Posts made by Pundemonium

The Tl; Dr;

.Net wasm development has been maturing nicely over the past couple of years so I thought i'd take it upon myself to get my creeps able to C#.

After an initial PoC with mono I have the begginings of an initial swing at full C#ness up on github here here (not much to see, watch this space) built on top of blazor.

Initial Findings

So far I have experimented with three approaches, none of which are particularly screeps-ready.

Mono Wasm

• Build environment can be be finnicky to set up
• Build output (.net libraries + mono 'runtime') is above 5mb
• Allows use of the .net dlls (broad strokes here) practically as is

My first working versions have been with the base mono wasm implementation.I expect that as time progresses mono will in the short term create the most usable 'core' for screeps as they continue to work on mixed mode (AOT + IL side by side) compilation.

Blazor

• Built on top of mono
• Easy to set up & deploy
• Makes major assumptions about deployment environment (hint: it isn't screeps!).
• Requires custom harness that ignores almost everything that blazor puts on top of mono.
• Continued blazor development keeps causing issues with the harness.

Why use blazor if i'm going to ignore most of it? It's a (mostly) stable target that tightly integrates with the core visual studio experience with very little extra footwork needed. I messed with mono/corert/blazor simultaneously but blazor is simply the first past the gate to work on the public (ptr) server.

I do however expect the blazor team to keep optimizing for browser deployment which will only make its use harder for screeps. For example I've already discovered two removed methods that I relied on removed as they push towards release (Removed for good reasons, mind) .

CoreRt Wasm

• Aims to be AOT optimized - theoretically better size & speed
• Furthest from feature complete in regards to wasm
• No inbuilt javascript interop as of writing
• I found it easier to set up than the mono wasm stuff, but it has more steps
• Currently less overall documentation for wasm builds

Long term this is probably the most promising for .net wasm in general not just for screeps. Their work is mindblowing* but utltimately I just ended up happier for something easier to use and build against (blazor)

*Not to say that mono/blazor isn't fantastic, i'm just most familiar with the inner workings of corert... not that that is saying much.

@kyralee Then NPC orders can be re-introduced if/when such a thing becomes a problem.

--My 2c of a lot of the talk i've been seeing on this:

I like the idea to remove NPC buy orders for minerals. There's certainly enough player demand for T1 minerals it seems. If NPCs only buy these new materials player demand will only go up for T1 minerals.

In particular to the concerns i've seen from players on Shard 3: I'm currently in the process of setting up an outpost on the shard to access the market and I know i'm not the only one. Regardless, my code can't tell the difference between an NPC order and a player one without looking at the room and it should be the same with your code.

The concerns i've seen for inflation are not unfounded but there's enough knobs on this system to tune it so that doesn't happen. Ideally this should be self balancing.

Finally processing these materials are far simpler than labs, and if i'm understanding it correctly even powerless players have access to T0. I just don't see how newer players are excluded from this at all.

@davaned Walls out seems to be a becoming more viable approach in order to protect the sources; the most dangerous unit (power creeps) can just duck into a room for a single tick anyway.

Ultimately at the highest level of attack/defence I expect more complicated strategies such as moving towers and multi-room spawning of defenders (for flanking) to be practically necessary.

But I hardly take part in combat so what do I know.

The most basic (and seemingly common) way to achieve it is to have one creep (typically a healer) constantly moveTo a leader creep (typically a dismantler or attacker).

The most complicated way that I currently have in my code is a 3x3 grid with pre-coded movement patterns. Eg. Turn, swap rank etc.

@davaned Power efficiency is a tricky thing; I can't make use of all powers all the time even with an infinite amount of ops.

Specifically:

• Operate_Tower requires a target (or need for inefficient repair)
• Operate_Spawn requires filling infrastructure (and a demand for more creeps).
• Fortify is only of use in a siege
• Disrupt_(X) ditto
• Shield is more generically useful but I cannot see it being used whenever it is off cooldown.
• Others, depending on economic situation and code maturity

I feel that 'sometimes' powers like these are great powers to level to keep a creep 'ops neutral'. Even a creep that only generates ops most of the time can gain value from being level 25 thanks to powers like the above.

