Vapourware Zodiac Legion - X-COM and dungeons

[Jinn]

I prefer numbers and bars over tokens pretty much any time. Much more quick to register their actually quantity and meaning for me than a stack of tokens, even if there is a numerical representation next to them.

 

[Galdred]

I am currently reworking the pathfinding in the game:

Previously, pathfinding was just dijkstra (to know the distance to all reachable tiles), and A* (to get the quickest way from A to B).
However, both ignore the fact that the game is turn based; which has two important consequences:
The first one is that the cost may differ, if you wase movement points because you needed 4 to cross a swamp tile, but you only had 3 left for the turn.
The A* algorithm will just ignore it, and count that as a 4 MP tile.

More importantly, choosing where to go is a decision for each turn, that depends on more factor than how fast the unit can reach its final destination:

So each turn, I would like to balance the following factors to choose the destination of each AI agent for the turn:

1. **The tile must be within movement range** (obviously). Unspent movement points provide no benefit, so the movement cost itself is only a binary value. Some terrain types cost more than other.
   

2. **the risk of getting there** (going near enemy can trigger attacks of opportunity, walking into enemy line of sight can trigger overwatch fire)



3. How closer to its long term objective



4. How good the position is to attack/hamper the player team.



3) is got easily by running dijkstra from the objective, using whatever combination of movement cost + risk is appropriate.

4) I compute this using an utility function

My problem is combining a) and b)

I first thought of running dijkstra with a cost limited to the movement range, and only use the subset to run another dijkstra with risk.

But if I do that, I risk having paths that could be reached within the movement range, but will take longer to get to if I follow the path that factors risk.



So I would need a function that does the following:

For each tile in movement range, gets the minimal cost/safest path to get there without spending more than M movement points.



Is there a "standard" way to do so, assuming the number of movement points is relatively low (12 or less)?

And my own answer to my question (it came to me shortly after having written the question):

I was thinking about running a floodfill algorithm, with the following alterations: Each tile would store a table instead of a single "cost" value.

The table would be store:

[Mv_Cost_To_Get_There] = danger level

Starting with the origin tile, we visit each neighbour, and to adjust their cost/risk matrix if we find a better way to get there.

Each neighbours for which any table value has been updated would then be re-added to the queue.

Whenever a tile is added to the queue, we also add the first "mv cost index" modified as priority (lowest goes first).



For instance, we now visit tile (2,1).

Its matrix is:



MV Cost Risk
1 False
2 False
3 4
4 4
5 3
6 3
7 3
The cost to get from tile (2,1) to tile (2,2) is 2

tile 2,2 is as follow:



MV Cost Risk updated Risk
1 False False
2 False False
3 False False
4 6 6
5 6 4
6 5 4
7 5 3
If (2,2) was already in the queue, we check its priority index: if it was higher than the first we changed (5), we update it to this value.

Otherwise, we add (2,2) to the queue with priority 5.

I should probably add a column for the previous visited tile to the table.

The safest move will always be the one that expands all movement, so we can just take that and add this: "danger cost parameter" to each tile when deciding to which of the tiles in range to move.

 

[Galdred]

Another thing that is being reworked is the way Zones of controls and attacks of opportunity work:

I am not a big fan of free AOO, it makes movement too limited in combat, and makes taking perks that avoid AOO mandatory (Battle Brothers is a good exemple of that).

I like the way it is done in D20 actually, but I don't like the interface implications of having to choose between using all of your move for a 5 foot step, or trying to pull back and eat an AOO.

Also, it should be relatively easy to pull back when supported by friend, and very hard when surrounded by enemies.

So:

Zones of control:
They work a lot like in Blood Bowl:
characters that are adjacent to a tile, but not in someone else's zone of control provide support for or against the player in the tile (or about to move to the tile).

Support also helps in melee, and works almost the same way, but some weapons (polearms) provide support 2 tiles away.
Defensive support requires either to be in melee range of the attacker, or to be adjacent to the target, and at the same range, or less, from the attacker.
Defensive support also requires a large shield to work (if not in melee range of the attacker), but can work against ranged attacks.
Some perks also allow to give offensive or defensive support even when in an opponent ZoC.

Now, the way it interacts with AoO:
When a player moves away from a tile in an opponent ZoC, he needs to pass a test (melee or agility), whose difficulty depends on the difference between offensive and defensive support in the tile he wants to reach, and the melee skill of the best enemy in melee range of the origin tile.
If he fails, he remains in the tile, cannot move anymore for the turn, and gets an "imbalance" debuff that will make hitting him easier during the opponent's turn.

