Risk vs Reward

Obviously, this is a very complex subject, and one that people do a disservice to when they simply cry “Vanguard is op! Pls remove it!” In order to discuss the issues surrounding this, I’m going to posit four hypothetical systems by which someone could construct an RTS, and then analyse what they tell us about upgrades, side-grades and the ancillary gameplay around them.

So let us consider the following scenario. You are making an RTS. You already have a system in place by which players can gather resources, and spend them on units. The only things left to add are the combat units themselves. However, like Uber, you are on a budget. Unlike Uber, your budget is not in the millions, meaning you only have time and resources to make three units. No more. How many different ways are there that you can come up with a system to diversify those units? I can think of 4:

1. Rock-paper-scissors
You start with the basic unit of a tank. You then create two other tanks with identical movement speeds, DPS, health, etc. However, tank A does double damage against tank B, which does double damage against tank C, which does double damage against Tank A. This is where most RTS games draw their inspiration from. I hope I don’t need to explain why this is not the road that PA is going down. To put it simply, this RTS game only works based on the premise of knowing your opponent’s army composition, and any decision made in the absence of such information is little more then a gamble. This doesn’t offer the kind of depth that we are looking for in an RTS and if we only had three units, this would be a terminally dull game.

2. The Bigger tank
You start with the basic tank. The next unit is still a tank, but it is twice as powerful, and has twice the cost. The final tank is twice as powerful again, and twice the cost again. What we have here is a flat, but neutral balance. An arbitrary unit of cost still provides an arbitrary unit of power.

There is one difference here, in that most RTS games follow the convention that units do the same amount of damage regardless of their health until they are destroyed. This means that the biggest tank does have an advantage, in that it can comfortably destroy one enemy tank in an army, whilst still remaining as powerful. Thus there is an advantage to bigger units. Let us assume that this phenomenon (which I will refer to as “Concentration Factor”) can be balanced around by making bigger units slightly weaker than a linear projection from cost would imply. Overall, this “bigger tank” system is again serviceable, but it doesn’t really add interest as there is no functional difference in just upscaling numbers (assuming CF is accounted for). We still really only have one unit. My main point in describing this is to show that a “bigger” unit does not necessarily have to be “better”, just that size and concentration of firepower is another vector for unit differentiation.

3. The Better Tank
Again, we start with a basic tank. The next tank costs more, but is proportionally more powerful still than that cost would imply. It is a better tank. The same again (but more so) for the final unit – the best tank. This system may or may not account for concentration factor. This is the straight upgrade system.

Here, it becomes important to understand Metabolical’s risk vs reward points. The risk of constructing a unit is embodied in its cost, while the reward for completing it is embodied by its power. However, that risk to reward ratio becomes smaller with the better units. There is a dominant strategy in this kind of RTS: “progress to the best unit as fast as it is safe to do so”. At any point in the game, there is a tier that you should be using and failure to correctly estimate this will lose you the game. This is a more interesting strategy compared to the other examples, as it does require some degree of assessment by the player. The player must make a judgment call on whether or not it is “safe” to progress to the next tier of tanks. This does create some interest, but at the end of the day, it’s still not that much. I can still sum up the overriding strategy in this game in a single line, with only one variable based on player assessment. It’s better than the previous descriptions, but by itself, it’s still not the best. Let’s come to the final system.

4. Apples and Oranges
The three units are a tank, a mobile artillery piece and a scout. They are not comparable.

How on earth do I determine the optimal strategy here? Do I destroy my opponent with a column of tanks? Do I hold back and shell him from a distance? Do I raid and irritate him with scouts? This system is by far the most difficult to optimise because the externalities of any given situation will determine not just the best unit, but may potentially allow multiple solutions to the same problem.

This is by far the most engaging experience, because it is the least “solvable”. It has the highest likelihood of throwing unique situations at the player that they must solve on the fly, rather than forcing them to hone their execution of a single pre-determined task. It is also an experience which many feel is not well catered for in the current RTS market, and one that PA has an opportunity to capitalise on.


At the moment, PA is a hybrid of type 3 and type 4. Some units are just better versions of their inferior cousins. Some are radically different. This is where we come to the crux of the matter. I do not think anyone here has a problem with units which are very powerful when used correctly. What I think people have been trying to articulate is that units should not be more “efficient” for their cost across the board, compared to other units. For example, let’s look at the basic tank (whatever it is called these days) and compare it with the leveller. In the current released version, the leveller is very much cut from the same cloth as the basic tank, but simply upscaled and improved. It costs more, but it has longer range, more DPS and more health. Uber is trying to balance this by upping the cost relative to basic tanks. However, all that this can do is take us from a type 3 to a type 2. Without being more ambitious, we are still stuck either with a “no real difference” type of diversity, or a “use the better one as soon as it is safe to do so” diversity.

Let’s now look at a comparison of two tanks from TA. The core had a basic tank called the “raider” and an advanced tank called the “goliath”. Despite them both being tanks, they operated very differently. The raider was surprisingly quick, averagely armed and well armoured, although quite pricey for its tier. The goliath cost the moon, had a beast of gun, an excessive amount of armour, but the speed and manoeuvrability of Mount Everest. It was a very different kind of tank to its distant basic cousin. It was very much a case of type 4.

This is what I believe the community is trying to get at. Risk vs reward is a perfectly good game mechanic, but it is ultimately quite solvable compared to the creation of completely different units. One could very easily make the leveller a very different unit, by changing just a handful of numbers. One can even still leave it as a very expensive unit, which can best any basic tank in a one-on-one fight, as long as it isn’t just better on a metal-for-metal basis at fulfilling the same role.

At the moment, PA doesn’t have a huge number of units. That’s fine, as we are still very early in its development. However, that means that every unit which is an upgrade is a waste of an opportunity to be a more interesting sidegrade.

First of the unit cannon is already in the game. Its the T1 Tank. Second of all, what I would do.

I would either attempt to kill my enemy as quickly as possible with basic tech or if I am confident enough that I will not be interrupted or I could hold the fort I would tech to Meta and get the I WIN button. Otherwise I would play T1 game until I either won or reached a stalemate of some kind when that happens I would again tech to meta. Once I am meta tech I would forsake t1 all together and never look back. If Meta tech is indeed so rape I would never try to run T1 with it unless it somehow suplements the meta tech (I build 2-4 T1 factores for spinners, combat fabbs and stingers, but that is all). As for the running costs. I would obviously build the fabber first and build meta tech eco but after that I would undoubtedly rush for a game ender.

This is highly hypothetical. This game here doesn't even support the usual itineration of this final answer which would be that this makes the game like TA or other games where you have a superunit and the counter is sniping it or you have armies of utility units and you snipe super units and chip away until you win. Basically, the result is, a super unit wins if it is made, and a utility army wins if it prevents super units as one prevented superunit and the enemy can no longer fight standard units.

Except, this game has planets. That unit couldn't travel planets. Picking that unit up probably isn't possible, if it was it still would get shot down by fighters or anchors. It really is a catch 22. That would only work on per-planet fights, and this sort of game would have worked in the mid 2000s as the gimmick would have been appreciated as at least two-stream balance, where you either build superunits or build the tools to snipe one that comes up, like Red Alert 2 and superweapons, or Tiberian Sun much the same way, or C&C 3 Tiberium Wars and their expirimentals, I loved those games more or less.

Sadly, I would like many flavors of army ultimately make this game work, rather than army tools vs a superwin unit. This game shows promise in that direction. Gil-e when balanced is still one option to kill a commander as long as they needed something else to kill buildings in bulk or whatnot. Levelers getting multiple angles are another good way to kill a commander. Laser sats. Bombers. Kestrels. Artillery. Nukes. Halleys. So many endgame tools. They just need to be balanced. Like MadSci said, type 3-4, except type 3 isn't taken to such an extreme to be a superunit but a variety of superunits. This build forsakes the prior tier but the next one sounds like it encourages it again so we are in a good place don't get me wrong.

