Diversification of Technology

Alot of talk has been said about how current technologies would be better if you had them in more diversified roles, like a railgun tank supporting other heavier vehicles from a distance.

The diversification/specialization choice isn't hard to do if you make it something that commanders decide and is a real decision. What I mean is, right now you go for one path and you are done. Diversification is mostly inefficient, because by the time you have heavies the game is basically over, and spending time on diversification will usually make you worse off.

I believe most the people here remember the tested research-change modification, I believe it was made by Drag. It introduced exponential increases in cost for additional technology levels. This would increase diversification, but each tech tree already did everything.

What if instead of that we followed the 3-tier design. Only this time, each different branch had a point of focus, like it does now, and disadvantages. You shouldn't be expected to be able to do everything with a single tech tree, but it also shouldn't be best to specialize in it exclusively. For example it could be something like:

Physics1:Single-Shot High Damage, LR, Weapons (LR cannon)
Physics2:High Heat causing, High Damage Weapons (Plasma)
Physics3:Long Range, very slow, very high damage(nukes)
Chemistry1:High explosion, heavy (HE rockets)
Chemistry2:Good Ammo, heavy, high explosion radius (HE cannon/bullets)
Chemistry3:High DPS and Explosion, Heavy (Incendiary?)
Electrical1:High RoF Weapons, Accurate.(Chain-cannons? Good MG?)
Electrical2:Super-High Accuracy Weapons (missiles). Good DPS
Electrical3:High DPS, Heat, accuracy, and weight (Railguns)
Mechanical1:Half Cost, Time to research, and weight weapons, generic
Mechanical2:Half Cost, Time to research, and weight weapons, generic
Mechanical3:No Weapons at all
Bio1:No Weapons at all
Bio2:Good Damage Over-Time, multi-use, good support(Bullet/cannon delivery)
Bio3:Awesome damage over-time, damage, support(Missile/Artillery delivery)

Armor:
Physics: Solid mid-heavy-weight armor (tech 2) Strong against angles, weighs more than average
Chem:Cheap light-weight armor (tech 1) strong against fast attacks, poor armor per plate
EE:Cheap, heavy armor (tech 1) very fast to repair, generates heat when hit
Mech:Slightly expensive, awesome armor(tech 3) A bit of everything. (cheaper than it is now)
Bio: Solid mid-light-weight armor(tech 2) Regenerates, weak against bio.

Engine:
Physics: Good all-around engine. Best horsepower. Expensive (tech 3)
Chem: Worst all-around engine. Great cooling. (tech 2)
EE: Fastest engine. Huge problems when hot (tech 2)
Mech: OK all around engine. Slightly better than chem but poor heat.(tech 1).
Bio: Slight improvement over standard. Does not overheat (tech 1).

This leads to each of the categories having their disadvantages compound. For example:

Physics:Average, always slightly overweight
Chemistry: Heavy Weapons and inefficient armor=poor defense.
EE: Armor generates more heat, more heat damages engine.
Mech: Everything is light and weak, lots of wasted space when alone.
Bio: Damage doesn't stack. Weapons hurt own armor most of all and have good splash.

Research levels would cost more and more, but all items unlocked in them would cost the same 60 seconds and 500 res. All weapons in a category would be unlocked or armor or engine.
Level 1: 400 Res, 60 seconds
Level 2: 800 Res, 180 seconds
Level 3: 1600 Res, 300 seconds


This is just a working idea though for how we could get both specialization and diversification working together. Feedback?

 

Interesting idea, but because of its sweeping changes to somewhat fundamental pieces of the game, I'm not sure many people will like it.

Still, I'd like to see further exploration of this

 

Mostly right now I am looking for feedback on what concepts people do or do not like. The point is to try to find a balance between taking a single tech path to victory and having diversification forced on people.

 

Yes. I like it.

 

Good to see some discussion on this. The research system I always liked the most was a combination of Drag's split-up tree and my suggestion of unique armor abilities. I'll write out a quick explanation (for the fifth time or something, recycling text from previous posts):

Firstly, here's Drag's split-up tree. I'll be focusing on the armors in my explanation:



Right, so the concept of unique armor abilities is simply that: every armor has a unique ability that gives it a certain advantage for a certain playstyle. So I'm not talking about "+20% resistance to missiles" or something, I'm talking about cloaking, health auras, shielding, etc. The main reasons why I feel the combination of Drag's tree and this concept will be the easiest to balance and most interesting to use are the following:

1) With the implementation of unique armor abilities, armors will be chosen not as much because of their ability to counter certain weapons, but because of their usefulness in a certain playstyle.

2) Placing weapons, armors and engines as proposed by Drag's design (Click here and read the bold parts in the quote for some more info) will allow for more flexible playstyles. If you place armors and weapons (or even engines!) in a single tree like the old research tree, you effectively limit your team into one single playstyle or manner of doing battle. This is a gigantic bitch. Don't believe me? How many of you think going full chem or full mech eng is a great idea? Without realizing it, players are already trying to mix and match playstyles from across trees by combining armor from one tree with the weapons from another and the engine from yet another. This proposed system makes it easier to do so, and removes the need for having to know the tree inside out to even get the most basic research paths right.

