Letter to Zock, Balance Team and the Forum in general

[JoonasTo]

First of all, Zock, I thank you, and your team, for what you have done. I've honestly enjoyed the properly done balance changes a lot. Not just the recent ones, but things that happened before too. SCUs, Aeon frigates, etc. From the recent changes the T2 suit nerf in combination with the low level survivability upgrades that are in beta feel the best thing I've seen in FAF balance since the resto nerf.

I'd also like to thank you for finally realizing the mistake in hover balance and raising the speed back. Even if it took a while, it happened.

Now I'd like to discuss which lead to this mistake in the first place and what I have seen happen on other occasions too.
The balance approach of FAF to singular, highly effective tactics and strategies has often resembled a knee-jerk reaction. That is, it has been radical and over-the-top(1,3 speed reduction for hover out of 4,3 33%, this was too much) to make sure that the tactic, or strategy, is "fixed". At worst, it has been cumulative with other, simultaneous bug fixes, which have been nerfs(hover hitbox bug.)

This is a great example because it has even more characteristics of a bad change. The speedy wagner, that was already capable of easily dealing with t1 PD, unlike the blaze, was left untouched. It indirectly buffed the unit, which was already very strong, as a raiding unit on maps where this was applicable. The seraphim already had the ever usefull zhtuuey, which was also left untouched. The former made cybran(and seraphim) stronger, in comparison to UEF and Aeon in island raiding. At the same time it further widened the gap from Seraphim to UEF and Aeon in naval lock situations while bridging the gap from cybran to UEF and Aeon, again, strenghtening Cybran.
The end result, aside from the obvious result of weakening hover to not make it viable alternative to spam only hover instead naval, was quite a lot different.

1. It made the t2 hover almost useless vs navy because the change was too large and it had the cumulative target bone nerf applied at the same time.
2. It strenghtened cybran, and to smaller extent, also seraphim, in relation to UEF and Aeon.
3. As a result of the above two, it weakened ecoing and defending, less so for Seraphim but especially so for Aeon and UEF.

The first shows an overreaction to the problem at hand, to fix it quickly(tm), instead of actually balancing it to viable alternatives with more care.

The second shows short-sightenedness concerning the ripple-effect in a bigger picture.

The third was a favourable result but the magnitude of, especially, the first and the second outweighted it.

I hope we can all agree on this much.



I also hope this can shed light on why I disagree with some past changes so adamantly. For example, the strat bomber energy cost increase.

100k to 144k energy(NEARLY SAME AS RAS!), or 44% increase was, and is, NOT an acceptable solution to the problem of rushing strat bombers on t2 power income. We don't have strat rushes on t2 power anymore, AT ALL.This is not the point of balancing and improving the game, I'm sure we can all agree that the point is to have more meaningful choices.

The problem here was that the strat rush, while still having decently high risk, had game-breaking reward. The whole enemy team's powergrid was wiped and it was over. Just from one player risking their game for it this was obviously not balanced. But like with the hover, this one was an over-reaction as well. It led to the strategy completely disappearing from the game. And that was not it's only consequence. Once again there was a ripple effect to something else.

I'd like to talk about the airlock situation now. You manage to rush t3 air before your opponent and gather an asf force large enough to wipe out any opposing inties, and the few asf your enemy possibly has out. You manage to camp his air factory and force him to build sams. Now you wish to take advantage of your well-executed, high-risk, strategy, that you either sacrificed a teammate or your eco for.

What do you gain? You gain safety to your remaining teammates from air snipes and practically unlimited scouting, barring any aeon cruisers or sera sams. What else? You wish to build strats of course. Now you go and queue up your two strats for harassing. This means sacrificing even more of your eco AND risking your air control due to allowing your enemy to catch up. In the old 100k strat balance, this meant 4 asf(roll-off time reduces the 5,0 to 4) and 3 t2 mex. But with your early advantage in ASF and the damage you expect the strats to inflict this is OK.

Now with the 144k strat balance the exchange for 2 strats is 7 asf(7,2 minus roll-off time) and the same 3 t2 mex. Now early game airlock is usually 5-10 asf. It is obvious that the enemy having 7 allows him to contest your air-superiority(especially in his territory with radar!) and definitely allows him to shoot down your strats with relative ease. Suddenly the option of going strats for harassing isn't at all attractive. Unlimited scouting is very nice but teammates can be secured from airsnipes with inties, and especially swifties, as well. Sacrificing eco and going for an airlock is also less attractive as a result, especially as aeon.

So the end result was:
1. Eliminating a strategy completely, simplifying the game, reducing depth
2. Reducing the possible reward for an advanced, already risky strategy, making it less favourable, again, reducing meaningful choices and depth.
3. As a result of the two above, ecoing up and defending was once again made stronger.

