Thinking about blobbing

[KDR_11k]

So one common issue being talked about is blobbing and that it's bad. Maybe I missed key discussions but I feel like there are some key considerations missing before we can even go on the search for a blobbing solution.

First off, why do players blob? That's easy to answer: Because there's no real downside, only some arbitrary special cases where your blob will get murdered right away (Fortress, Eye). Because the combat physics are so simple (you shoot, you hit the target you were aiming for) it's pretty much a numbers game: Who can output the most DPS and withstand the enemy DPS the longest? And because it's simple, one eye opening fact was that in game balancing the effort of the player is one of the cost variables along with the explicit costs that the game actually knows. A fleetball set to FRD requires the least effort on the part of the player and usually works, it's inefficient (because the ball splits up) but the cost of replacing those dead ships is usually lower (only some resources which you tend to be floating anyway) than manually sweeping enemies up with the most effective ships (costs a lot of player attention).

Second, what should players do instead? That I can't answer. The simple combat system pretty much means that more ships directly equate more DPS and HP which also equates more survivability (because enemies get to fire fewer shots) so using fewer ships makes you weaker and lose more ships. And because of the combat system the interaction between your units and the enemy usually amounts to slugging it out (trading blows until one dies). What can a player do to change the outcome of a battle other than adding more ships (maybe of a specific type or something)? Since micro is specifically discouraged there's not much else to do.

Third, why is blobbing bad? It's boring and simple but is there any alternative that's more fun? Does sending a smaller force (essentially a smaller blob) actually make anything more fun? What can you do to make the simple interaction offered by the game more fun?

Most of AI War's strategy happens on the cross-system scale, which planets to attack, which to take, which to avoid, etc. How to make your way to the AI homebase. The system layout may matter for raids where you want to do only a small amount of damage with an expendable force but during an assault you're going to fight everything anyway and most of the time the distances don't matter much (except for a few key targets like fortresses). And in these assaults the fleetballs come out. The objective is to kill everything so committing your entire army makes sense.

In other RTSes the cost of units is much higher. Maybe not in the immediately visible sense but if you're fighting an active opponent (another player or an AI that acts like a player) your main cost is how much time you take to build your force up and how vulnerable you are to raids since an active opponent will build up quickly too and wasting resources would make you fall behind in the arms race. In AI War that arms race obviously doesn't exist, the AI is almost completely passive in that regard. You don't have to operate on a tight schedule if you want to survive because the AI's actions are so slow that you have time to build whatever you want. That's the usual RTS campaign issue, against an enemy that does not grow (granted this AI grows to SOME degree but all planets have ship caps and the AIP growth happens primarily at the player's pace, not the game's) your best bet is to grow as much as you can and then use overwhelming force.

The only RTS campaign I remember that avoided this was Perimeter and that's because Perimeter had an active AI in the campaign that built its forces up at full speed so you had to act quickly to not get outpaced.

So in short: As long as there's no major time pressure every unit is cheap. Cheap units get massed. Being efficient in AI War is not about only using the force you need but only applying force to the parts you have to. Applying force is done best via fleetblob, even if most of the units do only pitiful damage against the target that's still more damage than they'd deal if you didn't bring them along. And losses can be replaced.

If we want to prevent blobbing we need to think about what exactly we want to encourage. Sending a fleet of starships instead of your fleetball like you'd do in response to an eye isn't exactly that much different.

[chemical_art]

Hard to say, I take blobbing to the extreme.

Bring in everything. The starships, the traingle ships. Get some cheap, high cap ships, get a enclave, and make every battle like the battle of Stalingrad where essentially weapons roll right off the factory into battle. Only stop bringing in the sledgehammer if needed for defense or if there is an ai eye.

And it works.

[Draco18s]

Yep, same here.

I tend to hold golems in reserve (depending on type, and how strong my economy is) just so I have a big unit to use in case of emergencies.

Also because they cost two arms and a leg to repair.

[Cyborg]

I prefer softening up the enemy before bringing in the troops. On games where you have multiple theaters of war, you really can't bring everything unless you have a great defensive set up or spire units. One of the counters for blobbing is the eye. There are much more elegant solutions for those planets.

Yeah, the final assault and all that you are putting everything in there, but I don't think this is every scenario. Sometimes I dance units around to avoid defensive blasts (exploit range limits), and if you play with shield mechanics it can be even more interesting. I rarely move around one giant fleet on a planet when I need multiple targets down fast.

Maybe you are just noticing that your highest DPS is putting all your ships in one spot? I really don't think it's the answer to everything.

[kasnavada]

There already are some mechanics to try to prevent blobbing, like when the entire planet becomes "alerted" and ships freed when a blob large enough is in a system, or the eyes... The "new" guardians, which explode like a rocket by taking out a part of your blob also count.

It's enough for me to change tactics when assaulting worlds to limit losses. Some raid, some blob.

[Cinth]

I blob because micromanaging all those ships is really a pain in the ...