Selling ops would be a good way for newer players to get credits too, as apposed to simply selling that power. So I don't expect ops to be terribly expensive. (If an op is worth more than the minerals it can generate i'll damn well sell them)

Finally while I can appreciate that YMMV (not everyone likes spreadsheets) I like the added challenge of balancing ops use & generation with the powers I want to use.

There is simply no need to send energy across the map. Buy local or just set up an ORDER_BUY.

If you need to estimate credit/energy overhead costs to maximise profits you can use the energy sending cost and multiply it by the cost it would take you to buy more energy (Highest buy order + 0.001 in my case).

The market is fine IMHO; just one more programming challenge to overcome.

@smokeman said in Ramparts should be weaker than walls and they both need a nerf:

Hence, my base position that the max settings are 10 times too high.

So some people are unwilling to make the investment is the reason we should set the cap lower?

Wherever you set your cap (or lack thereof) for walls is all about your priorities. I'm starting to be more energy hungry and leaning on my active defensive more so I simply don't want walls that high and have the luxury of thinking about where I want to set them.

The argument that seems to be being made is that rooms are too hard to take, with walls maxhits being the most visible cause due to the players you are facing. Rooms being hard to take is currently being addressed in other ways (controllers got the dev spotlight recently).

Another real problem I feel (due to the opponents I tend to face and where my offensive code is at) with taking a room is that as long as you can get energy into the walls it's an uphill battle on top of all the other mechanics that favour defence, wall maxhits included.

Nukes help since you can wipe out a room's creeps but as it stands now to attack a room you effectively attack however many rooms are supplying resources to that room, all the way up to the alliance level. Perhaps even up to the shard level thanks to the market and the deep credit reserves some players have.

But that is me just shining light on yet another mechanic that works in favor of defence. Every player you talk to is probably going to have different opinions about what makes defence so strong.

So a few things:

I've seen higher level players opt to not upgrade past 30mil because of the heavy investment 300mil takes(1). That is exactly what is going on; up front cost for a payoff down the line. The thing is they have to upgrade every tile to 300 mil. An attacker only has to punch one hole.

Is that balanced? I don't know but it's important to keep in mind.

Nurfing walls/ramparts will do nothing for the imbalance between new players and old since there is already very little difference between two players in regards to walls (sans time upgrading) when taken on their own. Active defences and tower placement are king.

It will make a room take less time/investment to crack for sure but i'm not sure how much easier it will be to crack any given room unless you render walls/ramparts worthlessly weak.

Nerfing can also make it harder on newer players. Anything less than what will go down in a few creep lifetimes puts a massive amount of pressure on active defences to pick up the slack.

Hell I would argue that even if they where logged on at the start of an attack they couldn't write the code fast enough against a coordinated(2) attack. This is of course assuming they have the time to get the walls upgraded, newbies and all.

Put it at anything more than a few creep lifetimes and the players with well written defences won't even notice the difference against all but the most determined of attacks. And that's if they even bothered to build their defences all the way up anyway.

Changing upkeep amounts is the only workable suggestion i've seen so far. But strictly speaking a lot of defensive layouts have a lot of 'fat' that could be trimmed(3) and I feel there are unexplored strategies that could prove even more optimal than present(4).

We could just end up back where we where in the attack/defence with a new, dare I say, scapegoat. The problem is systemic and while walls are a part of the problem they are not the problem.

Use the extra GCL you have over them (due their massive standing investment) to level a couple more nukes at their room or something.




(1)Starting to lean towards this camp myself. But interestingly my code starts to break down for varying reasons if I where to try to use it.
(2)I'm talking your typical boosted squads. No matter how you slice it defensive code takes iterating and quite likely, boosts both take time. Many players simply do not work on defence until the enemy is at their gate.
(3)Solid lines of ramparts work well for some things but are not the most efficient design for energy use IMHO.
(4)Eg. New bunker designs, moving towers for walls out, exploiting unreleased mechanics etc. Actually I'm convinced that some combination of walls out & walls in simultaneously is the most optimal defensive (if not economical) solution.