This time the first was caused by the overreaction to the problem(44% increase) and the changes made to strat bomber flight dynamics. Inties being able to kill something so ridiculously expensive(FULL ECO of T3 Pgen and two T2 pgens for 42!! seconds) just made the risk-reward ratio skewed.

The second was also caused directly by the cost increase, but in a different situation than intended. Again, the bigger picture.

This time also the third result was unwanted, as it lead to a more static game, instead of a dynamic one.

There's also a fourth, more hidden result.
4. It buffed all land experimentals, and to lesser extent, navy

This is due to the increasing the cost of countering them with strats, the only flying normal(not experimental) unit that can deliver some form of DPS while the experimental is being guarded by masses of T2 MAA or some amount of T3 MAA.

The jury is still out on whether this is a beneficial result to the state of the game.

I hope this can convince the balance team to look more into rationalizing this change, like they did with the hover one.


I have observed a positive lessening trend in over-reactions over time and especially in the recent balance patches. For this I am grateful and hope it stays that way.

There are also different kinds of issues which I will lighten my views on in the next post however. Not strictly balance, but game design also.

[Zock]

Thats kinda what the beta phase of the patch is for, it would be great to have this kind of posts before the patches go live, and where its must easier to adjust things.

In any case I don't think its possible to always make perfect changes. The game is extremely complex, and can have many unforeseen sideeffects for any change, especially since you have to consider 1v1, teamgames, different skill levels, all the different maps and much more..

Balance is a process, and having to readjust and finetune, or even revert some changes is not the exception but something rather normal. This doesn't mean it shouldn't be tried to be avoided as much as possible, but what you call "overreaction" is to me just a side-effect of working with imperfect information. In the best case, all of this finetuning would happen during the beta period, and the changes can be made "perfect" before being released. But in a normal case, having some changes being to small, too big, or being bad in other ways, can't be avoided.

If there isn't enough feedback or testing in the beta phase of the patch, the chance increases that some bad/too raw changes go through, no matter how much we try to avoid it, or how much we test ourselfs. It's not possible to test for all possibilities with the few people of the balance team. In this case, if i remember correctly, there wasn't any negative feedback on either hover, or strat change during the beta. There was negative feedback on the hover change later, which lead to re-investigating it, seeing that the critic is reasonable, and now, correcting it.

But i do think personally, if this one or two changes are the only things that need to be adjusted in hindsight, it is a really good margin. Many games with professional balance teams, and a much bigger amount of test games, have to do far more and bigger corrections on their first changes.


Now about the actual points..

I'm not gonna talk too much about the hover, since we are already changing it, even though i think there is much more playing into this change than you're stating, and i wouldn't 100% agree to the argumentation (Sera gets better than aeon because of zhueeys, what about auroras?).

With the bomber i agree that removing the option for a risky strat rush is not a nice thing, but i am not sure if there is a way to keep it without bringing back it's devastating effects on teamgames where a couple of seconds in airtiming could decide a whole game. And i don't think the assumption that strats are useless now is entirely correct, i've seen them used with, and without success plenty of times, they are very risky now, but can still be very rewarding. A single strat can kill way more than 3 t2 mex, and killing a couple of t2 PGs can be devastating too..This doesn't mean the situation is perfect now, or that the energy cost is not too high, but the situation is much less obvious than with the hover speed in my opinion.

By the way, i have no idea what "flight dynamics" you're talking about, am i missing something?

In any case, the last word on strats and t3 air is not spoken. As written at the time in the patchnotes at the time, this was never intended to solve the problem completely, but to "hotfix" it, until a real solution has some space in a patch. When that is happening, the cost will probably be reverted. Having strats as such a risky unit (in the early lategame game at least), but, if successful, just as powerful as before isn't a great place for such an essential unit. For now it is better than having a lot of teamgames decided by stratrushes though.

I'm open to reduce the energy cost a bit, but i'd like to hear more opinions.

[Morax]

[JoonasTo]

Yeah, I am going to get into the current patch eventually. Before that I want to lay the groundwork for where I'm coming from and how I see things. Of course no patch can be perfect out of the gate, it is finally looking like you guys are having the time and will to adjust things so it gives me more motivation to post these.

As to what comes to making it as balanced for as many possibilities as you can, there's a very simple solution. You don't. If you look at any RTS out there, none of them are perfectly balanced between different maps, or skill levels. More often maps are balanced to the game, than the other way around. There's really no reason why this shouldn't be the case for FAF. Sure you want general balance on navy, land and air maps but not balance on Seraphim Glaciers, Setons and Bermuda Locket. There will always be differences between factions on the maps, and this will lead to different playstyles against different factions. This is the faction diversity that you long for. It also creates a lot of different strategies and tactics that introduce depth into the game. Rarely, if ever, is there a single perfect solution against every playstyle and every faction, and in the cases that there is one, it is more often map imbalance, than game imbalance and you should treat it as such. This is why vetoable maps in ladder would also be nice as it is very hard to predict such things in advance.

