"Research" System in 4X Games

Started by
25 comments, last by Orymus3 10 years, 10 months ago

This makes me think of your idea in a new light, and one that is even more intriguing than the original idea. It's not the research that matters (although you could easily have a research component if you wanted) or even your production capacity, but your production facilities that are the key. So if you can make Laser 5, it might take a significant amount of resources and time to retool each individual factory to make them. So even if your society can pump out Laser 5 units pretty quickly, you have to make a substantial upfront investment of time and resources to begin producing them at all; it's a major decision to put your Laser 5 tech into production. That means that you would want to have an overarching strategy for any tech you want to actually use.


I had never actually seen it that way, but I believe you are right. This is also somewhat captured in VGA Planet's approach. Making the switch up from 4 to 5 for example, would cost the equivalent of building 5-7 actual beam weapons. Upgrading thus becomes a cost of option where you're willingly choosing to halt your weapon production to get better weapons. It isn't done artificially with a time delay, but with an actual setback in terms of resources.


This sounds fun to me. On a shorter timescale it might be more fun to use the population as a limited resource. You have 1,000,000 individuals, maybe 35% of which are prime military aged. You can draw from the population to crew your ships and fill out your military infrastructure, but every soldier you lose represents a serious loss that will take a long time in-game to replace (if you ever can). The population doesn't grow very much, but your forces can absolutely shrink.

I'm growing to like this idea as well. I'm still carefully analyzing it to insure it wouldn't mean I need to take away a key component of the gameplay that I would've liked to see implemented. So far, the biggest threat is this:

I need the players to start with large populations and cannot emulate a system where you start off with a few colonists and need to carefully make decisions to increase their birthrate. That said, the system would get there if you spend your populace needlessly as you'll quickly find yourself in a spot where you have too few of them to continue expanding your fleet, and then you'll need to figure out aggressive ways to increase population (cloning perhaps?).

All in all, it feels like an organic limitation to rampant end-game fleet sizes that generally become hard to micro-manage, and given the scope of the project, it feels like a wonderful idea to keep the focus on fewer ships anyway.

Slavery might even become a key component of the gameplay. I had initially anticipated for it to be only minor...


I'm really warming up to the game you're describing. I think it has tons of potential to be novel and fun while still playing like the 4x games I already love.

Most of the credit still goes to Tim Wissman for establishing VGA Planets which remains my base reference. That said, I'm glad to see this spurs some interest :)

Advertisement

This is an interesting idea. I've always been a little bugged by the research systems in most 4x games. Real-world R&D does not really depend on a static function to produce research results and there are far more "dead ends" than useful products created.

Another model to use might be the real-world ship "arms race" beginning some time after the Napoleonic Wars and ending with WW1. Nations were scrambling to try to figure out how to build the baddest and best designs with emerging technologies. IIRC, La Gloire was already obsolete when it was launched, made so by HMS Warrior. So, you could conceivably achieve an incremental breakthrough and someone else's huge success could render it useless.

The other side of the coin is that the "arms race" doesn't necessarily factor in things like crew skill and training to the degree it should. After all, the Soviets had better tanks at the beginning of WW2, but deployed them poorly and had not developed their armor tactics as well as the Germans did. Maybe a de-emphasis on weapon quality in combat design and an increased emphasis on how well those weapons are used would also make sense.

This is an interesting idea. I've always been a little bugged by the research systems in most 4x games. Real-world R&D does not really depend on a static function to produce research results and there are far more "dead ends" than useful products created.

I was actually having a conversation with a coworker the other day, as he was telling me he had just read an article comprised of 10 new inventions that will change our world.

I replied to him saying that probably none of them would ever hit mass-production, and would therefore disappear no matter how appealing they were. Since no real tech ever gets used until a large portion of individuals see the use in paying its initial cost (which eventually finances means to reduce production costs by increase automation) a lot just won't see the light of day.

I've always been much more interested in techs that have been discovered 10 years ago but that, somehow, no one has put to good use yet, than the ones that have just been discovered.

Another model to use might be the real-world ship "arms race" beginning some time after the Napoleonic Wars and ending with WW1. Nations were scrambling to try to figure out how to build the baddest and best designs with emerging technologies. IIRC, La Gloire was already obsolete when it was launched, made so by HMS Warrior. So, you could conceivably achieve an incremental breakthrough and someone else's huge success could render it useless.

