Saktoth is moving house and going to uni and wont be around as much for a while. 1faction is basically finished model wise, though it could do with more work on gameplay, that is non-essential stuff. If there is anything specific you need/want me to do you'll have to tell me. Before its proper launch however. 1faction needs a GUI, a frontend, a lobby, easy, accessible 1-button singleplayer content (just chickens, or AI's, or ideally, missionbuilder missions), a name and branding, a website, and an overall smooth and consistent experience.

Maps: Backgrounds need to match the map, rather than the sky.

Nanotower/Overdrive rings should highlight units in range.

The constripe border is busy.

buff merl?

nerf lugar?

Chicken, unassistable factories (otherwise conventional econ).

nerf halb

• Google: http://trac.caspring.org/changeset/5164 should be fine now. Probably UP but that's ok, noone likes hovers
• Saktoth: Nope, its still holds its own and can still hit gunships/raiders which is more than you can say for the anaconda.
• KR: In closed SP testing (yeah yeah I know, should've done those before committing buff), post-nerf Halberd loses quite badly in straight up fights with Reaper, Bulldog, Ravager, Stumpy and even Anaconda. Anaconda performs significantly better than Halberd against Reaper and Crabe. On sea, vette whacks Halberd, though annoyingly cruiser loses to it for cost (cruiser probably needs looking at). Halberd has a really poor HP/cost compared to other assault units. I agree with Google, it's probably pretty UP now.
• Saktoth: Did you /godmode /spectator micro both of them, with the halberds retreating? Did you take into account the fact Halberd can hit AK's etc 100% and gunships sometimes? Cruisers main strength is range, weight and antisub capacity its probably OK.
• Google: That's the strange thing, even pre-nerf Halberd did not have stats any better than Anaconda for cost. But when you add those stats together you've got a high weight class skirmisher with much more HP than other skirms(disregarding cost) with reasonable dps/cost and the ability to hit fast things. It's a combination of double stats(cost, hp, dps) not equaling same power and the unit being almost made to be OP. It's like the old Croc, when it comes to unit balance we've either got to chose Assault or Skirm not both.

grenade dcl? dumb idea

Shields start charged? More 'rush to hold off arty', less 'maek eplic proec'?

Underexplored roles and stat combinations

There are a much wider range of stat combinations possible than CA currently uses. A part of this is sticking to the riot/raider/assault/skirm/etc roles too religiously, we should rather think of units as a range of overlapping RPS structures and abilities. A rocko and a samson while both skirms do not share all the same counter mechanics: The rocko is countered by the roaches, the samson counters roaches. The Samson deals with LLTs and Antiswarm in an Artillery behaviour, the Rocko deals with them in an Assault behaviour.

Even using the basic stats of HP/DPS/RoF/Speed/Weaponvelocity(accuracy)/AoE there is a lot of room for unique units. These can possibly be applied to current ununique units or new units (Hovercraft im looking at you).

Raiders

Raiders are pretty homogenous. Fast, low HP high DPS prettymuch defines them, the exception being the Zipper (medium HP, very fast speed, low DPS) and the scout/raiders which tend to have low DPS too. Short range is also a must on fast units, due to kite. But other common stats, like a high accuracy and RoF are not necessary for raiders. In fact, an inaccurate low RoF unit that can hit and-run buildings and slow units would make a good raider.

There are barely any short ranged or fast units that are also inaccurate or have low RoF. The Pyro doesnt lead, but this is frankly a bug. The Panther has a bit of reload time, and is fast, but it has medium range.

A part of this is how hard it is to use low RoF units, short ranged units will hit the first thing that gets in their range. But since this is so short, it should actually be easier in many respects to micro the reload time on such units than longer ranged ones. Think a grenade-shooting weasel with a slow velocity, high arc, and long reload. Something that can strike a target, as a raider, run to reload, then make another strike. Or do sweeps, or use its frontloaded damage to make quick attack runs on high value targets.

Equally, something like a melee-equipped, accurate, but low RoF raider, or a medium RoF stumpy-like weapon which is nevertheless inaccurate.