 

[Galdred]

 

[Galdred]

Btw, regarding the current state of the game, I spent a lot of time reworking the UI.
I still need to add some combat preview, and some non obstrusive tutorial for a few things (flanking, reinforcements...).
Then work on some levels, and it should be good enough for a linear demo.

What tutorials would you recommend me to "research"?
I usually prefer to go completely blind than to endure some more explanation of how to use the Middle Mouse Button to rotate the camera and such (and I miss this time we used manuals instead of suffering through these tutorials...).

 

[Darth Canoli]

I usually skip the tutorials altogether, games like Wizardry 8 never had any tutorial, just a manual you should read and everyone got it.
It'd be great if the tutorials were optional.

Of course, if your target audience is the dumb masses, the middle mouse button tutorial is mandatory.

 

[Galdred]

Actually, I'd rather just skip the middle button tutorial, but I'd like to explain how flanking, actions, zones of control, attack and defense support work in the game so that it doesn't feel like clicking things randomly when trying the game

 

[Darth Canoli]

Actually, I'd rather just skip the middle button tutorial, but I'd like to explain how flanking, actions, zones of control, attack and defense support work in the game so that it doesn't feel like clicking things randomly when trying the game

What about a couple of pages in the manual ?
Or (and?) a couple of static pages in-game explaining how it works ?

 

[The Red Knight]

In-game help/manual like this one should be enough: (@0:09:30)

(paper manuals were fun to go through but reading books on computer screen feels like a hassle so I tend to skip on wordy pdfs unless I encounter some problem or want to find more detailed information on mechanics after figuring out what I could myself, and pressing a button to open it in-game is preferable to alt+tabbing between the game and the manual)

You could add to that a few dismissable popups with a few lines of text and a picture or animation if you need to communicate specific mechanics or less obvious parts of the UI: (@0:04:50, three popups in 2 minutes)

 

[baud]

What The Red Knight said. Also making an in-game manual accessible from both the main menu and in-game could be a good idea. For the manual, don't forget pictures to visually explain the concepts.

Also, another idea, since it's about gameplay concept: the first time (first few times) the flanking, zones of control, attack or defense support occur in-game, have a small popup appear in a corner to explain. Perhaps even make it clickable so that it opens the in-game manual at the right page. And provide an option to remove them.

 

[baud]

 

[Galdred]

You missed the polemic about the former teleporter design being the same as the SS black sun of castle of Wewelsburg Castle!

But the artist changed it immediately after being told about that by someone on Twitter.

The sad thing is that it was probably just a "cool ornament" that had no particular significance (it was inspired by merovingian decorations), but is has become a symbol afterwards.

 

[Galdred]

I am adding a few power effects (and UI templates) to the game:
Knockback (both for melee attacks and telekinesis), AoE Knockback (both as a "air blast" spell, and as add on to spells, like enhanced fireball).
I now need to add friendly targeting (for telekinesis and in case you really want to push your comrade 1 extra tile with a large hammer swing to the back. We did that a lot in Descent: Road to Legend in my group, and it was a fun risk to take).
Given that Telekinesis will have a "push character" and remote use skills (like lockpicking, or activating a lever), I think "force choke" would be in order (I needed a damage spell that would ignore armor anyway, so it would work).
Then I think I will be almost done with the effects (I'll still need to add Teleportation, a "phase attack" or whatever you call the anime effect that makes a character ends in someone back after a strike), a stun, and the exposed effect (when you fail your check to move away from an opponent), then I will move to content and geoscape.

I also need to add combat preview to the UI (and an UI tour to give some explanation about the controls).

Regarding content, I need to add some more opponents (currently, there are skeletons legionaries and auxiliaries, and armored revenants).
Then make the level easier to build (it is currently a pain to do them with Tiled, and I don't plan to add a custom level editor).
I will also add some more environmental props (traps mostly), and maybe short range teleporters.

As for Covid, it doesn't affect me too much direclty: l was already working for home and avoiding human interaction before. But it is much harder to work from home with kids.

 

[Zombra]

I prefer if you can choose [initiative] order yourself, so if I were to make it stat based, I would base it on the commander's tactics skill, and not on the individual units.

Cool idea! I envision the interface asking me "What do you want your team's turn order to be?" Then I set my preferred marching order, then based on my Commander's Tactics roll, it will scramble the order I set (a perfect success meaning it gets to stay exactly how I set it).

