Friday Facts #309 - Controversial opinions

[Sir_Wulfrick]

Twinsen,

To address your thoughts on blueprints.

I understand your reasoning behind wishing to only have blueprints as a MODed feature, allowing (especially a new player) To stamp down a whole factory from a blueprint (I've seen such blueprint books) Defeats the whole point to Factorio.

I think this is typical "designers overcontrol syndrom" (or whatever you want to call it). I don't think that's a good way to design games. To me the blueprint system is one of the deciding features that makes factorio so unique where every other game (minecraft, etc) constantly forces you to build the exact same layout over and over and over again. By hand. I can't describe the agony. Just look at the number of "early bots" mods to get a feeling of just how agonyzing it is to manually build the same thing more than once.

Blueprints (plus robots) perfectly solve this problem. But if someone decides to use it to copy large factory bits from the internet why is that wrong? Maybe they're having more fun with other bits of the game. Maybe they're learning from others? Maybe... they're even participating in a forum thread to find an optimal solution for yet another self-imposed problem. There are so many reasons *for* blueprint sharing. And the only reason against is "but i want them to play MY way!". And besides removing the library wouldn't even solve the problem. It'd just lead to people posting pictures instead of strings (many already do), and other people rebuilding those pictures in their games. Or just start sharing savegame files of empty maps with blueprints. As a designer you have to accept that people want to and will enjoy your game in ways other than you originally intended. You can try to "guide" people, but you shouldn't put them on an iron leash. But this is clearly something you already know, so maybe you just forgot about it .

I've been playing factorio for a while but decided to register on the forum to comment on this point. Can't agree more with the above. The paragraph about blueprints in the controversial opinions article came across as being seriously condescending. Factorio is by and large a single player game. Obviously there's co-op, but for most it's a SP game. Of course the dev team can set up a framework within which the game is played (i.e. the game mechanics) but outside of that, stop trying to dictate how people play single player games! If people want to insta-blueprint something, they should absolutely have the freedom to do so! Writing things like (and I paraphrase here) 'this is how the game should really be played' smells strongly of elitism. You're not a *real* factorio engineer unless you play the game our way. Nope, it's up to me how I want to play within the sandbox created by the game mechanics. To be sure this isn't how I personally chose to use blueprints but if people want to play that way they should be able to do so.

More generally.

I don't feel that bots are at all OP. They're laughably inefficient for long range logistics and belts are far superior for medium range transport. Short range high-intensity complex production, boosted by beacons is where the bots really come in to their own, for example final assembly of satellites.

The only other point I have is that I like inserters the way they are now. Granted it can be complex to calculate their throughput but given the scenarios in question that's absolutely realistic - of course the throughput would be dependant on the inserter type, item supply rate etc. I sympathise with the performance argument but the proposed behaviour feels like an unnecessary simplification.

[meganothing]

The detrimental habit is when someone new to the game starts to use designs from the internet before even having thoughts of his own about a solution. It is not about players who already built a solution and now try to optimize it. That is my point, and that is also why I think achievements are too late. You yourself bring the example that achievement hunting normally starts after a first play-through. A no-imported blueprints achievement is useless because it is too late.

Thats not really a habit, thats just a style of play. Sure it isn't the way I (and most likely you) find to be fun. But it may be the way someone else does. Having an achievement that requires a full play through with out imported blue prints would require the player to play from start to launch building everything by hand. For some players the presence of this achievement would be the only reason they ever do this. For other players, they would get this achievement on their first play through. And unless you are using a specific seed with a specific bp, you likely are just getting modular bps (one for a yellow belt filling furnace, one for making a circuit assembler, etc...). They then still have to manage integrating these piecemeal bps into their specific map. This integration may just be the level of complexity they are comfortable with.

Quote developer (shamelessly copied from AndrewIRL's post, not sure who from Wube exactly said this): "It's our job as developers to incentivise the player to play the game properly.". If you don't agree with this, fine. Then we would need to go back to fundamental game design to argue the case.

