given the vast range of abilities and load outs the various armies can take, really all you can do is give general guidelines.
 
Of course, I came in at the very beginning of Second Edition, so for me, that's just how it's always been! :wink: Necrons have changed lore several times now, though, from what I understand. And Custodes used to just be the Emperor's personal bodyguard. Now they're their own faction of Super Marines.

I just remembered another sort-of co-op game. 1000 pts of IG and 500 points of Marines against 1500 points of whatever. The Space Marines deploy in a 24" area centered 2/3s of the way down the long-edge of the board. The other force deploys along the short edge closest to the Marines, and the IG deploys along the other short edge. The marines then need to hold out until the Guard can arrive.

More of a specific PvP scenario than co-op rules, I guess.
 
I've often wished I had the time to properly figure out some good solo rules for many of our games. Either just good rules or tie it all into a good bit of software. In terms of rules I've often heard good things of Adjutant Introuvable as a control system, but I've not tried it myself. It was produced (I think) by someone from the https://lonewarriorswa.com/ site.
 
My own game I've been working on slowly (main problem is while it is designed for solo play with multi-player options hoping to come into it, I can't just keep testing it myself, and don't really have many people I can ask to test it.. might have to make some hints ^_^.. also, Mission designing.. I did a first test one which plays well.. I did a second one which... I think is having problems which just REALLY slow it down. Need to come up with atleast two more mission drafts to use to get a really good test to see how well the system does)

erm.. my own game has the 'enemy' being controlled in a very basic way but it works well (I think) for what the game is. It mostly uses types and flow charts. NPC's there main type, then have one of two (at the moment) of subtype (Patrolling or Guarding), and then depends what mode they are currently in within the current game (It's a stealth game, so they can be Alert or unalert). Though the flow charts are kinda small does there isn't really a whole lot they can do ^_^
 
ManicMan said:
Though the flow charts are kinda small does there isn't really a whole lot they can do
Click to expand...
That sounds like for the best. Constraints are likely beneficial because it is easier to define the challenges & how they can be overcome.

I am of the mind that the fewer rules you use the better. Solo rules for Space Hulk are relatively simple. Genestealers can only attack in close combat. The board restricts where they can move. Their goal is relatively consistent - kill the terminators.

What would I want from a solo war game is that it be intellectually stimulating in the same way as solitaire card games or newspaper puzzles.
 
ManicMan said:
It mostly uses types and flow charts.
Click to expand...
Seems everything in game AI boils down to Monte Carlo tree searches in some fashion :) I wonder if you could adapt an existing "war game" (not like they haven't been around a good while for "warhammer") engine to work well with a physical game, the nuances of feeding in the data I guess would be the biggest pain. Beyond some very basic familiarity with the broad techniques it's not something I've ever really had to delve into.
orkwrangler said:
What would I want from a solo war game is that it be intellectually stimulating in the same way as solitaire card games or newspaper puzzles.
Click to expand...
I think that's a good nuanced point. I'm much less bothered if I win or loose a game, it's much more about how much fun it was or how stimulating it was. Probably why I prefer the older more story focused approach of everyone here.
 
Too many of the old warhammer computer games were designed to be 'nothing like the miniature games' (I've tried to play Shadow of the Horned rat... but I've not been able to get in to it with all the changes, like making it real-time.. and then reading the intro of Citadel about how they didn't want it to be just a video game version...

