Internal Playtest – Balance Build, Social Features Reveal, and More! – 4/4

Anchors not moving? You know what would look cool on the immobile anchors?
Actual anchors, like a space elevator.

Sorry, I just don't see a problem with an exponential economy...
Maybe it's just me, but I can't see how those changes make you come to T2 economy faster?

Current experimental cooldown is 4 sec. We just used the ammo system like the bombers use because we want there to be a visible indication whether you can fire the Uber Cannon or not. But it's not very clear still so we're not satisfied.

We raised the relative cost of T2 factories to make it a bigger risk to rush T2, and to give us some economic room for the "Tier 1.5" orbital. If it takes you longer to get there, then if you rush it you have a greater risk of your opponent coming in with a dangerous T1 army and destroying your stuff. So T1 should become more relevant. But then if we still want you to build additional T2 factories, we needed to scale the economy after that accordingly. This experiment was intended to test that theory in game. And T1 certainly felt more relevant in this game.

I don't see how this causes snowballing at all. You can afford more factories, but you can't necessarily afford to run them, because everything has the exact same relative cost within T2 as before. If anything, the concern should be that they are harder to run because your existing T1 economy is less helpful as you enter T2.

The in game twitch controls have been in for a while.

Gunships missing was not an intentional balance change, and may be in the live build.

The orbital factory where you can queue up orbital stuff you want to build feels really good.

There is no difference mathematically between scaling all the T2 stuff the way we did, and lowering all the T1 costs including the T2 factories, and lowering the T1 metal extractor. We could have lowered the T1 metal extractor to produce 2.3 metal/sec and made everything 1/3 as cheap (or about 1/4 now), except that explaining 2.3/sec instead of a round number is lame.

Maybe the thing to do here is talk about principles rather than implementation. I bet we can agree on that anyway. (For more on principled negotiation, I recommend the book "Getting to Yes").
1. T1 units/phase in the live build feel underwhelming and less relevant.
2. Part of why is because of the difficulty using them effectively to assault a base or kill a commander.
3. Part of why that is hard is because the uber cannon can wipe out so many units, and towers are strong.
[Incidentally, we did those things to pull away from the situation where people we just make a death ball of dox and kill everything. So it was more like an over steer. Over steering is good because now we have specific examples on what is too far one way and too far the other, and that helps us find balance.]
4. Building all your orbital with fabbers is hard to see and doesn't have the command and control advantages of factories.
5. When you're coming in to crack a planet (if for some reason you don't crack it with an asteroid) it's hard to create an on ground beach head from space.
6. One of the reasons that is true is because a fully developed planet can be patrolled by a lot of units making it hard to get a teleporter down, especially if it takes a while to build with only 4000 health.
7. Another reason is that the only orbital to ground weapon is a sniping weapon that fires infrequently for large damage (but that has a role).
8. When you are taking over territory in T1, you can often just build a proxy base of factories, but at T2 it feels too expensive to build 3 or 4 t2 factories by 5-10 metal spots.
9. Role diversity is low. Dox, Unit Cannons, Levelers, and slammers all feel like they server very similar purposes: to provide consistent straight line/accurate dps in battle. Diversity might look like different units are better at kiting, assaulting defended points, soaking up damage, killing bases, killing units, etc.

There's tons of stuff Scathis and I talk about with regard to balance, and these are just some of things, but it's 9:30pm and I am an early to bed/early rise kind of guy. I leave it as an exercise for the community to figure out why we might think the principles above might start to get addressed by the experiments we've been running.

Thanks for explaining this to the folks who didn't really understand where you all were going with this.

A few questions I have -

Regarding 1-3:
You mentioned oversteering the balance - when is the correction going to hit? Are we going to try a middle ground? Like, say, 150 for x1 laser, 450 for x2, and 1200 for x3?

All the orbital Issues in general:

Why not add an AoE satellite, like a Microwave death ray thingummy? I mentioned it in this thread.
It would certainly add some interesting diversity to the orbital-> ground game. Of course, you could add in a *frigate* type unit that can both bombard the ground from orbit and combat avengers.

