Introducing a new Dev

But this is also a function of the incredible low health air has, making ants able to oneshot them. If air has more hp as has been announced this will change greatly.

Additionally they could change the pitch speed of the gun to be slower while leaving the yaw speed the same, making it harder for them to track flying units without changing their effectiveness against ground units.

Anything except for walls is currently rather low on health...

Well you have to admit that the amount of health walls have makes up for the lack of it on every other unit. On average.

Well i think balancing metal so it actually limits you, and then balancing that on the pace of the game, and balancing that to cap how much an economy gives you an advantage, would be a great investment of time.

I like the idea of less metal spots, little less mex income, lot less t2 mex income where it is only a third more than t1 at most, don't ignore my post cuz i use t2 as a word cuz you know what's going on, and units might should have more build delay on factories, so heavily assisted factories are second fiddle to a new factory altogether and hopefully factory assisting will be added.

I'm always happy to see a new dev on the forums. I may not have been too active recently, but I expect things to lighten up here the day after the Beta goes live, so i will be around plenty to discuss balance.

This is extremely heartening to hear. I'm not a fan of build-order style games, and i look forwards to seeing how your balance concepts will play out.

In support of this, I only have two "high-level" balance observations. The first is that individual units should not do too many things. In order to promote the gameplay you want, units need to function almost like simple tools in a toolbox, or individual commands in a programming language. Simple in isolation, but providing a lot of emergent complexity in how they can be combined. For example instead of having a fast unit that can cloak for raiding purposes, have a fast raider, and a cloak generator, as these can be combined with other units in many more ways.

Secondly, i would avoid proscribed unit synergies. For example, in Red Alert 3 there was a soviet tank called the hammer, and an anti-infantry vehicle called the sickle. These two units were designed to naturally work well in a synergy with each other. However, doing this limits their usage elsewhere, and starts to promote a fixed strategy, which can only be improved upon by better micromanagement. There's nothing wrong with synergies in general, such as having units which buff others in specific ways, or weaken enemy units so as to allow other units to finish them off. However, these synergies shouldn't be designed from the start as specific optimum plays.

You're probably already aware of these, but I thought I would get them out for others. Looking forward to see how this develops.

Just to be clear, what is a "build order style game"? Is Starcraft II a build order style game? SCII has build orders but there is no single best build order. To me this type of build order is an unavoidable side effect of being unable to see your opponent. Anything you do before contact with your opponent is unable to be interrupted and may as well be done with some per-conceived plan. Any plan will have a build order which optimally executes the plan.

Easier early scouting reduces this effect.

Starcraft is definitely very Build-Ordery. Just watch one of the replays or a let's play, even from players that don't claim to be very good. When I hear phrases such as "i'm gonna start with a 2-rax, but my scouting saw a 7-pool, so i'm gonna change up to a BC-rush" my inner RTS fan crys.

Build-orders are not a natural consequence of this genre. They are consequences of a unit and structure progression which rigidly enforces particular synergies. At the moment, PA has one or two optimal build orders to get a player's economy of the ground. I hope this will be fixed with the implementation of the "egg". You folks did a similar thing with Zero-K by presenting a flat tech tree, and a free first factory, which is another way of doing it. There are ways of avoiding the build-order trap, but it require that the player is initially presented with plenty of viable options, which only increase as the game goes on.

Increased options just potentially increases the number of viable builds. It might be possible to not increase the number of builds which are strictly bad but it sounds unlikely.

Could you address my point that if you have not yet seen your opponent you are basically playing a singleplayer optimization game? Any plan that you start the game with will have some optimal method of implementation. Matching the plan with the implementation is a build order.

I have highlighted my key issue here. The fact is that in many SC type RTSs the player goes into the game with a set plan that they wish to carry out is one of the parts where I take issue. Why are you forming plans when you have no idea what the enemy is going to throw at you? Early objectives should be to scout whilst remaining as formless and undedicated to a particular strategy as possible (I believe this doctrine stems from Sun Tzu). If you want to pick a particular direction to go in, then this should represent a gamble, as you don't know what the enemy is going to do. You are gambling flexibility for specialisation, and gambles do not always pay off. In a game which emphasises build orders, each move you make moves you down a fairly inflexible path, which builds a momentum of it's own until the player is hamstrung when they try to adapt. The only way to improve becomes their quality of execution of their original plan, not adapting a new plan. TA offered an alternative to this, mainly through a flat initial tree, no upgrades, and the ability to assist. For example, if I'm primarily making K-bots, and then I decide that I need to switch to air, I have a number of things which allow me to make such a switch. Firstly is the fact that the basic air factory is immediately available to the commander and basic construction units, so I don't have to start climbing up a different part of the tree. Secondly comes the lack of upgrades, meaning I haven't thrown a lot of resources into a strategy I'm no longer following, which lessens the chance of making a sunk cost fallacy. Finally , if much of my build capacity is mobile, then i can quickly shift it to a new plan, perhaps with only a small penalty in overall efficiency.

