I'll put a little bit of background to this(which people can probably relate to)... Over the summer I went to several small LAN parties of 6 or so people, and our game of choice for Real Time Strategy was the old TA, it's an old favourite, and we played about 7 or so games back-to-back, and were getting pretty tired of doing the same repetitive thing over and over again, with each game lasting about 3 hours, and the first 45mins comprising of setting up resources and whatnot. And with the final game coming to a close we shut down our machines, got stoned, and started rambling about game duration. We realised that for an RTS, the game seldom lasts for more than 3 and a half hours TOPS, by that time if the map hasn't been stripped of resources, or people have billions of farms in AOE trading food for other resources, then the AI has given up, and the game has become stagnant. I decided to have a go at trying to figure out what makes the game 'dry-up', and how to combat these symptoms effectively to increase gametime in multiplayer RTS games. The first one which I think is pretty obvious is the limited number of resources available to a player, I dislike finite stockpiles of resources in a game, it make's it seem like a nasty kludge to stop the game becoming bogged down with large numbers of units and ultimately collapsing the game's economic structure down to 'building-queue central' and 'click-o-rama'. I'm not trying to further develop the solutions as to resolve this as yet, but expand the scenario of gameplay rot. In the situation that the player's are against AI in any team configuration, the AI always 'gives up', at least in the games that I've played. C&C:RedAlert was the worst for this, you almost had to help the AI along later on in the game in any way possible, ie, imposing stupid rules like 'no certain types of weaponry' or '2 ore refinerys only'. This situation almost probably comes about due to case 1, the resources. To ensure that games last longer people begin to impose soft rules during the game, ie, ones that are not enforced by the server, such as, no attacking until 10mins have passed, some may argue this as a strategy, but IMO rushing with an MCV full of engineers is not really strategic, cunning.. maybe. This works as a sort of balance for people slightly slower with the controls, and also to ensure the game lasts longer. What would it take for a game to last in excess of 8 or so hours? - A save option - Larger maps (discuss? Possibly over-expansion on a tile based RTS) - Slower building times (boring? Quick kludge?) - Infinite resources (requires 'throttling' and careful balancing in beta testing, ability to ensure that resource trading and production is stable, or at least cannot spiral out of control. Warehouses are a real-world solution which can be incorporated. Resource shipment to factory?) - Careful introduction/limitation/removal of 'super units' (The cliche here is that the first few tech levels incorporate many units of varying strengths and weaknesses, until tech level x is reached, where the players can purchase uber weaponry which takes ages to build, slow as hell, but can smack all units into oblivion. examples: Krogoth in TA, Trebuchet in AOE2. It can be argued that these units are used to provide ways to break a stalemate, in the case of impenetrable base/town/city defenses) - Not just battling (The name implys Strategy, but not necessarily restricted to battles. Incentives to develop advanced technology for weapons and in refining of more advanced resources, this is without a doubt insanely boring for some hardcore gamers, but some people love micromanagement and economy). These are just a few ideas I'm throwing into the ball pool, hopefully people will be able to add more, and criticize on the points that I've made.

I definitely like the idea of better/different uberweapons. One of my major gripes about most of the RTS games I've played is the lack of *scale* - for example, in Starcraft a battlecruiser costs 10x as much as a marine, has 10x the hitpoints, more armour... but only 5x the firepower, at less than half the rate of fire, and it doesn't *look* that huge on screen. I want a game where you can have (say) four different zoom levels, each one twice the magnification of the one before. Zoom in, and all your flying uberweapons disappear above you and have to fend for themselves, or zoom out but lose the ability to give your ground troops any but the most high-level orders ('everyone in this area attack in this direction' type things).

In terms of making the game last longer, I'd say you need:
* Larger maps
* More complex tech tree, and/or simply a more interdependant tech tree to increase the time required to climb it
* Shift balance of strength from units to structures. It frustrates me that, for instance, a barracks, your infinite source of little soldier guys, can't itself shoot back when attacked. Of course, you then have to come up with some creative solution to prevent offensive towering or its equivalent. How about having your harvester units deliver the required resources directly to the building being constructed, rather than back to your town hall and then having them magically transported?

