Early game modular armor rebalance

Place to discuss the game balance, recipes, health, enemies mining etc.
User avatar
bobingabout
Smart Inserter
Smart Inserter
Posts: 7351
Joined: Fri May 09, 2014 1:01 pm
Contact:

Re: Early game modular armor rebalance

Post by bobingabout »

Yeah, my opinion on the personal roboport is mixed.

The ONLY reason why I feel the personal roboport should be an early game items is because a logistics network is locked behind yellow science.
Without a logistics network, it makes making a roboport network kind of useless.
but construction robots are on green science, which means you basically put a roboport down where you need something built, put a chest next to it, and all the items you need for that construction operation in the chest.
But a personal roboport exists, you might as well just use that instead while carrying the needed components, and it performs the same task.

now if they put the Logistics network (we're talking requester chests here specifically) on the logistic robots technology (I did this in my mod) then Logistic robots instantly become useful, so setting up a roboport network becomes something you would do, so your construction materials are already in the network, might as well expand it to cover your construction site, now it's fully automated.
you don't need a personal roboport anymore unless you're going out to build a remote base.

To note: I consider the player logistic requesting system a luxury.



summery:
The way the game is now:
Construction robots are green science
Player logistics are green science
Portable construction roboport is green/blue science.
Logistic networks are yellow science.

The way I think it should be:
Construction robots are green science
Logistic networks are green science
Player Logistics are Blue science
Portable construction robots are blue/purple science.

Keep in mind this is my personal opinion, and I look at it from the perspective that the more something moves (like the player), the harder it should be to interface with it, so stationary objects like provider to requester chests, with known fixed positions should be the lowest level technology possible.

But since the devs insist on locking the logistics network behind yellow science... to compensate for the otherwise uselessness of logistic robots, the pointlessness of even making a roboport network without this... everything else should be on the same tech level (green) as the roboport itself, so you should be able to make a full player based (Logistic request/trash, and full portable roboport setup) on the same tier as the robot unlocks... green science. Which means using green science alone, you need to be able to fully kit out your modular armor to support these construction roboport systems.
Creator of Bob's mods. Expanding your gameplay since version 0.9.8.
I also have a Patreon.

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

the pointlessness of even making a roboport network without this
But a roboport is extremely useful for using blueprints and repairing walls right away. Even speed runners are making a detour for roboports just to get that extra build power going.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

I think robots should be available early and should just be nerfed. They should take time to do each thing. When they are used to mine trees and rocks, or collect entities, they should take as long to mine it as a player takes without a tool. When they mine rocks, they shouldn't be able to carry more than 1 stone back to the port. Logistic bots should be deployed to gather more stone off the ground. When they place entities, there should be a small amount of time it takes to complete the placement process.

They should have a collision box and not be able to crowd together in huge swarms of bots. There should be a limit to how many bots can fit into an area.

With a large number of bots, this will still greatly accelerate your work. But with only a few robots while it might not be much faster, it's still useful to automate things.

Also something I want to tackle: bots should not immediately try to repair other bots. When a bot gets damaged, it should immediately pause what it was doing and attempt to flee the source of damage, and only after a certain amount of time has passed should another bot be deployed to repair it. This could be a setting on the roboport.

User avatar
Ranakastrasz
Smart Inserter
Smart Inserter
Posts: 2124
Joined: Thu Jun 12, 2014 3:05 am
Contact:

Re: Early game modular armor rebalance

Post by Ranakastrasz »

Early available, yes. The bluebuild and similar mods are too useful, and early game robots is just reasonable.
Collision boxes. No. Woulds make.sense, but too expensive programming wise to run.
One item limit. Makes sense. One as cap for construction robots. Maybe make the rest drop on ground and be marked for decon. Makes sense, but unessesaey gameplay wise. Although disassebling chests is already similar, removing contents and all.

Robot damage ai. Yes please. Could tie into combat regen mechanic. Cant repair unit until x time after combat.