But HeroQuest was very faithful. Space Crusade was... I think faithful to the game but not the missions. I haven't played it for years but I remember Aspects of War being very faithful (it was Turn based with the same rules and stats from the miniature game, yay)... for more modern ones? no clue.. would be more faithful to more modern rules. I'm sure I remember A version of Space Hulk which was faithful but I've yet to be able to find it again. They like to make it real-time ¬_¬;
 
I don't think I ever actually played any of the warhammer ones. Space Hulk yep - played a lot of that computer game. My "battle game" of choice around that era was Myth The Fallen Lords. However I suppose I was more meaning that mass battle "AI" systems have existed for a long time and I wonder how much of that could be leveraged for a fully automated opponent.
 
mm.. the main problem is when it comes to terrain. and re-actions. If it's turn based, it's easier but still have the problem of keeping in the right in-forces while allow it to change to suit the players tactics. There was a nice article some time back which I like to re-read once in a while cause it's useful, as such things are, from a small time games company where one of the rules of horror games is the player is the enemy. They will find exploits to remove the 'fear' cause that isn't a positive emotion. It's the same with many things.

People find the game too hard? They were notice and discover the games weaknesses. The great game 'Trespasser', at the time had alot of problems cause it was kinda too advanced for it's own good. For one thing, the Dinosaurs you were up against couldn't enter buildings. They wanted to allow them but the small corridors and rooms created problems with the path finding, and with the time they had left to make the game, they had to just disable that feature. So now you have the situation that instead of hiding, running, using what resources you have, you can just always enter a building, jump up and down in front of them saying 'nah nah, you can't catch me'... mmm.. come to think of it.. as the game is more based on the novel (it's a Jurassic park game), even with the movie tweaks, hiding in the building while baiting raptors is very much what did happen in the book. Inside the Visitor Lodge (which didn't appear in the film apart from a map reference)... Though that kinda lead to the death of Doctor Henry Wu... always keep track of how many raptors are there when you are trying to play cat and mouse with them.

anyway. ALOT of video games, people hunt for the exploits. find way which cheat the system. Due to how a system is kinda fixed, there will always be methods. Speed running (which is something I really don't like at all) shows us how people don't care about the games and find what ever bugs they can to skip all the work so they can get the shortest completion times.

Making a physical version which doesn't end up having exploits which completely ruining it for anyone that noticed them? not too easy. I've played some which use pretty nice adaptive systems, but in the end, they get pretty boring because of the adaptive system. They become very predictable, even if they are meant to change to different ways you play.

And sometimes, you end up with the mess that was 'Alien Side step'.. There is only one way to win the game at all, and it's just all about doing the same thing again and again and again and it's dull.

That said, it's like Chess logic. Some computer systems are very complex, but only work because there is such a limited number of moves you can make. For example, there are 21 opening moves in chess. Only 21. (some people don't think resignation is a move. To that I just say is "A strange game. The only winning move is not to play." [pretty much my favourite movie (though it's close cause there are a couple of others which rotate with it), and the only movie I own on Laser Disc cause, I thought it would be cool ^_^]. Not playing is a valid move, but it's not an opening move. Resigning is.

Anyway, you would run into the problem with re-enforcing the enemy (computer player)'s objective. If their goal is to say, capture your base's flag, then they need to be told to pretty much always head towards that goal. But in real-life, sometimes you need to retreat to regroup. Working out when that is best (I guess, checking the size of player forces, and strength, compared to enemy forces, checking if regrouping would help etc) and when to stop doing that (sometimes you need to feint or you can need to change strategy half way though doing the action.

It's not really like the original star wars films which had Stormtroopers working to a very simple logic. "If out numbered, run for help. If have superior numbers, attack" with the added "If boss says attack and will kill you if you don't, attack".
 