I do raid to remove the threats to the blob though... raiding is usually done with Raid SS, enter system, pause, set attack points, resume and watch the Raids do what they do best. heh, what would be awesome is if I could set waypoints across multiple planets.

[TechSY730]

The only way I can think of to "nerf" blobbing, short of radical changes to how AI works to give it ways to control game pacing, is to increase the effectiveness of an individual unit (mostly mobile military), but increase the cost of an individual unit by even more.

Yes, this would make "refleeting" even more time consuming than it is now (possibly long enough to span several waves before you can rebuild enough, letting the AI "bust through" and win), but that would a discouragement to blobbing as you wouldn't want to run that risk. Also, as each individual unit is more effective, you can accomplish more with non-blobbed strike forces.

However, this would probably require tuning of AI offensive actions to prevent the game from becoming nigh unwinnable even on sane difficulties, plus, "blobbing" would still be effective if you have the economy to spare. It would just be harder to have that economy to spare.

Heck, you could even leave unit stats as they are now, but just across the board increase costs.


However, would this really be a good change? The player would still control the pacing mostly, you could sidestep the "lose everything, have nothing for too long" risk of the increased costs by just keeping a portion of your fleet on defense, and "blobbing" with the rest of the fleet.
Plus, it doesn't really give any more rewards for micro, short of potential savings of losses (which would matter more due to increased costs). Yes, your individual units might be stronger, but so would the AIs, thus, military wise, you are basically in the same place as before.



As for the pacing thing, Chris has posted an explanation of why the human has control on the wiki, Why Does The Human Player Always Have The "Tempo?"

[Cinth]

Ugh. Just no. I despise micro... forcing players to micro is just wrong. It isn't fun.

I pretty much ditch starships once I can get enough higher mark fleetships unlocked because they take to long to replace. Making fleetships cost more will make them feel like starships.

Also in high AIP games, there comes a point where the game gets so bogged down that several seconds can pass before you can even react to or give inputs. Fine control at that point is completely lost.

[Wingflier]

Third, why is blobbing bad? It's boring and simple but is there any alternative that's more fun? Does sending a smaller force (essentially a smaller blob) actually make anything more fun? What can you do to make the simple interaction offered by the game more fun?

Personally, I have no problem with "Blobbing", but I don't think it should be the optimal strategy.  If your entire force and right click a destination all game, that should be an option, but it certainly shouldn't be the BEST option, for obvious reasons.

One thing that makes AI War great is that it doesn't have as much emphasis on micro as many other RTS games; but at the same time micromanagement IS an important part of an RTS.  If micromanagement weren't a part of your RTS, then there would be no reason not to make it a TBS.  With blobbing in place, you could realistically get away with about 30 actions per minute or less.  To put it into perspective, most professional Starcraft 2 players have an APM of 200 or higher.  The game shouldn't be based on how fast you can click, but it should keep you busy enough that you're not twiddling your thumbs, waiting for your units to do their jobs (in my opinion).  I think the mark of a good strategy game is to always keep the player busy, but not overwhelm them.  With the current blobbing system, I find myself waiting on my units an awful lot.

It's also really boring and takes away from the game.  Can you imagine watching Star Wars or any other Sci-Fi flick again and instead of all the Fighters, Bombers, Frigates, Cruisers, and Battleships moving at their own speed and having their own mission, they were instead all moving into 1 massive, slow blob, firing at everything from a range?  I mean it's just not a very appealing game mechanic to have honestly.  I'm not saying we need to turn AI War into Homeworld 2, but there should at least be SOME incentive to micromanage your units if you want.  I'm also not saying that blobbing should be taken out of the game, but it shouldn't be the BEST strategy either.

My personal belief is that blobbing is the result of the homogenized nature of the Triangle and the Triangle Counterparts.  There's really no reason not to lump your Fighters, Bombers, and Frigates together and just send them in a massive ball to your target(s) (unless it's a Fortress).  Adding bonus ships into the mix doesn't really change the formula either. 

My analysis of the problem is that the Triangle ships are not unique or individual enough to really warrant using them specifically for what they supposedly counter.  There's very little incentive to send certain Triangle ships on their own individual missions for maximum efficiency, because even then they would probably do worse than if they just stayed together as part of the blob.

I think that if we changed the Triangle, making the Fighter a fast, short-ranged powerhouse, which was strong against most targets, made Frigates kill them quickly, and made Bombers the counter to Frigates, the entire dynamic of the game would change.  I've mentioned this before in my Mantis Report, but in my view, Fighters would be extremely fast and powerful, but also very fragile, dying to Frigates quickly.  Therefore, making them part of a blob would have disastrous consequences for the player.  However, not using them at all would be wasting a huge source of DPS.

http://www.arcengames.com/mantisbt/view.php?id=9122

In other words, the Triangle needs to be less homogenized, and the stakes need to be higher.  There needs to be incentive to take your Triangle apart and use them for what they are best at.  Bombers will ALWAYS have a purpose because of their specialized roles against shields, structures, and golems - that's not a concern.  However, as long as Frigates and Fighters are second rate, they'll always just be better as a boring fire support - a cog in a wheel. 