I think robots should have a changed and uncharged item. Roboports and personals spend energy to convert them. They can launch only if they are charged. Once they land, their remaining energy is added to the roboport. (And is used to instantly recharge them, minus the difference)

Robots dock instantly, but still spend time recharging outside if they have lots of tasks. But mostly they get primed inside roboports.

Also roboports may need to have one slot limited to charged and one to discharged to avoid deadlocks. (7 stacks of uncharged? Cant charge because no room for the +1 robot)... Unless it charges a whole stack at once anyway.
My Mods:
Modular Armor Revamp - V16
Large Chests - V16
Agent Orange - V16
Flare - V16
Easy Refineries - V16

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

I don't see much point to bots repairing bots at all. Not only does it create a snowballing disaster of sending even more bots into danger, there isn't even a point to it. Bots already get a free heal inside the roboport.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

Ranakastrasz wrote:Collision boxes. No. Woulds make.sense, but too expensive programming wise to run.
Biters have collision boxes and huge swarms of them seem to operate fine. Total Annihilation operated similar or greater numbers of collision boxes at once (hundreds, easily), way back on 1998 computers and it ran fine. That game took collision boxes off of aerial units but I figure by now it should be fine, also TA had terrain height which could have also been another consideration, but Factorio does not use terrain height. If in adding collision boxes it forces people to use fewer bots, that will probably help their FPS.

bobucles wrote:I don't see much point to bots repairing bots at all. Not only does it create a snowballing disaster of sending even more bots into danger, there isn't even a point to it. Bots already get a free heal inside the roboport.
I had no idea they do that, maybe because I've never had one dock while damaged. No need to have them repair each other when they can get a free repair upon docking.


I'd like to see bots be larger and more expensive, and have collision boxes. Construction bots should operate at the speed of the player using iron axe, but they should additionally take time to place objects. Logistic bots would carry several items at once much as they do now, but the stack size might need a boost--I don't even know how much they carry. They should be slower than inserters and belts, because they are easier. Then there can be special personal bots which operate from the personal roboport and share your shields--so when they take a hit the hit goes to your shields. If your shields are out it hits their HP. These personal bots could be modified with suit modules which affect their attributes:
* a module for increasing their build/gather time and enabling them to have full construction bot capabilities
* a module for enabling them to carry more stuff and enabling them to act as logistic bots
* a module to increase their speed/range
* a module to increase their resistances
* a module to enable them to carry ammo and shoot at enemies

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

Hey now, let's not rehash the whole bot debate. That one has already be beaten to death and back. :lol: Stick to armor and armor modules.

Early game modular armor and modules do have some balance issues that would benefit from being smoothed out.

- The absolute biggest balance snafu is with pocket solar. PSPs are too weak and ridiculously overpriced. The balance is so bad that players rush for pocket fusion and actually save themselves a lot of trouble by doing so. They need the most dramatic change. My preference is to give 25-50% more energy per tile, and to upgrade the art to a prettier 2x2 design. Keep the recipe unchanged to bring the obscene cost down to reasonable levels. So:

Code: Select all

PSP
2x2 size
5 solar, 5 steel, 1 red circuit
50 - 60kW energy
- A mid tier induction style charger would be extremely handy and help bridge the tech gap between solar vs. pocket fusion armor. Induction chargers are a strong defensive type of buff that can have a positive effect for PvP balance. Something like this would be effective:

Code: Select all

Induction charger
Lots of copper, red circuits, steel/iron
2x2 size
60kW-100kW energy (when in power)
- Personal exosuits available too early? They are a high energy consumer that demands pocket fusion to really shine. Tech probably belongs at a later tier, they are available deceptively early. Exosuits aren't really needed early-mid game because they ruin the point of using stone paths and personal vehicles/trains to gofast.