And you should not pay too much attention to balancing for the low or even average rated players, like me. As long as there isn't anything game breaking, the overall balance will always be better served by balancing for the best. They are the ones who will figure out the counters to the units the fastest and as long as information flows freely down the ratings, the problems will solve themselves. FAF still has some work to do in this regard but it's getting there and if the player base starts growing, it will naturally create more information flow. EVERYTHING looks overpowered to a player who doesn't know how to do it, or counter it. Their perception of the situation is very, very flawed and the best way to fix it is not balancing the game to them, it is giving them the information they need to solve the situation themselves.

As to what comes to the quickfix method of the strat among other things, yes I know it was the case, what I am saying is I hope the lessons have been learned and there will be no more of these.This is what I hoped to get across with the first post. Not so much the actual values(after all, I'm only an average player, not to mention who hasn't even tried to win seriously in a long time) but more on the way the balancing is done and identifying the good and bad habits I've observed.

I am planning on making a part two but that will probably be someday during next week. As mentioned, that will touch more on the game design choices FAF has made. You will probably find that my view there isn't exactly as rosy as this.
If you want something in the mean time, check the T3 factory buildtime increase and ask yourself why wasn't T3 acu buldtime increased to match this. There's a clear balance point to be made there between Ravager creep vs T3 mobile arty counter. Or EXP rush route. You know the deal by now.

Oh yeah, with flight dynamics I meant the physics of the strats. When they were slowed down a bit to allow for easier kills with inties.

PS. Aurora's can't kill PDs, Zhtuuey's and Wagners can. Aurora is also relatively ineffective vs navy compared to the zhtuuey in a naval lock situation.

[Zock]

Hi,

First to say, i completly agree to most of what you wrote! I'm curious to what you next points will be, despite, or especially, when they will be more critic.

But a few things:

And you should not pay too much attention to balancing for the low or even average rated players, like me. As long as there isn't anything game breaking, the overall balance will always be better served by balancing for the best.

Yes and no. While i think classic "faction balance" should be done mostly for high rated players, lower rated players and games need to be taken into account too. High rated players are 0.1% of the players base after all, usually whats good for them is also good for others, but not always. E.g. if there is a unit, that is easy to use, and extremely strong, and where countering it requires a lot of skill, good players won't have issues with it, but it can have a huge negative impact on the game for worse players.

The best example are the recent bomber changes. First bomber was not too strong. Good players didn't have problems to counter it, even though opinions if its a nice part of the game or not were split. For them, there wasn't much need to change it. But since you needed to know/do more things to counter it than to use it (and someone who defends against it needs to learn a million of things, while someone who just bomber cheeses needs to learn just that), it was so frustrating to many players that they just stopped playing. The idea of losing to a single unit in the first minutes, even if they are trying to counter it, wasn't really what they expected from a strategy game.

Without new players, FAF will eventually die. Considering them when doing changes, and looking for ways to improve their quality of (game-)life is something i find very important for the future of FAF. This doesn't mean we have to make the game more simple or easy. Also new players start playing this game for its complexity and depth after all.

This isn't so much disagreeing with you, but just giving me an opportunity to share some of my opinions about balance, since some people wanted to know more recently.

As to what comes to the quickfix method of the strat among other things, yes I know it was the case, what I am saying is I hope the lessons have been learned and there will be no more of these.This is what I hoped to get across with the first post. Not so much the actual values(after all, I'm only an average player, not to mention who hasn't even tried to win seriously in a long time) but more on the way the balancing is done and identifying the good and bad habits I've observed.

I have to say, right now, i'd probably to the exact same change again. It solved the problem, it had some negative side effects too, but i still don't see any better alternative. I'd hope for more testing and feedback on it, so the actual numbers can be finetuned more (and they still can), but i have to admit that i don't see anything wrong in the approach. Doing a more fundamental solution than a hotfix wasn't really an option, as much as i'd have liked it, as it would have blown the seize of the patch out of proportion, and the first patch was trying to avoid too controversial changes (that all changes that go beyond small fixes are inevitable are).

At the time of the change, it was impossible to know if the cost increase is good, too much, or not enough. The taken number looked to have the best chance to be "just right", so we took it. Now, we can see that it was enough, because it solved the problem. But i wouldn't be able to say if less of a cost increase would've also solved the problem, but with reducing the negative sideeffects (if its even possible to have strat rushes being more viable than now without becoming OP again). From what i've seen, it looks rather good now though. If there are many people with the impression that strats are too weak now, we can try to reduce it a bit and hopefully test it in the beta to see if that will bring back the problem or not, before we'd implement it.