But if you agree, then not all "styles of play" are equal. Some styles of play are used by new players and they only find out about it being detrimental to their enyojment of the game when it is too late. A good example is save-scumming in RPGs, another is looking up solutions or strategies on the internet in adventures, strategie games and puzzle games. Seems obvious, but people make those mistakes all the time.

A game often can't prevent it and often shouldn't. You gave a very good example, Factorio may be too complex for some players without internet help. But having some speed bumps to disincentivice players of such practices is not preventing people from doing that style, it is making sure that they conciously decided to adopt that style (by doing some extra work like installing a mod) .

Similar to keeping the rat poison on a shelf high up in a far away corner of a store.

[eradicator]

Yet another additional thought: The whole idea of "limiting blueprint import" seems based on the assumption that without it any player will nessecarily develop the skills to solve all the problems and gain a greater "feeling of accomplishment" (aka "fun") than they gain by "copying the solution".

I think this assumption is false not only concering the "fun" part (see my post here), but also concerning the skill part:

Not everyone has the nessecary logical planning skills to "solve" factorio. Some people might not have them *yet* (i.e. children, people new to games, people new to the genre), and most people will never reach the planning skill level required to build a megafactory, or become a "circuit magician". Yet all of these people may still enjoy factorio, and all of them will probably learn *something* from the game. Not everyone needs to learn *every* lesson told.

Further there are constraints, like disabilities, carpal tunnel syndrom or - much simpler - available time, that can make it impossbile for someone to ever get the "full experience". To these people the blueprint system isn't a cheat, it's what makes it possible for them to play the game at all.

And back to the "fun" bit... Most simulation games so far (Sim City, Railroad Tycoon, etc...) don't simulate anywhere near as much of the "factory" bits as factorio does. To someone coming from one of these it might be *more* fun to "select a circuit factory from the build menu" and connect it with trains/belts/etc to the rest of the "town".

Looking at this and at how much effort you devs have lately spent on making the game *more* accessible to new players i would've - if asked - found it more likely that some future version has curated example blueprints for every recipe built-in, than that you're still internally considering to kill one of the best parts of the game.

[gorothdablade]

Quote developer (shamelessly copied from AndrewIRL's post, not sure who from Wube exactly said this): "It's our job as developers to incentivise the player to play the game properly.". If you don't agree with this, fine. Then we would need to go back to fundamental game design to argue the case.

But if you agree, then not all "styles of play" are equal. Some styles of play are used by new players and they only find out about it being detrimental to their enyojment of the game when it is too late. A good example is save-scumming in RPGs, another is looking up solutions or strategies on the internet in adventures, strategie games and puzzle games. Seems obvious, but people make those mistakes all the time.

A game often can't prevent it and often shouldn't. You gave a very good example, Factorio may be too complex for some players without internet help. But having some speed bumps to disincentivice players of such practices is not preventing people from doing that style, it is making sure that they conciously decided to adopt that style (by doing some extra work like installing a mod) .

Similar to keeping the rat poison on a shelf high up in a far away corner of a store.

There are 2 key concepts in your quote. "incentivise" and "properly"

incentivize is generally considered a positive reinforcement or reward based action, though you are right that putting road blocks on alternatives does indirectly incentivize the correct path. There just are going to be way too many alternatives to put road blocks on. Putting an achievement in place would be a way of rewarding (some) players for trying things the "proper" way. (I say some because not all people care about achievements)

The question then becomes what is the "proper" way to play. If factorio was a jigsaw puzzle, BP imports just allows for the number of pieces to scale to a persons preferred play style. No bps is the 10,000 piece puzzle for experts with too much time on our hands (like me). With BPs, you get the option to scale it down to a 5000, 1000, or even 500 piece puzzle. Sure there are a few people who will find a way to make it a 100 or 10 piece puzzle. We are all still doing a jigsaw puzzle, we just are setting it to our perspective skill levels.

[Lastmerlin]

Well, I'm too late for the party again

I will post a selection anyway:

Inserters should not chase items:
Unsure, but not indifferent. Usually, I am not too keen on the optical qualities, and I am always pro any UPS-friendly changes. However, I cant judge, if it will feel right. Perhaps test it internally? Of, if it is a huge effort (which is probably is), just leave it alone. This would be no mistake.

