The function of automatically building mass storage around the miners.

Turinturambar

no, since its allmost never used and cause actual damage if you accidentally click on your pgen

Jip

Understood

arma473

@tagada said in The function of automatically building mass storage around the miners.:

Imo, remove the assist feature for engies and just make each engie receive the proper set or commands, the only use case for it is the ACU so if you can add the assist functionality for it it would be useful, if it's not possible then so be it, you can assist manually with your ACU.

You also need the assist functionality when you select engineers from different factions. For example, if you have 4 cybran engineers and 2 uef engineers.

Tagada

Good point

Broker

@jip said in The function of automatically building mass storage around the miners.:

I think the consensus is clear, but what about these situations:

What if you have engineers of multiple factions (say, full share after gifting)?

What if you have your ACU in your selection?

What if you have Mantis, a Siren (!) or other units that can assist in your selection?

Tagada mentioned about the latter, what are other opinions?

edit: an initial version can be found here on Github

These tasks are very easy to solve.

If there are engineers and a commander in the group, then everyone receives the order to help the mine, but only engineers receive the order to build storage.
If in the group there are engineers from different sides (for example, 2 uef and 5 cybranes), then everyone receives the order to help the mine, but the storage is built by those who are more. those. cybran.

No problems and orders for assistance.

Always happy to help.

Jip

I'd like to note that there is an update to the pull request on Github, see also the description of the this message.

Jip

The following pull request on Github has been merged to address this feature:

https://github.com/FAForever/fa/pull/3484

For the full changelog so far, see this file:

https://github.com/FAForever/fa/pull/3481/files

The final behavior for capping structures as it is programmed right now:

Extractors

2 clicks + shift to mass storage an upgrading t1 extractor
2 clicks to mass storage a t2 / t3 extractor
3 clicks + shift to mass fab an upgrading a t2 extractor
3 clicks to mass fab cap a t3 extractor

Other structures

2 clicks to mass storage cap a t3 fabricator
2 clicks to t1 power gen cap a radar
2 clicks to t1 power gen cap a t2 artillery
2 clicks to wall cap a t1 point defense

Assisting

When all engineers are of the same faction, they can all build the same storage. No assisting happening.
When you have engineers of two factions, one must assist the other as they can't build the same storages.
When you have engineers of one faction and units that can't build the storage (kennel drones, ACU) then they must assist an engineer as they can't build the storages themselves.

I've increased the click count to allow you to just repair without capping, I always found that to be very annoying myself. The requirement of shift is now more consistent: if you want to apply the capping right after a structure finished upgrading then you need to hold shift.

Please test the feature on FAF Develop and report back in this topic. You can switch branch when hosting a game:

TheWeakie

Would it somehow be possible that if you have engineers from different factions selected you can start s building of one of their factions and the other would assist? It can be very annoying in teamgames with fullshare when you have t1 engies of different factions everywhere

Jip

@thewheelie That is how it is implemented at the moment, I quote:
When you have engineers of two factions, one must assist the other as they can't build the same storages.

Unless you tried and it didn't work

TheWeakie

No thats not what i mean. I mean being able to build any building, not just storages

Jip

I don't understand how that correlates to your previous message. Can you write down a specific example scenario?

FtXCommando

Have 3 t1 engies, aeon/cyb/uef respectively.

Select them, cycle to build a factory, nothing pops up

Need to select a single engie, tell it to build then select the other engies and assist the fac/engie

CheeseBerry

@jip Just tried it out on faf develop and I think the downsides of multiple clicks for mex storage largely outweigh the upsides.

Having to click on my mexes multiple times to perform a single, very common action, doesn't feel like a good change at all.
Even two clicks are too many imo.

If repairing a t2 mex without putting mass storages around it is a frequent occurrence, why not just use the repair command? Having it bound is a good idea anyway, as it is very useful for engie micro.

For capping arty, radar and t1 pd I'd probably prefer one click as well, as we already have a dedicated repair button, but we don't have a dedicated cap button.
Especially for the radar two clicks don't make much sense, as I have never seen a radar that wasn't either at full health or dead.

Accidentally surrounding an arty with pgens or a pd with walls when you actually wanted to repair could be quite bad though, so I I'm less sure about wanting only 1 click to cap those than I am for the mexes.

That being said I love the idea of being able to cap arties and radars similar to mexes!
Similarly, mass fab capping a t3 extractor without needing a template is a great addition as well.

I have two possible suggestions for capping:

"Combat buildings" need 2 clicks to cap, while all others need 1 click, as they do now.
All capping is only done by using shift, and normal assistance/repair functionality happens when not using shift + click

Also, can we add the ability to queue capping for structures that are currently being build from scratch?
That'd be useful in quite a few situations, e.g. after getting dropped to rebuild mexes, or when building a couple capped t2 arties to stall a navy loss.

Spelled out this would result in the following:

1st Suggestion

"Combat buildings" need 2 clicks to cap, while all others need 1 click.