So i'm not sure what should be the "lesson learned" here exactly.

Oh yeah, with flight dynamics I meant the physics of the strats. When they were slowed down a bit to allow for easier kills with inties.

Care to point out when and where this did happen?

If you want something in the mean time, check the T3 factory buildtime increase and ask yourself why wasn't T3 acu buldtime increased to match this. There's a clear balance point to be made there between Ravager creep vs T3 mobile arty counter. Or EXP rush route. You know the deal by now.

Its a good point to take into account, but not a good point to warrant a change imo. Ravager creep is still counterable just fine, exp rush is generally mass and not BT limited, because they have a ridiculous BT anyway (which will be changed in the t3 patch), and generally the BT increase for the HQ has very small effects, since its so easy to even it out with more engies. And generally, change should be kept as low as possible and only done if there is a good reason for it (e.g. a problem, or a clear improvement), not because there is some interaction with another change. Sometimes this is necessary, but if you start doing it before making sure that it is, you'd also have to take into account that changing t3 BT will effect comdrops, and ACU tml rushes and countless other situations, and then make changes there too to keep "everything in the same relation to each other". Needless to say, that's impossible.

In short: T3 acu BT wasn't changed, because there isn't any problem with it, and changing it doesn't seem to make the game better in other ways either.

Regardless, it is important to bring this points up and consider them, because there could be problems with not changing it together. We try to do this as much as possible, but there are always special situations that you can miss. Thats why feedback and testing is so important. Especially feedback that points out and informs of such possible problems.

[JoonasTo]

As I said this is going to touch more on the game design and how certain changes affect that. To do that we first need to establish some facts about Supreme Commander and FAF game design.

First, Supreme Commander was not made to be a competitive RTS and it shows. Multiplayer focused? Partly but the campaign was clearly the selling point of the game for most of the audience. This reflects badly on the game the higher you climb the skill ladder. The basic multiplayer works very well, 50 apm players with good connections won't even notice anything before getting into late-game naval. Pick someone from the 150 apm range and they can immediately tell that there's such a thing as being too fast for the game.

What's worse is that the command lag in the game isn't consistent. Get someone with a bit worse connection into the game and what used to be 500ms lag can easily be double that. This way one player's issues affect everyone's performance. This is due to the peer-to-peer nature of handling the network connectivity. Were it a centralized solution we would have that single player lagging himself but the rest would get their commands through with the same speed as always(and FAF wouldn't exist, yes we know.)

Then there's the game's pathfinding. What can I say here? It's worthless. Send a single unit on a move order far enough and it will not find a straight path even if there's nothing in between to block it. For some reason nobody thought of adding triangulation to the code(or maybe there's just a simplified chart across large distances for performance' sake.)

Not to mention unit responsiveness. Give a hover unit a move order backwards and it will first hover forward before actually going back. Or give Tempest a move order and after spamming the move command for four seconds it might actually start moving and not forget it instantly.

These are not the only ones but I'm sure the point is made by now. The game simply wasn't designed for high level play. There's so many issues with it that knowing and being good at preventing issues like these is almost as important as knowing how to actually play the game in getting better at it. The more you play the more prominent this becomes because you need to spend proportionally more time and attention fixing things that shouldn't be an issue in the first place. This leads to less and less enjoyment as you get better at the game(and I admit, is partially why I don't tryhard in this game.)


The next things we need to establish are the skill floor and ceiling of Supreme Commander. In theory, the ceiling is unattainably high. There's hundreds of units that benefit from individual micro plus countering your enemy's unit mix on the fly, adjusting your builds and eco to the map and the reclaim resulting(or, just as importantly, not resulting) from battles is more than any human can ever achieve. Add to this the multiple snipe opportunities appearing every game and it creates a lovely environment filled with variables to keep you constantly challenged. To quote TLO(not direct, check FAF launch video for that): It doesn't matter how much apm you have, there's always more to do in SupCom.

In practice this rarely occurs because of the problems with the game design mentioned earlier. You simply cannot do too much at a same time and the longer the game goes on the worse this becomes as the situation naturally evolves to become more complicated and the engine starts to lag more. This leads to a surprisingly low practical mechanical skill ceiling. I've had to wait seconds for my commands to go through.

Thankfully, there's more than just mechanical skill to most games and Supreme Commander is no exception. There are other things to do than army(or engineer) micro, like the aforementioned adjustments to eco, build or unit mix. The skill ceiling in these is not quite as high as it could have theoretically been in mechanical skill but it is still very high due to the sheer amount of different units in the game and those differences actually matter(you can beetle snipe with a single UEF T2 transport but not with a Cybran one, now you know.)