Teleporters are hard to put down because people don't attack soon enough. If you know your enemy is on another planet, you need to send a fabber there ASAP. Forget consolidating. It should be your number one priority, unless you want to nuke them quickly. Otherwise, it becomes nuke/orbital spam.

As for 9:

...you just used the name Unit Cannon to refer to the RoflTank? I don't know how I feel about this.....

And they feel similar because they are direct upgrades. Mind, Levelers aren't great against Dox and other swarm units. But if you put 5 levellers against 10 ants, the levelers win every time.
Slammers ARE a direct copy with better stats. No doubt about that

Also, I'd like to point out that the addition of units that can kill anything in seconds, no matter their range, is really killing the attrition game. Why go for map control when you can t2 air transport harass/snipe with a vanguard at 10 minutes?

I remain skeptical about the t2 eco changes - but I really want to try it out first before delivering a verdict.

'night Meta

As soon as you get T2 eco up, you can afford a massively huge number of T1 factories. If the amount of eco produced by T2 is 10 times that of T1, then you'll be able to afford 10 times as many T1 factories as before. How does this not lead to snowballing?

I frankly don't think that this is a bad thing. Why should T2 be somehting you can spam everywhere? It was supposed to be a supplementary force for your T1, never your primary force if you intended to win.

That's how it is right now, and that's fine, although slammers are too much like doxes and levelers too much like pounders, but that's a seperate balance issue. Doxes are better at kiting and killing units, pounders are better at soaking up damage, assaulting defended points, and killing bases. That's great and is not an issue. We don't want that to change.


These balance efforts are simply going in the wrong direction and are trying to adress nonexistant problems without fixing the actual ones.

I don't disagree this goes towards fixing that particular problem. But from what I'm reading, people are concerned about a different issue - that of what happens to the first person who gets to T2, and how much of an advantage they get in a very short time after that. Having a very fast 4x boost to economy means lagging behind your opponent in teching by even a short time creates a potentially massive difference in your economies. I've spoken before about "fragility" in games, and this is a similar concept to @brianpurkiss' recent thread about players being able to recover - big jumps in economies make this very difficult if not impossible, and therefore makes the game "fragile" around these events and who gets to them first.

I'm not an active player, so this is mostly untested theorycrafting, but has the following been considered:

• For the moment, ignore any T2 units that are direct upgrades of T1 units.
• Make T2 metal extractors only produce as much as T1, or maybe only slightly less, but cost a more than T1 (to limit speed of upgrading all existing mex's). This still results in a doubling of your metal economy.
• Reduce all remaining T2 mobile units to roughly equivalent cost of T1
• Reduce cost of T2 factories to that of T1.

That makes the jump to T2 less jarring on your economy (no immediate massive jumps to metal income), and on the unit side, it gives you access to units, at roughly equivalent cost, that are specialised. T2 no longer becomes rare, but instead becomes situational.

The reason I'm suggesting this is because much of this is possibly caused by the current presence of T2 units that are direct upgrades of T1. Taking them out allows you much more options for what to do with T2.

As I said, I'm not an active player, but this is my 2c

Oh, and increase unit HP

Wholeheartedly agree with this post.

[Question to anyone] Can anyone actually give an example of one of these "dangerous T1 armies" that can make a dent?

Also I find it interesting, but somewhat discouraging, that T1 is actually balanced towards merely a "rush" role and everything else in the game is balanced to mauling T1 like it's nothing more then an inconvenience; including the commander which is just more or less a walking almost free T1 defensive nuke.

It's quadrupled now?
Wow, I still don't know about that but... I guess we'll only truly get it once we play it.
Who knows? Maybe Uber really is onto something here, I hope they are, but we'll see.

You go T2 eco first. T2 eco is the primary reason for going T2.
Imagine if you couldn't make any combat units from T2; if all it did was give you T2 eco. Now your income is several times higher than another player if they wait a couple minutes longer to build T2 eco. You can now spam T1 far more than they can, and chances are they won't be able to catch up, ever.
Even if T1 is super viable, if rushing T2 becomes ineffective, this kind of economy balance will still throw the match greatly in favor of whoever gets T2 eco first, even if they didn't rush (heck, even if the other player rushed and was thwarted). Better T1 units solves the rush, but it doesn't solve the bigger problem.