3) In general, this proposal will shift the focus from "group of best combos in the game" to "group of best combos for that round depending on the enemy tech and your team's playstyle". For a more elaborate explanation, see the last paragraph of my quote (quote can be found in link above). When properly balanced, the chosen research path will rely entirely on what the team does with it and no longer on the individual merits of a few strong research items.

4) More communication between the commander and his troops will be necessary with this proposal, perhaps even some prep-time to discuss the most interesting path for the current round based on what his team likes and what the commander feels will be the best tactic RTS-wise. Addendum: I believe this will be the most important change to allow players to use this system effectively.

Brainstorming sessions with Drag, Dizzyone and CoffeeBurrito gave a few quick ideas for unique abilities:
*For electricity, reactive armor would have a 20% chance of firing an arc of lightning to any nearby enemy tank projectile, destroying it.
*Physics would have very weak armour, but have shields that can protect infantry aswell.
*Bio gets regen or a healing aura.
*Chemistry is completely undetectable by radars, and infantry near it will also not be detected by cameras or other means of detection.
*Mechanical Engineering would receive cloaking. When still, it would be near-invisible, but when moving it would be obvious.

 

I will see your tech tree, which I argue hampers strategic diversification and raise you one.



Your ideas permit tactical diversification, but it is basically an unfocused mashup because of the way it is done. Because you get everything with every tech path and can pick the best engine and armor for the situation specifically, there is little reason to properly diversify. There are no costs and benefits weighed, it is more like a shopping cart with you can grab everything you want before running out the store.

Yes, in this version the best things are tech 3, but they are not the best by such a wide margin that they are the only choice once you have achieved them. Instead the point is to give a logical place to expand if you already have everything you want. Poor weapons alternatives mingle with the advanced vehicle designs in mine to give people a cheap alternative.

Likewise, I agree that giving the armors special effects is a good idea, but special effects within the confines of armor.
Reactive heals quickly
Regen regenerates
Absorbent is good against fast things
Composite weighs almost nothing

These are 'powers'. I admit, I do not believe they have been focused on highly enough, but these are the sorts of powers that armor should have. I don't want my sheets of metal to be zapping random things as they pass by, I want them to be able to take a hit.


In the above tech tree I have proposed, I am pretty sure people have a damn good reason to diversify in almost all cases. At the very least they are always getting to tech-2 mechanical.

On a side note, I would like to propose another tech path that deals with infantry, defenses, and the like, but that is for another time.


PS: Yes, bio seems a bit poor. I had intended it to be unique by having the tech tiers cost half as much and take half as long, but the individual techs at each level take twice as long.

 

I dislike Beerdude26's research tree especially because of this point.
There will be always one best weapon, armor and engine for each map. And with this system you can get them very easy.
The current system adds much more tactical variations, because it costs time and ressources if you want to mix different trees.


And if the physics tree will become a little bit weaker, then the current research tree is very well balanced. It could take really long and many releases to balance a totally new research tree.

 

Ideally, there will only be good combinations that are exploited correctly. Weapons that are good against tanks, will falter against infantry and buildings, weapons that are all-round will be decent but fail against dedicated anti-tank weaponry, etc. Combine this with other stats such as reloading speed, heat produced and unique abilities and all the other crap we've stacked up along the years, and we should have some pretty interesting stuff. In any case, I feel that this should be given a try.

 

Yet this has been given a try in testing. Again, I will state that the problem with your design is that it is completely non-committal. Regardless of their choices players gain access to everything they would want. There is no reason to diversify because diversification is already given to the player with no cost, and further diversification is pointless and redundant.

Why should I get high explosive cannons or rockets, when I can just go up to the next tech level and get fusion rockets, cannons, and artillery that does the job even better? I am not even making a choice I have to mix-and-match with. The problem is that people will like all of these 'options' better in general even if it dilutes strategy in the long-run.

You get alot of unique stuff, but you just sorta get it. No hard decisions, no real strategic diversification.

 

While you are correct that Drag's proposal often already had everything in one tree, that was because you would select a certain tree with a certain armor and a certain engine based on the playstyle your team thought would be the most succesful in a round. I guess if you wanted you could see my proposal as the armors and engines all being kind of the same, with only their special abilities defining them.

EDIT: I'd also like to point out that the testing often consisted of tank battles with a lot of high-tier weaponry, armor and engines from several trees researched. I'm afraid that skewed some of the results.

Now, I can understand why you feel that this is rather a poor choice, and I can understand your reasoning in giving each tree a certain weakness and a certain advantage (or several of each). What I'm afraid of is that research will no longer really depend on tactics or what weapons the player likes to use, but what the enemy uses. I fear that research will become a neverending game of cat-and-mouse, with the team that made the first move being a step ahead. In the end, you may end up with a lot of stuff from across a lot of trees, most of which isn't that good anyway.