Lucky for us that doesn't mean that the skill floor would also be high. There is no need to know that you can catch a Striker with a Thaam to play the game. You do need to know basic eco management, true, but that is expected from an RTS anyway so the new player coming into the game isn't too burdened by it(let's face it, if you're looking for SupCom you probably know what an RTS is already.)

The mechanical skill floor, likewise, is also very, very low. You click on an engineer to build a factory. You click on the factory to build a tank. You click on the repeat button. You set a rally point in enemy base. You have just given all the commands necessary for your continued attack into the enemy and your engineer probably still hasn't finished the factory. Supreme Commander is the mechanically easiest RTS I've ever played. And this is a GOOD thing. It is very easy to pick up a friend and have him play it with me(or come back after a long period of inactivity.)


Third necessary base to establish is that we have more resources than just mass and energy in multiplayer. I'm sure Zock doesn't need a reminder but someone else reading might. There's time(build-, travel-, roll-off-, etc.), intel and build-power in game. Outside of the game the resources dependent on the player are apm, attention, knowledge, adaptability, stress(or rather resistance of it) and time.

Also one very important meta resource that often gets overlooked is opportunity cost. This is why you rarely saw T1MAA before. Even if it was cheap as grass with roll-off-time included its opportunity cost was simply too high. As players get higher in skill they also learn how to manage all of these more hidden resources too.

If you want to use the first bomber(pre-nerf) as an example we can continue with that. I agree that it wasn't too strong. In fact, I'd argue that unless the map is specifically set up for it, it's a very weak opening in 1v1. Let's look at different levels of the tactic:

You rush a T1 PGen to be able to pay for the bomber, then you build an air factory and a bomber. Depending on the faction(airfactory engineer roll-off time varies) you are now horribly behind in having your first engineer out and your eco is spent. You sent the bomber to attack enemy expanding engineers followed by power and expect to get results that put you ahead of your enemy despite of all this. Your enemy sees the bomber with his scout, queues up 3 T1MAA and shoots it down. Maybe it gets one engineer, two if you are lucky. You are very faf behind and lose.

So you learn how to micro(hover) the bomber and factory lock your opponent while killing his adjacent power. You actually manage to kill the power and all expanding engineers before he builds a T1SAA but because of all the apm, time and attention you need to give the bomber you are now even further behind in your eco than in the non-micromanaged version. That is okay though, as your opponent is even further behind.

So in reality the opportunity cost for first bomber in 1v1 is much more than just the later expansion and worse eco from the build, there's a lot of lost efficiency from all the player resources required. OH BUT WAIT, didn't you just micro the bomber the whole time and your opponent didn't micro his engineers? That would imply you have better apm than he does, but of course his apm is down a bit because the situation is more stressful for him. We can still expect an equally skilled player to at least dodge with the engineers long enough for forcing you to abandon either killing them or killing his power. So in reality the risk/reward for the strategy is rather poor on most maps. In combination with the huge opportunity cost it simply isn't worth it against decent players without some kind of map gimmick or mind game.

But if decent player have no problems countering it, why do worse players have a problem with it? Are they actually matched against superior players? In a way, yes. This is what is(was) happening. The player doing the bomber rush had more knowledge of that specific strategy than the player failing to counter it. For some reason there was a block in information-flow down to the worse players in how to counter the first bomber. I'm going to quote Blodir(paraphrasing from last weekend's meet) here:”I don't get why it's an issue, it's so easy to counter, you just build 3 T1MAA or 1 T1SAA.”

There are a couple of reasons why this isn't the view of the average low rated ladder warrior. He might actually not know that he needs 3 T1MAA to kill the bomber. In that case the reason he's stuck there is because he is unwilling to find out how or lacks the ability to realize that after a few lost games. More likely he is just too low level player to grasp everything that amounts into making the first bomber a successful tactic. The facts are the same but his perception is flawed. This is a common problem in balance and there isn't a good solution to this. It doesn't really matter how first bomber is changed, it will still seem imbalanced to the uneducated as long as it's an used tactic, all tactics and strategies with lots of advanced resources will. If you remove tactics/strategies, which include significant cost of advanced resources, you will simplify the game and lower the skill ceiling.


Which leads me to the conclusion of the start of this post. You should always aim for as low skill floor as possible, to incite new players, while keeping the skill ceiling inhumanly high, to keep high level players from reaching it. Skill floor too high will keep new or returning players from ever playing again and too low skill ceiling will make good players stop playing. Nobody wants to play a perfect game, everyone loves to strive for it.