It is in the current build. Gunships have a very hard time hitting bots (they seem to shoot where they are, not where they will be), which is why stingers have become a hard counter to them (along with a lack of target priority and it being difficult to attack specific units in a blob manually).

You are absolutely right here, of course. That change would have the same effect. It would still give the first player to get T2 economy with 24 metal/s extractors a crazy advantage over any player still running off 2.3 metal/s extractors, even if they only get there a minute or two earlier.

I agree with most of your principles here, but there's more to the issue than just those things. I'm not sure if you agree that T2 economy throws the match one way too abruptly in the live build. That's something I think is a big problem in the current build, and it looks like it remains so in the experimental.

He wasn't suggesting they are currently happening. He was suggesting that putting too much effort into rushing to T2 (under the new balance) could get you overwhelmed by a T1 army. This is actually a good thing - rushing should be a risk, and defence vs expansion vs teching vs attacking should be a strategic choice. It's just I think that solving it in that way opens up other issues.

@aevs - if you are under pressure from a rush, then maybe building t2 "anti-swarm" units would be a good idea.

It makes me very happy to see that from the Devs As well as the comments about the ubercannon.

I think that is necessary for Doxen and Unit Cannons. It's nice that they perform similar types of role, because they are available from two different factories. As it is, with a bit of a tweak to further cement this, Doxen are better at atacking undefended areas, and UC's are better at defending areas. There was a reason (and it persisted for quite a while) why people preferred to build vehicles over bots - because while vehicles moved slower, the engagement favoured vehicles over Doxen.

I wasn't suggesting lowering the income from the t1 metal extractor, just the drain on the economy that t1 factories provide.

I still maintain that deathballs of Dox do badly against lines of tanks, because of the range advantage. It's something I've experienced quite often. It's something I experienced in my human game of PA - I was clearly ahead of my human opponent, yet my balls of Dox were simply dying on his line of turrets and line of tanks.

I personally think that a principle that vehicles are very good defensively, and offensively over open terrain; while bots are very good offensively, and defensively in urban contact, would give the two land factories greater definition. It is quite likely that this build would turn into racing the t2 factory vs. rushing your opponent with t2 units. And I think that having a factory that builds more defensively orientated units, by default, would be good.

That is my objection to the Doxen Grenadier. I don't feel that the cost-benefit analysis weighs up, given that we are losing reliable straight line damage.

I actually have no problem with the Levellers/Slammers are direct upgrades to Tanks/Dox. I have an issue with the range upgrade, because that is the differentiating factor.

To clarify, let's compare 58772 values. Levellers did 500dps, had 625 health, and cost 7 times as much as Ants. Therefore 1 Levellers worth of Ants did 294 dps with 875 health.

It takes the Leveller 7 seconds to kill all 7 Ants, and it takes the Ants 3 seconds to kill 1 Leveller, assuming they both have the same range. In reality, Levellers and Ants have different ranges, so Levellers win, even assuming that the Leveller doesn't kite.

However, it takes the Leveller 10 seconds to kill a 5000 health building, and it takes the Ants 17 seconds.

I think that is enough of a differentiating factor.

I think it is fine to define the roles as basic unit is better at focus firing, basic unit is better at killing many low health units, swarms of basic unit performs better against high-damage single target units

Advanced Unit is better at operating alone, Advanced unit is better at attacking high health units, Advanced health is better at "tanking" low damage attacks, and advanced unit is better at soaking up splash damage.

I.e. The Leveller would be better at killing a Slammer, while the Ants would be better at killing multiple Doxen.
Ants would be better at attacking many single laser turrets, while Levellers would be better at attacking a single pelter.

Now place Shellers in the mix - Levellers are good at killing Shellers, which are good at killing Ants, which are good at killing Levellers.Assuming Levellers and Ants have the same range.

If Levellers and Ants do not have the same range (Same deal for Slammers and Doxen), I think the advanced unit should be slower than the basic one.


Err guys... has anyone seen @ledarsi around the forums lately?

P.S, The reason I was saying 58772 was because I was under the mistaken belief that the Leveller did 125 damage per shot, not 500, as is the case in current build..