Blueprint import/export should be a modded feature
I dislike it. I never downloaded a blueprint from someone else, but I still dislike it, because I love how easily you can transfer your own blueprints everywhere. There are steam Users who insist on vanilla play, and having it as a core feature is really essential here.

Miners shouldn't output directly to belts
NO. Miners and smelters are already super boring and a chore. The inserters are the pain in manually building the smelter lines. But at least you can set up the miners quickly. Please dont change it.


Pipes should work like electricity
I like it. Usually, I like complexity. Thats why I play Bobs&Angels, Seablock, 40 addition Science packs. Guess what: Fluid dynamics are irrelevant 90% of the time. Pipes are either full or empty, thats it. But when they get relevant (like high troughput Angels Chemistry) you realize, they are horrible to debug, unintuitive and utterly frustrating. The only problem is: How can you keep fluid wagons from becoming irrelevant. They are the only fun challenge I have found in the fluid dynamics so far.

Extensive Combat/Biters stuff
As default option, I hate it. I don't care about combat, but often play with friends, who want a bit of combat. Current default settings are perfect for this - they are about 20% combat, so one player could concentrate on fighting, but other can also concentrate on building. This is how I see Factorio: Mainly a building game with some combat as deterrent and spice. Not a 90% fighting game. However, occasionally doing a real fighting game can be fun. So, as a option I approve it.

Robots should take up space and time
O NO! Robot hate and belts vs bots again. I bet that whoever proposed that never did something UPS-critical to propose something like this. Moveover: Construction bots are the point where free designing really starts. Nerfing this because you cannot balance out one single combat style (turret creep) is completely silly.

Items should have volume and mass
NO! Best way to make a nice game more annoying.


Adventure mode
Sounds like something. that is best accomplished on a larger scale by mods. Small scale goodies to find (similar to large rocks) could be nice.

Mining furnaces and assembling machines should return the ingredients for the in-progress recipe
I am surprised, that it is not the case. It does not matter in most cases. However, the example of kovarex process really hurts. On the other hand, cheating items should not be possible either. Compromise: Only return items, if the recipe does not allow prod modules?

[lapantouflemagic]

hi guys !
second or third time i post on these forums, i did only one full playthrough until now, and i really like the game !

so i'm just giving my grain of salt, as well as some feedback : I really think that the idea about having gantries to build large stuff such as locomotives is good, keep it in mind for factorio 2 if you someday do that =)

on the "biters should be more aggresive, i kind of agree : I dropped my first freeplay game because... well i didn't really feel any challenge (i think i left the default world gen settings) there was a nest here and there, but i really didn't feel threatened.

so what i did was tweak the world gen settings so that biter are way more numerous, evolve a bit faster and spread more (i don't remember the precise settings, my save in attachement). In addition to this, i cranked up moisture really high, as well as tree coverage and cliff continuity, i also increased a bit the pollution emmitted to compensate a bit the amount of trees. and last but not least, i made the ore deposits more frequent but more shallow to "force" me to extend faster and more frequently as well as fight the aliens more. (exept for oil and uranium, i made them rarer but richer, so that i have to explore a lot further away in alien territory)

overall, during the first two thirds of the playthrough, i'd say it made the game much more engaging, typically since there was biters everywhere, i didn't want to deplete my main iron ore deposit, so i tried to grab ore that was furter away first (i mean, not first, first, i'm not crazy), which means i had to make turrets and bullets before spamming science, which i think was more interesting.

also, because of all the trees and the green lush vegetation, the map was much more beautiful than my first playthrough, which was mostly sand and arid steppes (if the world seems ugly ans sterile, you don't really feel bad for crapping all over it, that's kind of a shame), and the density of trees meant that navigating with the small car was much harder. And finally, having the cliffs make natural defenses was much more interesting tactically, i could identify areas that i could bring entirely under my control if only i could wall off a few entry points, that was a fun project to have in-game (i appropriatedly called that "operation retake wall maria")

however on the long run, (before i completed my wall) since biter recreated the nests i eradicated litterally every 5 minutes or so, it became tiring to have to patrol the unprotected areas, and i really wished radars had a much bigger coverage, also when i reached the yellow science packs, i really would have needed to scale up my production of base ressources and electronics, but in the end staying in my tiny country and waiting patiently was much easier than trying to conquer new lands.

so i'm all in favor of impenetrable forests, mountains, cliffs, peninsulas, rivers, you name it.

oh, also, i didn't use delivery bots much because i kind of felt it would have been too easy. i mean, factorio is basically a puzzle game right ?