Currently we have multiple issues impacting the execution part of the skill ceiling, pulling it down lower than it really needs to be. As such our first priority should be to never make changes that reduce it further. Or if at all possible, fix our problems that keep it low.

On the contrary, our general skill ceiling is still relatively high because of the vast amount of knowledge one can gather and use from this game. It should come logically that while keeping the skill ceiling high from other areas of the game is always to strive for, there is need to try and raise it here.

Both general and mechanical skill floors are really low at the moment, and again, THIS IS GREAT! But only because we still have a decent player base. The higher skilled the lowest rated players of the community become the higher the effective skill floor for both general and mechanical skill will become. This is natural evolution for an aging game but it is never preferable as it will drive away possible new blood and create a self-sustaining problem. At the moment the best option for this is better education for new players. It's not perfect as it aggravates the problem but it's the best we've got. Even still, there is no need to try and reduce the skill floor as it is already amazingly low.

It is in the light of these simple rules that all changes should be weighted, while keeping in mind the issues we have and not forgetting to also account for the advanced resources.


It is also in this light I am flabbergasted about some of the recent balance changes. But I reckon this is enough of a read for one post so with the basics out of the way I will write more about them in the next part.

[IceDreamer]

There's a lot of people on about lag like that, but I'll just pop in here that through 10 years and 5 individual PCs, I have never seen a click take more than the allotted 500ms to go through. Drop out entirely, yes. Missed packets or whatever, yes. But never slow.

About first bomber being too strong - Too strong or not is really besides the point. The point is that it's not fun to get hit by if you've never seen it before, or even if you have and aren't clued in or fast enough to respond. At the end of the day what counted against hoverbombing was extremely simple: Players were leaving the game community because of it. Call them noobs, scrubs, unadapting, whatever you like. They were here. Now they're not. That's not an acceptable gameplay situation.

[Zock]

Problems about lag seem very exaggerated - it is very rare that someone goes over 500 ping in 1v1, and the only time i've experienced your mentioned problems is when i played setons. Don't think i had ever a unit losing it's orders unless it was stuck. Pathfinding has its bad moments, but most times it actually works decently. It is indeed a part of skill to manage your units in a way that minimize potential path finding issues with them, but it being "almost as important as knowing how to actually play" is far from the truth. It's not great, but far from making people hit the "skill ceiling" or the game being unfit for high level play in any way. No one is even close to any kind of skill ceiling, and at least for me i'd know very well what to do with 150 apm..or more.

You do need to know basic eco management, true, but that is expected from an RTS anyway so the new player coming into the game isn't too burdened by it(let's face it, if you're looking for SupCom you probably know what an RTS is already.)

This comparison is quite dangerous. Managing the economy in SupCom is massively more difficult than in other games, and the main reason why this game is so hard to learn. In other games, building power won't cause you to power stall even harder, just to let you realize you made way too much power afterwards, because you made power until you have enough and then stopped..just to suddenly get all the extra power, that the engies used so far to make the other power. Or that whole flux economy thing, thats amazingly simple in theory, but incredible hard to manage well.

What you say is still true, and learning how to manage your eco properly goes beyond any "skill floor", but it's important to note that this is not like other rts. Eco management in SupCom takes up a much greater part of the game compared to other rts, you can go as far to call it an eco management game rather than a strategy game, and you wouldn't be too wrong.

I'm going to quote Blodir(paraphrasing from last weekend's meet) here:”I don't get why it's an issue, it's so easy to counter, you just build 3 T1MAA or 1 T1SAA.”

If this were completely true, it wouldn't be an issue at all indeed. But what you actually need to do is more like this:

-Build some AA immediately when you see the bomber. Not a problem for blodir, but a huge problem for newer players, who are usually busy with things and don't see the bomber once it comes in vision, and then need to realize what is happening and remember how to react to it, then often need a few seconds to get to their factory and find/remember the AA icon, and by this time the bomber might very well already killed their power..even if they reacted "perfectly" and "just make some AA". This was the most frustrating part of it. People had the feeling they lose the game to bombers, even when they tried their best to counter it, even after learning how to do it.

-Learn to spread your pgs, which isn't too hard, but one more thing to learn regardless. And something completely unintuitive, you are rather inclined to build them together to profit of more adjacency bonus. It makes sense that you have to protect your engies from raiding, and that you need to make AA against bombers, but spreading pgs is not. In an optimal game, all things you have to learn should be as intuitive as possible, even though we're far from that.

-Now the AA isn't actually too helpful without a scout, so gotta make that one too. You probably didn't quit the game yet, this seems fairly reasonable to counter a bomber, even though just AA could really do a bit more..Oh, and put the scout into the correct spot. And better keep moving it to keep vision on the bomber as long as possible.