- Personal roboports use too much energy? At least early on it is difficult to use the mk1 without cheesing or rushing pocket fusion. Robo internal storage is probably too large, switching them out will annihilate your battery supply. Energy consumption can be vastly reduced by shrinking the coverage area and reducing the coverage scaling, but buffing solar and giving a mid game energy boost will also help a lot.

- Grid size is 5x5 and 7x7? So many modules are 2x2. This creates pointlessly awkward layouts with lots of dead space (PSPs are dead space) to function.

- Armor scaling too heavily set towards mk2. It's easy to skip mk1 for many reasons, but the biggest reason is that mk2 is so overwhelmingly superior. The mk1 armor has room for 9 modules, while the mk2 has room for 25. That's nearly triple! Mk1 armor can actually be upgraded by shrinking its grid size to a 6x8, which is room for 12 modules. This would make the mk1 much more useful and not so easy to skip.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

bobucles wrote:- Grid size is 5x5 and 7x7? So many modules are 2x2. This creates pointlessly awkward layouts with lots of dead space (PSPs are dead space) to function.

- Armor scaling too heavily set towards mk2. It's easy to skip mk1 for many reasons, but the biggest reason is that mk2 is so overwhelmingly superior. The mk1 armor has room for 9 modules, while the mk2 has room for 25. That's nearly triple! Mk1 armor can actually be upgraded by shrinking its grid size to a 6x8, which is room for 12 modules. This would make the mk1 much more useful and not so easy to skip.
Check out Ranakastrasz's mod Modular Armor. It solves a lot of power production problems by giving more options, alters the size of several modules so that you can make interesting fits--with multiple types of small modules for that dead space, and I'm guessing has reduced the power capacity of the personal roboport because I haven't been having power problems with it. The mod also adds 2 more ranks of power armor, and I suggested giving the Mk 4 a 10x10 grid and making the ranks only increase in grid size by 1x1 per rank. So the modular armor would be 5x5 (tiny, cause it's starter), then power Mk 1 is 7x7, Mk 2 is 8x8, Mk 3 is 9x9. The increase sounds small but isn't, as I'm sure you're aware.

5x5=25
7x7=49 (+24 // +96%)
8x8=64 (+15 // +31%)
9x9=81 (+17 // +27%)
10x10=100 (+19 // +23%)

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

So I've been playing with personal roboport ranges to see how they work. It looks like the scaling is a simple formula of counting the total tiles covered and adding them together. The Mk1 is 30x30 covering 900 tiles while the Mk2 is 40x40, covering 1600 tiles. Of course we don't care about the number of tiles covered but rather how far our bots have to run out and drain their batteries. The radius is the most important factor here, with a shorter radius giving superior bot speed and reduced charging. Without further ado:

Code: Select all

radius 10____15__________20
20(400)    30 (900)    40 (1600)
28(784)    42          57
35         52          69
40         60          80
45         67          89
49         73          98
53         79          106
57         85          113
60         90          120
63         95          126
66         99          133
69         104         139
72         108         144
75         112         150
77 (5929)  116(13456)  155 (24025)
You may notice I put a size 10 in there. I've been toying around with the idea of reducing the personal roboport radius by 5(both tiers). This reduces the robot range by 33% and 25% respectively. Less range means cheaper trips, which means the personal roboport uses less energy. As you can see the coverage area is still pretty impressive, and any normal blueprint can fit inside or require a short walk to reach. For reference the structure roboport has a construction range of 110x110.

I have very good reason to believe that shrinking the personal robo range will actually be a BUFF to the roboport, since you will be able to build things faster and with less strain on your suit energy.