[redis]

How about this simple nerf/improve bots idea:

Restrict roboports to be placed disallowing overlap of their range! You would not be able to place 2 or more roboports close to each other because of "interference".

This will nerf bots by reducing number of charging stations and lower throughput. Bots should have the lowest throughput because of their other abilities. They are meant to be used for "expensive" items. This will also cause a player to plan better how/where to place roboports to cover all squares bringing back some logistical challenge to the game with bots. Some research upgrades to roboport entities could be added to allow moderate increases in bots throughput.

I believe this should have been done originally to the game to prevent roboports spamming strategies. To solve the problem with large existing robo megabases you can add a menu option to enable placing roboports withing each other range or allow to create an easy mod.

Regards.

[Roberth]

You need to retire the we can not change it now, people are already used to it argument, if you genuinely believe a change makes the game better, not changing it is wrong.

I couldn't have said it better, this should always be a core idea when discussing new changes.

[Ivan.TaCake]

factorio--biter-corpses

factorio-1.jpg (329.29 KiB) Viewed 9050 times

At the moment the biters just run through the corpses without a speed penalty and you get this ugly mess of corpses.

For dumb biters, these corpses should act as a speed penalty. A speed penalty could force them to change direction with the goal of going high speed, rather than slow. So they would seek out terrain that allows high speed movement (non-corpse ridden ground; clean path to target that each biter identified).

Smart biters(?) -- could see the edge of a biter corpse as we the player see a concrete wall. So you could have a biter either run the perimeter of corpses trying to probe weakness, or possibly the biter could be triggered to perform a long range attack from the edge of biter corpses.

Maybe create a scout biter that sees the results of the attack and returns to the nearest nest and relays battlefield results to all other biters. So even if the corpses vanish, they know where previous failure in attack was. Maybe the biters determine that a massive and/or continuous swarm is required, so a nest or group of nests could enter such a mission.

Though...... these biters seem like climbers. If you guys could make those corpses not remain flat, but add elevation (implying that the biters climbed the corpse -- world war z style), this would obstruct the laser for some time and allow biters to move in closer with safety. This would force player to not only have laser defenses, but possibly also have flame throwers to burn biter corpses (fire will spread to other corpses) in order to allow clean ranged shots from lasers.

[BlueTemplar]

[Oktokolo]

The gif shows a medium Rampant raiding party.

[meganothing]

There are 2 key concepts in your quote. "incentivise" and "properly"

incentivize is generally considered a positive reinforcement or reward based action, though you are right that putting road blocks on alternatives does indirectly incentivize the correct path. There just are going to be way too many alternatives to put road blocks on. Putting an achievement in place would be a way of rewarding (some) players for trying things the "proper" way. (I say some because not all people care about achievements)

The question then becomes what is the "proper" way to play. If factorio was a jigsaw puzzle, BP imports just allows for the number of pieces to scale to a persons preferred play style. No bps is the 10,000 piece puzzle for experts with too much time on our hands (like me). With BPs, you get the option to scale it down to a 5000, 1000, or even 500 piece puzzle. Sure there are a few people who will find a way to make it a 100 or 10 piece puzzle. We are all still doing a jigsaw puzzle, we just are setting it to our perspective skill levels.

Again: Achievements are too late to incentivise, even if they were the bestest coolest incentive of them all. At the time most people look for interesting achievements they either have already copied everything from the net or solved the puzzles themselves at least once. In either case the achievement makes no sense anymore (except as a goal for people who collect achievements).