As another random gripe, I think the player shouldn't be able to see what a unit sees unless it's in sight of other units which are, eventually, in sight of the town hall. One thing that always frustrates me in multiplayer games is that you can't spot a scout and 'kill it before it sets off the alarm'.

/ramble

(trivia: did you know that the best human counter for carriers is stimmed marines? carriers can't focus well on multiple small targets, and the marines destroy the interceptors before they return to the carriers.)

Quote:Original post by Zanthos
We realised that for an RTS, the game seldom lasts for more than 3 and a half hours TOPS, by that time if the map hasn't been stripped of resources, or people have billions of farms in AOE trading food for other resources, then the AI has given up, and the game has become stagnant.

This is quite a generalization. I'm currently playing a game of Empire Earth with a friend against two AI opponents. The game has already taken more than ten hours, which is not uncommon for our EE sessions. Of course, EE has infinite resources, at least for all practical purposes, which does allow longer gameplay (even though the resources themselves aren't limited, there is a strict limit on the speed of gathering, which slows things down), and the AI cheats. But it's still a great game, in my opinion. Anyway, I'd say the actual time depends on the players' skills and preferences as well as the game mechanics.

Larger maps would be, in my opinion, a welcome addition to rts games, as it would make it possible to make the scale more appropriate. I have to agree with fractoid here, the sense of proportionality in rts games is awfully twisted. Now, I can accept the fact that building a temple using stone age methods takes less than a minute (no one wants to wait for that in real time, even though it could take a couple of minutes, at least), and other similar temporal oddities, but it is the spatial distortion that annoys me. Let's take the common "large islands" map type. They really are islands, and not continents, as ships as simple as triremes can fire their arrows almost in the middle of the islands. Having a slow, old, fragile priest walking across the map will take a few minutes at most, so the area is not too large to begin with. Still, the buildings occupy a surprisingly small land area. Comparing the size of the soldiers and the buildings, the legendary "wonder of the world" is actually the size of a largish house, and the actual houses are smaller than an outhouse. Heck, in some games some of the soldiers are larger than the buildings (WarCraft III, anyone?)!

It isn't even consistently used: in StarCraft you have buildings of the usual size (usual with regard to most rts games, that is), but then some levels take place indoors, the whole level being inside one building. Still all the soldiers are of the same size.

Firearms have a ridiculous range: a common rifleman can usually shoot at a maximum distance of measly ten meters or so. An artillery unit has a flabbergasting range of twenty meters.

It would be neat to actually have an rts with a sense of proportionality, where continents were really continents and you couldn't just walk from place to place. Clearly this would have major implications on gameplay, as the game world would have to be mind-staggeringly huge.

I do acknowledge that my argument is more on the gameplay-realism-axis, as the level of abstraction even in rts games is greater than the representation of the game makes you think. After all, the seemingly single soldier is still called a unit, which could easily refer to a whole group of soldiers, just represented as a single soldier for convenience.

I have started to find rts games less and less appealing all the time for another reason as well: in real life, in order to make a successfull attack, you would need at least four times more troops than the defending faction. In games, however, not only this is the other way around (at least it often seems that way), but in addition you have to control each and every defender yourself. Of course rushing early is a viable strategy when both factions have approximately the same number of troops.

Then again we do have games in which the tactical opportunities are much better, such as Sudden Strike, Desert Rats vs. Afrika Korps, etc. In Sudden Strike you can build a defence line over which simple rushing will not help. Soldiers are not larger than buildings. Artillery has a firing range of half the map width. The evident "problem" with these games, however, is that in being more realistic, they won't allow you to actually collect resources, build units etc.

Quote:Original post by Numsgil
Add an automatic handicap to the game, so that if a player is losing, their units become stronger. By the end of the game, their lone worker can take out 5 or 6 uber soldiers. I'm sure this element has been deployed in games before, and since I haven't seen much of it, I'm assumming it's an ultimately broken technique. However effective it is, it will help lengthen any game without changing the pacing.