Well, since this game will rely on very few ship designs for each species, I want them to remain relevant at all times, and I'd like to have a relatively small margin between upgrades where, say, upgrading from tech 4 to tech 5 beams won't make such a big difference (you won't suddenly start killing everything that moves) but allow you to kill a ship you could already kill perhaps 1 round earlier, reducing the amount of damage you'd take in the encounter and either allowing you to face a second ship and live or reduce the amount of economical cost it would take to repair your ship (thus improving your logistic network as a result).

A lot of weapon improvements generally head that way as well. For example, the (beloved) AK-47 isn't the best weapon in terms of accuracy and stoppage power (a M16A2 would definitely be better there) but it has the advantage of firing efficiently in pretty much any circumpstance (even when soaked in water or covered with sand). At the end of the day, it represents a considerable improvement over previous weapons, without it being significantly stronger. It can't quite compete with a tank's armor yet :)

The other side of the coin is that the "arms race" doesn't necessarily factor in things like crew skill and training to the degree it should. After all, the Soviets had better tanks at the beginning of WW2, but deployed them poorly and had not developed their armor tactics as well as the Germans did. Maybe a de-emphasis on weapon quality in combat design and an increased emphasis on how well those weapons are used would also make sense.

Quite to my point :) Since better weapons aren't necessarily much more deadly, I think it leaves the player with a number of decisions:

- Do I REALLY want this upgrade?

- How does 1% more damage really help me?

- In which situations can I make this an actual advantage worth investing 10% more resources into?

etc.

Gameplay comes with player decisions and options. Its quite possible that upgrading from level 4 to level 5, with your current fleet, opponents, etc might not bring any immediate gain aside from allowing you to tech up to level 6.

The best example I've seen of this is in the Starcraft series, when considering the weapon and armor upgrades.

An in-depth analysis comparing the amount of actual HITS it takes to kill each unit with or without the upgrade revealed several interesting factors.

For example, its quite pointless to upgrade zealots' attack power if they are fighting zerglings. But also, there were some situations where upgrading a stalker's attack to defeat marauders was not as efficient as upgrading the armor because it did not reduce the amount of hits it would take to defeat it, whereas the armor upgrade would allow the stalker to resist one additionnal volley, thus being able to strike one additionnal time. When you pay attention to the numbers, you can really seek means to min/max your decisions, and I don't want to take this away from the player by making each upgrade substancially better.

In VGA Planets, you had to build ships with purpose in mind. If you expected to fight a Carrier, it didn't quite matter the tech level of beams you'd use, so long as you used a lot of them. So you'd pack your 'sacrifice' ship with a bunch of level 1 beams to rid the carrier of most of its fighters. Your ship would die in the process, but would make the carrier defenseless against your second ship which would have much fewer beams, but higher level ones to quickly defeat the carriers (in case it had weapons of its own).

Player skill will be noticeable when they start making these decisions, and for them to be able to figure this out, I think one needs to let them make the wrong decision first (aka, cripple their economy by always maxing the tech they use, etc).

On a sidenote, you seem to have a fair grasp of military history.

My university field of study was military history. I sadly did not specialize in the same eras as the ones you've just mentionned (WWII and Napoleonic conquests) therefore I'm grateful for your insight!



One starbase is level 1 engine and level 10 weapons, the other is level 5 engine and level 5 weapon. I'd like the player to, somehow, be able to assemble a level 5 engine, level 10 weapon. Perhaps however, its best to keep the cost of option present here and simply not grant them this ability?

VGA Planet's solution was to build a level 5 engine level 5 weapon ship, move it to the other starbase, and retrofit it with local weapons.

The idea of moving components merely gives non-critical ships (freighters) the ability to do the moving instead.

I'm not hellbent on implementing this mechanic though.

I feel it should be automated somehow... I mean, there is no decision involved, just pure micromanagement. The optimal strategy is trivial, make 4 starbases, each maxed out with one component, then move your "in production" starship between starbases as in assemblyline.

OR

Maybe make it so the shop can be retrofit only once (means only 2 starbases can participate in building it so you need to build these as some sort of "puzzles")...

