StarDestroyer.Net BBS

yup, i use a mobile anti-aircraft defence, and a squadren of jets as well

ggs wrote:

PrinceofLowLight wrote:Though I agree the lack of SSBNs is annoying.

Its a game engine limitation.

The unit AI isnt smart enough to rise to the surface of the water to fire a weapon.

Ah, so that's why I couldn't find a user-created one even after a few hours of searching...

Darth Wong wrote:Preferentially trying to flank the enemy is tactics, not strategy.

Whoops, my bad.

Darth Wong wrote: Ideally, a group of units would (once grouped) have a collective AI rather than a group of individual AIs, and it would intelligently attempt to do things like flanking the enemy once it is engaged.

Meh. I don't know how that'd work with 'artificial intelligence' without screwing something up (attempting to flank when not wanted, etc.).

BTW, if anyone's up for some interesting reading on high-level Starcraft, check out this site.


http://hautamaki.isgsa.org/

While some of this is a bit dodgy, specifically the silly Bobby Fischer thing and the '10 ranks' in the intro, most is spot-on. The author made Canadian Team B, if I recall.

Ground Control 2, for all your combined-arms needs. Artillery, air units, infantry, armor (WITH separate armor values for front, sides, and back), various combinations, different subsections of said...

Give it a look. Also no real resources, basically control move of the map = gain points faster, have a larger army = gain points slower, nice little balance there. Awesome game, and hella fun multi.

EDIT: In its current state, it's got some balance issues, which makes it /most/ enjoyable when played with a group of pepole you know. But hey, that's what patches are for.

Oops...think I've been mixing Ground Control 2 for some other game...my bad

phongn wrote:The AI isn't even smart enough to use ABMs or nukes, for that matter

Yup.

The AI is fairly stupid. Believe it or not, the AI was tonned down in the v2.0 patch.

It really doesnt help that the pathfinding degrading is linked to the absolute number of active units and not the number of units the CPU could handle.

Stofsk wrote:So Submarines can ONLY fire torpedos? Not even those missiles from the 'missile' ships?

Nope. Underwater objects can not rise above the surface of the water to fire any non-water bound weapon.

The concept of what is underwater & whats above water isnt defined clearly enough in the internal simulation todo it.

The key problem with RTS is, I believe, the problem with time management. The rest is just small stuff compared to it. I think realism shouldn't be a part of most RTS, except quesi-simulations that few play even in the TBS world. Afterall, even if a vet-pikeman-mountain-citywalled and kill my mech infantry on defense, civ2 still rules.

In a RTS game, a player only have so much time. The key to RTS design is to make sure the player spends his time on interesting, thoughtful and "fun" things rather than boring, annoying or mindless things. A good RTS game focuses the player on tactical and strategic choices and gives them the free time to execute them, as opposed to more boring command of logstics, path finding, trivial unit commands (like ordering attack in old C&C games). Of course different player what to command different things, as some would want to command a hero in close combat while another perfer perfect placement of defensive structures and walls.

The reason why so called "tactics" are not common in RTS games is not because bad gaming conspiracy but the lack of time to manage them properly when juggling a resource expansions, 2 free builders, 4 factories and 1 recon group at the same time. A few seconds spend playing around with formations or flanking is a few seconds not spending on resourcing, upgrading or tending factories. Formations and flanking matter even in facing-less games like starcraft if only the time couldn't be spend better. In the "zone control" starcraft map where the only unit are marines and bunkers, with a bounties-based-resource system, formations and flanking isn't merely important, but almost absolutely vital and can spell victory and defeat in every game. The effects of flanking, as in extra firepower squeezed in and the "roll up the flank" automatic concentration firepower as a group targets a single marine on the edge of a firing line means that 2:1 or better kill ratios can be achieved with good formation and movement work with the game force. Of course, in regular SC and most other games, there are no time for killing a few extra marines.

The problem with micromanagement isn't with AI as well. If one wants to reduce management, one can simply simpify the system. This is already done in most games, with no armor difference between facings, no fire while moving, "hitpoints" instead of mutitude of damage conditions and weapon ranges shorter than LOS. The resourcing system have been simified for some games as well, like the minimalist "Z". Things like pathfinding can be fixed with more open maps, lower unit count, more agile unit and smaller units. If the Ai merely takes over human control while giving no new options, than it is really similar to simpifying the mechanics of the game and giving it more "toys" for the engine while not improving gameplay. (aside from immersion/look factor)

It appears that many RTS designers don't really have idea whats going on. RTS of the traditional "builder" type can see unit count grow by magnitudes and the potential micromanagement difficulty grow even higher (as they are proptional to unit interactions) and players are either bored by the early end of the build up when unit count is small (Aok feudal age...yawn) or overwhelmed late game (TA with 500 unit limit) or both.

------------------------------------------------------------------------------
As for strategy in RTS games, the problem is this:
If an intelligent person fights the exact same battle a million times, all (or almost all) useful tactics and strategies would be discovered. As soon as they are, they get documented and any newbie can simply follow the book and "win without thinking". Rush based game plans are popular because early game gives very few variables and the strongest strategies found for the period can be verified to be indeed the best, while they are generally easy to learn and a strong early game is absolutely necessary while the rush also limits the chances of a harder to play late game. Learning late game does not effectively give a player a win due to the number of varibles, but lacking a early game means a player would certainly lose, and thus every n00b and more serious players learns to rush. I believe there is no stopping rushes (as even super slow AoE/AoK is "rush"able, as first attack usually is defined as such), but give early game the kind of varibility and depth as late game with its large number of units and tech options.

---------------------------------------------------------------------------------
If you want realism, you should spend 30 minutes moving units a few hundred meters, 15 minutes pinned down by unseen MG fire while calling artillery, and 5 minutes watching the enemy gets obliated by tons of steel rain while your opponent calls you are "artillery l4m3" and 40 minutes finding that last man hiding in the forest.

Not that much fun.

no fire while moving

Arg, I hate that with such a passion!

"We're taking fire, but we're not doing anything because we're busy moving!"

What might be a good idea might be a standing order for selected units, where a normal move command allowed for units to track and fire upon targets that come into range (which, in the process, would slow them down as they take aim), and an "express move" would keep the units moving at maximum speed to their ordered destination. Why Blizzard has the default set the other way around, with "attack move" being a special command and not just a single mouse-click is beyond me. At least give us a menu option so we can choose for ourselves whether right-click = attack move or standard move!