We can add Eyes, and soft counters, and whatever other band-aid we want to the game, but as long as the Triangle (and their counterparts) continue as a homogenized, stakeless, blob-friendly mess, the source of the problem is never going to go away.  That's my take on it.

[Martyn van Buren]

I'm not really sure whether the essentials of the balance now need improvement --- I certainly enjoy the game as it is --- but less blobbing sounds like an appealing thing.  I was just wondering, what if the triangle weren't a triangle, as it were?  Supposing that bombers became something very fragile and specialized towards rapidly destroying large structures, and that these were given a bit more teeth, so the tactics game became about trying to get your bunch of bombers across the map safely to take out the target?  I think for that idea you'd tweak the multipliers so that either fighter- or frigate-type fleet ships could quickly decimate bombers.  In order to get to a guard post you'd have to find a way to cut a path through the AI's fleet ships and get your bombers in place.  It *might* create a non-blob style of play, although I'm not sure if it would be interesting or just infuriatingly fiddly.

[Draco18s]

I think that if we changed the Triangle, making the Fighter a fast, short-ranged powerhouse, which was strong against most targets, made Frigates kill them quickly, and made Bombers the counter to Frigates, the entire dynamic of the game would change.  I've mentioned this before in my Mantis Report, but in my view, Fighters would be extremely fast and powerful, but also very fragile, dying to Frigates quickly.  Therefore, making them part of a blob would have disastrous consequences for the player.  However, not using them at all would be wasting a huge source of DPS.

I don't see how this changes anything.

All you've done is changed the "primary" ship from bomber to fighter.  The reason not to blob?  Because they die fast.  Bombers already die fairly quickly if you try to blob against fighters.  The reason to blob?  Because of the high DPS against particular targets.  Bombers already do that.

[Cyborg]

I maintain this is not a widespread problem and not always the optimal tactic. If people can think of new tactics, that's great, but let's not get too crazy… the systems are so integral to the program, and it's a lot of fun right now. If everyone is going around blobbing and getting easy wins, please post here, but I don't see that at all.

[Wingflier]

I think that if we changed the Triangle, making the Fighter a fast, short-ranged powerhouse, which was strong against most targets, made Frigates kill them quickly, and made Bombers the counter to Frigates, the entire dynamic of the game would change.  I've mentioned this before in my Mantis Report, but in my view, Fighters would be extremely fast and powerful, but also very fragile, dying to Frigates quickly.  Therefore, making them part of a blob would have disastrous consequences for the player.  However, not using them at all would be wasting a huge source of DPS.

I don't see how this changes anything.

All you've done is changed the "primary" ship from bomber to fighter.  The reason not to blob?  Because they die fast.  Bombers already die fairly quickly if you try to blob against fighters.  The reason to blob?  Because of the high DPS against particular targets.  Bombers already do that.

You don't get it.

The counter to Bombers (the best Triangle ship currently) is Fighters. 

Fighters get cuddly hugged  by the campy Frigates, which you blob with your Bombers.  Therefore you encourage blobbing by making Bombers the primary damage source.

If Fighters are your primary damage source, and Frigates kill them quickly (even more quickly than they do now), then by blobbing your Fighters they all die before they get in range and therefore you're wasting a huge source of damage.  It doesn't work that way when Bombers are your main damage source, because their counters can never get in range.

[Martyn van Buren]

To refine the idea I tried to express above into a less nonsensical form, what if the AI just had more fighters on its worlds?  Players usually recognize that its a good idea to keep a fair few fighters on defense, since they're pretty good against the things that are going to blow up your forcefields and turrets; wouldn't it make sense for the AI to do the same?  And wouldn't it force you to be more careful about how you use your bombers if the AI took them seriously?  I feel like one problem might be that the AI tends to spread out its reenforcement choices so much among its different ship types that it isn't especially dangerous to throw a lot of your bombers at most of their defenses.

[Diazo]

Okay, I have this up on mantis also but the core problem I see that this change tries to address is that the bombers are the pure attack platform.

Fighters are a defensive platform but because turrets are a better defensive players build turrets for defense and bombers for attack and fighters get forgotten about.

Frigates are pretty balanced (hybrid), they have the range to be useful offensively but their attack bonuses make them more defensive.

As long as you can bait the AIs mobile ships into a friendly system seeded with turrets, fighters are always going to be weaker in terms of usefulness then bombers.

To correct this, how do you change the fighter to make it useful while on the attack in the AI systems without turret support? Answering that is the real fix to this question, just boosting the fighters stats will be taking it into the bombers role for attacking stationary targets.

For the record I do not actually have an answer to that question at the moment.

Also, for purposes of this post, Attack: Destroying AI guard posts and command stations.  Defense: Destroying Mobile AI units to stop them attacking your command station.

D.