User avatar
Deadly-Bagel
Smart Inserter
Smart Inserter
Posts: 1498
Joined: Wed Jul 13, 2016 10:12 am
Contact:

Re: Early game modular armor rebalance

Post by Deadly-Bagel »

bobingabout wrote:The devs have stated (in private conversation, but fuck it, it's not a secret) that Electric trains that draw power from a grid when the train is on a tile with a grid won't happen because the train would constantly need to check "Am I powered?" every time it moved (which is every tick the train isn't stopped at a station or signal, so might as well be every tick), which is an expensive operation... So, the player doing that, because he has some grid equipment installed is... unlikely. However, some kind of "Dock", so you have to go and stand on a specific tile of a new type of entity MIGHT be acceptable.
Have it only work when you're standing still. One check every time the player stops moving should be easy and very lightweight, even if a player manages 10 directional presses per second (not normal use-case) that's 1 check every 6 ticks. As a bonus it adds to the inconvenience of relying on your networked power to push you towards personal power generation, and batteries will become much more useful.
thereaverofdarkness wrote:
bobucles wrote:If anything the biggest energy drain on the personal roboport is the range scaling. The robo range gets too large too quickly and your batteries get MURDERED by the constant 50-100 tile journeys that you have no way of avoiding. Shrink the range scaling by half and roboport energy consumption will go down with it.
Or make them stop purposely trying to go as far and as random as possible, and have them prefer to hit up things closer to you.
I agree that would be really awesome and way more time and power efficient, unfortunately it's not going to happen due to the way ghosts are processed - basically there's a table of them (order is irrelevant to location) that iterates X items per tick to check if there is an available construction bot and item to fulfil it. This way only a few items are being processed per tick. To rewrite it the more useful way it would be extremely inefficient - either every single ghost on the map would need iterating in a single tick, or every single Roboport would need to do a check for covered ghosts (and good luck with overlapping ranges).
Money might be the root of all evil, but ignorance is the heart.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

Deadly-Bagel wrote:To rewrite it the more useful way it would be extremely inefficient - either every single ghost on the map would need iterating in a single tick, or every single Roboport would need to do a check for covered ghosts (and good luck with overlapping ranges).
It's never impossible. There's always a better way.

In this case I immediately see a possibility: personal roboports would perform a separate tick operation in which when you enter a chunk, place a blueprint in your own chunk, or remove a ghost from your chunk, your personal roboports (acting as a single entity) would update a list of ghosts in that chunk, and then when the personal roboport makes the choice to send out a worker, it runs a range check out to the range of the roboport build area or to the first ghost it encounters for which you have the materials to build it, whichever happens first. When a ghost is selected, it sends a worker to that one.

Problem: you can't build anything in adjacent chunks
Possible solution: load the ghost list from 1 or 3 adjacent chunks when you are standing close enough to them for your operational area to overlap into them.

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

Making the roboport range smaller does the trick just fine. You don't NEED a 150x150 personal robo range, and it's completely detrimental to their use anyway.
Have it only work when you're standing still.
That's pretty silly. There's absolutely no harm in allowing the player to connect to the power grid on the move. We're talking about a number of entities that can be counted on your fingers. There is no CPU crisis.

User avatar
bobingabout
Smart Inserter
Smart Inserter
Posts: 7351
Joined: Fri May 09, 2014 1:01 pm
Contact:

Re: Early game modular armor rebalance

Post by bobingabout »

bobucles wrote:Making the roboport range smaller does the trick just fine. You don't NEED a 150x150 personal robo range, and it's completely detrimental to their use anyway.
Have it only work when you're standing still.
That's pretty silly. There's absolutely no harm in allowing the player to connect to the power grid on the move. We're talking about a number of entities that can be counted on your fingers. There is no CPU crisis.
I'd still say it should only work if you're standing in a charging dock. it means it has to check once when you stop moving in a few specific spots. it also means you'll be heavily reliant on batteries, because you charge them, then run and do things. It's also a good reason to set the throughput high, it only works while you're standing in the charger dock.
Creator of Bob's mods. Expanding your gameplay since version 0.9.8.
I also have a Patreon.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

bobingabout wrote:I'd still say it should only work if you're standing in a charging dock. it means it has to check once when you stop moving in a few specific spots. it also means you'll be heavily reliant on batteries, because you charge them, then run and do things. It's also a good reason to set the throughput high, it only works while you're standing in the charger dock.
That's one solution, but it would also work fine to simply connect the track to the electric network. The train may be moving but it's staying in a given track system. And the track isn't moving.