As far as the tiers go, I always just felt they should just be additional tools selectively chosen to add to basic armies. You shouldn't need to rush t2 extractors, you shouldn't even need them in the short run. It should take a while to install t2 extractors, their cost should force you to turn off a few factories, and you shouldn't be able to turn the factories back on until you installed almost half your metal in t2. The only bonus in building them should be after finally installed, you can have even more factories going than before. If you can build another metal spot, it should always be cheaper than building a t2 extractor. As far as the factories, if you lost t1 factories it should be a travesty, you should worry about them as much as t2 factories. Many many t1 units should accomplish general destruction and self defense. Each t2 unit should only accomplish it's one role, and not be able to exclusively defend itself.

As far as all this goes, really t2 should just be tools. Leave mid-stat direct fire roles as t1, leave specific roles as t2. Make t2 cost more, but not for it's power, for it's role. Leave t2 economy as a eco-dump with marginal returns, people will still resort to them as a tool if they require those returns but it shouldn't be the better choice than getting a brand new metal spot. T1 doesn't have the tools to score commander kills besides numbers, which kills bases more than commanders. T2 has the utility to kill commanders, snipers and artillery and all, just not the numbers to kill t1 armies albeit some units might achieve attrition on larger numbers using smaller numbers.

I think you're doing it wrong.

There's no point in reading every opinion if you're not going to take them into account, try to grasp what the core idea of the other guy is in your own mental language and so on. A discussion doesn't need "professional readers." (Every opinion doesn't need a reply either.)

Don't take this the wrong way, not like pointing the finger at you or anything, but on forums and conversations in general reading/listening in the literal sense is a very poisonous (and pointless) position to take since most of the time it means you're just doing it to further your own agenda, ie. the whole attitude of "I read every word you wrote and slammed every one of them in my reply" instead of what it should normally be like:

• state your point/position if it's different
• clarify your position if it's misunderstood
• add to your point/position if you need to
• respect that others may have and are entitled to have a different position

Obviously read before you reply but how much you read is irrelevant. It's the responsibility of the one writing to present his point efficiently and make is post readable (I'm very much against the inversion of "understanding" whereas its the readers responsibility to decipher the arcane "truth" in posts).

I always assume everyone doesn't read everything, since that's fair to everyone's time.

I understand most arguments, but I rarely agree with them.

And I very rarely omni-slash anymore, neither does most of the forum, we have all become very respectful of each others points even if we don't agree.

We already do what you ask, but you do need to understand what an actual back and forth debate is.

Also what the internet is like.


A debate is always about trying to convince other people of your point.....because that is the point.

Just stating our own opinions would kind of undo the point of us even having a form, as it is as effective as writing our opinions on our walls at home.....kinda pointless.



An in regards to this topic, I have been VERY open to the idea of creating an interesting teching mechanic that also keep players from getting simply better stuff.

But I happen to disagree you your opinion and you lecture me.

Perhaps you should be understanding my opinion?

Hi, interesting analysis, and overall I agree with you. However your anaolgy concerning TA isn't really fair. Yes the Raider and Goliath were very different- however there was a direct upgrade in T2 for the Raider- the Reaper. The Reaper was a twin cannon, larger version of the Raider. Interestingly in terms of counter- Raiders were more usefull against the Goliath than Reapers were as the Goliaths main weakness was its turrent turn rate (and slow rate of fire). For the cost you might get a couple of Reapers or 10 x Raiders- in which case the 10 Raiders could swarm a goliath and attack it from all sides- preventing it destroying all the Raiders. In the same scinareo with 2 x Reapers the Golith had much more chance to destroy them as it only had to deal with 2 targets and the Reaper was slower than the Raider.

TA did have direct upgrades as well as diversity- and I don't see the issue there. I think that the changes to T2 will bring T1 back into play much more which is nice. The main thing this highlights though is how many more units TA had at the T2 tier in particular compared to PA.