If "no pbs" is the 10.000 piece puzzle for experts, why was Factorio such a success before A17 (or whenever bp was introduced) ? DId we all play the most complicated version first and then went all to an easier game? Do megafactory builders do the 10 piece puzzle now? Your analogy breaks horribly here.

Again, I'm talking about the first playthrough(s), were the puzzles are finding ways to get something working (like a sorter, a production pipeline or a big reactor). Eventually most people have solved each puzzle to their personal satisfaction (or just looked up/copied the optimal solution from the net) and inevitably look for different goals. This could be achievements, mega-factories, special scenarios, mods. I think this is also the first time a new player can really judge on which scale he wants to play Factorio further on. The player who knows he just wants to play a train simulation before even having begun playing Factorio is surely a big exception.

[Beily]

So, if you ask what i think about it...
1) Inserters
I don't think this changes will be bad, but i can't tell that they will be good, i think inserters are fine like they now.
2) Blueprints
I love to create blueprints and if it really boring blueprint(like for railroad) - i will any way put it inside new game, by UI or mod, nvm.
3) Biters
I don't think biters must be agressive, i think they must be smarter when you dig their planet. I mean, on ~4 hour of game zerg ruch of biters looks boring, so i like when 1 biter moving to electic power line and kills one pole, that makes them looks smarter.
4) Miners
I think they fine like they are now.
5) Boilers
I love new boilers and steam, don't touch them.
6) Pipes
Yes, i know that they are trouble for calculating, but they now looks more realistic and i love that.
7) Adventure mode
No com.
8) Bots
It really looks like nerfing, adding mass to items that will slow bots - good, making them slowly picking up items - bad.
9) Items
Deep, really deep, i don't think game needs it right now, or maybe, it can be working in some "realism" game mode?
10) Hotkeys are needs any way, i hate to use mouse to do all staff in games.
Sorry for my EN, i'm not frequently write in EN. :3

[posila]

Controversion Opinion of my own: Get rid of "achievements" again. All they do is cause people to whine when they're disabled due to mods/commands/editor/cosmic rays. Ofc they also whine when certain achievements get easier/harder due to changes in game mechanics. The dev time wasted on these psychoactive carrots would probably have been enough to complete spidertron *and* the space platform extension.

I would go the other way and removed separation of modded and vanilla achievements. Also running commands wouldn't disable achievements. It turns out, if someone wants to get achievemnts by cheating, there is really no stopping them, but the cheating prevention mechanisms seem to make lot of people unhappy.

Yet another additional thought: The whole idea of "limiting blueprint import" seems based on the assumption that without it any player will nessecarily develop the skills to solve all the problems and gain a greater "feeling of accomplishment" (aka "fun") than they gain by "copying the solution".

I think this assumption is false not only concering the "fun" part (see my post here), but also concerning the skill part.

In short it comes from "Given the opportunity, players will optimize the fun out of a game." (https://www.designer-notes.com/?p=369, https://www.youtube.com/watch?v=7L8vAGGitr8) In general, players will tend to play in the most efficient way, even if it's not fun to play that way. And I'd say importing optimized setups some else made and just placing them down is god damn efficient way to play Factorio. Do players actually overuse it and rob themselves of challenges they would have more fun solving on their own? I don't know.

As for your assumptions about assumptions ... I'll paraphrase meganothing's argument: Factorio was successful even before blueprint import/export was added to the vanilla.
I think the core idea is that every problem has some simple, but probably inefficient, solution that alsmost anyone can figure out. And a solution player made on their own does bring feeling of accomplishment as opposed to placing blueprint from the internet (but this is probably much more complex psychological question and depends on context and some other things). I would also argue, that even copying a setup from screenshot has better learning value than just copy&paste and place blueprint.

In the end, people can play the game however they want, as long as they have fun. Not having fun is frowned upon.

that you're still internally considering to kill one of the best parts of the game.

None of the ideas presented in the FFF are being "internally considered". It's just ideas/thoughts of individuals in the team, with which some other (but not all) team members may agree. (That said, it doesn't mean the ideas can't be internally considered in the future; but I wouldn't worry about blueprint import being removed, I don't see that happening)

[DarkyPupu]

I would go the other way and removed separation of modded and vanilla achievements. Also running commands wouldn't disable achievements. It turns out, if someone wants to get achievemnts by cheating, there is really no stopping them, but the cheating prevention mechanisms seem to make lot of people unhappy.