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

it means it has to check once when you stop moving in a few specific spots.
Checking for the player is such a minor concern there's no reason to base balance reasons on it. Checking for trains is a potential concern because they're huge and there can be lots of them roaming around.

I am happy that there is only one kind of train. It is absolutely more than enough and means you don't have to rip your train network apart for tech reasons. I'd be happier if the train could use liquid fuel directly and get rid of the silly nature of solid fuel all together. it would also be nice to rip out old fuel to upgrade it. To me that's more important than electric trains.

User avatar
thereaverofdarkness
Filter Inserter
Filter Inserter
Posts: 558
Joined: Wed Jun 01, 2016 5:07 am
Contact:

Re: Early game modular armor rebalance

Post by thereaverofdarkness »

bobucles wrote:I'd be happier if the train could use liquid fuel directly and get rid of the silly nature of solid fuel all together.
Train engines should come with a liquid fuel tank that you have the option to use.

User avatar
bobingabout
Smart Inserter
Smart Inserter
Posts: 7351
Joined: Fri May 09, 2014 1:01 pm
Contact:

Re: Early game modular armor rebalance

Post by bobingabout »

bobucles wrote:
it means it has to check once when you stop moving in a few specific spots.
Checking for the player is such a minor concern there's no reason to base balance reasons on it. Checking for trains is a potential concern because they're huge and there can be lots of them roaming around.

I am happy that there is only one kind of train. It is absolutely more than enough and means you don't have to rip your train network apart for tech reasons. I'd be happier if the train could use liquid fuel directly and get rid of the silly nature of solid fuel all together. it would also be nice to rip out old fuel to upgrade it. To me that's more important than electric trains.
I was actually talking about the grid interface, but, same applies...

what about multiplayer? on an active multiplayer game it's possible to have more players than trains. I mean, you do remember the game that had over 400 players connect to it, right? I'm not saying that's a common thing, but I probably wouldn't want to break that claim just for a feature that you're probably going to stop using when you get fusion reactors.
Creator of Bob's mods. Expanding your gameplay since version 0.9.8.
I also have a Patreon.

bobucles
Smart Inserter
Smart Inserter
Posts: 1669
Joined: Wed Jun 10, 2015 10:37 pm
Contact:

Re: Early game modular armor rebalance

Post by bobucles »

on an active multiplayer game it's possible to have more players than trains.
Players are tiny, slow moving box entities. Trains are massive snake entities made of multiple fast moving boxes that could be pointing anywhere at any time. A single mobile train has more power network hitbox checks than half a dozen players.

You keep saying "what if" but it's right there. It's nothing. Player checks are literally a drop in an ocean, even when you reach completely unrealistic player counts. And no, 50+ players is not realistic. 5 is realistic. 10 is realistic. 20-30 is an absolutely crowded game and well within the reality of what players deal with. Checking the hit boxes of 30 players is no more demanding than checking the hit boxes of half a dozen medium trains, in a game that checks hundreds of trains without issue. Saying there are CPU concerns is not a valid argument.
just for a feature that you're probably going to stop using when you get fusion reactors.
If fusion reactors were cheap I'd accept that a mid tier energy module was fruitless. But fusion reactors are expensive, and they absolutely have no way of being used in a PvP type of scenario. The tech and resource cost is simply too high for the compressed time frame. Fusion reactors also don't make sense for a campaign that tries to have a respectable balance curve, because they completely break the energy growth curve. Solar is pitifully weak and fusion is overwhelmingly strong. It's not wrong to want a middle ground option.

Induction power is the ideal choice for a mid game suit energy module. It provides the energy that mid game suits desperately lack and it does so in a way that is highly slanted towards defense and base building. That's exactly the sort of stuff players worry about in the mid game.

Post Reply

Return to “Balancing”