 

[Galdred]

I prefer if you can choose [initiative] order yourself, so if I were to make it stat based, I would base it on the commander's tactics skill, and not on the individual units.

Cool idea! I envision the interface asking me "What do you want your team's turn order to be?" Then I set my preferred marching order, then based on my Commander's Tactics roll, it will scramble the order I set (a perfect success meaning it gets to stay exactly how I set it).

I am still unsure about whether I should change the initiative system. Currently, the prototype favors alpha strike a little, but given that most of the combat happens at melee range, going first usually doesn't mean that you'll be able to sweep a room.
Also, tanking doorways is relatively efficient until I can teach the AI to telekinesis pull the defender in the middle of its guys (or to AoE ), or to knockback him away.
But if I move to a tactics based system, you would have the turn order set as you do at the beginning of the mission, but a command check would provide command points that could let you :

swap some of your characters in the turn order, delay some of your activations (relative to the opponent), or chain some (instead of alternating them).

I think it's bad but they had to do it because their soldiers have no attributes, they had no choice.
Please, get a full character creation system with attributes and perks on top of combat skills.

Why not taking a look at the Colony Ship demo for some inspiration instead ?

I need to take a look at Colony Ship indeed. I liked the way weapons differed in Dungeon Rats.
Regarding attributes, I'm trying to find a balance so that they make it easy to determine who's good at what without having too many of them.
I am aiming to have something relatively similar to Blood Bowl. Currently, there is:

HP, Movement, Stamina, Melee skill, Ranged skill, Will/Magic skill, Evasion, Armor, but most of the stats/skills that only serve for a small number of characters/situations exist as perk (for instance, a tower shield provides a Ranged Defense perk, and a Defensive support perk).
But what I dislike about initiative rolls is that you have little input on who goes first (you can delay action, but I don't really like this mechanism, especially on large maps).

My problem is that initiative makes even less sense in large maps (where not all characters are present in the same room). Also, the goal is to de-emphasize alpha strike, not to tie it to some stats.
Also, I prefer if you can choose the order yourself, so if I were to make it stat based, I would base it on the commander's tactics skill, and not on the individual units (and have the leadership roll affect how "slots" are shared between both teams).

Well, you have to pick a system that makes sense for your game, as long as it's not the DoS/Chimera system, it'll be fine.
Also, as everyone said, i prefer the turn per team system for huge groups (Fantasy General/Warbanners style) then again, if you think something that works better for you, it'll probably be fine once vetted by monocled beta-testers.

It worked rather well in Final Liberation actually (and its TT inspiration, Epic 40K), but you could choose each turn who would go next (and there was a tactics skill to retain initiative, and maybe to skip turn in Epic: Armageddon).

I really enjoyed the multi-player mode of Final Liberation (maybe because i was nuking everyone ... Ah, good times ... ) and yes, it worked well but everything has to happen lightning fast if you choose this path.

Indeed, things used to die quickly in Final Liberation. I have seldom played the SP, but I remember having 100% victory with the Imperium, and 0% with the Orks.
Good times indeed
You are right in that it works because the game was very lethal, but the map size relative to character movement is similar to what I want to do (traversing the map is a bit faster in Zodiac Legion, and there are usually less units to move).

 

[Darth Canoli]

But what I dislike about initiative rolls is that you have little input on who goes first (you can delay action, but I don't really like this mechanism, especially on large maps).

Difficult to help from the outside, i just know the chimera system is cheap, it's fun for a couple of battles and then extremely tiring because you're always doing the same thing, trying to prevent your opponents from acting.
It's a different genre but Lords of Xulima does that too with the stunning/freeze system but mostly stun because it stacks so you stun an opponent for 7 seconds, your next unit act 3 seconds later but adds 4 seconds more of stun (so it's still 8 seconds) and so on and your target never acts ...

If there's enough distance and as you said most of the combat will happen on close combat, maybe the regular initiative system based on attributes (+ eventual perks) isn't so bad, specially if you add your command bonuses to grant freely to boost a particular unit's initiative.
If it works for the whole team, maybe use reserve bonus points, P&P-like you can use for initiative or other purposes (defense, offense, damage, healing ?)

Indeed, things used to die quickly in Final Liberation. I have seldom played the SP, but I remember having 100% victory with the Imperium, and 0% with the Orks.
Good times indeed
You are right in that it works because the game was very lethal, but the map size relative to character movement is similar to what I want to do (traversing the map is a bit faster in Zodiac Legion, and there are usually less units to move).