Yes, that !!!

And please also remove the limitation for servers which are running H24, which makes modded achievement impossible to play with. There are a ton of cool achievements we could add, but with that limitation it's not possible to play with them. We would see more achievements as a team work than individual so it's no problem if someone connecting would get one not being here at the right moment.

Maybe a server setting ?

[bartekltg]

In one of the discussions was proposed a more or less easy fix for robots.

>There is no reason to mess with belts at all when bots can 'teleport' unlimited resources between storage chests.

Chests can get a limit on how much robots per second can access it. The other bots are just waiting, a similar solution is already implemented for charging in roboports.
The throughput for a single request chest can be set to yellow belt (or less) x Worker robot cargo size bonus.

The harder, but not necessary part is that some consideration is needed to limit the number of bots sent for one type of items. For example, no more than a number related to distance, velocity, and throughput, so a constant supply line wouldn't be overcrowded. Or the chest sent another bot only if the chest didn't get an item in the last x ticks.

> Laser-turret-creeping enemy bases is trivial because construction bots can instantaneously place enough turrets to completely overwhelm all the biters.

Wait. Didn't the long time for initial charging the laser suppose to fix it?

[mrvn]

In one of the discussions was proposed a more or less easy fix for robots.

>There is no reason to mess with belts at all when bots can 'teleport' unlimited resources between storage chests.

Chests can get a limit on how much robots per second can access it. The other bots are just waiting, a similar solution is already implemented for charging in roboports.
The throughput for a single request chest can be set to yellow belt (or less) x Worker robot cargo size bonus.

The harder, but not necessary part is that some consideration is needed to limit the number of bots sent for one type of items. For example, no more than a number related to distance, velocity, and throughput, so a constant supply line wouldn't be overcrowded. Or the chest sent another bot only if the chest didn't get an item in the last x ticks.

> Laser-turret-creeping enemy bases is trivial because construction bots can instantaneously place enough turrets to completely overwhelm all the biters.

Wait. Didn't the long time for initial charging the laser suppose to fix it?

Chests already have a limit for how many bots can access them. The limit is just rather high. At some point they simply are full or empty.

Unless ... you have a buffer chests and you get bots inserting and removing items in such a way that the chest is never full or empty. Can that happen?

[redis]

Adding limit how many bots can access chests does not solve the problem. You can create more chests instead. Here is another better solution for bots.

Allow ROBOPORTS to be placed only with NO OVERLAP (or minimal overlap) of their range. You would not be able to place roboports very close to each other because of "interference".

This will reduce number of charging stations and increase distances to fly to recharge lowering throughput. To keep it possible to improve bot throughput more research bot upgrades could be added like more cargo, charge time and speed . If this is not enough then also more bots could be allowed in the roboports. Moving roboports further apart will reduce number of bots flying and and improve UPS. The biggest issue is spamming roboports strategies allowing to create clouds of bots with unlimited throughput in the late late game.

To support existing bases this needs to be done as optional feature, which could be disabled with a checkbox or mod.

In one of the discussions was proposed a more or less easy fix for robots.

>There is no reason to mess with belts at all when bots can 'teleport' unlimited resources between storage chests.

Chests can get a limit on how much robots per second can access it. The other bots are just waiting, a similar solution is already implemented for charging in roboports.
The throughput for a single request chest can be set to yellow belt (or less) x Worker robot cargo size bonus.

The harder, but not necessary part is that some consideration is needed to limit the number of bots sent for one type of items. For example, no more than a number related to distance, velocity, and throughput, so a constant supply line wouldn't be overcrowded. Or the chest sent another bot only if the chest didn't get an item in the last x ticks.

> Laser-turret-creeping enemy bases is trivial because construction bots can instantaneously place enough turrets to completely overwhelm all the biters.

Wait. Didn't the long time for initial charging the laser suppose to fix it?