-Now your base is save, but what about your engies? You probably need some more than 3 AA now, more like..6. (4 would usually do if you do it perfectly, but you don't). Don't forget to keep the scouts close. And you better learn how to dodge with your engies until the AA aarives. If you don't have a scout close you have to dodge before you can see the bomber based on your guess how fast it'll turn. If you do have a scout, based on this triangle and your knowledge how far away from the engy it'll release the bomb, how fast it flies, and how early your engy needs to move, you can dodge. At this point we might ask if this is reasonable to expect from new players to learn, to not lose the game to a single unit. Short answer, and the main reason why we changed the bomber: No. Longer answer: Some people think micro has no place in a strategy game. I couldn't disagree more. Micro is great. But micro should ideally be optional, give small advantages and decide very close situations, but not decide the game by itself. It should be something to increase possibilities for better players and to make the game more interesting. If someone wants to learn how to micro a bomber to get more out of it - cool! If someone has to learn to micro engies to not die to said bomber - not cool. Not at all, not only because new players already have a million things to learn, but also because having a whole game decided by micro and a single unit is not what they expect from an rts, especially not from supreme commander.

-Oh, and you might have to dodge with your AA too, or it can die to the bomber. Hope you have your scout close.


In my lessons, the amount of time i needed to teach people to not die to first bombers compared to the time i needed to teach them how to first bomb reflects that. People had a way harder time countering the first bomber than doing it. And that were mostly 1000+ rated people, some even around 1500, so not even close to new players. If this people already had so much trouble, i don't want to know how much trouble completely new people have (unfortunately i do know, because i gave lessons to a few too and could see it first hand - and this were not stupid people, or such that were too lazy to learn how to counter something).

Something that needs more time or effort to learn how to counter than how to do it is already bad, it's not as obvious, but it's just as bad as something needing more mass to counter than to make it - but if that's even something that can decide a game on its own, its a huge problem that is dangerous to the future of FAF. Without new players, FAF will die not later, but sooner. New players don't have a lobby for their changes here, they won't vote for someone who speaks for their interests and they won't complain if changes or no changes are done for them. Changes like this are often unpopular, because people are used to the issues and found their way around it, so they don't concern them any more. Engymod was a similar change, and just as unpopular in the beginning.

And that is just one of the problems the old (first) bomber had.

Players were leaving the game community because of it.

Here it should be mentioned that this were an unusual large amount of players . There are probably some guys who get killed a few times by labs and leave, instead of learning to protect their engies. There are people who will leave over balance changes. There are people who will leave over no balance changes. People leave the game all the time for a multiple of reasons, that's not warranting a change yet. Only if the number grows, it's an indicator that there might be something wrong.

I agree to pretty much everything else though - especially that better education for new players is badly needed. Could go more indepth about skill ceiling (and how no one is even close to reaching it, making a discussion about lowering it purely theoretical, yet it is important to keep it high - as was also taken into account with the bomber changes, and one of the harder parts to accomplish) but this is long enough too.

[JoonasTo]

I completely disagree about supcom economy, but maybe that's because I've played far too many strategy games before. Not just what you'd call RTS but grand strategies ala Paradox too. I find supcom economy refreshingly simple in comparison, there aren't even any supply lines!

Also, almost all RTS have a flux economy. Or more specifically, supcom has a tick economy too, the ticks are just smaller than in most games.
StarCraft has a tick economy, there is no difference between building more mexes and building more workers for your mineral line. It is the exact same result, you get a flux income from it. The only reason why it feels different is because you do not have infinite queue for your barracks there. You can even spend resources you don't have if you queue up multiple buildings in a row, just like supcom. It also has one additional resource in population that you need to manage, that isn't the case in supcom. Or having to uproot all your mexes and collecting all your engineers to fight the enemy T1 rush. And then there's the Zerg. Having a combined resource of production you can set aside for engineers, mexes or combat units, but not all at the same time can be quite a hassle. You also need to sacrifice one mex and one engineer for every building you make.If you didn't plan those PDs ahead of time, your economy is dead in the water the time the attack is over.
Now economy stall works a bit differently there as your production just won't start without the resources required so it doesn't affect the production speed of a single unit, like in Supcom.