Extractors

1 click + shift to mass storage an upgrading extractor or a t2/t3 extractor being build from scratch
1 click to mass storage a finished t2/t3 extractor
2 clicks + shift to mass fab cap an upgrading t2 extractor or a t3 extractor being build from scratch
2 clicks to mass fab cap a finished t3 extractor (need 2 clicks to avoid overlap with the storage cap command)

Other non-combat structures

1 click + shift to mass storage cap a t3 fab in construction
1 click to mass storage cap a finished t3 fab
1 click + shift to t1 pgen cap an upgrading radar or a t2/t3 radar being built from scratch
1 click to t1 pgen cap a finished t2/t3 radar

Combat Structures

2 clicks + shift to wall cap a t1 point defense in construction
2 clicks to wall cap a t1 point defense
2 clicks + shift to t1 pgen cap a t2 artillery in construction
2 clicks to t1 pgen cap a finished t2 artillery

2nd Suggestion

All capping is only done by using shift, and normal assistance/repair functionality happens when not using shift + click

Extractors

1 click + shift to mass storage an upgrading extractor, a t2/t3 extractor being build from scratch or a finished t2/t3 extractor
2 clicks + shift to mass fab cap an upgrading t2 extractor, a t3 extractor being build from scratch or a finished t3 extractor

Other non-combat structures

1 click + shift to mass storage cap a t3 fab in construction or a finished one
1 click + shift to t1 pgen cap an upgrading radar, a t2/t3 radar being built from scratch or a finished t2/t3 radar

Combat Structures

1 click + shift to wall cap a t1 point defense in construction or a finished one
1 click + shift to t1 pgen cap a t2 artillery in construction or a finished one

Jip

@CheeseBerry These are good suggestions - I'll make sure shift is required (as otherwise it overrides the queue anyhow, as WhenDayBreaks PMed me) and reduce the capping of non-volatile things (e.g., storages, pgens, walls) to 1 but keep the capping for volatile things (fabricators) at 3. Unless people disagree. The reason is simple: to counter lag

Also, can we add the ability to queue capping for structures that are currently being build from scratch?
I already tried this and sadly the answer appears to be no - the command issued doesn't provide the blueprint information that I need to determine whether it can be capped or not.

I think it would be possible if we'd determine the capping behavior on the sim side instead of on the UI side - that would require quite some refactoring and if I'd do that, it would be for the patch in January.

@FtXCommando I see - not sure if that is possible (yet). It is at least more than this particular feature (capping) is about. Could you make an issue on Github?

One problem I have with just clicking once is that you can no longer queue repair orders without accidentally capping things (when holding shift).

Regardless - here is a pull request on Github with the adjustments. I am eager to hear other opinions.

CheeseBerry

@jip If using 3, instead of 2, clicks to mass fab ring a t3 mex decreases lag (I assume because then you don't give nonsense mass fab construction orders to t1 engies?) then I'm all for it.

It's a rare enough occurrence that the difference between 3 or 2 clicks probably doesn't matter all that much.

CheeseBerry

@jip Also, maybe a dumb question, but is ringing t3 mass fabs with storages even efficient? Cause with some very basic back of the envelope math it seems like building another t3 fab is just straight up better.

From testing:

Mass Cost of a t3 fab: 4000 mass (m)
Energy upkeep of a t3 fab: 1500 energy upkeep (eu)
Mass generation of a un-caped t3 fab: 16 mass per second (mps)
mass cost of a mass storage: 200 m
energy upkeep of a mass storage: 0 eu (obviously)
Nr. of storages required for full adjacency: 12
Mass generation of a caped t3 fab: 21.5 mps

A simple model for energy upkeep to mass cost is to use the conversion: 2500 eu = 3240 m, because that's the price and production of a t3 pgen.
Hence 1 e upkeep = 1.3m.

(Note that this is the highest cost energy can conceivably have in the t3 stage, i.e. the best possible case for capping with storages.

In a real game, energy is probably a bit cheaper in the lategame, due to pgen adjacency and stuff like RAS, SACUs, you pausing your (air-)facs and teammate overflow often giving you more energy than you need 'by accident')

Putting the above together:

Efficiency of an un-caped T3 fab:
16mps / (4000m + 1500eu)
= 16mps / (4000m + 1950m)
= 2.7e-3 mps/m
Efficiency of a caped T3 fab:
21.5mps / (4000m + 1500eu + 12*(200m + 0eu))
= 21.5mps / (4000m + 1950m + 2400m)
= 2.6e-3 mps/m

Result:
Even in the best possible case, capping a t3 mass fab is slightly less efficient than just building a new one.
If you value energy less than what I assumed here, and imo you probably should, the case for capping gets even worse.

If my assumptions and calculations are correct, I think we shouldn't add a t3 mass fab cap command to the game add all. It would just incentivize bad play.

Jip

@CheeseBerry It doesn't decrease lag, but it prevents lag from accidentally causing you to queue fabricators while you can not sustain them (yet).

I think the typical scenario for a t3 mass fab is not to have it surrounded entirely by storages, but instead you cap those that are at the 'edges' of your grid for a small bonus.

CheeseBerry

@jip Well, the result doesn't change, regardless of you using 1 or 12 storages next to the t3 fab.
Storages are, if the above is correct, only efficient if they touch 2 or more t3 fabs

CheeseBerry

I should probably note that I don't have a strong opinion about including or not including a t3 mass fab command in the game.
The efficiency difference is small, so even if one or the other is 'better', it's not a huge deal either way.

Also both of the above options are less efficient than just building a t2 mass fab (3.0e-3 mps/m, at worst), which is also much more flexible imo. As such, I can't remember the last time I even built a t3 mass fab, hence the command probably wont affect my gameplay at all.

My only concern is that if we add a command like that in the game, we basically say that using it is "correct" as otherwise, why would it be there?

Jip

My only concern is that if we add a command like that in the game, we basically say that using it is "correct" as otherwise, why would it be there?
That is why not all suggestions were included. I am personally not of the opinion that everything has to be optimal - it just has to be decent. I'm eager to hear other opinions!