Returning to a build-order game, in those limited times when they can decide to adapt, they are often forced simply to pick a different "optimal" path. I have seen SC games going on where players are still honing their build-orders well past the point where they have scouted the enemy. And these are the players who are considered good. It goes beyond a simple optimisation problem here, and becomes a core pillar of how the game is meant to be played. It only becomes at the very high levels of play that a player has learnt all the tools they need to creatively generate a new strategy. This is what I would like to see promoted all the way down the skill levels.

I agree with what you just said. I think we have different definitions of a build order. I would have called the thing you are talking about something along the lines of "strategic flexibility" or some similarly unwieldy term.

I now see that "plan you start the game with" is ambiguous. By this I meant your plan for how you start the game. As in you think "This is what I will do for the first 3 minutes", as in you kick the game off by implementing this plan. Having the flexibility to do many things after you scout your opponent is good and should be part of 'most' optimal builds.

I concur. I'm probably using build-order in a slightly pejorative sense here because... well i can't think of a better term for what I witness in a number of games.

To contrast, I do genuinely believe you folks have done a good job with Zero-K when it comes to this. In my (limited) experience Zero-K is at it's best when played in teams. When the game starts, a good team will normally divvy up responsibilities - one player will take air, another will sit on the front line with vehicles or shield bots, and so forth. As a result, although individual player's options may be limited in what they build, the whole team tries to maintain a decent amount of flexibility in how they manage their options. A team that goes entirely for planes is asking for trouble. Occasionally one gets stuck with the odd moron who constantly spams "ALL IN SUMO RUSH" to the chat, then rage-quits when he is ignored. Such plans may even be mildly entertaining. Once.

After that, it's just silly and asking for pain. And gnats. All of the gnats. Ever.

aside from the mentioned advantages of a higher HP to DPS ratio, this is also something worth considering when looking at the general scope of the game.
With multiple planetoids as a playing field and options like high player numbers the probability of multiple concurrent battlefields is far higher than in a conventional RTS game. Even with an UI that allows to keep track of things (in whatever useful way), the ability, in terms of having enough time for it, to switch between battlefields is largely increased if the battles take more time and aren't just fire and forget.

Thanks. We've tried to develop some idea of 'roles' in teamgames. It took some work to make it work for 1v1 with the issue of Pregame RPS.

I've linked the thing you described to the longstanding idea of Pregame RPS. Anything you do before scouting your opponent is not affected by your opponent's actions, so there is the possibility for RPS type situations. Pregame RPS exists when there are common openings which counter other common openings. This is something to be avoided.

I don't know firsthand whether SCII has much pregame RPS but given it is played at a high level I would be surprised if it does. Surprised because I don't think people would play that sort of game, especially not at a high level.

I'm no expert, but my viewing of various replays by what entertaining casters I can find suggests it does. I suspect that at higher levels, this changes somewhat as some builds would be more flexible than others. The majority of SCII games i have seen tend to be quite short. It seems that as SCII is designed to be playable in 15-30 minute games, RPS style strategies off the bat are an appropriate way to limit play time. Either you win, or you lose, but you don't drag out the game. This makes sense for a game which is designed to cram all it's engagement into short periods of time, but its not a game style that I enjoy playing. I don't mind watching the odd game, but playing it just makes me grumpy.

Not sure if it's the best way to balance things but with regards to AA and especially mobile AA it would be nice if there range got increased to a bit past their field of vision. And if the tanks had there chance or hitting air units reduced to ~10% either by a simple prob calculation or by reducing the angle they can fire up at.

Ideally it'd be nice to see mobile AA have enough range that you can include them in with other mobile units and if you have radar coverage they have a chance to shoot the units down a bit further out than they currently do. If you combine that with a nerf to Ants & Levelers ability to shoot air I think it'd make things more interesting. The AA range boost (but not vision) would allow them to engage aircraft from further away so even if they're in a middle of a swarm of units they've still got a chance of hitting the Aircraft before they come in range of the ground units. Otherwise we need some sort of formation that allows AA to be placed on they outside to intercept Air...

Thank you everyone for your responses. It is unlikely everything with be addressed for the release of Beta, but here are some of areas that I have identified as being most important to the forum community.

Damage to HP ratio too small.
Direct overpower upgrade from ants to levelers bad
Direct underpower upgrade from hummingbird to Peregrine
Economy jump from basic to advanced too big
advanced economy just being upgrade boring/doesn't provide strategic decisions

I can say I agree with all of these to some extent or another and will be presenting your comments/concerns with the team.

Hmm, you say the jump is too big, and also say that 'the jump' is boring and isn't a strategic choice.

Interesting. So... do you like the jump at all?

Great stuff, thanks for the feedback.

Some sort of economic jump is necessary to fuel the requirements of waging a war across an entire star system. What that jump is will depend on how multiple planets play out. The goal is to provide enough resources to allow for end game play, while still making build decision meaningful.

Make everything do a little less damage or make everything have a higher HP across the board. Seeing a basic bomber take out several tanks in one shot is ridiculous. Seeing a tank kill another tank in almost one shot is also ridiculous. Units should have more staying power in my opinion. It makes for more interesting battles when units can take more than one shot of damage.

Part of this might be due to the fact that units clump so close together (like in supcom2).