EDIT: Also sorry that I didn't respond in the previous message, I hadn't noticed your reply

 

Alright, I will give you a scenario then.

Each technology grants a tactic.

Each tactics beats two others, ties against 2 others, and loses against 2 others. If the tactic is advanced, it will tie against the one it loses to, beat the ones it ties against, and crush the one it already beat. This is a gross oversimplification of what occurs, but it is functionally an unusual game of rock paper scissors. This scenario is basically what you fear.

The problem is that by diversifying and taking two different technologies, you would then gain immunity to being crushed by any given play-style. Additionally, you would be giving the commander the option of choosing his team's play style, to a limit extend. Your tech path merely enables the commander to make the right decision or the wrong decision for a given situation, rather than carve out a particular strategy against an opponent.

i.e. in yours, you could choose weapons that do slightly more damage, lets say versus weapons that have more ammo (physics versus bio). The variations between the two groups are functionally relatively minor. Except for this group ability, and you can choose to have fast engines and expensive composite armor (because you have alot of money).

Compare this to:
You choose to have your guys specialize in hit and run tactics to cost your enemy alot of money and wear him down. You choose physics for this because of the high immediate damage values involved. You have can choose to diversify into chemistry during your early game, because it will increase your damage output early on, but instead decide to put all your resources into getting nukes as fast as possible, attempting to surprise your opponent when they come out faster than expected. This is an unexpected move, because your missile technology is key for NF, but you are willing to suck a little mid-game for the late game supremacy.

I don't see those complicated decisions happening with yours. People get best choices and they don't have to make difficult long term/short term decisions.

 

I find it rather hilarious that your scenario is pretty much exactly how I'd like Empires to be. I guess we'll need to sit down together and explain all of this to me in more intricate detail.

So you feel that the trees should be mixed by the commander to reach the playstyle he thinks will achieve victory? (just a question)

Another question, do you feel hard counters are necessary in this system or do you think soft counters are a better approach?

 

I am pretty sure I just described why I believe that is the worst thing possible, but just to re-establish.

Commanders already do little enough, but their role is one of strategic importance. They should exist to define strategies and help carry them out. Paths like this downplay the strategic elements to a simple upgrade system that includes little to no depth or compelling strategic choices.

Err.... and beer posted right before me, this is to flasche.

I am editing up a response to beer now.

 

The problem is that it isn't actually diverse, it is fake-diversity.

You really just have 5 researches, and your choices are about how to divert between them. Comparing it with the current system alone, it is approximately 1/64th the total possible choices in the first 10 minutes, and the choices that do occur mean less. I will show you what I mean for changing what is 'best' in a bit, right after I finish my numbers. It should take a day or two.

 

Some basic notes I'd like to point out.

First of all, hard counters only really work in RTS games. In a FPS game there is a huge component of player skill, and being frustrated that your gun will 90% of the time lose to the other guy's gun is NOT a good place to put your players. Thus, soft counters (but not so soft they do nothing) are the way to go.


Secondly, we need to try to keep things as unified and simple as possible. The current tech tree system is very broad and complex and newcomers have little to no idea what's going on. So, let's try to make the different trees similar in some ways in terms of structure to make life easier.


Thirdly, we need to ensure that due to the lag of the research time that early choices based on little to no info of the enemy won't completely doom them. This ties into soft and hard counters. See StarCraft gameplay. There are plenty of choices the players can make based on reasonable information, but if they want to play a gambit (like proxy gates, or rush DTs, etc) they need to be able to scout and know what their opponent is doing. Even if they accidentally pick the wrong strat, they aren't completely hosed and can try to come back in the mid game.

This also plays into information flow. It has to be reasonable to, with proper play and investment of time, know what your enemy is up to. I'm not talking about every scout can check the enemy's current tech, but things like clearly marking different weapons/armor using particle effects/textures, a system that helps explain basic counters, and so on. We can't rely on guessing based on refire rate and how much that cannon shot hurt.


Lastly, make sure everything has a purpose on its own. Sure there are synergetic combinations, but everything must be useful even on its own. Nothing should ever rely on another tech to make it useful. Sure it would be more useful if you have this other tech to go with it, but researching it shouldn't handicap you. This is a huge problem now where noob comms might try to rush nukes without getting Heavy Tanks that can shoot them. Using a simple requirement system to prevent this kind of wasted research would be beneficial.

 

yes.. Yes.. YES! *orgasm*

...

But the whole armor special abilities seem silly. Cloaking? Arcs of electricity from REACTIVE armor? That's not how reactive armor works! Reactive is just a shitload of brick looking armor you put on vehicles.

 

Okay, who is arguing what here? All of you seem to be blending together just enough to have distinct arguments of the same color. I am very confused...

Edit: Reactive = Ultra charged plates of metal that, when penetrated, discharge and vaporize the incoming projectile.