Which shard is affected?
All shards

What happened?
Related: https://screeps.com/forum/topic/2285/after-intershard-travelling-placed-on-a-portal

A creep was placed on an active portal to shard 3 after moving through a shard1->2 portal.

What should have happened?
Creep should be placed in open space.

How can we reproduce this?

Have a large number creeps navigate to another shard. This other shard needs to have two close portals.

It is possible to see me attempting this at https://screeps.com/a/#!/history/shard2/E10N30?t=10111162

You can see my creeps making it to shard 3 here: https://screeps.com/a/#!/history/shard3/E10N40?t=20560

@shibdib The cool-down doesn't effect boosting. I take advantage of that in my own unboosting.

Ultimately it's 'free' minerals if you can code around the limitations. In particular most lab setups have two labs that never act as a target for reactions that can be used as well.

Does such a dis/advantage need smoothing out?

I am strongly against basing anything around holding onto a room long term for the simple fact that it can be quite crippling to hold onto a room that you simply don't need.

I'm also not convinced a survival leader-board is all that informative and feel it might even be counterproductive. New shard opened up? Boxed in by the next dissi? Too bad you have a 3 year survival high score!

@tigga So that a player that is still working out how to even farm power can be kicked over without hope of counter by a player who has had a chance to figure out how to get at creeps that can literally cripple (or in extreme cases, shut down) a room?

This has to be opt in.

@wtfrank said in Warfare Ranking:

I don't understand how this capped at 0 thing would work to avoid abuse. If everyone starts at 0, and no-one can be sent below 0,

As an example: Everyone starts with (assuming killing a rooms is worth at most 1000 points) GCL*500 + 10% of their ranking last month. Done; Players not actively participating are only worth a fraction of their empire (but can't opt out entirely) while those who participate are on the hit list for the others that do participate.

finding someone who doesn't care about the ranking

I'm sure they will care about you coming and kicking over their rooms though.

or finding zombies to mess with.

You can only kill a zombie once.

"damage to structures and creeps in controlled rooms"

does not always correlate damage done to an empire (just energy spent) while RCL damage is directly correlated to wasted potential due to less spawns, less labs and yes, energy spent on that RCL.

And it would be simple to track and well correlated with military combat.

Do we want to track throwing stones ineffectually at one another or do we want to track murder? Anybody can throw a few creeps at a room. I want a warfare ranking to measure as close as possible the damage done to another players empire. Not energy, not creeps, not structures empire

You can't truly hurt an empire by killing a few creeps. You hurt an empire by razing a room to the ground while optionally signing their controller with profanities.

@atavus said in Warfare Ranking:

4. To avoid abuse, the awarded points will be subtracted from the defender (capped at 0)

This also rewards the defender in a round about way since they don't lose points. While rewarding defence is a good idea I feel that other ways to reward defence are going to be problematic for one reason or another. I especially like it because the highest ranked players will effectively have a bounty on their heads while also making the ranking dramatically more dynamic (want to be on top as #2? attack #1!)

In any case subtracting points seems like a good means to prevent abuse / weaker players being hunted. It does make me think that this ranking shouldn't be entirely monthly either but have a partial carry over from the previous month.

That said there is a potential for abuse for this scheme: unclaiming a controller when the walls fall so as to not lose points. The question is will it be worth it to give up a (for example) RCL 8 room (and the energy cost to upgrade it) that easily?

Finally a key point for why I lean towards a more complicated 'key structures destroyed' system along with 'driven to rcl 0' rather than anything based on attackController() is because it
a) often isn't practical to get a short lived claim creep to where it needs to be.
b) encourages/measures making it as hard to stand a room up again as possible; this is war not a slap fight.

