Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mephansteras]

Based on mainiac's post I had an interesting thought, although it's a bit designer intensive. And best for Sci-Fi games where we don't really know the answer yet.

Basically, have several 'potential' ways of doing each tech and have the game randomly decide which one is going to work. You dedicate funding to, say, Engine research. The game then reports that scientists are spending time researching 'Engine type A' or 'Engine type B'. After a while (depending on resources and chance) they'll either have a breakthrough and you get the new tech or they realize that that route won't succeed and switch research over to another type. Eventually you'll get it, but you might have several failed starts. Or you could luck out and get it with the first avenue tried.

Hmm..you might even have an advantage in picking the right path if your base sciences were advanced enough, giving a reason to spend money on basic research instead of focused research. It'll never be guaranteed, but it'd be a way to normalize things a bit.

Overall, this puts research at a much more random pace then most strategy games, where you can often calculate pretty much when you'll get the next tech based on your research capabilities.

Thoughts?


---

On the subject of muskets - initially they weren't much better then crossbows, no. But they had a great scare factor to them, for both humans and horses. Especially since most armies were made up of peasants who weren't big on fighting to begin with. So they had a morale advantage that made up for their technical faults.

[deadlycairn]

Wonder what would've happened if we'd continued advancing crossbows?

I came up for a system for a little poisoner minigame that basically gave certain substances different qualities, which would then react differently in different situations. Thus the player has to actively research and experiment to create an ideal poison More suited to a roguelike or action game though.

[Servant Corps]

deadlycairn: Then how is it fair? Why should we randomly flip a coin to decide if Crossbows or Guns or Clubs should be used, and waste valuable resources if you decide wrong?

Far better to just have the game randomly decide and take the choice out of it. Let the researching be done by the researchers, and let the "taking-over-the-world" business be done by the "taking-over-the-world"ers.

[deadlycairn]

The system was done with plants. The possible effects were randomly given out to plants, along with a random description, then by boiling, crushing, injecting, mixing etc. you created your desired poison. Give it to your victim, examine results, rinse and repeat until you are happy with the poison you've made. And I did say it was better suited for a roguelike or action game, or even just the minigame it is, where the player has a more active involvement.

Don't see where your randomly flip a coin part comes in, must only assume one of us has misunderstood the other.

[mainiac]

Basically, have several 'potential' ways of doing each tech and have the game randomly decide which one is going to work.

Yeah, that's the core concept.  After that, it's just fiddling around in the margins to give balance and flavor as it applies to the specific game.

There's pluses and minuses to having it be different each game what techs do and don't work.  I'm for having the techs, once researched, be the same every time.  Otherwise you might need to constantly be checking the tech tree, trying to remember what the heck your tech does this game.  Until research is progressed  labeling them like Mephansteras suggested ("enginer A", "engine B") would be a good idea.  Or the names could describe progress (a "feasibility test of car engine" becomes a "test design of car engine" then "prototype car engine" until you finally know what you have.)

One thing I didn't make clear is that I imagined a limited list of technologies with both the worthwhile and worthless idea's on it.  So research is never completely wasted, as you have narrowed the field down and have a better shot next time. 

1:  Acceptance rate:

Bingo.

Imagine a Civ game where the rush to gunpowder failed because muskets weren't actually more powerful than crossbows?

If the technology system allows historical paths be unsuccessful because something else becomes dominant, I think it shows the tech system has massively succeeded and it opens new avenues.  Imagine playing a global strategy game in the 20th and 21st century.  Except gas engines never took off, everyone uses bio-fuels refined with coal energy.

I don't think the musket to crossbow analogy is the best though.  Those two technologies were discovered more then a millenia apart and it's crucial to realize that muskets were a result of experience with cannons.  Also, there's the whole issue of metaleurgy advances being needed that were because of church bells not weapons... Overall, this sort of tech system is probably better for strategy that's either after 1800 or takes place over only a couple hundred years in a preceding era.

A possible way to allow technologies to fail to take off is to keep in mind how well a technology is "engineered."  The more you've seen it and built it the cheaper and/or more effective it becomes.  I think alternative energy sources are a fine example of this situation.  We know that some of these idea's are very profitable... for 30 years from now.  But today they are competing with idea's that have a 100 year head start in engineering.

Another concern does arise though, that players might metagame to irrationally stunt an avenue of tech research to give good techs that are available later a shot over techs with fewer requirements that might out engineer what they want.

More suited to a roguelike or action game though.

I know a system like this sounds pretty fantastically complicated, but that player attention needed could be minimized.  For it to work correctly, most of choices are taken away from the player.  It's only when something becomes a strategic concern that the player meddles in the tech development.

[Il Palazzo]

I'm totally against detailed tech research, unless it's really small-scale, like in deadlycairn's minigame idea. But in grand-scale strategy games, generalizations are simplier, and so, easier to implement. They're somewhat like ascii graphics, in the sense that you don't need a hundred people working on inventing new wacky names for every little tech, while giving generic name like 'cannon V' or 'advanced infantry' lets the player's imagination do the rest. Naming and researching every srcew and every test stage is meaningless, mindbogging micromanagement and gives rise to silly situations, like "Oh, I've got 'super-duper plasma gun', now I'm researching' 'super-duper pg with blinking diodes'".

There are some games that did the research(and/or tech implementation) aspect almost right, namely: MOO2, Europa Universalis and Hearts of Iron series. If one combined these three, he could get a system close to ideal(in my opinion, of course).

MOO2's best innovation(as somebody mentioned before) was the percentage-to-breakthrough system. I'd go as far as allowing a small chance of succes right from the start(logarythmic scale, perhaps? so no 100% chance either). Also, It'd be probably a good idea to hide and randomize the tech cost.
This way, you can have accidental discoveries; prolonged funding of dead-end ideas(each turn you're unlucky); the common-sense rule that the more you work on something the higher chance of achieving the goal is also satisfied; if costs are randomized and hidden, you don't know in advance which tech will be easily done just because you've played the game before(there could be a general indication like 'easy' or 'difficult' etc.), also this would simulate:

Acceptance rate

.

EU's foreign influence on your research is a great idea. Basically, if you're bordering a country with higher tech level, you get free research points. More if your culture/religion is similar, if there's high advancement levels disparity, and if there's more than one such country. This simulates the "we need bazookas to counter their tanks" concept, how one invention stimulates another, as well as the free flow of ideas(with religious/cultureal restrictions). Would work well for any kind of 4X too.
For added 'realism' one could get extra bonus 'science points' to combat techs when at war(or better when actual combat took place) and reduction in civilian tech bonuses(closed borders etc.)

While EU did away with complex tech trees altogether(just 5 or 6? techs to improve), HoI had a bit more 'juicy' system. You could develop(it's IIWW) generic types of units e.g. infantry, tanks, fighters, short and long range bombers etc. While some of them had names to match the histiory, they were simple upgrades of the previous ones. Basically, you had (almost)every type of unit, available from the start, which you just improved during the game.
This upgrades system is woth noting: you spent resources from your pool on a day-by-day basis to refit old units, so you didn't have to disband the old '36 infantry. You'd just let them fight and, providing the supply lines were maintained, bit by bit your army would get modernised.
Ships, being large hunks of metal, were unreffitable, these had to be scrapped, but were still useful even when outdated, as a ship is a large investment of time and resources - by analogy, this would work great for 4Xs

China's economy today is an excellent example of the variation problem(...)

this can be easily done too(it had been done before). Your teritory/planet has varied levels of industrial capacity/infrastructure. You can't build you newest tanks/spaceships where you've got no ability to do so, but you can easily commision a starport, or improve said infrastucure to change that over time. And this could cost more/decrase happiness to reflect the cultural inertia of a given society.

[Antioch]

The research system of MoO 1 was far superior to that of MoO 2. In Mo1 you were forced to divide research across the different fields (computers,weapons,shields,bio, engines and construction) because you got interest on the research already spent, thus it was far more efficient to have the research equally divided then to first finish tech 1 and then tech 2. This was combined with the percentage chance of breakthrough you also find in MoO2. Other innovations were the fact that technology could be stolen quiete easily trough spies or conquering planets with factories (it made developed planets really important to attack/defend) This caused all technology to become way easier accesible to other factions once it was researched. (creating new strategies, focussing on other fields and try to steal technology). The game also had random techs so there was not one winning research path. Also you often had the oppurtunity to research a cheap - little benefit tech or a higher tech wich was way more expensive but gave much more bonus. This caused you to decide if you want to invest in a small term gain or a long term

[Granite26]

Hell, I'm pretty sure that in most categories they failed against a crossbow, but they were easy to use. I may be wrong though, but I always thought muskets were pretty faulty and inaccurate, as well as slow reloading.

I was really thinking 'Sid Meyer's Civilization' for this example.  IIRC Musketeers were available as soon as you grabbed gunpowder, and became the new favored infantry unit.  Playing for military conquest insisted you get there ASAP, because it was such a huge leap in technology.

As far as constantly refering to the stats, I'd find that more interesting than the inevitable reduction to heuristics that follow from a known tech tree. 

I've long had the idea of a Blizzard style RTS that balanced itself...  20 units per team, and they got more or less powerful dependant on the outcome of other games.  (You've got 3000 people playing Race A, and they never building Unit B?  Reduce it's price or increase it's power until it's used proportionately... ideally each resource-buck should have an equal chance of going to any given creature in the race.)  The same could be applied to research.  Revalue the research so that for 1000 people given the same 4 starting choices, approximately 250 pick each choice.  (A choice with an obvious right answer isn't a choice, otherwise it's just a series of A or B decisions you have to memorize to get to the real game, and that's lame)


My biggest issue with modern 4X research trees is how fast you blaze through them...  You should be able to fight a major war with only one or two game-changing techs coming out in the duration.  (And it should be EXTREMELY hard to maintain a high technical advantage....  Trade should automatically trail your partners by 3-5 levels below you in all areas)

[Servant Corps]

2:  Unknown advantage.  Variable advantage rates for different technologies.  Imagine a Civ game where the rush to gunpowder failed because muskets weren't actually more powerful than crossbows?

This was what I was referring to deadlycarin, and I don't like it because I have to think if I should flip a coin if I should research Crossbow or Muskets, since in 50% of the time, Crossbows are superior to Muskets, and 50% of the time, Muskets are superior to Crossbows. Add in more complexitiy, and then I'm going to have to randomly choose between the techs because there is nothing concerte, the advantage rates are 'variable', and left to chance.

I don't want randomness. I want the game to randomly choose a tech for me to learn, not force me to randomly pick a tech and hope I picked right and win the Lottery by researching some stupid technology that would be awesome.

[mainiac]

China's economy today is an excellent example of the variation problem(...)

this can be easily done too(it had been done before). Your teritory/planet has varied levels of industrial capacity/infrastructure. You can't build you newest tanks/spaceships where you've got no ability to do so, but you can easily commision a starport, or improve said infrastucure to change that over time. And this could cost more/decrase happiness to reflect the cultural inertia of a given society.

I gotta keep this short but, no that doesn't indicate the spread of technology well at all.  Deploying new techniques isn't just about a central industrial beureu deciding to, even in a communistic country.  To go with HOI, which you mentioned.  Anyone playing as the US 36 starts the game building factories and getting the industrial efficiency techs because they can't go to war for a while.  Then, once war looms, they stop expanding their economy and start building troops.  That's the opposite of history, industrial growth was practically non-existant in the depression and low wages from high unemployment meant efficiency was less important then before. 
The war time boom meant there was opportunity for industrialization (i.e. industry technologies) to spread again and technical efficiency was pursued with new vigour.  Adoption of a technology is about more then devoting resources to it and it's even possible to go back.  In an extreme example, an Indian industrialist started a profitable textile industry before it's time (early 18th century I think, waaay before India industrialized).  But after his death the industry disappeared.  The pain to the local economy caused so much resistance that no one else could make the industry succeed.
But setting local production limits does not model why technical ability varies across a country.  This is significant for strategy games.  What's available in one theater can be very different from another theater, even single battlefields often feature a variety of the same tech.

Yeah... I suck at keeping things short.  It was so straightforward in my head.

[Asehujiko]

2:  Unknown advantage.  Variable advantage rates for different technologies.  Imagine a Civ game where the rush to gunpowder failed because muskets weren't actually more powerful than crossbows?

This was what I was referring to deadlycarin, and I don't like it because I have to think if I should flip a coin if I should research Crossbow or Muskets, since in 50% of the time, Crossbows are superior to Muskets, and 50% of the time, Muskets are superior to Crossbows. Add in more complexitiy, and then I'm going to have to randomly choose between the techs because there is nothing concerte, the advantage rates are 'variable', and left to chance.

I don't want randomness. I want the game to randomly choose a tech for me to learn, not force me to randomly pick a tech and hope I picked right and win the Lottery by researching some stupid technology that would be awesome.

Instead of a lotto, the "better" weapon should be based on what the player(s) is/are doing. Somebody tech rushes for gunpowder displays some promising results with a prototype and naturaly everybody will start copying that. If you have an advantage in crossbows in the meanwhile, you can continue to develop that and use your superior technology to defeat them quickly as it will take them alot of time for them to get muskets ready for mass production and they have to use underdeveloped crossbows and faulty prototype blunderbusses instead. However, if somebody refines making crossbows to an artform, they may be subject to an arms race for a very long time and gunpowder will have alot of catching up to do and may not even be viable by the time the next "revolution" comes along in the form of energy weapons or whatever and thus ignored completely. For a real life example of that, have a look at the gyrojet weapons and how they lost out to the m82.

[Granite26]

2:  Unknown advantage.  Variable advantage rates for different technologies.  Imagine a Civ game where the rush to gunpowder failed because muskets weren't actually more powerful than crossbows?

This was what I was referring to deadlycarin, and I don't like it because I have to think if I should flip a coin if I should research Crossbow or Muskets, since in 50% of the time, Crossbows are superior to Muskets, and 50% of the time, Muskets are superior to Crossbows. Add in more complexitiy, and then I'm going to have to randomly choose between the techs because there is nothing concerte, the advantage rates are 'variable', and left to chance.

I don't want randomness. I want the game to randomly choose a tech for me to learn, not force me to randomly pick a tech and hope I picked right and win the Lottery by researching some stupid technology that would be awesome.

That's an oversimplification.  In the Civ example, you're researching new technologies up the tree.  You'd never be forced to choose between muskets or crossbows, because crossbows are part of the prerequisite paradigm (the middle ages).  The question isn't 'crossbows or muskets', it's 'will spending a lot of time and effort researching a new weapon give me more advantage than staying with crossbows and spending that effort researching rubber or building more cathedrals.

The difficulty it's trying to address is that you shouldn't know how effective a given technology is going to be.  Muskets may or may not be better than crossbows (even in this forum, real history is a matter of debate), but machine guns are certainly better than repeating crossbows.

[deadlycairn]

Not necessarily - we never put enough research and effort into repeating crossbows

[Sean Mirrsen]

Machine guns use gunpowder to throw the projectile. Repeaters would throw bolts with their own force, and use gunpowder to amplify damage. They would eventually evolve into quite deadly weapons. If crossbow research was still active when electricity and magnetism were discovered, we would probably have railguns by now.

[deadlycairn]

Damn - maybe DF is an alternate present?