Chests already have a limit for how many bots can access them. The limit is just rather high. At some point they simply are full or empty.

Unless ... you have a buffer chests and you get bots inserting and removing items in such a way that the chest is never full or empty. Can that happen?

[Azzinoth]

@Twinsen @wheybags @TOGoS

I know I'm late to the party, but I actually agree with most of your proposals. I understand that some of them are controversial, but honestly people are always against change. Maybe they will like it when they actually try it.

If you don't plan to implement them into the base game, my idea would be that each of you guys release a mod with these changes. I mean as devs you could make the necessary changes to the api to make your mod work, you could even implement them as hidden features in the game which are just enabled by your mod (for things that aren't easily moddable). There are mods which do some of the things that you propose, but they aren't as well integrated into the game as you could do it.

Inserters should not chase items
I have hundreds of hours in factorio and I have build some fairly big bases, but I have never had any problem with performance. I guess my bases don't qualify as megabases then. The point is the game is already so well optimized, performance is a non issue for most players.

Blueprint import/export should be a modded feature
I do like the blueprint library for my own creations. I never just used someone elses creation in my game, I only got inspirations from pictures.

Weapons shouldn't lock on
Biters should be more aggressive, and probe your defenses
Clearing bases should not leave you safe
I guess people who don't like the combat aspect of factorio disable biters anyway. People who do play with biters probably would like a more elaborate combat system. I would definitively love something like that.

Miners shouldn't output directly to belts
Thats a minor issue to me and I don't really care.

Boilers shouldn't have a water output
I don't care again.

Pipes should work like electricity
Again, performance has become a non issue for most players. But if you say "work like electricity" you mean like "electricity" in factorio. Do you know about the electronic–hydraulic analogy? It means you can model fluid circuits exactly with the same math as used in electric circuits. There are very efficient algorithms to simulate electric circuits in steady state. In the game it's not exactly steady state, because input and output can change. However you can simulate it as semi-steady state where the values go from one steady that to the next steady state continuously while the input changes. This approximation works when the reaction time of a system is much faster than the speed at which the outputs/inputs change. I think it would work good enough in factorio, because in a game it doesn't have to work completely realistic. So I would propose to model pipes as an electric circuit. The resistance is proportional to pipe length. You would only have to model junctions instead of every pipe, but the model can still output current through each pipe.

Adventure mode
Thats something I always wanted.


Items should have volume and mass
I agree. You also shouldn't be able to assemble nuclear power plants or tanks with your bare hands, and it should take time to build them too. Currently you are able to build almost anything with your bare hands. Thats boring. This is a reason why the lazy bastard achievement is so much fun.

Robots should take up space and time
You could also give robots a higher stack size and reduce robot count for it. One new robot is equivalent to five current robots for example. Five robots tucked together would have five times the stacksize, five times the electricity cost and five times the material cost, but they would have the same velocity. Now replace them with one bigger robot with these stats and you have the same troughput but five times less calculations.

Power-user hotkeys
I agree, and according to the poll on reddit, 90% of users agree too. The other 10% probably didn't understand the question.

Mining furnaces and assembling machines should return the ingredients for the in-progress recipe
Thats such a minor issue to me, I wouldn't even waste my breath on that.

[Light]

Controversion Opinion of my own: Get rid of "achievements" again. All they do is cause people to whine when they're disabled due to mods/commands/editor/cosmic rays. Ofc they also whine when certain achievements get easier/harder due to changes in game mechanics. The dev time wasted on these psychoactive carrots would probably have been enough to complete spidertron *and* the space platform extension.

I would go the other way and removed separation of modded and vanilla achievements. Also running commands wouldn't disable achievements. It turns out, if someone wants to get achievemnts by cheating, there is really no stopping them, but the cheating prevention mechanisms seem to make lot of people unhappy.

I strongly dislike that creating a new force instantly robs everyone of their ability to get achievements if that's their thing. It's a necessary function that shouldn't affect them to begin with.

If you don't blanket remove the achievement disability with commands, then the list of what can disable them should probably be reconsidered. Right now it seems to encompass almost all the commands by default when there needs to be a few more exceptions.