However, as the ratio of Levellers to Alleged Tanks in current build is 1:3, that's neither here nor there, because the ATs do not have enough damage to kill the Leveller.

Metabolical, I will try to explain why we believe there may be a snowballing effect.

We can all agree that the more economy you have, the faster you can build. This is kinda' obvious. Furthermore, the more terrain you 'control,' the more mass points you (should) have access to. This results in positive feedback- the more you have, the more you get.

Some amount of this is unavoidable, and indeed, is probably good. However, it /appears/ as if it may be heightened too much in this patch, and that may even be the case even if T2 mass-extractors take longer to 'pay' for themselves.

Why?

Because now, you're able to have a /much/ larger economy than your opponent without actually expanding.

Let's pretend that in the 'old' patch, T2 was twice as 'good' as T1. If I start upgrading my metal extractors, I'll eventually get twice as much metal income as my opponent. Obviously the amount of metal my opponent gets will eventually be upgraded as well, so it'll probably never be fully /twice/ as much.

Let's then take this new data. If I get to T2, the maximum is now four times as much, which is significantly larger. Furthermore, it requires a much greater investment. That first extractor I get up is much, much more important than the first T2 extractor I got up under the old system. Furthermore, if I manage to get down a few of my opponent's T2 extractors, it becomes /much/ more difficult for him to rebuild than it was under the old patch. Why? Because now jump-starting a T2 econ is more arduous than it was previously, /and/ your opponent has a greater advantage than he would under the previous patch. This leads to a 'snowballing' effect, where the player who gets a T2 econ first, and maintains it, has a much greater advantage in this patch than he would in the previous.

Obviously, there are so many variables that it's very difficult to say if this will be how things play out. However, I'll be surprised if this is not the case.

That is literally what I said wasn't my point.
Please read my post again:

I'm saying that the balance of the T2 rush dynamic won't solve the issue of T1->T2 eco. Pretty much in line with what raevn posted.

T2 eco buildings should simply have built in defenses and marginally more eco generation.

Did not read anything about turrets - but with a 4*T2 economy you can simply spam walls of turrets/pelter.
So turrets may needs the 4* price also ... but how to afford them with T1 economy. In the end this may all work somehow but only balance everything with mass sounds like ...


"if all you have is a hammer, everything looks like a nail"

(maybe introduce some "special case" economy which works like the "special case" armor system to factor out the corner cases instead of going nuts with the mass)

This is why I have firmly been on the side of not streaming or sharing these experimental balance changes.
You guys are getting all bent out of shape based on balance changes that are not final and no one on this board has played. Even talk from some people who don't even currently play!

There are a lot of subtle changes that have gone along with this. 'Snowballing' is not an effect of these balance changes. If you play a game and feel like your opponent won by 'snowballing', you need to consider different strategies. Maybe you left them alone too long? Maybe you didn't scout their base and see what they are up to?

What the 4x change does is it increases the time it takes to get to T2 and allows a richer T1 interaction.
There are now, finally, an interplay of strategies of when to upgrade to T2, when to hit your opponent in order to slow them down, how to play at T1 and possibly win at T1.

Is there substantially more economy at T2? Yes, absolutely, and that is a good thing. It's good that the game gets bigger, numbers get larger, power and strategies you bring to bear are more vast and powerful. Yes, you can build bigger T1 armies, but you are limited by the speed at which they can roll off the factories. The changes we've made to the defenses make the later game defenses much better against the T1 units but not as good as the T2 units. It takes a lot of time to upgrade your economy to T2. If you do it too early, you hurt yourself. You sacrifice time to attack with T1 units and risk being attacked by trying to rush to T2, opening yourself up to losing early.

When/if we push these changes it will be another 'Check your premises' time. Lots have changed. We've let the current balance settle and now we're fixing the issues we've seen come out of it.
Keep in mind: This forum is but one of many sources of feedback we have. What's 'known' here is not on other feedback channels and vice versa.

The main things we are improving, to reiterate what Metabolical said. are:

• Make T1 play relevant and more interesting.
• Make orbital easier to use from a UX standpoint.
• Increase the diversity of the units.
• Make mixed groups of units even more important than they are currently.
• Make upgrading to T2 a riskier but more rewarding move.

These changes are going in that direction. It is a good direction that will result in a better balanced game for everyone.

A couple of things you guys might not have noticed.
NONE OF THIS IS FINAL:

• Single barrel turrets now have the range of the Unit Cannon. Doxes have slightly shorter range but can shoot over walls, making them effective at taking out turrets if you put Infernos in front of them. Without that, Unit Cannons can assault the wall in front of the turret without taking turret fire. It's slow, but safe.
• Double barrel turrets can now shoot Unit Cannons if a wall is in front of them but are out ranged by Levelers.
• Triple Barrel Turrets vs Levelers work like singles and UC's, range-wise. (So at T2, the triple barrel turrets can keep a LOT of T1 units at bay, but have range issues with T2 units)
• Orbital Landers are cheaper and faster. Built by an orbital factory means you can put out a lot of them, allowing planetary invasions using these as transports. Imagine a Vanguard drop from another planet.
• Having Anchors as geo-sync satellites enables you to hold an area of ground under it, creating a beach head and allowing unit invasions. They are now the orbital turret they were intended to be.
• Bots have not sped up, but tanks have.
• Mines build really fast now.
• Now that there's more space, metal-wise, between T1 and T2, combat fabbers occupy a middle ground allowing them to repair units even faster.
• T2 Metal Extractors don't bring in 4x the metal from the current live-balance. They bring in approximately 3.5x, but all building costs have been 4x'd.
• Anti-nuke silos now use much less metal and energy to process, but keeping the same build time. This allows you to let them go on their own a bit more and still feel relatively safe.
• Every orbital unit metal cost, health, damage done has been adjusted. Orbital is where the biggest change lies.
• All artillery-style damage has changed it's damage model. They do a lot of damage at the impact site then fall off from there.
• Air has slowed down overall.

If/when we push out these changes, I encourage you to try the changes without so much theory crafting. Theory crafting leads to short sighted decisions. There's a ton of nuance and subtlety that you will miss and then you'll lose games.
And when you lose games you aren't awesome.
And I want everyone to be awesome.
Because we're shooting for awesome.

I'm with you @frobb that the whole idea of number fidling a nonexistent concept into a "game" is kind of iffy. To make an analogy it feels the same as making a game with just a tank factory and a tank and somehow filding with to get "depth" out of that equivalent to a game that actually has more variety than that. With things like "looking for min and max extremes" being particularly worrying since while I believe in evolutionary development this is more in lines to just having the game be based on some procedural generated random balance: throw in units, then generate numbers until it meets some criteria

That being said I'm willing to sit on it for a while and see what happens; though can't say I feel too strongly about T1 even with their recent changes. What will most likely happen is we'll see turrets with walls and combat engies behind them for instant repair.

I'm with scathis. If t2 is so expensive, and you must build t2 using t1 before you just have t2, then it will take 4x t1 metal to do it and 4x as long. A good player could try to react to t2 upgrading enemy before. Now you can react to it 4x more slowly and still catch it.

it does pace, force longer t1 usage, make t1 cost effective, makes first t2 a longer to build project, and increases risk in teching. Worse case scenario, people attempt t2 rush in new patch, just to amass t1 units perhaps besides kestrels or gil-e or seldom vanguard for drop, which is still better than current balance? Best case scenario is t1 is safest way to combat through most the game unless another planet is captured to use to collect metal points.

my 2 cents, is some units needing balance weren't touched hope they get one though, and my honest opinion is t2 income coulda been little under 3x and price 4x but that's presumptuous when scathis will use the scientific method and test independent variables one at a time.

actually, every book gives you better detail when read twice, reread scathis post, he actually mentions t2 price rose 4x and income 3.5x, air slowed, vehicle sped, turrets lost just enough range to have key strategic planning for each one but all overall nerfed.

There's no need for you to get bent out of shape about people speculating, it's all there is to do. If you want proper feedback, release the changes (note this is not a request for you to release it, just a statement of the obvious). Some opinions are vitriolic but I don't see the problem.