My balance concerns - Not really about balance but how it's done.

Zaphod, somewhere around here this is a signature dedicated to you about the zaphod limit. I think that is very appropriate to mention here, and using starcraft 2 (!ghasp!) as an example of how balance works is also appropriate.

So, storytime. I dont know if anyone follows starcraft 2 at all, but I have a perverse interest in occasionally reading into all the little gripes and current 'drama' in starcraft balance. I dont actually play starcraft very much, though.

Lately, people have been raging at this little portable mine unit called the 'Widow Mine'. Some people say its WTF IMBA OP and others are like 'Meh. Its pretty good'. How? Well, for low level players, its a devastating weapon because of its APM requirement to deal with is way higher than its APM cost to implement. For high level players, however, this is not so much of a big deal. They have a much higher pool of possible actions per minute to pull from, so its comparative impact isnt nearly as much. Instead, its other balancing factors, like how it inflicts friendly fire and tends to telegraph movement come into play.

But only if you have the APM to get there. It may be more or less balanced for pros, but for average joes, its completely and totally gamebreaking. And anyone who says this is micro, it wont matter in PA, is dead wrong. What is macro is starcraft will become micro is PA, and 'attention' will be just as critical of a resource as energy or metal, if not more so.

All this to say that balancing at the pro level and balancing for the not-pro level is often times like balancing two different games entirely. This is evidenced in lots of games, from super smash bros to Halo and League of Legends. I think PA wont suffer from this disconnect between the metagames of different skill levels as much, but it still needs to be addressed carefully. Balance is hard.

We had the same problem with the Mercy in FA.
Long story short, even it wasn't a problem for good players, they were also able to realized it was one for new/less skilled players, and helped fixed it (without any nerf or buff actually, it was a hitbox problem).

In this case, I think it's a problem of caring about the game VS caring about your level of play.
It's a very good point, I should have mention it in my OP.

@stevenrs11 I totally agree with you.

There is balancing for the skill floor for the lowest common denominator.
And balancing for the skill ceiling for the competitive pros.

Both must be accounted for.
Generally, devs can handle the skill floor balance pretty well from community feedback.
But with balance at the skill ceiling, the only ones who face this daily will be the competitive gamers.


Putting it simply.
Blizzard should be taking the time to listen to Bronze league players to ensure the game is fun and balanced at the entry level.
But they should also be listening to their eSports superstars and masters league player. Because if something is IMBA, they will find it.


Zep knows what he's talking about.

I am a personal believer that there should be no skill ceiling or floor at all. My ideal game requires no "skill", in that everything you have to learn is the result of conscious mental faculty, as opposed to unconscious routines and "muscle memory".

Now obviously I am a super-minority in this radical point of view, but being an RTS player it can't be that foreign of a view. The primary issue with actually designing a game around being focused on thought and prediction instead of skill is that the game balance isn't logical. Unit power doesn't scale positively over time, the economy doesn't require significant amounts of input from the player, and most importantly, "High-skill" players do not have the final word in what is balanced and what is not.

I think Tribes Ascend is a good example of how a game can have ludicrously good gameplay, but be so hard to "learn" (read: adapt your instincts to) that all but the most hardcore, dedicated players can actually play the game. Everyone is always too discontent with getting killed/loosing all the time to play with this elite caste of veterans.

Oh god I remember the Mercy. That was some mean ****. Perfect example though, totally stoppable for someone over a certain skill/awareness point, but ridiculous for anyone below it.

I am constantly amazed at the ability of people to deliberately miss a point or ignore what is right in front of them.

Ok I'll bite. What is your point?

As far as I can tell you've basically said that FaF is bad because it lacks good AI support and the players of FaF don't like to play with AIs in their 1v1s.

If you ask the top 10 players of a game and they all say X is imba after carefully considering it and testing it out over many games I think it can be believed that X is imba. But only at their level of play. The example with SC2 is pretty good: A newbie might not be able to abuse the imbaness of X at all because you need to be too fast for it and never build it. From personal experience in SC2 I fully agree that the gameplay differs greatly and the "problems" a player has ("damn X is imba, Y is too weak" > get better > "X is so weak, but Y kills me everytime" > get better > ...) do change drastically with their skill level.
So in a perfect solution you indeed need to balance the difference skill levels in a way that they all work out. This may be harder in SC2 as it can be crazy APM dependend, but PA will also have problems with it, just like FA.
Timevans with his AI issue is a perfect example for a person who is unhappy with the balance (***rush problems solved by AI) at his skill level.