There is also a lack of any pure-DPS raiders, ones that kill quickly but are useless vs anything that can defend itself- most raiders are also good combat units. The Pyro used to be but it got nerfed due to all its secondary abilities. There also arent many high-HP raiders, or high-weight raiders. The Panther is prettymuch the heaviest raider. The Zipper and Egg have higher HP (as 'blockade runners') and weight but they can be hard to use due to their special abilities (zipper also sucks generally).

Gunships

Quant said the only way to vary gunships across factions is with HP and DPS differences. I wholeheartedly disagree with this! The following basic stats are underdeveloped or totally unexplored in Gunships:

1. Low RoF. Ive gone towards rectifying this with the Black Dawn. It should play like no other gunship now, because they all have rapidfire weapons.
2. Range differences. Though we have the hugging issue, if we can fix this (probably engine side) we can have range differneces in gunships. Still its perfectly possible to have a turreted gunship that hovers rather than dances, and doesnt chase. This would allow it to exploit its range. Evil4Zerggin: Range is a tricky thing with gunships--back in the day there was a huge row with 600 range Vindicators. Although we could go the opposite direction and have a gunship with extremely short range in the vein of Red Alert's Kirovs.
3. Static vs hover. Krow does this but it can easily be explored in other units. The problem is that for fast units, it really really does dodge a lot of shots. Evil4Zerggin: Perhaps widgetable.
4. Slow Gunships. Again, Krow. I dont know if we want geniunely slow gunships but its just another stat to play with.
5. Weaponvelocity/accuracy. The Blackdawn used to be unable to lead its target and jam its firing, but this is NOT the same as being inaccurate, its a bug and aiming issue. If we want genuinely inaccurate gunships we can do it via the physics.
6. Special abilities, like shields (to protect other gunships and even ground units), cloaks etc.

Evil4Zerggin: I do think that more gunships would be nice. Unfortunately, it doesn't quite work like land in that gunships can go anywhere and are difficult to apply fine micro to, which removes some of the positioning details with land units.

Midknight: we have no effective "anti-raider" gunship. Banshees are more effective vs. air, rapiers just generally suck, I've had a pene kill my blackdawn, and a group of flash can kill a brawler. So far, only the bladewing has come close to this role...

Maybe a dedicated anti mobile gunship, which could be deployed over an area, say, if the enemy is rushing your ally, who made only 1 llt. :)

Saktoth: People whine rapier is OP. The Bladewing being anti-raider is a HUGE PROBLEM. In fact, probably the #1 problem with the bladewing. The problem with robust anti-mobiles that go at the ridiculous speed of gunships is that they shut down raiding completely. This is not what we want. Gunships are already pretty resonable at quickly responding to a deep-striking raider force, much must usually be done with your own raiders. There is a very good reason that riot units are so slow, to give enemies the oppourtunity to get around/past them. If a geniune anti-raider goes faster than raiders, it also begs the question: Why not also just use it as a raider itself? Ill let you in on a little secret. Riot units operate mostly by just having ridiculous, OP, crazy DPS for their cost. Its only that all their other stats (range, speed, HP) are crap that stops them being OP. The other stats that make a Riot unit are a high RoF (generally) and AoE, and we already have gunships that do those things.

Midknight: hmmmm... I'd say, give the anti-raider gunship low hp to prevent raiding, maybe a design that allows really fast strafing, but slow movement? Also, imo a riot gunship doesn't need to be fast, it just needs to hit units accurately (maybe high los and weapon range?). the banshees accurate, but doesnt have a powerful enough weapon to do stuff like this. It's other strengths far outweigh this particular use. The situation that made me aware of the lack of anti-mobile gunships was a supreme lake game i was playing, where our ally had lost his outpost on the little strip of land connecting the 2 shores. I sent 3 rapiers, one rezzed brawler, and 15 blastwings (not clumped) to hold them back until turrets were made, but they were almost instantly. As to the bladewing, I'm working on an indirect counter to them (insert maniacal cackle here).

Saktoth: I just dont think a dedicated anti-mobile gunship is desirable at all. The current gunships fill that capacity and if they are underperforming, can be changed (they might fly a tad low and close). What were you trying to hold off, how many, what cost was it, did they have AA? If you mean 'before turrets were made, but the turrets were made almost instantly' then uh, riot wont help you. As for working on stuff: share with the class, now.

