In Total Annihilation, land movement relied on a complex set of physics. While there were as many combinations of movement attributes as there were units, a few common trends emerged. Certain movement types were made to excel on certain types of terrain: Code: Units with high tolerance for slopes work great in hilly terrain. Units with high turning rates were more effective when obstacles got in their way. Units with high top speed and low acceleration work great on open flat terrain. Units that shared a movement type very often shared similar movement attributes. This led to generalized trends like "K-bots are excellent in hilly and forest terrain" or "Vehicles excel on large flat surfaces". How explicit should these "biome bonuses" be in PA? Should units be designed around the idea that their turn speed and suspension height are defining factors? Don't forget that the pathing engine and AI are dramatically changing. The idea of nightmare hedge mazes or spam clicking up a hill could very well be a thing of the past. Which traits should remain? Which should go? Should the complex arrangement of movement stats be replaced with a more straightforward "excels in x biome" system? For example Starcraft 2 had its own "creep" biome, which gave a speed boost to all Zerg units. Edit: Tweaked the options to be a bit more clear. I think.
I think PA's scale means it makes more sense to have areas be painted onto planets rather than be defined by small features like a field of rocks. That might still be a factor, but "rough" terran should probably be independent of features on the surface. Different textures to indicate the terrain type would be a simple and quite serviceable way to differentiate land types. Bumpy, rocky texture means wheeled units will have trouble, while legged units excel. Sand, grass, stone, metal, there are any number of possible types to implement. The question is, which ones should be added, how significant they should be, and in what capacity? Note that these properties are in addition to local features like obstacles and elevation. There can still be hills, cliffs, canyons, mountains, rock fields, etc. But a grassland with a wreckage field could perform differently than a marsh with a wreckage field. I think that there should be essentially four broad land movement categories. Legs, wheels, treads, and hover, with hybrids/mixes possible. Legs are the slowest, but the most versatile with respect to terrain- legs work in even the roughest terrain with minimal slowdown, and can scale steep slopes and navigate obstacles effectively with high acceleration and turn rate. Wheels are the fastest in good conditions, but steep slopes or very rough or difficult terrain types such as rock fields are completely impassable. They are greatly slowed even by mild terrain difficulty, and have very poor acceleration compared to leg-based systems even in open plains. The cheapest mode of locomotion, all else being equal. Treads are between the two, able to cross limited rough terrain, but they are greatly slowed. Very severe terrain is still impassable, but treads are a large versatility improvement over wheels, with lower top speed. Hover is a mode of locomotion for land units that ignores terrain, but is typically expensive and with low absolute performance. Most notable for their ability to float over oceans despite being land units. Then, a regularly applied set of rules are defined which determine how these locomotion types perform in a given biome. For example, sand might slow legs, wheels, and treads by 50%. Flatland might not slow anybody. Slopes of a certain steepness slow wheels by a certain amount, and treads by a somewhat lesser amount. Slopes of a certain, higher steepness, are impassable to wheels, but remain passable to legs, with reduced top speed. Now, I am not suggesting total standardization. Each unit's own top speed, acceleration, etc. is still independent. But normalized rules about how different modes of locomotion interact with the terrain are a good idea. I would also say there is no point in having bonuses- only a maximum value and deductions for penalties. Mathematically identical to having bonuses, but simpler.
I think this is a great idea. Adds a certain depth to it that, I think fits PA. You've got all these different planets with all the different biomes so using the normal tech tree and having to go through it for every planet would be silly. Adding bonuses or penalties for certain types of units for different terrain can totally work. How i see it: you have a tanks with treads and an assault bot that uses legs. The bot will be able to use some sort of jump mechanic to scale small but steep walls where tanks would slide back down, and the bots may have trouble moving through a swap because it puts all the pressure on the two feet while the tank has a bigger surface area to spread out the pressure its mass causes. Other things you can get into to make it even more complicated. Generally (or how I see it anyway) Tank like platforms have the ability to host a big turret or a number of smaller types of weaponry. The 'walkers' (as I decided to call them) are more used for flanking unsing rapid fire smaller caliber weapons. So the ability to scale steep cliffs with the 'jumping' mechanic from the previous paragraph the can set up ambushes in canyons and the tanks will use the superior firepower as a more direct form of attack. Another thought about the different platforms-weapon combination is how it might work. What I mean by this is you can still have a 'walker' type AA bot on two leggs but the the 'tank' like platform should have the bigger guns. Let's take an in between. Have a more spider like platform with 4 legs instead of two you'll be able to have better stability for AA turret while maintaining some of the mobility of the walker type platform but the tank like platform might still be more accurate because of the treads. Maybe, if I had more time I'd write more but I have to be off. Small disclaimer: all this stuff is how I'd like to see things
The only bonus terrain should serve is to protect your base from enemy fire if you build a base in a hilly/mountainous region. I don't want to see silly buff that don't have a place in a game like this it requires to much micromanagement
yeah that's one thing I thought off after I hit that submit button. Either you go all the way with the bonuses and stuff, but do they really belong in this game? I mean, destroying of planets, why do small bonuses matter on these units? With the emphasis of the game being MASSIVE BATTLES (right?) it might seem like it defeats the purpose of all these small bonuses. Abilities like jumping on rocks or over cliffs might be a bit too much, but I think that biomes should serve a purpose rather than just look pretty. Maybe not as deep as different units behave differently in each biome but maybe general buffs or debuffs or both? Some examples to clarify: Say a desert biome, has a big flat salt plain, almost no elevation, means you'll be able to see as far as you possibly can. Say a dense forest inside you won't be able to see much but neither will you be able to see very far into the forest and having to tear down trees to move through them might take time (unless I got the scale wrong and it's only the trampling of the forest but they see over the tops of the trees) Swamps will hinder movement a bit. Maybe you can use it as a trap. Sacrificing some of your units as bait to lure him into an artillery barrage? Ice might cause units to either crash into them or maybe not go as fast so they can brake in time. Smaller things that might cause tactical decisions depending on what type of planet you're fighting on.