Anyway, the question is, why the player would want non maxed out starbases? What is the incentive to make lower quality starbases? Do certain ships require different sets of components (therefore different starbases being optimal for their construction)? What's the player's choice involving constructing starbases?

- How does 1% more damage really help me?

I would say, the minimum increase should be 25%, otherwise it is not noticeable and therefore this whole research and upgrading weapons is just a scam :D

Stellar Monarch (4X, turn based, released): GDN forum topic - Twitter - Facebook - YouTube

It assesses # 1 in that it would be possible to think of a game turn as 1 day or 1 week

it would seem to me that the proper way to do things would be to start by defining the time and distance scales for the game (simulation): how much real world time = 1 turn. how much real world distance = 1 unit of distance in the game. from there, it all downhill.

While its quite possible that one of your starbase quickly becomes maxed out with high-tier components, it doesn't help you defend the outskirts

sounds like what you're talking about is more like construction than research. making incremental improvements to the capabilities of individual installations. and the duckets to do so, like everything else, comes out of the same warchest as your fleet, etc.

Norm Barrows

Rockland Software Productions

"Building PC games since 1989"

rocklandsoftware.net

PLAY CAVEMAN NOW!

http://rocklandsoftware.net/beta.php

Anyway, the question is, why the player would want non maxed out starbases? What is the incentive to make lower quality starbases? Do certain ships require different sets of components (therefore different starbases being optimal for their construction)? What's the player's choice involving constructing starbases?

Assuming a sizeable investment is required to upgrade the starbase, I can think of many reasons, namely, the inferred role of the starbase.

A starbase close to an enemy border would act as an outpost, building ships with better weaponry, regardless of engines for example.

A starbase on a pacific or unexplored outskirt of your empire would want to help the scouting effort by increasing efforts towards getting better engines and sensors, regardless of weaponry and hull sizes.

A starbase closer to the center of your empire would probably want to support the economy by building top-of-the-line freighters, boosting hulls (and possibly engines) and ignoring everything else.

Finally, your capital ship starbase would probably max all of them and produce the fer-de-lance of your fleet at tremendous costs.

For every investment you make, you burn resources that prevent you from building ships. Even worse, as the upgrade is performed on the starbase, you cannot build ships from this starbase. I believe this adds a lot to mindlessly upgrading everything you can. Upgrading your starbases all at once to the new tech might leave you unable to respond to a sudden enemy incursion, therefore planning will be of the essence.

I feel it should be automated somehow... I mean, there is no decision involved, just pure micromanagement. The optimal strategy is trivial, make 4 starbases, each maxed out with one component, then move your "in production" starship between starbases as in assemblyline.

For the sake of simplicity, I believe I'll go without the shipping component parts. You build what you can from the starbase. This will strenghten the role of the starbases within your empire and force you to make just the investments you need.

I would say, the minimum increase should be 25%, otherwise it is not noticeable and therefore this whole research and upgrading weapons is just a scam

Consider the below table:

Tech Type Mass Damage Crew Kill Mines MC Dur Trit Moly 1 Laser 1 3 10 1 1 0 1 0 1 X-Ray Laser 1 1 15 4 2 0 1 0 2 Plasma Bolt 2 10 3 9 5 2 1 0 3 Blaster 4 25 10 16 10 12 1 1 4 Positron Beam 3 29 9 25 12 12 1 5 5 Disruptor 4 20 30 36 13 12 1 1 6 Heavy Blaster 7 40 20 49 31 12 1 14 7 Phaser 5 35 30 64 35 12 1 30 8 Heavy Disruptor 7 35 50 81 36 17 1 37 10 Heavy Phaser 6 45 35 100 54 12 1 55

You'll notice many oddities for sure, yet, there are several strategies that can be employed with these 10 weapons alone.

First, notice the basic logic. Every now and then, you get to a 'well-rounded' weapon, then, a crew-killing one, and then, a hull damage one.

Notice also that some techs seeks to reduce mass at very limited actual upgrades. Notice also the varying costs, and how oddly they scale.

The most interesting are 3 and 4, Blaster vs Positron Beam.

16% increase in hull damage.

10% decrease in crew damage.

25% decrease in mass.

400% increase in Molybdenum cost.

20% increase in Megacredits($) cost.

Would you upgrade from 3 to 4?

It probably depends.

PROS:

As you pointed out, the 16% damage increase isn't stellar. But on some ships, it actually does make the difference. Ofttentimes the difference between a ship surviving or losing an encounter with a similar-sized ship can be shifted by less than 10% increase in damage or armor. Now, I don't need to explain the advantage of a surviving ship, no matter how damaged. You can just tow it back for repair and bring it back into the fray for a ridiculous portion of the original cost in resources and time.

Also, its quite possible you can sport more than one beam on your ship, and having the increased damage on every shot, if you have more beams than your opponent, continues to increase this advantage. While a ship with fewer beams might need to consider getting a better level beam, having 4 or 5 of these can really stack quickly.

If you are building a reconnaissance, scout, or skirmisher/dogfighter, or even a ship that is meant to go deep and harrass the enemy freighters, you'll applaud the reduced 25% mass. This will make your ship able to go faster with smaller fuel costs and will help with your mobility. Imagine you could get all of the components with such a bonus: you'd actually gain an overall 25% mobility, and that is an advantage that is noteworthy if you intend to play the guerilla tactics game.

CONS:

In some cases, the added 16% won't be that useful. Either your ship uses beams for the sole purpose of killing smaller crafts and making way for the big guns, or your ship size is large, and you know that in order to defeat larger ships, you'll need a much bigger increase than that.

If you intended to capture an enemy ship, it will not longer be efficient because you can't kill the crew as efficiently. In fact, chances are that, by the time you kill the last crew,

there will no longer be any ship to fire at anyway.

The increased Molybdenum cost (which is the rarest resource) is frightening. Chances are you can't willingly choose to spend so much for an only 16% increase in damage.

What I like here is that you can measure the pros and cons of every single beam in there, and realize that many of them will remain relevant for long periods of the game.

Would you be surprised if I told you that the levels 7-10 beams are far from being the most used beams in the late-game? They are present, but in much smaller numbers than you'd expect. Some of them are even frowned upon. Though they bring more firepower, it oftentimes is firepower that isn't necessary (especially when fighting a carrier). So this gives the option to players to ruin their economy simply because they haven't taken the time to consider their needs.

it would seem to me that the proper way to do things would be to start by defining the time and distance scales for the game (simulation): how much real world time = 1 turn. how much real world distance = 1 unit of distance in the game. from there, it all downhill.

Initially, I wanted to use Newtonian physics, but in the end, I'd rather do what's fun than what's real. In other world: balancing and fine-tuning with user tests to insure is as fun as it can be and feels right to the players, regardless of whether this is real or not.

sounds like what you're talking about is more like construction than research. making incremental improvements to the capabilities of individual installations. and the duckets to do so, like everything else, comes out of the same warchest as your fleet, etc.

Precisely.



Assuming a sizeable investment is required to upgrade the starbase, I can think of many reasons, namely, the inferred role of the starbase.

A starbase close to an enemy border would act as an outpost, building ships with better weaponry, regardless of engines for example.

A starbase on a pacific or unexplored outskirt of your empire would want to help the scouting effort by increasing efforts towards getting better engines and sensors, regardless of weaponry and hull sizes.

A starbase closer to the center of your empire would probably want to support the economy by building top-of-the-line freighters, boosting hulls (and possibly engines) and ignoring everything else.

Finally, your capital ship starbase would probably max all of them and produce the fer-de-lance of your fleet at tremendous costs.



For every investment you make, you burn resources that prevent you from building ships. Even worse, as the upgrade is performed on the starbase, you cannot build ships from this starbase. I believe this adds a lot to mindlessly upgrading everything you can. Upgrading your starbases all at once to the new tech might leave you unable to respond to a sudden enemy incursion, therefore planning will be of the essence.

As a designer I can see it working and even being enjoable to some people. As a player I'm unthrilled and would probably pass on a game with that description (mostly because it assumes there are hundreds planets I need to micromanage and crawling slow fleets that become outdated once they reach frontline - not my style of playing).

Would you upgrade from 3 to 4?

That's what I would consider a scam :D I have invested in the technology, it was hard and expensive and now they are telling me to consider if it is worth implementing that technology or not... I understand the decision could be if I want to retrofit the OLD ships, but for new ships it should be no brainer, put it on all new ones since it is a superior technology that give you an instant benefit (that was the reason why I researched it in the first place).

This simple makes the whole research soo weak I wonder if it makes sense to implement that feature...

How about you enable ALL technologies from the start? And the ony decision is about upgrading your infrastructure?

Stellar Monarch (4X, turn based, released): GDN forum topic - Twitter - Facebook - YouTube

As a designer I can see it working and even being enjoable to some people. As a player I'm unthrilled and would probably pass on a game with that description (mostly because it assumes there are hundreds planets I need to micromanage and crawling slow fleets that become outdated once they reach frontline - not my style of playing).

You are entitled to your opinion, this is obviously niche gameplay.

With that said, the scope of the empire management wouldn't reach these unbearable limits typically associated with 4X games.

For example, most games of VGA Planets end up with players controlling FAR less planets than other 4X games. You could technically have a dozen important ones and a few abandoned outposts used for sensor ranges only.

Also, fleet doesn't generally become outdated. Ships created on turn 7+ are generally what you'll need up to the end of the game. You still have to gauge your production of large scale warships vs freighters vs smaller scout/harrassing ships though, and will generally need to refresh your fleet with every critical loss, but the essence of the game remains.

That's what I would consider a scam I have invested in the technology, it was hard and expensive and now they are telling me to consider if it is worth implementing that technology or not... I understand the decision could be if I want to retrofit the OLD ships, but for new ships it should be no brainer, put it on all new ones since it is a superior technology that give you an instant benefit (that was the reason why I researched it in the first place).

This simple makes the whole research soo weak I wonder if it makes sense to implement that feature...



How about you enable ALL technologies from the start? And the ony decision is about upgrading your infrastructure?

Perhaps you were confused, but that is precisely my intention: all techs available from the start and only gauge infrastructure.

The premice here is that I wouldn't expect new techs to come-to-market within a few years of space travel. As a result, I'm taking for granted that all of these techs are ready-to-use, but require tangible investments to upgrade the infrastructures.

Thus, there is no research investments, just infrastructure.

Its also, actually, a much simpler system to implement than the usual research tech tree.

Oh, now it makes sense :)

Yes, it could work, I see no obvious big problems here. Still, that's not my favourite thing to play (but I would probably try it out of curiousity).

My biggest concern is how you market it, maybe using the "4X" term is not the best in your case since it's quite far from a traditional 4X game... Maybe call it "space warfare logistic sim something" or "turn based space RTS"? I mean, if it appeals mostly to SC fans and VGA planets fans try to name/phrase it so they get it's a game for them (and they not necessarily might be fans of 4X). Plus, if you market it as 4X you will get poor reviews because it might be considered "a 4X that lacks research part and other obligatory elements".

Stellar Monarch (4X, turn based, released): GDN forum topic - Twitter - Facebook - YouTube

it's quite far from a traditional 4X game

Actually, its closer to the origins of the genre smile.png You have to remember that VGA Planets, one of the original titles, was actually following this model.

It does have all 4 Xs (eXplore, eXpand, eXploit, eXterminate) possibly in larger quantities than modern 4X games which focus mostly on the 4th X.

The intent is to put more emphasis on the other 3Xs. More focus doesn't mean more scope though. Just because the game plays with a fraction of the units commanded in other 4Xs doesn't mean its any less of a 4X.

With that said, as I've previously mentionned, I'm not hellbent on it being part of the 4X continuum. I just want it to be fun and mind-breaking smile.png

Maybe call it "space warfare logistic sim something" or "turn based space RTS"?

I've been googling for space warfare logistic sim and tossing economy keywords in there as well, to no avail. It may be that its not a very "marketing-friendly expression", or that there is visibly a niche that needs taking care of. I'll see how I'll market it when I get there though.

For now, its really just a game about ships, planets, economy and warfare.

a 4X that lacks research part and other obligatory elements".

I don't believe research is an obligatory element. It certainly wasn't this way back in the original 4X titles, and the fact it became a cheap "reflex" to associate the two shouldn't lead to a bad review. Afterall, a review is about the fun, not genre-consistency (at least it should be).

That said, I'd also be interested in seeing a game that really redefines the way we look at research in games. I've seen games where discoveries were more random for example, and it was interesting. I'm just not too fond of the current mainstream application of the system, quite possibly only because I've seen it too often.

This topic is closed to new replies.

Advertisement