In Sudden Strike I once hid in a forest area with the last few of my soldiers and could snipe the enemies while they tried to reach the forest over a clearing. No "adrenaline mode" was required, just good old tactics. Why this was possible was because you could see out of a forest much more easily than in the forest from outside (which makes sense). My opponent finally started making artillery strikes in the forest at random and eventually I lost. Of course, such tactic could be used at any point of the game, but it is easier to hide a small number of soldiers at the end of the game than the large armies in the beginning.

On the other hand, one reason to not add this "adrenaline mode" would be the fact that most players don't like to hunt for the last few enemy units anyway, as the victory is ensured anyway.

Quote:Original post by Numsgil
Increase the HP of buildings. Alot. A whole lot. That's sure to lengthen any game.

Not only this, but make most buildings impervious to most light weapons. It pains me to see a huge castle built of stone to catch fire and eventually collapse after a couple of dozen of swordmen have blunted their swords on it for a few minutes (or shot with rifles at it enough times, using more bullets than what most countries could ever even afford). Having to bring in demolition experts (or what have you) would be appropriate and could lengthen the game.

On the other hand, why do we always need to destroy all the enemy buildings? Why can't we just walk in, declare the buildings our own and be done with it (as in Cossacks)? The buildings have done nothing wrong! End building genocide! Why, O why do we have to ...

Oh well. I could go on and on raving maniacally about all these issues on rts games, but maybe it's time to stop now, before it becomes truly a religious sermon. Thank you for listening and I apologize for the inconvenience...

Quote:
The ever changing mission requirements will keep the players fighting each other, but at the same time keeping them doing different things. The time limits on the missions keeps it exciting.

I've just been chatting with one of my friends about this idea and he says it sounds a very interesting way to fuel the game. One of his suggestions was having rewards like technology/research advances for the first person to take and hold x position for y number of minutes.

Kazgoroth: My aim wasn't to target the dull-ness of repetition with regards to what game's we were playing, but with regards to the repetition of setting up and establishing your base/forward command post/settlement in the first x number of minutes in a strategy game. Well done with regards to your TS modification, it seems that you have managed to hit the nail pretty much spot on with the extended duration of the game. I like the specialization of units with no 'heavy' units or uber weapons, and can assume that you have no more than 3 tech levels? That seems like a sure way to ensure that no players become too advanced over others, whilst retaining the diversity of the game by introducing new tactics other than new units.

Quote:
Uber-weapons tend to be bad for game duration in most (but of course not all) cases. Normally one player will attain and/or deploy/use these before any others, and as a result gain an advantage that can take a rather large effort to balance out again. Once advantage is tipped heavily in a particular player's favour, the victory of that player becomes fairly inevitable, as typically the advantage continues to grow, while the other player(s) have to work very hard simply to regain thier previous position.

I definitely agree with uber-weapons being detrimental to the games fun(unless you're the player with a dozen missile silos), and this was recognised in RedAlert2 with the option to disable them.

Quote:
Add an automatic handicap to the game, so that if a player is losing, their units become stronger. By the end of the game, their lone worker can take out 5 or 6 uber soldiers. I'm sure this element has been deployed in games before, and since I haven't seen much of it, I'm assumming it's an ultimately broken technique. However effective it is, it will help lengthen any game without changing the pacing.

This idea, although it has balancing properties, is fairly flawed if we have any concept of army morale, heavy losses would make troops much less efficient, and would not make settlers suddenly drop their spadesand become elite marksmen. Units could become fanatics when their morale drops below a certain level say 95% below max, possibly at which time they are removed from player control and go on a suicide rampage.

Quote:
I quite like how TS (and most of the other C&C games) handled this actually. Resources on most maps are unlimited, in that the fields from which you are harvesting them regrow over time.

It seems pretty sensible to assume that for a game to last a long amount of time(whilst have space available on the map for something other than the resources), that they should be infinite, and acquired from something such as:

- Drop pods : sent in by aircraft according to when resources are needed, ie, when a tank is to be manufactured, all the resources required are dropped near the base, and carrier units can transport these resources to the factory).

- Mines : sprinkled around areas of geological interest would provide players with for all intents and purposes, an infinite supply of the said resource.