I was thinking something along these lines. But it's probably a bad idea to memorize some kind of movement elemental chart. IMO the right idea is that a unit has some kind of base speed, and it is considered to have movement bonuses in ideal biomes. That way every movement bonus is treated as a unit perk, which is bought and paid for. For example, a K-bot has above average movement speed anywhere. Be it sleet or mud or sand or stone, a legged bot is a jack of all trades that excels nowhere. A climbing type might have an ability to climb "steep hills" where other units can not. This is good for units that benefit from high ground (artillery, snipers, raiders, etc.). K-bots often have climbing traits, but it is by no means required for every one. A vehicle might have average movement speed, but it could have a 50-100% speed bonus on flat terrain biomes. This could get insane on metal worlds(+50% flat, +50% metallic), which are practically super highways for tanks. An ensnaring terrain slows every unit the same way (mud, rough terrain, etc.). A jeep/buggy type unit may ignore ensnaring penalties, making them good for units that depend on all-terrain roles (like scouts and AA). For example a mountain climbing unit might have the ability to climb steep slopes, and is additionally not slowed down by it. Stuff like that. The idea is that a typical unit has zero, one, or two terrain bonuses that explicitly state where they work best. I don't think it needs to get more complex than that.
bobucles, isn't that how things essentially worked in TA? Zero-K emphasized it a bit, to make the distinctions more prominent, but otherwise it was the same system. Personally, I prefer to keep the distinctions as they were, and avoid getting too fancy, lest the units become to reliant upon you babysitting them and keeping them on the terrain where they have the advantage. Remember, the units should have multiple situational advantages. So vehicles may be less mobile in rough terrain, but a player may still choose to use them anyway because they need their hardiness for a particular tactic.
Take a look at the mars pathfinder images. It got 6 wheels and it is designed to move on rocky difficult terrain
Indeed, TA had a number of movement attributes that created emergent terrain niches. However, a lot of it was due to "herp derp" pathing that screwed some units more than others. They'd bump into rocks, they'd bump into terrain, they'd rattle down forests like it was some kind of roller coaster ride. On the whole it was wildly inconsistent. Not everything will translate well into a game with superior pathing. Units will glide across terrain with ease and find few troubles with obstacles. Units will separate as some work their way across terrain better than others. Even worse, an oddity in terrain can create a solid barrier for some units but not others. A more explicit system may be needed for PA. The main debate is really how to present the clarity of these movement differences to the players. Using something like land shape isn't terribly clear as any part of the map could have any arbitrary shape. The nuances in terrain will be difficult to spot while still having their impact on units. Depending on emergent properties isn't very clear either, as the unit's qualities are not obvious from a pile of stats.
I agree with bobucles that a more standardized, simplified system would be much better than the madness that was TA's individual unit performance scheme. The "pile of stats" approach needs to be made more transparent and simplified. However as long as it is made clear to the player, characteristics and mechanics using unit distinctions are good. For example, barriers that are impassable for one unit, but passable for others are fine. What if a player can build tank traps that are passable for bots, but create a hard, impassable wall against vehicles?
What would be the point? Please quantify your reasoning. Are you assuming bots are universally weaker? Would there be an inverse trap for bots?
This lie again? You've been spreading this across the forum. It's like saying terrain blocking fire only matters if there's no line of fire check. Otherwise why make terrain block fire at all if units are going to be smart and move to gain a clear line of fire? Surely it has nothing to do with introducing strategic elements. If you continue posting this, you will just continue being wrong. All I need to do to tell something is impassible is vicserally look at the terrain protrusion...you know...use basic human cognition. A Spring-esque terrain passability overlay wouldn't hurt either.