If you want something that has stalling economy just like supcom, you can take a look at Command and Conquer. You try to run too much production at the same time all of your queues will stall, just like supcom. The game probably has the harder economy to understand because it's flux(or really miniscule tick) economy just like Supcom but supcom is nice and constant, C&C isn't. It's a variable flux economy and it's a huge pain to manage for a new player in comparison. In supcom, you can just look at the mass bar and see, Ah, I have +4 mass and +20 energy income and no expenditure, I can just make one factory making tanks on repeat and not stall, ever, how nice. If you look at your one harvester and one refinery in C&C it tells you absolutely nothing. Try making a couple of tanks and you will crash completely midway through. Then someone nice might tell you that having 2 refineries and 5 harvesters allows you to run 2 normal production lines at 100% efficiency, assuming optimal refinery placement and some harvester micro to begin with. That is, before you run out of initial resources on your field and have to rely on the regrowth that will require just one harvester and one refinery and won't even result in full efficiency for that single one.

Then there's stuff like Settlers.

Just a reminder, building more wood or stone production in Settlers WILL stall you even harder.
Building more tiberium refining in C&C WILL stall you even harder.
And yes, even building more workers in StarCraft will stall you even harder, it just won't affect the single unit production speed.

So yes, there's economy in supcom,
and no, it isn't really all that special everyone makes it out to be.


If you want lag in 1v1 just play vs someone from oceania or japan or whatever. It really isn't all that hard to get over the 500ms ping that way. Of course in team games, the chances are a lot larger and lategame there's performance issues with the engine as well. It just compounds.


Pathfinding does work decently if you don't give longer move orders than ½km. Or command more than 5 naval units at a time. Or try to use beetles. Or try to use a tempest. Or try to move through a base that has engineers in it.
The point is, it's easy to play with the bad pathfinding at maybe around 1000 rating level, every game above that, you need to adjust your play to avoid it's issues. And the more you have going on at the map, the higher percentage of your effort goes to avoiding the issues with pathfinding, bad unit behaviour, etc.
Don't get me wrong, I've played games with bad pathfinding, I've played StarCraft:BroodWar and the pathfinding there can be pretty horrible sometimes(dragoon micro deserves a category of its own). But that came out in 1998. Supcom is ten years younger and the pathfinding is worse.


PS. I think you're way overestimating the value of an pre-nerf bomber without micro. The aiming and firecycle on those really was quite bad. I don't particularly have any gripes with the changes per say, you just presented a great example so I continued using it.
And yes, I know BRNK still doesn't know how to deal with bombers.

[JoonasTo]

Automated overcharge is technically a nice feature but it has some problems game-play wise.

If you are just retreating from a bunch of T1 tanks and know you'll survive you can sacrifice some energy and commander HP for extra APM to spend elsewhere, sure you'll overcharge a few single tanks but that's fine if you can micro your economy back home or raid enemy eco on the other side of the map. It's a fair trade to make.

However, WHY does it prioritize higher tech units by default? In the T1 spam phase manual overcharge doesn’t waste energy or damage on single tanks unlike the automated overcharge so there’s a fair trade-off to be had there. When T2 starts mixing into the group this is no longer the case. It always targets the highest value unit you would target anyway(aside from the rare only 100hp left case) and it does it better than a human(see the reason in the chapters below.) Besides of completely countering certain tactics(beetles, or any early T2 unit mixed into T1 spam for that matter) it also removes player involvement. From this point on in the game, it’s a no-brainer to have the automated overcharge turned on constantly. It is just too good attention and apm wise, there’s no reason to pass up on using it.

There's also a feature included with the auto-oc that is actually caused by the game 500ms inbuilt lag making the auto-oc have around more 500 more DPS than manual overcharge, see Heaven's video here. This should be changed. At the very least, the manual overcharge should be as good as the automated one.

You can also fire automated overcharge while moving. Either give this ability to the normal overcharge or disable this from the automated overcharge. There is no reason to have it work this way while the manual doesn’t.

Now automated overcharge actually being less efficient would reward player interaction. So you could just turn on the auto-oc when you know the shields will hold against the ML, no problem. But let's say you are unsure or you know the outcome of the fight will be very close, then you would want to spend all of your attention on surviving the encounter and overcharge manually. We don’t want to change the actual damage value on the overcharge because they use the same reload timer and it would feel very bad disabling automated overcharge just a little bit too late to not get the full damage from it. It would also further complicate the game and mean you would have to learn new value for what units you can kill and what you can’t. Increasing reload time after an overcharge would also be bad because then if you need to use the manual overcharge and are too late in disabling it you are punished for it with a longer reload.

What we should do is change the aim time on automated overcharge so it’s worse than an active human doing it. This way, you don’t get punished for not using it, because the reload time and damage will still be the same and you can even have it turned on and fire it faster manually if needed, at targets of your choosing. Let's say automatic overcharge has two seconds aim time before firing so the manual overcharge would have around 1000 more DPS. 1000 DPS for all of your attention is a pretty costly trade but at times it could change the outcome of a fight and give you a decisive advantage over your enemy. This would be worth it. It would be a meaningful choice to make.