Actually, the orks were not that bad, i've had some success with them even if they're somewhat inferior, some overlooked units were really dangerous, like the long canon on wheels which was firing a salvo, when it connected, it could take down a leviathan or at least offer him to a death-blow.

I also remember a kind of vortex, low "hit rate" but when it did, the targeted unit was in trouble.

 

[Galdred]

The firescorpion in action:



I am in the process of importing all creatures. I have finished importing the hybrids (beastmen, minotaures), and I just need to patch the flying animations for the flyers like the gargoyle and demon (it is also used for dodging).

 

[Galdred]

I have been busy overhauling the level building system to make it a bit simpler, and making a tutorial to let other build their own levels.

My plan is to go X-COM/XCOM2 when it comes to dungeon generation:
Each map section would have some parts fixed and possibly other to be filled by a smaller map preset. I think it offers the best compromise between replayability and having sensible map layouts.
Something like that:

 

[Darth Canoli]

My plan is to go X-COM/XCOM2 when it comes to dungeon generation:
Each map section would have some parts fixed and possibly other to be filled by a smaller map preset.

I think it offers the best compromise between replaya
Something like that:

Did we lose half a sentence in the alien dimension ?

 

[Galdred]

Did we lose half a sentence in the alien dimension ?

Thank you, I went back there to bring it back.

 

[Galdred]

Sorry, it's been a long time without updates!
I have been working on the kingdomscape:

The provinces need to be redrawn, because, as the writer/designer told me, they should follow rivers a bit more... That may also make connections a bit more interesting.
I have started integrating both parts together, but there are still a lot of bugs happening when playing two maps in a row (especially when both battles use the same mission parameters).

I also added a combat preview:

The damage preview is still missing.
It displays the critical to hit number (7), the to-hit number (10), and the threshold value (the number of hits required, depending on the target defense value: 4 here).
Is it too crowded?
I guess there should be a more elegant way to show it without brackets (like 7 on the left, and 10 on the right?).

I think it is in the beginning, but as it would be used a lot in the game, it should be relatively easy to grasp.

I decided against showing percentages, because they don't map too well to the game system.

 

[baud]

Sorry, it's been a long time without updates!
I have been working on the kingdomscape:

The provinces need to be redrawn, because, as the writer/designer told me, they should follow rivers a bit more... That may also make connections a bit more interesting.
I have started integrating both parts together, but there are still a lot of bugs happening when playing two maps in a row (especially when both battles use the same mission parameters).

I also added a combat preview:

The damage preview is still missing.
It displays the critical to hit number (7), the to-hit number (10), and the threshold value (the number of hits required, depending on the target defense value: 4 here).
Is it too crowded?
I guess there should be a more elegant way to show it without brackets (like 7 on the left, and 10 on the right?).

I think it is in the beginning, but as it would be used a lot in the game, it should be relatively easy to grasp.

I decided against showing percentages, because they don't map too well to the game system.

I don't think it's too crowded, but I think I'd center each number in its bar (10 in the middle of the red bar, 7 in the middle of the orange bar)

 

[Darth Canoli]

The damage preview is still missing.
It displays the critical to hit number (7), the to-hit number (10), and the threshold value (the number of hits required, depending on the target defense value: 4 here).
Is it too crowded?
I guess there should be a more elegant way to show it without brackets (like 7 on the left, and 10 on the right?).

I think it is in the beginning, but as it would be used a lot in the game, it should be relatively easy to grasp.

I decided against showing percentages, because they don't map too well to the game system.

What about putting the rectangle tool-tip 1 cm lower so it's under the sword instead of behind it ?
It would certainly feel less crowded then.

I agree with baud on centering numbers.
Some tests might be necessary to see what looks better.

 

[Galdred]

I haven't changed the tooltip yet, but I re-added status effects, and tweaked the way they work:
There is no more set duration. Status effects remain each turn unless the character passes a stat check (or fail for a positive status effect).

It makes its a bit less predictable, but it also makes it more about the stats of the target. Someone with high resistance will shrug poison off much faster than a frail character, even if he fails the initial test.
And a sturdy character will benefit from regeneration much longer than a frail one.
A normal failure (or success for a buff) will reduce(increase) the threshold by 1, while a critical one will leave it unchanged.

They are checked at the end of the character turn, except if they were applied during this turn (so that they will always proc at least once if the initial check is passed).

 

[ERYFKRAD]

Nice