Balancing Bots vs Vehicles - Two example models

Yeah I don't like the following:

It sounds like people want bots to be tough for their metal and tanks to be tougher but not tough for their metal.

Example: Bots might cost 30 metal and have 50 armor and tanks might cost 100 metal and have 75 armor. Maybe this is drastic but you get the idea. A tank is still tougher than bots, but bots are better at taking hits for their metal cost. Generally, people will want their tanks behind their bots if they have them available.

This could definitely lead to a very strong bot presence early game, because force will play a larger role than speed in your main battle lines.

I've posted this before, and I'll post it again. Bots should be able to SURF VEHICLES!

If there is no such thing as a population cap, or some kind of upkeep, there wouldnt be many advantages to having tanks instead of bots. Rather, the tank should be more cost effective in open combat, while the bots are more flexible for some purposes, like manouvrability or a different dps/health ratio.

Movement differentiation I completely agree with. Bipedal locomotion has inherent differences to wheels / tracks and is an obvious way of differentiating the 2 types of unit.

Accuracy differentiation I don't agree with. We already have targeting computers on IRL tanks which allow them to be accurate (for a given value of accurate) on the move. PA tanks are being run by an AI, and therefore already have a targeting computer. If anything, a bipedal combat unit would have more issues dealing with recoil on the move than a tracked one, making your accuracy based balance mechanic a little counter-intuitive, in my opinion.

Scale wise, T1 bots and T1 vehicles are far closer than IRL infantry and IRL vehicles. Although looking at the whitebox pics any bot is small compared to the T2 tank

I'd be all for making the T2 tank (assuming it stays at that scale) able to drive over small bots; that would also function as part of a balancing mechanic between bots and vehicles, if your cheap T1 bots are at risk of flattening by large vehicles. I'd also be happy for bot weapons to be implemented with a smaller range than vehicle, (with a justification of needing less recoil), and bots to be implemented with less health (with a justification of the motive systems being more complicated). All of those would function as a balance method with a believable in-game justification.

In TA, bots are also normally given better incline mobility than vehicles, i.e. they don't slow down as much when going up ramps (and can go up steeper ramps). They also generally have better acceleration characteristics, but a lower top speed. This gives bots a mobility advantage in difficult terrain (e.g. heavy forest) where they can navigate around obstacles more easily than vehicles, but a disadvantage in clear terrain where the higher top speed of the vehicles can be used.

From my experience it isn't that micro heavy. If you want your units to be stationary when they fire you give them a Attack-Move command and they will automatically stop to fire when in range of the enemy.
Stutter step requires alot of micro as you can stop to fire and then move while the unit reloads. However if the time it takes for the unit to accelerate is long compared to the time it takes for the unit to reload this is doesn't really give any significant advantage to stutter step micro.

Stutter step imo only works effectively if acceleration is very high.

For example, in StarCraft2 (besides the fact that units can only attack when standing still which obviously increases damage done infinitely) units just accelerate instantly, making it viable.

If vehicles take 5-8 seconds to get to top speed, stutter step wouldn't work at all. Since in the first few seconds you wouldn't have moved very far at all.

In the first example, I gave vehicles reduced accuracy while moving (and bots consistent accuracy) because it's similar to that IRL. Most military vehicles have suspension designed to absorb heavy shocks, not to maintain firing aim. And I know that some modern tanks can fire on the move, but I think even they require a constant velocity and level ground to do so. Vehicles simply don't have a great deal of stability on variable terrain. Bots, on the otherhand, could, particularly if they were designed with additional joints than what simple design entail. Think of a human's design, except with pelvis being two separate bones, and you have a design that would provide very good stability when moving at a slow or medium pace, even over rough terrain. The only reason people are terrible at firing while moving is because it's tough to keep our eyes and the gun's sights aligned. If our eyes were mounted directly on the gun (like a bot in TA), it would be much, much simpler.

I generally dislike direct comparisons between bots and infantry. Infantry are only used IRL because they are cheap and (relatively) easy to maintain. If tanks were as cheap, plentiful, and could run on 3,000 calories a fay, you'd probably never see an infantry again.

That said, I hope bots are carefully balanced, and that they feel different from tanks.

Any decent modern tank can keep its aim on the move. That's why they have turrets.

I don't think you've thought about the mechanics of bipedal locomotion. For walking (low speed) you spend a lot of the stride with one foot off the ground, balanced solely on the other. For running (high speed) you spend a significant amount of time with both feet off the ground.

Not a particularly stable platform to deal with weapon recoil whilst moving IMO. You want a tetrapedal, or preferably hexapedal design to allow at least 3 points of contact at the moment of firing to maximise stability while in transit.

I'm prepared to allow a whole load of suspension of disbelief for the rule of cool, however making a bipedal unit more accurate while in transit would break just a few too many rules for my comfort

It should be doable without too much hassle. TA provided some groundwork on differentiation and unit balance between vehicle and bot, and SupCom showed that you can even lump them together under one factory and players don't have an issue with it. There's obviously the capability to have them complement each other without too much overlap, if a bit of thought goes into the design

That is a good point, however I wasn't talking about recoil. Keeping the sights aligned for the first shot is every bit as challenging as doing so for subsequent shots. The reason I consider bots more stable firing platforms when moving (for weapons with low recoil) is they have more joints between the turret and the ground to absorb variations in terrain. Vehicles usually just have on set of suspension, but bots have toes, ankles, knees, and hips to absorb that and can potentially have far greater stability (again, for low-recoil weapons).

However, in the OP, I said vehicles would be more accurate and longer-ranged when stationary, to reflect what you said about vehicles being very stable, even with higher caliber guns (when not moving).

I see what you're getting at, and I'd like to agree, but I'm having a hard time justifying it. I guess I'm too tied up in the realism, when we're talking about a game that's only going to bear a passing resemblance to realistic

In the end these are all simply justifications for why certain balancing mechanisms are implemented; in my experience the "best" games have enough of a believable premise for these mechanisms that they become self-evident to the player during gameplay, rather than requiring rote learning or being counter-intuitive. I suspect I may be applying too strict a definition of "believable" in that sense, since I was going to replying detailing how an AI controlling a tank turret would have a simpler task than one controlling a bot gun (only 2 joints to consider vs all the locomotive joints). I'm pretty sure that the relevant compute powers of the AIs is basically irrelevant at the point you have a several thousand year unending war though, so as long as the distinction between bots and tanks feels natural enough, I don't think the actual physics of it matters particularly :mrgreen: