Slippery slopes and intuitive games

[Warlord Zsinj]

SC was not better balanced, just differently balanced. It was almost perfectly balanced, for sure. But that's also because it was a lot simpler, and hence had a far easier premise to balance. Hence different. They did not have to consider height, acceleration, turn rates, splash damage, etc.

And the classic response...

[KDR_11k]

Orf course then we have to ask, since the goal of an RTS is to be a game, not a simulation, then why would you add so many factors to it that you cannot balance it anymore and the game completely derails?

[[Krogoth86]]

Well that's easy to answer imo:
Different games have a different degree of complexity. Making a game more complex makes it more interesting to play (especially when experienced players fight each other) but at the same time also make it more difficult to balance...

With that said in the end all RTS more or less are complex and as the name says "real-time versions" of chess imo...

[Warlord Zsinj]

That's a slippery slope argument KDR!

... hur hur hur

If you want a game that entertains you irrespective of simulation, why invest in expensive computer hardware altogether? You could achieve your aim with a chess set, or with a basketball

[KDR_11k]

It's not a slippery slope because I'm not arguing that something is inductive, I'm saying that the difficulty of balancing TA is not a natural law but that it was created by the conscious decision to add that many variables and units to the game. Starcraft set itself a goal (three fairly different factions with ~20 different things each) and reached it well (everything's pretty balanced). TA set itself a higher goal (250 units, lots of physics and stuff) and reached it badly (lots of useless or overpowered units, many ways to exploit everything making the game no longer resemble the original design at all).

Lesson: Set your goals where you can archieve them well. The basic RTS game design has proven itself, it is yours to screw up.

EDIT: Getting off-topic though as this isn't a thread about general good practice but a very specific issue.

[Pxtl]

Well, you forgot to say the most important part... are those games fun? :)

Hell yes they are. Cosmic Encounter is pure awesome, because there is almost no chance of elimination - a player may be down-strength if he has lost too many units, but actually being beaten down to the last man is rare. So, you have a war game where everybody is in from start to finish, and alliances flow like butter on a frying pan.

Critical Depth is cool because the power-ups don't unbalance the game, since the power-ups also make you a target.

[manored]

So... I had a good idea :) A good idea that someone else already had tough :)

[zwzsg]

Orf course then we have to ask, since the goal of an RTS is to be a game, not a simulation, then why would you add so many factors to it that you cannot balance it anymore and the game completely derails?

The idea between TotalA and SupCom is to make a sandpit with lots and lots of possibility, to build an environnment rich of interaction, and wait for the balance to emerge. It worked for TA.

Also, when adding so many factors, the idea is to add factors that are intuitive, logical, easy to see and grasp. Like in TA. Not like in Starcarft. Then you have a lot of factors, but it's still easy to comprehend.

For instance, having curved ballistic shells that can hit anything in their path and that are slow to travel adds some more to the interaction complexity, but is easier for a human brain to learn than instahit shots with hard coded special damage against certain units. The balancing effect is the same, make it more effective against slow and large unit, but it is both easier to understand and more rich in possibilities.

[Forboding Angel]

An interesting thing is the fact that Evolution, while it looks very simple from the outside, is actually a very complex game. The trick for me was to hide the complexity from the user so that it flows more smoothly.

I prefer games that just about anyone can jump into and have a good time. As Evolution's focus is not to specifically market to this community per se (because the majority of ppl here play TA games and I'm fine with that) so being easy to jump into is very important. EE and Gundam I believe suffer from this. I don't that that it's a bad thing, and they are both really fun, but it takes a bit of time to understand how those games work.

Interestingly enough, myself and lurker are working on revamping the resource system and the way that it works to something that is very complex internally, but to the end user, it will be easier to manage.

I think that things like that help out people and are capable of keeping people playing for longer periods of time.

This is all a viewpoint tho, there isn't really any way to back it up with some sort of definite FACT, but it seems to work well, and aligns with my views on how a game SHOULD be from my own perspective, which once again, is neither right, wrong, or based upon some sort of infallable evidence.

So far it is working well though.

[zwzsg]

Don't worry! I have the solution to the slippery slope problem!


About every RTS I played or know have their skirmish/multi game start the same: With a single construction unit (mobile or not). In Command & Conquer you start with a construction yard. In Warcraft II you start with a town hall, or a couple peons I forgot. In Dawn of War, you start with a lone construction dude. In SupCom, you start with a commander. etc...

It's been so rarely tried otherwise that everybody seem to have forgotten that there is no reason for game to start that why.

I long for a RTS where instead each player start with already a well fortified powerful base. Something like a geo, 3 moho mexx, a dozen HLT, as many L2 flakk, twenty LLT, an annihilator, 4 storage of each kind, repair bay if it existed, a handful of nanotowers if I didn't depised mod which use that, etc... However, no mobile unit, save maybe a couple light scout and lowliest cons just so there are orders to give and exploration to do to occupy the player from start. Maybe one or two factories, maybe not so as to not influence the player into going one tech tree or another. Every kind of short range radar, watchtower, seismic senor, decloacking device, so that unit can be seen and killed a little bit befor they reach the inner of the base. And walls all around. Have it all tightly packed atop cliff. Give exactly, or almost, the same to the two opposing players.

I talked in TA unit names to make it simpler to convey my point, but of course translate it to whatever mod / game package. In more generic term, have the player start with a small area - heavy power base that has:
- Enough ressources producer to make up for half the player ressources in mid-game
- Lots of short defense to make it 1) Looks like a fortress 2) A suicide to attack in the first 15 mins. Make it sure every kind of attack is covered. For instance, it would be a pity if having forgotten anti-air or anti-cloack made it so a plane in the first minute could deal huge blow to the base inner guts.
- One medium turret to cover like 1/4 of the map. Something like a slow projectile, very little explosive power guardian, that make it so any enemy structure in the 1/4 of the map will eventually get killed when found, while letting mobile units, or short-timed raids, or huge assault, relatively unharmed. Come to think of it, the annihilator with its instakill fast shot was a bad idea, guardian would be more like it, but maybe with a little more range, and way less power in their shot. But not a bertha, they have too much range. Could also be some sort of merl-like unguided rocket.
- Everything else that makes a functionnal base, save 1) the mobile units inside 2) the "teching up" structures. You'd start with some advanced structure (like ressources producers, maybe shields), but without the cons or factories to make the cons to build them.
- No army. None to very few mobile units, and no attack units but only exploring unit and "getting the teching up started" unit.
- Being given units at start also can help getting never-used units actually used. Ensuring variety instead of the ever flash rush. All those units that are too expensive for their use, well, if they are given at the start, then they will be used! Same for under-used tech tree, by giving the never-built factories for them at the start.


That way, the base given at the start buffers any early mistake, and ensure no player gets mortally crippled right at start. But as the game starts longer, the relative importance of the initial base compared to the player-built stuff diminish, until the initial base is half destroyed but the owner doesn't care because he built a 4x more powerful base sprawling the map.

The heavy wall and defense would buy newbies enough time to have fun toying with stuff, unlike current *A mods where a newbie can get killed before producing his first mobile unit.

[Pxtl]

Shameful admission: I have barely touched NOTA. But I have dim memories of NOTA having a commander-tower. Couldn't that "big starting base" simply be a single huge commander tower?

[Sleksa]

The heavy wall and defense would buy newbies enough time to have fun toying with stuff, unlike current *A mods where a newbie can get killed before producing his first mobile unit.

Or it could only prolong the already decided game

[Forboding Angel]

That is a good idea zwzsg. An excellent idea in fact. I'll bug lurker about it and maybe we can make that happen in Evolution. The only thing is that you wouldn't start with mexes (because the placement of them would vary).

And sleksa, you can't idiot proof everything, but this would be a heck of a good start. And have you ever heard of modoptions? You could put it in there and assuming someone who isn't an idiot did it, you would be able to turn the option for prebuilt bases on and off.

[zwzsg]

Well, when a newbie plays a pro, the game is decided before it starts. So if you don't want to "prolong the inevitable", the only way is to have the lobby declare a winner from just looking at ranks.

But the point is that by giving a fortress the newbie, he can still builds a few units, kill a few attackers, try one thing or two, before dying. As opposed to today where even a decent player can't attempt one single thing against a superior player. I remember many games of TA where my veh plant was bombed, flash rushed, samsonred, before I could build more than three units. I felt hopeless and had no fun, and only my ethics prevented me from rageequitting.

Or it could only prolong the already decided game

The problem with slippery slope is not that a decided game last too long, it is that the part where the losing player is so utterly dominated he has no fun is too long. For instance when I play Deci on EE, the outcome is decided from before the game starts. Yet, even if I lose, as long as I built many units, have reached tech2 and have managed to do a couple suprise side attacks, I'm happy.

With my proposal:
1) Each player strenght stays relatively equal regardless of skill during the first 10-15mins of game. Player skills starts mattering only later and progressively as game time elapse.
2) Each player, even complete idiots, can put up a nice final fight. Assaulting a fortress, or defending one, is fun, even if outcome is known.
3) Rushing and first blood still matters (to take the ressources in the middle), but not to the point of deciding the outcome (your fortress has enough ressources for you to build a force enough to retake the middle, assuming you are more clever in your unit choice and tactics.)


Forboding Angel: If no mexx, then fusion + adv metal maker (or rather, a custom, un-buildable structure that produce M for nothing). The idea is to have the base produce lots of ressource, so failing your initial extension is still a small disadvantage, but not a fatal one. And also so that rigth at the start you can quickly build lots of units, without staying limited to two jeffies for the first five minute.

[bobthedinosaur]

i like the sand box analogy. its pretty accurate, for any one thats ever had a sad box and played war in it.

[Erom]

A "fortress start" mutator for BA or CA should be a (relatively) simple task- I say we try it out and see if it's fun!

EDIT: Obviously, for simplicity, this would require maps with nice wide flat areas at the start, but I'm sure those exist / could be made to try the concept out.

[bobthedinosaur]

what about the no resources, just prefabed armies. like in the total war series?

[Erom]

I think both could be done by re-writing Trepans(?) Pre-Deployment mutator...

[bobthedinosaur]

luas?

[Erom]

Yep.