- Supply routes : similar to drop pods, but via road. These, unlike air supplies, would require defence along the route, and could carry much more than just resources, vehicles and troops should also come in via these supply lines, downside to this is that an opponent could simply cut off the route leaving the other player stranded without supplies, but this is a valid tactic, and shouldn't be forced out of the design.

- Harvesting : introduced in early RTS games, heavily armored vehicles would wander out to mining fields and collect various resources of high value, to be traded in for credits which would enable the user to construct units/buildings.

There are probably many other ways resources can be accumulated in the game, some finite, some infinite, but the aim of this isn't to discuss resource acquisition but to identify how to keep resources flowing in an extended game.

Quote:
On the other hand, why do we always need to destroy all the enemy buildings? Why can't we just walk in, declare the buildings our own and be done with it (as in Cossacks)? The buildings have done nothing wrong! End building genocide! Why, O why do we have to ...

One way this could be combatted is by having non military units and buildings not count towards whether a player has been defeated or not, TA had the commander which via checkbox enabled a player to be defeated on destruction of this unit, decoys were available to make this more difficult for the opposition. Although, this may be too easy a way to kill a player off in terms of an extended game. In Settlers4 territory is secured via guard towers and castles, non military units and basic settlers are not attacked, instead once a guard building is captured the land changes ownership).

Quote:
Let's take the common "large islands" map type. They really are islands, and not continents, as ships as simple as triremes can fire their arrows almost in the middle of the islands. Having a slow, old, fragile priest walking across the map will take a few minutes at most, so the area is not too large to begin with. Still, the buildings occupy a surprisingly small land area. Comparing the size of the soldiers and the buildings, the legendary "wonder of the world" is actually the size of a largish house, and the actual houses are smaller than an outhouse. Heck, in some games some of the soldiers are larger than the buildings (WarCraft III, anyone?)!

Units with unrealistic range really annoy me, the component of inaccuracy is usually omitted and instead a hard unit range is implemented (ie, a hit or nothing at all). The scaling of the units and buildings is probably to do with the fact that the people are going to be too small to manage effectively, so instead they are scaled much larger to make them easier to click on and identify their type.

I'd like to open up ideas on this:

Quote:
* More complex tech tree, and/or simply a more interdependant tech tree to increase the time required to climb it

An idea was presented by my housemate that involved minimal interaction with the tech tree, but ultimately can change the way a player progresses. This would involve selecting goals or paths for the research to head towards, take this example:
"The principles of mining have been unearthed by your villagers, do you wish them to pursue: a) Advancements to this mining technology, b) Mathematics and ballistics, b) Explosive weaponry"
The player can select what he wants to head towards, and determined by his team, handicap, etc, how long this will take. And does not have to bother with researching individual technologies. This could possibly be broken down into categories, ie, Weapons, Economics, Structural and Defence, each with their own seperate research trees.

Back on to what Sandman said with the missions, I'd like to try and expand on that a bit more, as it seems a very good way to increase the excitement and intensity of a game that could have been going on for well over a day:
- Race to complete missions with same object target: First player to capture x town/village/building and defend it for preset length of time
- Race to complete missions with no object target: AOE2 build-a-wonder, although this is more of an end-game target, and not so much a randomly generated mission
- Sabotage: Possibly involving sabotaging a co-operative player by stealing resources and/or units, this could make breaking and gaining alliances very interesting, for example Player A and Player B versus Player C and Player D... Player A gets a mission from the server:
"We must break free from our petty alliances, there exists technology much more advanced than our own from within the realm of (Enter Player C name here). Your mission is to retrieve the blueprints to (xyz) from this building (marked on the map)." The reward would also be listed. Now I've already thought up a way in which this could falter for example Player A tells Player B that he has to sabotage his plans, in which case the sabotage goes ahead and they share the result. Possibly make it so that the players cannot re-ally when an act of sabotage is carried out or make it so that the morale of units is lowered(possibly resulting in revolt) when they re-ally after a sabotage has taken place.
- Make Alliance: The aim of this mission would be to establish an alliance with an opposing player(assuming it is all versus all), and then proceed to destroy a structure of importance. If you wanted to be sneaky, you could send both players in quesion Make Alliance mission directives, with the clause that they must both destroy a command building.