Do you prefer removing Novax omni so it isn't an fast, unkillable counter to Cybran cloak at the front/in their base or would you keep it because it's cool to have such a rare interaction?

@maudlin27 I think you're underrating the shield by rating it purely on its stats/mass for a few reasons:

1. A big shield HP pool is super valuable to the UEF roster due to their reliance on Percies which are very vulnerable to AoE from Ythotha, Megalith, Fatboy, and SACUs.
2. Percies are also vulnerable to overcharge which is imo a bit too early to be building shield SACU vs ACU, but Sera SACU can have OC that the shield would be good against. The shield generator also does not die to a single OC like other mobile shields.
3. Due to overspill shields have a cap of 3.3x their HP in terms of how much damage a shield ball can block, so having 2.2x the HP of Sera which is 3.3x UEF and 2.85x Aeon lets even just the first shield have much more effective HP than stacked mobile shields. Sure you can micro shields in/out or off/on to mitigate overspill, but vs burst damage there's no time for micro, which plays especially nicely into UEF's roster countering the air snipes that Fatboys experience.
4. The single large bubble takes less damage from AoE than stacked small bubbles. For example billy nuke definitely would not go straight through and kill the shield generator, while it can 1-shot giant stacks of T2 shields and 2-shots stacks of T3 shields due to how AoE interacts with multiple small shields defending each other to deal extreme damage.
5. The SACU also has buildpower, can be upgraded with sensors, and has tons of HP letting it survive to recharge instead of dying like normal mobile shields.

Have you considered using /lua/ui/game/shapes/lineshape.lua? I'm seeing that you're copying over the worldmeshes used in the height indicator which seems a bit inefficient seeing how many are needed.

no
There have been attempts at drawing it using lua code but they had poor performance and are not terrain sensitive.

Develop and FAF Beta have been updated with the following changes for testing, and we have also provided some questions to guide your thoughts in gameplay:

Please provide replays with your feedback if possible.

Czar Buff
Cost slightly reduced and AA missile firing angle limits removed.

Does its cost better match its anti-ground performance without being overwhelming in air-to-air utility/performance?

GC Claw Nerf
DPS and range heavily reduced but units can now be shot while they are in the claws.

1. Is GC still too powerful against T3 units by itself?
2. Is it too powerful accompanied by DPS from the surrounding army?
3. Do shots targeted at units in the claws miss too often when the GC is moving?

Beetle rebalance
Cloaking removed and speed reduced in exchange for HP and regen.

1. Can both sides interact with the beetle in a fair and enjoyable way, especially in T2 stage?
2. Is it too weak, too powerful, or too difficult to use?
3. How powerful is it when dropped?

Sniper rebalance
Range significantly reduced (below T2 PD range) in exchange for movement speed and a large cost reduction.
Ability to prioritize ACUs is removed.
Seraphim snipe mode speed increased and damage model adjusted for shorter reload alongside the main changes above.

1. Can non-sniper armies interact with snipers in a fair and enjoyable way?
2. Are the lower cost cost snipers counterable if they are rushed?
3. Do the snipers provide their user a fair reward for their resource and APM cost?
4. Are mixed armies more viable than pure sniper balls?
5. Are Percy/Brick counters for Sera/Aeon more accessible to lower rated players?

SACU Rebalance
Quantum gateway cost, buildtime, and adjacency buffed.
Base SACU cost reduced.
Energy and buildtime costs significatly reduced for most presets.
Costs further reduced for the engineer presets.
Energy upkeeps reduced.
SACU RAS upgrade increases death nuke damage.
UEF Bubble shield prerequisite of the personal shield is removed, and the shield is rebalanced with mobile shields in mind.
Added RAS for Seraphim SACU and removed increased base production + cost.
Overcharge upgrade cost massively reduced.
Shield moved to tele/OC arm (tele nerf and Rambo preset rework).
Rambo preset changed to Nano + Shield + Sensors/Gun Range (non-energy-based lategame single-target damage rambo).

1. Do SACUs come out and counter T3 or ACUs too early?
2. Is UEF bubble shield viable with its new stats considering the utility of the SACU underneath?
3. Is overcharge cost too low, especially after the first SACU dies and you can reuse the energy infrastructure?
4. Are any SACU presets snowballing too strongly into the lategame, overpowering even their T4 counter?
5. Are UEF Rambos a viable high-cost direct-fire option for UEF, who have no direct-fire T4?
6. Is the buildpower of combat SACU too high?
7. Are there any overwhelming/unreasonable faction differences with SACU gameplay?
8. Do engineer presets find use cases as a dense buildpower option?

The full patch notes can be found online or in the game lobby's patch notes button.

https://github.com/FAForever/fa/pull/7153

@magge I'm writing some changes to add back the delay before killing acus when someone disconnects.

Since disconnects cause acus to explode instantly and a disconnecting player disconnects from everyone else, their sim instantly explodes everyone and thinks the game is over so it tells the client to tell the server that their replay is a valid game. Really bad bug we haven't gotten around to fixing.