In TA T2 Vecs (core) we had the Reaper (heavy tank- all rounder), Goliath (super heavy tank- good at taking on defences and tanking damage, vulnerable to being swarmed), Mobile Artillery (good at range, vulnerable up close), the Diplomat (long range missile unit- very strong against single static targets, can't hit moving targets), the Copperhead (mobile Flak) the Hedgehog (mobile anti nuke) The Informer (mobile t2 radar) The Deleter (mobile radar jammer) and the Croc (Amphibious tank- in between the raider and Reaper for strength).

So in total we have 3 types of T2 tank for different roles, 2 long range units (again with differences depending on situation) and a slew of support type units. I really think PA could stand to having its unit roster fleshed out a bit.

I never said that TA didn't have upgrades. Only that it had some good examples of comparible roles which were actually quite diverse. Even the example you give of reaper versus raider still felt a lot more different than PA's basic tank versus the leveller. I'm not here to suggest that PA should be a shot for shot remake of TA, only that TA teaches us some lessons that is worth learning.

One of those things is that "apples and oranges" is fundamentally more interesting than "the better tank". TA could afford to waste one or two units that were quite similar, simply because it had an awful lot of units. This highlights that there really shouldn't be much focus on upgrade units, until a lot more roles are fleshed out with diverse units.

Depends. What if meta factory was specialised? That way it would not be more risky or rewarding, but it would still be useful. Imagine that the meta factory build infernos and the regular built just tanks. Obviously not everyone would use the meta factory but a lot of people would because it adds to their play style.

Since you obviously know and understand our concerns I will simply state that it really helps that you respond here and explain your view on the matter.

It removes that feeling of talking against a wall that you often get with other (usually larger) companies/developers and creates more confidence in that the issues we see will be resolved / not be present.

I am still worried about the large increase in income but I will now mostly wait and see what you come up with in the next balance patch, since voicing the same concerns over and over again does not even feel helpful if you know they have already been heard

Keep going and no matter what solution we end up with, it will be better than what we have now

I've clearly arrived very late to the party here, and all the arguments I thought of in my initial blaze of anti obsolescence rage have been put forward already. (Thank you for your page 3 post @brianpurkiss, I can't quote on my phone)

The one thing I would like to say is that I hate this idea of a 'T2 army' and a 'T1 army'. Discussing which of an equivalent metal cost T1 and T2 army would beat eachother pisses me off, because there shouldn't be armies composed entirely of T1 or T2 units. T2 units should complement T1 units. They should fill different roles. The 'reward' of going to T2 isn't more power. It simply allows more diverse and complex strategy. Games should be winnable with only T1 UNLESS your opponent beats you with superior strategy. And only then.

This has all been covered already, and I don't have any arguments to add but it makes me so mad I had to comment.

#nounitobsolesence2014

Thank you for your post. This sums up my opinion on the matter.

Although I haven't made comparisons of mixed armies for a long long time, it's always on the back of my mind. It's why I think it is relevant to say "t1 units are killing T2 units" - you need to protect your valuable t2 units by using them with t1 units.

What large increase in income?

I really like this point. So I highlight it.

If the above concept would be implemented in the definition of T1 and T2, we would be closer to a game with endless options for victory, depending by the actual strategy put in action by the players.

Instead I'm still a bit afraid that the game is moving towards a simple set of solutions for a victory: Inflated T1 vs HighTech T2. Which the second clearly advantageous in most scenarios.

Well I think the point your missing is that t1 and t2 can mix- the point to the comment about quantities of t1 vs t2 only pertains to the T1 tank in relation to the Leveller as with these changes that will be the only direct upgrade for land.

As I mentioned earlier- now you have a choice of realistically building lots of t1 tanks instead of a few t2 with similar effectiveness. This then leads to the question- what's the best use of my T2 factory?

Well in this case I'd argue you'd be better using the T2 factory for production the support units (e.g. shellers) whilst producing T1 for tanks and aa. Its similar with bots (where if the dox becomes an anti wall 'grenadier' then the stomper is no longer a direct upgrade).