MidKnight: @what were you trying to hold off:

!-----------|

!------------LAKE

!base------------------ !------------------tons of dead turrets---------my gunships vs. 5 stumpies and a flash----------------wreckage----gigantic army of ~20 warriors, ~30 glaives, ~20 stumpies, ~30 flash, ~4 janus

the rest of the army weren't in firing range and were trying to get past wreckage, my gunships were vs a few stumpies and flash, my blastwings attacked the big army

@share with the class: have you not seen this yet?  http://spring.clan-sy.com/phpbb/viewtopic.php?f=9&t=17235 I consider this a license to spam #ca with the link :P !

quantum: I don't think the only possible way to vary gunships is by DPS and HP. I'm saying that the current T1 gunship differences don't matter much, the Rapier and Banshee are nearly the same unit. Changing DPS and HP is one way to resolve this, but I'm open to other options.

We are sticking to Raider/Assault/Skirm/etc too much. At this stage, they are more useful as description than for design. What units need is to have pronounced strengths and weaknesses. No jack-of-all-trades units! In this we agree, if I understand correctly.

The main problem, imo, is that balance changes tend to be strong-point nerfs or weak-point buffs (because balancers don't have game design in mind when they do the changes, often), so our units are becoming less diverse over time, with average stats. Also, unique features tend to be removed for balance's sake, but none tends to be introduced.

In addition, I think that there should also be more tolerance for slightly unbalanced features. Mathematical balance with units that have the same stats and very diverse but broken units are two extremes of a continuum. We are setting our bar too much towards balance imo.

Saktoth: For a unit that is faster and has more range than other units (The old AK, the mumbo, etc), there is no choice but to nerf one or the other. These abilities, in combination, mean that the unit can beat an infinite number of units slower and lower ranged than it, given ideal conditions. Sometimes it is the outlying stat that is wholly responsible for the unit being overpowered. Even if you nerf it to be ridiculously weak in its other stats, it will still be able to kill an infinite number of enemies- just more slowly and at greater expense.

'Tolerance for unbalanced features' will lead to a homogenity among units not unit variety. Why? Because people will only use the most unbalanced unit, and nothing else. See the flash in BA, BA has a lot of unit variety yet to no effect, because of a unit monoculture. Balance must be taken hollistically. We must ask 'Are there units that can beat this unit?' 'Are there units this unit beats?' (and also circumstances where the unit is better or worse). The bot factory especially has extreme problems with fast + ranged units (even the samson), making single units counter the whole factory. Perhaps this a flaw with the bot lab specifically, but there is a problem here.

quantum: I'm not advocating tolerance towards large imbalances. Only slight imbalances. Most units aren't nerfed because nobody is building anything else. They are nerfed because the unit is probably OP although people are not really certain, or because it's too strong against one class of unit which is commonly used anyway. The number of 30% or even 50% changes show that balance is really not something fine-tuned. Changes with small balance benefits but big diversity costs should not be hastily committed. An extreme example would be eliminating skirmishers because they can kill inferior costs of their counters with 0 losses easily.

Saktoth: Most 30-50% changes are either changes to design rather than balance, or are in reaction to other changes, other than a few outlying units (the skuttle, for example, will need to be meganerfed, but most people havent caught on yet how OP it is- thought this is actually only because the walker lab used to cost a bunch). Only balance can preserve diversity: not single case balance, but a hollistic approach to balance. Specifically, a unit cannot be a counter to many more units than it is countered by, especially if it counters several whole facs and is only countered by a few units in specific facs. This is especially true if it is a hard or absolute counter (such as skirmishy units can often be, as you say, taking 0 damage) to a huge range of units, with only a small range of soft counters that can beat it.

Balance

Bladewing still a problem?

Most heavy units are probably OP. Stronger anti-heavy counters, or just nerf heavies?

Sumos role is meh.

More grav guns!

Hammer was overnerfed?

kbot lab should be closed at back

land firing crasher? (core kbot needs range and accuracy!). But then what to do with AA?

flash regen (too much).

roach (less AoE imo, maybe more HP to compensate so its easier to use).

Wolverine range/weight

Modelling stuff

Profile over detail

OP?

UP?

IDEAS

• Subs as own tech tree (Probably not).