If there are more than 20 desync messages sent from any players: https://github.com/FAForever/server/blob/1b98cdae23350c92fab5203b4e6ac9bf52198e6a/server/games/game.py#L477
So I guess the unpredictability comes from the player count heavily affecting the result. An Astro game will send 6 desync messages every 5 seconds while a Dualgap game sends 12.

Yes there's a button to select the external factory unit of the carrier. You could UI mod stop/soft stop to cancel the factory's queue but then you would be come unable to cancel all orders for your carrier.

• Another aspect of it being a T4 is that it is the lowest energy ratio air T4, so a land player making it can do so very easily on top of the low buildtime and not requiring tech. It's also cheaper than a Megalith so you could arguably even build it instead of land units if your air player coordinates their micro and strategy with yours; that is to say, it is extremely accessible as an air T4.
• For your DPS/HP stat comparison you missed a few critical factors: it doesn't lose DPS as it loses HP and it has AoE DPS. If you use f(n/cost, h, d) = h * d * (n^2+n)/2 to calculate the "power" per mass of the units (yes I tested this works before with GC vs percy testing), Wailers only need to take 1.55x damage from AoE or Ripper needs to deal 1.55x damage with AoE for Ripper and 29k mass of Wailers to have equal final damage outputs. Considering that high-AoE, high unit count T2 mobile flak is the main counter to gunships, Ripper has a great role in countering that counter.
  • Veterancy is not as easy to compare as DPS/HP, but it's also a large factor in the continued survival of the unit. Wailers can repair for free, but Ripper is far better at getting veterancy due to being a single unit and T4s having much lower vet requirements than units. That veterancy boosts hp by 10% and also heals by 10% every vet. T4s also get much more regen per vet than lower tiers. Long term, you're guaranteed to lose Wailers but rippers are few in number so they're much easier to micro and keep alive, farming that vet heal/regen value and if you're tanking with healthy future rippers then the old one is just free 100% Ripper DPS.

@IndexLibrorum said:

The recommended direct counter is groundfiring the stacked subs with ships that have splash damage, such as destroyers.

Only exodus has enough splash out of all the destros to groundfire subs (1.4 instead of 1), and even so it's barely enough so I wouldn't recommend it. Battleships with 2 splash radius are the usual groundfire unit, although by that game stage it's a bit irrelevant.

G2 - https://replay.faforever.com/27084307

You had 1 t3 sera shield barely covering your scathis and it seemed like it was already damaged. You got dropped 2 ahwassas dealing 8k crash + 1.1k initial bomb damage = 18.2k dmg. Then there's 9900x2=19800 remaining delayed damage on the bombs.

Didn't watch czar replay.

I think you're more complaining that ahwassa has too high alpha strike than anything related to suiciding them. If you ever talked about bombing and then ctrl k for the wreck to fly under a shield and kill it to let the bomb deal 9900 dmg that would be about the crash damage.
For mitigating crash damage: shields absorb crash damage up to 20% of their max hp (ahwassa 8k -> 3800 when going through sera t3 shield, shield bar takes 4200 dmg), and you can spam shields around your target to reduce the chance of the remaining crash damage's splash reaching your target.
The reason ahwassa can have the high alpha strike is because it's bad against shield spam since it has a long reload and you need to micro well to fire on time with that reload. It's also a bomber and the most expensive air t4 but also lowest hp. So you should spam T3 shields in a line parallel to the bombing path to stop ahwassas easily.

You could hook OnProductionPaused and OnProductionActive of the MassFabricationUnit class for fabs, MassCollectionUnit for mex, or Unit class for both mex and fabs (might mess with RAS upgrades though).
There you can access in self.Brain a shared counter of units and a list of units. You would adjust the counter and list, calculate the correct mass production modifier using the counter, re-create the Buff Blueprint using the new value (make sure it uses "REPLACE" for stacking), and then go through the list of units applying the new buff.

If you want an example of a game with non-linearly non-exponentially scaling economy there's zero-k: https://zero-k.info/mediawiki/Cold_Takes/5_-_Making_Metal
Basically you have to use a square root on mass production to counteract the exponential scaling of reinvesting eco.

For FAF you should use the documentation in the intellisense docs:
https://github.com/FAForever/fa/blob/9868135d041a49e4e48afce1ae8dcbd8d5c814bc/engine/Core/Blueprints/UnitBlueprint.lua#L81
Setting up intellisense is easy, it's just VSCode and the FA Lua extension. It will show tooltips for documentation, add autocomplete, and help navigate the code. Further instructions for a dev environment are here: https://faforever.github.io/fa/development

There's cinematic camera mods but the main issue in replicating the effect imo is you can't get a blur effect.

RFA does show enemy range, make sure you're pressing the shift key while mousing over the enemy and have the correct settings enabled.

For anyone working on the UI side you can get the playable area with SessionGetScenarioInfo().PlayableRect. There is also a user sync event with the key NewPlayableArea when the sim updates the playable area.
Though this was added in March 2024 so you have to fall back to size[1] and size[2] for old replays.

It's a faf bug caused by uef engi drones being near a structure when it completes or dies

Yes, others will still see the pings.