So yeah... balance is hard and Uber should listen to the top players. But not _only_ to them. It's more like that they need to balance it based on the top players feedback but without destroying the balance for others.
I'd say the feedback of the top players should have the highest priority though, as they know most about the game. So they have the most knowledge about how the balance works and can balance the game best at all skill levels. If they want. Ofc many high level players might prove incapable of understanding the problems of casual players. It's easy to go "you just need to do this and that" and ignore their complaints.

I think in the case of balance in high level vs low level play we should draw a distinction between the 'practical balance' which occurs for a certain mechanical skill level and the 'idealized balance' of a game which is how the game plays if you assume no execution barrier.

Highly skilled players do not play the game perfectly but they are the most likely to have glimpsed at the idealized balance through the exploration of the practical balance at their level. I agree that it is bad to focus on idealized balance at the cost of practical balance for less than highly skilled players. Instead of creating a different game for different levels of play I think we need to bridge the execution gap between high and low APM players such that unit interactions do not change significantly with APM.

For example if a unit can kite another unit then it needs to be balanced with this in mind otherwise high APM players will exploit this. But if it takes a lot of APM to perform the kiting then the unit will be underpowered when players with low mechanical skill play (all other things being equal) because they will not use the unit as it meant to be used. This is part of my rational behind automated kiting, it allows the units to be balanced with kiting in mind so no large advantage is given to players which are able to perform kiting micro.

The trouble with Faf is the same as nearly all other RTS online the games, options and tactics are limited in scale. The reason I think this happens because of performance ceilings people tend to load smaller games. Lan or Hamachi gaming is on a far bigger scale in my case.

I'm pretty sure I've coded a "Host" button that allow you to play any size of map you wish

Its not that simple.

I don't follow, in my experience with SupCom the main factor that limits a game's 'scale' is the Synchronous nature of the engine, not the way the players are connected. In theory from my understanding if you were using the same computers in both a LAN and 'normal' online setup they should be able to handle the same 'scale' game.

Mike

Care to explain?

First up, I have to say I'm constantly amazed by the quality of FAF. It's not just that the game is balanced: it looks balanced to me, but I wouldn't give my opinion much weight on this front, and, y'know, T3 Seraphim subs.

The balance of the game actually does something much more inspiring: it reveals the potential of the original GPG design. It has, as far as I can see, faction diversity, a solid pool of viable strategies, and a corresponding responsiveness to player decisions that makes the gameplay expressive of the players themselves. This is a substantial achievement, and imho Zef and his team have realised GPG's original vision. GPG provided the ingredients, but there's a real sense in which the FAF team, with their patience and commitment, have built the game.

We are a long way from even seeing that final game for PA. I honestly think that there is no possible way that Uber will achieve that level of polish on pure gameplay terms by this time next year, let alone by Christmas. But I think asking that kind of polish of Uber is entirely missing the point.

You absolutely do need high-level players battering each other into submission (preferably recorded in replays) to be able to run iterations on the game's balance; but you also need the rest of the playerbase to be constantly dreaming up crazy new strategies and occasionally surprising the pros with the more viable ones, even if you don't actually spend too much time listening to their opinions. Yes, balance works differently at different skill levels, but it's far better to balance from the top down than the bottom up. We learn from the pros a lot more often than they learn from us.

What makes a game balanced is one thing. What makes it balanceable is very much another. Blizzard couldn't balance SC2 if it didn't have a ton of folks, from casters and pros all the way down to bronze-leaguers, playing game after game and passing on feedback. Likewise, I honestly doubt that, given the small competitive playerbase, Zef could have balanced Supcom in much less time than he's actually taken.

So long as Uber are prepared to work closely with the community after release, or at the very least hand that part of the game over to the balance modders after an appropriate period, we won't have a problem. The thing about games being balanced is that it can take a while to work out whether a game is actually unbalanced. And provided we continue playing after release, and Uber release regular updates that reflect our growing knowledge of the game, our problems should only hang around a little longer than it takes us to realise we have them.

No the Lan is faster and more stable than online, Mike you should know something as basic as that if your any good that is.

People online either have poor internet or tiny computers or both. They never want the 20 minute start and never opt for maps like world dom.
My friends refuse to take out the sorian 2.1.1 and therefore we cannot play Faf. We would love to come on and have people schooling us but without that mod we would have stopped playing forged alliance a long time ago.

Did you read what I said? I said that using the same computers there is not difference between the scale of the game they can host regardless of the connection type. Technically a connection doesn't actually slow down the game, at least not in the same way as sim slowdown.

The Sim will slow down just as fast in a LAN as it would online assuming you used the same computers in both setups.

Mike

But we have sorian integrated in FAF. It's the latest version, not 2.1.1, but I still don't know why the older version is better.
If you have solid arguments, changing that is trivial.

@timevans999

do you really think that the outcomes of 20nr games with AIs should have any weight when it comes to balance?