I only really have a model for WFB solo, and it relies on interpretation of a "vibe" description after a quick RNG vs stats check (run through and looked up on excel) rather than a more clear cut if-then direction, but FWIW..
Every unit gets an entry on the sheet with its Int/Cl/Ld, basically to see how Smart (Int), Controlled (Cl) and Brave (Ld) its actions are that turn. Run a quick 2d6 RNG check against each stat, giving 3 Y/N pass results, for a range of results from YYY through YNY (etc) to NNN. Each of those possible combinations has a name, so YYY is Inspired, NNN is Disastrous, YYN is Smart Defensive (since the Int and Cl were a Y but Ld was a N), NNY is Unwise Aggressive (failed Int and Cl but passed Ld) and so on.
With every unit's 3 stats entered and the RNG and resulting Y/N string spitting out the final lookup text response (eg "Inspired" or whatever), it's very quick each turn to just refresh RNG and have the 'vibe' for each unit's actions that turn spat out without having to consult flow charts or roll lots of individual checks for every unit.
Where it falls down relative to what you seem to be after is that you have a description of how smart/brave etc the action is, but how you interpret that is down to the solo player.
Typically I'll treat Inspired as the unit is allowed to peak into my head as if an opponent had figured out my plans and react in the most intelligent way possible, while Disastrous would be something like moving in the way of a friendly unit's attack route, blocking missile fire, ignoring that obvious flank attack heading its way, picking a terrible fight - the options are very varied and situational. The more nuanced results - say the difference between Smart Offensive (YNY) and Unwise Aggressive (NNY) might be in the former the unit ignores an obvious bait offering and heads off towards a more strategic goal (war machine, flank, objective), whereas in the latter it will just charge the bait unit.
It works for me as it's very quick to generate the results but it obviously doesn't give specific actions - personally I prefer that aspect.
 
Oh, and a word to the unwise :)
Failing the Int component doesn't mean stupid as much as it means lacking strategic thinking. If a unit has a good charge option, pretty much any result will probably see them take the charge - the difference is that if the Int component is passed, that decision is made with reference to the wider strategic position, as if the player has though about it, while if the Int component was a N, the same charge probably happens, but as if the player hadn't considered the wider picture - this difference might change how you viewed a resulting pursuit / hold actions that turn though. A Disastrous result might mean not deciding to declare a charge into an easy combat (narratively think waiting for backup or reforming rather than seizing the advantage), or it could mean charging when they really shouldn't.. That Int result for the first Y/N bit of the result is the most important thing, and this does skew this model into giving better outcomes for high stat units, which arguably is narratively satisfying - should I expect an NPC Orc or Skaven army to act as consistently as a High Elf one? Anyway, enough jibber jabber from me on this.
 
Those extra psychology stats are pulled from WFB 3rd ed - I play a homebrew of mostly 6th ed, but there's all kinds of other DNA in the mix (3rd, 8th and TOW in particular), so I'm happy taking the occasional punt for units that don't have a direct parallel in 3rd ed. The actual spread of values in those stats is pretty narrow tbh, so having to make them up from time to time wasn't an issue for me.

Deployment is a bit convoluted tbh - I do initial deployment for both armies using a 'dice out of the bag' selection method (coloured dice in bag equal to number of units in each army, pull out dice in colour of either side A or B, place one model from a unit marking it's front/centre location on table, rinse, repeat) and basically try to deploy in a sensible way as if I were that force. Once that's all done the solo element kicks in and the NPC side rolls a 3d6 check against Ld+I - each point of success allows repositioning of one unit (max of half units in army). Before repositioning any NPC units I then run that Int/Cl/Ld check against the general's stats and use that result as the basis for ALL that side's repositioning - again, the interpretation of that is incredibly situational and so hard to generalise about, but with a specific deployment and result in front of me it's generally easy to see to how to make that deployment more eg Unwise Aggressive (say, move a screening unit, or push a unit further forward).
Once that's all done, fully deploy the units at their marked points, and it's done.

The 'dice out of a bag' is the same deployment system we use in normal play too, less the option to reposition units - I just find the alternating deployment too mechanical and it gives quite an edge to armies with more units - it always struck me as counterintuitive that a large unwieldy horde army can effectively outmanoeuvre a smaller force when really the reverse should be more likely. It can still happen by the power of randomness, but at least it's not nailed on, and you do get those games where you pull one of your own dice out 4 times in a row and have to double check that your opponent added their dice.. :)
 
You must log in or register to reply here.
Back
Top