There is nothing wrong with having a basic and advanced tank. TA had 2 tiers of tank- so what? The key is having the other units as well.

If anything I think the upcoming balance changes will simply invalidate the leveller as there are more useful things to build at t2. Personally i think the issue here is that PA could really use a bit more variety in units, especially at tier 2.

I get that what's trying to be depicted here is the beheading of the ever popular level unit roster philosophy on the forum. (At least I hope so, else I'll sound out of context from here on...)

My take? well I've voiced it a considerable amount in many-a-thread and here it is again : YES! hell yes, people have to get real and realize there will be units that are BETTER than other, that can crush a great number of a lesser unit before being overwhelmed.

You'll have that or you'll have your one-unit unit roster.

I'll have the big unit roster and FUN please.

I'm really tearing apart inside at the idea that the meta tier was only an image to illustrate a propos because god does it make sense especially for the renowned aforementioned 2-week 40p game.

How the hell do people keep wanting to stuff a plane unit roster with perfectly equal units in a million-unit game like millions of indistinguishable souls captive of the hell of boredom?

ahhhh here's where people try to weigh me down with (I'll just pop a note here to say : theory crafted) balance arguments. "Oh noooes! how ever will this work?", well lets set aside my ever faithful argument of: "well... I know one game were it worked ..ahem! ..works fine." because hey, people want novelty, even when it comes to new way to have their theories torn down.... : It depends. First off I'm not going to waste my time doing that on a 400R, single-planet system, there are certainly faster ways! but If they game is of a type that will likely last over 2 hours (time to pop another short note : exactly the type of time frame you'd start envisioning building a Paragon as valid.. oh.. whoops) then yea I might start to think I have to use this tier at some point in the game. The game setup is sure to last beyond three hours then by Poseidon I'd be a fool not to try and rush it as soon as I can whilst holding the enemy at bay.
Would I use it for the I win unit or the neverending stream of units? .......depends! I'll have to use my strategy, and assessments skills to determine the quickest way. Or maybe my escalation of commitment weak strand to choose the wrong or slowest one. Certainly varying the circumstances each are as valid.

oooo, tough one, didn't see it coming there. Now I'm hesitating.

LOL, yes.

It's pretty clear you don't understand the counter argument here. Most people seem to be against a much better T2 economy and unit effectiveness per metal.
I don't have a problem with a unit that's 20 times as effective as a tank in the same role, so long as it costs around 20 times more than a tank (it's a waste of a unit though, because it could have had a completely different role). The problem with more efficient units is that if it costs far less, it makes the tank obsolete. If it performs a different role but is still more cost effective at that role, strategies involving it will be more effective and it can still indirectly invalidate less effective units if their roles cross at all (even if they're only as similar as land vs. land).

Now, either way you're changing from a game about armies to a game about fewer more expensive units, unless you boost the eco accordingly, in which case you have a binary option that pretty much determines the outcome of the game. That can be balanced, it can be unbalanced in one direction or the other (as meta's post points out), but it's a bad mechanic IMO because of how binary and how important it is.

balanced != fun

I just wonder. Everything is fine now. But why CANT the leveler just have a lower Damage per Second than an ant, and a higher Damage per Shot than an ant? That simple change would make levelers utility, but not standard, because in long fights you are losing more of these expensive units against cheaper units. You would still use them in snipes, invasions, and strategic placement. I would imagine with formations, most people would stack levelers behind ants because their range overlap to make the damage devastating.

I don't care much for basic and advanced. I would be more inclined to not use t1 and t2, if the terms were diverse and specialized. Div. bombers inc. Spec. fighters on patrol. Ect.

T2 economy theoretically being able to output a few times as much resources as T1 economy.

Yes I do and I know what I'm talking about, this "level unit roster idea" has never been playtested, it really should, just for the sake of having proof once and for all that it's atrociously dull, not to mention an off-rocker idea to boot. I can see why people saw a semblance of that in TA but this idea is NOT what TA was. Ta made a unit valid versus another unit if said first unit was in much greater numbers and made both worthwhile to have by having the group of units and the single 'more resiliant' unit be relatively equivalent in cost. having a mix was even more interesting.