Room to RCL0 - Which player got the killing flow on the storage etc.? Or even the most damage to player structures? (This also means 'ownership' of a kill could change hands if a a second player worked to keep a room flattened)

Controlling reserved rooms (without mining) isn't particularly related to economy at all unless you need to hold them militarily. But honestly holding an empty room isn't worth the effort or the CPU.

Reservation counts also works majorly against alliances who's members tend to be shoulder-to-shoulder a lot of the time (relative to the amount a player has reach to reserve/sign) and doesn't account for war-games at all. On the other hand it also majorly works for alliances that can afford to bully and diplomance players out for the space too.

Imagine all of YP muscling everyone off of shard1 just to reserve (but not make use of) all of that extra space. Its an awful vision, and not just because i'm on shard1.

The nail in the coffin for the reservation idea for me is that it doesn't even measure military strength in any meaningful capacity outside of the ability to pick your targets carefully. Respawn in an area with a lot of new players and push them all out. Suddenly you're #1 on the scoreboard for the foreseeable future with little extra military investment.

Maybe that's too extreme, maybe all players end up with the same amount of reservations per claimed room... but that makes it even worse as a counter when we have GCL & rooms claimed statistics already.

You wouldn't have to create a formal concept alliances; it could be as simple as two players (The 'owner' and the player set to receive the buff) touching the flag and calling the same method.

The point is that if an alliance gets a flag and only one player can benefit from a flag at a time there is going to be some serious problems arising from a possible perceived shared ownership of that flag.

My problem is that number of reserved rooms is not direct measure of military might and does not directly encourage or measure player conflict; it is merely a measure of expansion which is already inherent to other metrics. I'm also not sure if convincing people to have larger footprints (without making use of the rooms) is such a good idea.

Or that it's even worth the 0.2 cpu across many, many rooms for players above GCL 30 who have massive CPU constraints already. Mind I probably shouldn't speak for them. But even for me i'd much rather dump that CPU into another couple of remotes and do that much better at a more useful GCL/Power ranking.

There are a lot of mechanics that have been proposed elsewhere that could perform the function of encouraging player conflict and many could also be used to measure a warfare rank. eg. duel claimed controllers, flags, etc..

It's also possible to simplify the rules I proposed to 'number of rooms driven to RCL 0' ('Rooms destroyed') . Everything else is to deal with problems that for all I know might not actually be problems.

Room kill attributation can be done via attackController() or, better yet, who did the most damage to nonrampart player structures within the room. Yes, it could be complicated if multiple players attack the same room... but is it really a war at that point? Does that even happen that often? (Most alliances will attack multiple rooms rather than all piling into the same one)

In any case I agree attributation is quite a complicated issue.

I think any method that we could come up with could be farmable in some fashion. Creep parts killed, boosts wasted etc. The point is to make it not worth, or at the very lest inefficient to farm.

Without resorting to mechanics that are not currently in the game I therefore propose something along the lines of 'RCL damage'.

Every room taken held, and driven to RCL 0 while also having their storage, terminal and spawns destroyed counts as a number of points relative to their starting RCL. Capping the same room in a month from the same player earns fractional points. Forcing a respawn is worth a lot of (a pool of) points given to the room-takers of the respawned player.

Finally, lower GCL players are worth dramatically less points than larger players to discourage hunting of noobs...

Actually it'd be quite possible to give points based on relative warfare rank if you wanted to get fancy. But that would have a lot of implications in that warmongers would be inherently worth more points and peaceful turtles less.

Lesser restrictions based on the above could be used; but make it more abusable too.

I say this because just counting claimed/reserved rooms doesn't seem to particularly encourage a change in player behavior.

Now for the thinking behind some of the specific rules:
-Terminal and storage in particular allow for rapidly getting a room back to rcl 8 if filled.
-Spawns are mostly to make it extra inconvenient to get a room back up and running.
-Pool of points on respawn make it less enticing to gang up on a single player in an alliance setting.
-RCL 0 is because you want to measure real, lasting damage rather than an RCL 8->7 inconvenience.
-Less points on retaking has multiple reasons, but if you have to retake it I argue that you havn't done enough damage to your opponent the first time to earn the points the second time.