Even though Supcom added two techs (counting experimentals) it stayed in the same exact strand. Just at a bigger scale and with less "theoretical balance" more risks taken in terms of balance. Heck some things were even outright plain imbalanced! ...I mentioned above the Paragon, was it balanced? Heck no!! what does infinite mean? that a thousand, a million or billion or a trillion or whatever you want to throw at it will never measure up.

was it the winning unit every game even though people knew this? Certainly not.

You could say that a crushing majority of games where a player pumped out that unit successfully where won by that player. It could still be destroyed, the player could still make dumb mistakes. But unless you're like @Martenus you'd avoid calling out GG untill you've finished the job.

That being said : a crushing majority of those who attempted to build it rather than the "weaker", "crappier" units lost in the process. or had to give up that silly strategy.

It's still a valid strategy, just not the one you win a bigger percentage of your games with.

And this applied to all 400 of the darn units going gradually downhill all the way down to the t1 artillery which had gotten you near 90% of your wins in Supcom. well... all depending on your preferred map, player count ect...

Gee, they used tiers and strictly better units they must have been terrible gameplay variety-wise....

Now that I have actually forded the river of this 5 page topic I can now meaningfully contribute to the current debate.
At first I was all for sidegrades, as early as possible but I think people opposing that are right in some degree. Having a single tech level would be boring and we could come against some major balance issues like the vanguard+pelican drops at 4 minute mark or where the advanced units simply aren't needed because oyu can just swarm them with basic units. There obviously needs to be some form of teching up, but then again the concept of Meta factory replacing current advanced factory really doesn't improve much. What it basically does is it makes a T1--(5 minutes)-->T2, where T1<T2, into T1--(10 minutes)-->T3, where T1<<T3. Basically the beginning and end point are more or less identical. Advanced tech is better than basic and that still invalidates basic which is what we (majority of community) do not want, the only difference is that T1 stays viable for a bit longer.
What I believe would improve the game is that Advanced tech should basically offer more options, that are only slightly better than basic but still keep basic a its backbone, while offering no economic advantages. Basically it should be a game of T1 and T1.5 Obviously this would render some of the units we currently have, duplicates for example leveler, slammer, inferno, hornet, peregrine, advanced combat fabrication bot and the battleship. Those should be repurposed into alternative roles or removed altogether. This is because the economic advantage is the biggest creator and drive in teching up to advanced and it also creates what, we have heard in this topic but I do not agree with, a penalty for not teching up. If you don't tech up it's your fault. You have left yourself to be exposed to nuclear bombardment and Economic disadvantage.
This statement also touches the second reason why you must tech up to tech 2: Countering, which has also been mentioned somewhere in this topic. Having a higher tech specific counters for higher tech unts forces the player to tech up, and further emphasises the huge improvement that advanced is to basic. There is nothing I can do to stop a nuke at basic tech, additionally there is very little I can do at basic tech to stop a death wall for 100 Vanguards, or a swarm of Peregrines and Kestrels. This would also be partially be mitigated if advanced was Tech 1.5 rather than Tech2. Partially because it means that while easily accessible and with less build up, there are still counters and defenses that require you to be Advanced to defend against them. This second part would simply be fixed by bringing those defense to basic tech. We have already seen this working with orbital, which was once considered T3, but has not been moved to T1.5 with umbrellas and orbital and deep space radar being T1.
So rather than having just one basic tech where everything is accessible and flat or having the mystical meta tech that is really hard to access but still a requirement we should have only a minor upgrade, that makes advanced an attractive option but not a necessity and we should remove the drivers that force us into teching up and punish us for not doing so.

Hmm tiering up with high risk high reward contradicts with the desire to have many tier2 factory's and many tier 2 units.

Also you have to accomplish the risk part as well and thats whats missing in the current released version of the game. (Not trying to go T2 asap is actually the risk now )