I beat space age! (long+opinions+feedback)
Posted: Fri Dec 20, 2024 1:18 pm
SPOILERS!! (obviously) Overall a definite 9/10, I'm here because of the small difference to perfection and because most "complaints" I have are just a few balancing issues (that maybe can be solved in 2.1 or something?) Except Aquilo.
The complaints are definitely complaints because I love the game and I want it to be even better.
The write up is addressed to "whom it may concern" which may or may not be devs and may or may not be other players. If I call something "bad", or a "bad idea", that's my take, my opinion and Wube / the lovely devs, who I greatly respect for making this game, are in no way under any obligation to do anything about it. However if it becomes clear that I missed something, or you have constructive tips for me that can help me enjoy factorio more, I'm curious to read your take on my take.
Summary
All the art, music, models, etc. are 10/10.Quality: wonderful chaos, just useful enough to pretend 3x(->4x -> 5x) your complexity is worth it. Peak self introduced chaos!
Vulcanus: cool environment, worm isn't too easy without uranium ammo, maybe a bit trivial with it. Good rewards, foundries are universally seful, artillery is the "solution" for hostile npcs, both on nauvis and gleba. Vulcanus may be bit low on build puzzle / mechanics in retrospect. But that's after unlocking cliff explosives.
Fulgaro: also cool environment, the idea that the problem is dealing with products in the product / recipe tree from the opposite direction + belt filters is different enough. good rewards, the recycler is awesome, electromagnetic foundry is also really really good because of circuit productivity. I didn't use the tesla turrets. Making the ammo and shipping that to different planets seemed too much effort.
Gleba: peak of the expansion for me. The bus / tree design didn't work and I had to learn everything from the ground up, excellent stuff! Still have some comments.
Aquilo: absolute low point, more on that later.
Space: neat, but with some issues.
Here is the galaxy thing: https://factorio.com/galaxy/Fluorine%20 ... ha3-4.C6T5 I did it as a deathworld run.
Gleba
Excellent switch up with the spoiling of ingredients, I think science packs could have gone without it though. I had the best time here, figuring out how to solve the new problem.So, my very few minor nitpicks:
The one thing I don't like about the puzzle is that you have to purposefully spoil things for carbon. It's weird that we're getting bacteria that make copper and iron but not that.
The rewards from gleba are nearly worthless. The biolab can't really be used on other planets.
The rocket turrets takes care of big asteroids and is just a game mechanical checkbox. You need them, if you have them, it's trivial. They are not good enough against stompers. My solution for those turned out to be to research lots of artillery range, which I find disappointing because it's the same answer as for nauvis and unlike on nauvis, I don't really have the choice to go for "conventional" defense, since the stompers are just too tanky. Imo rocket turrets lack range (2x, 3x) But that causes problems in space, because it would waste ammo. Maybe a "long range rocket" could be introduced but I don't know how the turrets were coded and if the range can be modified by the ammo type.
It's a bit silly my tinkering space genius robinson crusoe can't put more rocket fuel into a rocket to make it go further, and also that the walking mountains that destroy my factory from above, by stepping on it aren't easily detectable by radar.
The spider bot is useless because I have no purpose for it. Gleba is solved, nauvis is solved, it's the wrong tool for the job on vulcanus the other planets don't have enemies.
I'm not enjoying the spidertron fish joke. After setting up absurd quality BS, I want a "rare" spidertron. Having me go through another useless hoop of "fish breeding" that serves absolutely no point is pushing it too far.
I would like a circuit clock / access to time to be able to time things better. Seems appropriate if there is a time based spoilage mechanic.
Space
I was really pleasantly surprised that there was an efficiency curve for the thrusters! I would have really liked to exploit that, but in the end it seems to lack the controls to supply fuel, or not, how fast, etc. and it can again just be solved with more input.The platform building itself, the modular nature and the flexibility and the reuse of familiar mechanics is excellent!
Nearly all interactions between planet and space platform feel bad. It technically works when importing some massed planet exclusive resource, like tungsten from vulcanus. Or the science packs. But that's it, everything else is just the worst. The "automatic request building materials" wastes tons of space and delivers stacks of things I don't even want.
It's really not a good time to send things that don't fill entire rockets. That artificial UI difficulty clashes harshly with building new platforms on the one hand and the quality system on the other. The manual sending of items from space is causing so many missclicks for me, I save every time before I send something down.
If I had three wishes and a time machine that I could only use on factorio related things, I'd go back in time and tell the devs to just not do aquilo, at all, and build a better gui for that.
The math of loading rockets with specific things once, sorting by size (descending) and making sure there isn't wasted space, really isn't that hard, and would really like to see that in a version 2.1 .
I beat the game but I still haven't found a good solution how to manage the different asteroid percentages in different environments. It's strange that I can't configure the asteroid collector to not pick things up that I have enough of. I tried building a recycling setup with circuits but yeah, it didn't really work.
On the way to Aquilo / after the first three planets, things are awkward. I have a good chunk of science now, I can easily get nuclear power. But I also have heating towers, advanced asteroid reprocessing, heating pipes, steam turbines, but I'm just not allowed to make steam from ice and carbon, in space and turn that into power. Felt weird. I mentioned rocket turrets already.
Aquilo
Aquilo is essentially the last unlock step to end game and nothing else.I don't like 90% of it.
The only positive thing I can say about it, is that it forced to me build a cute little train system between the three resources and my space drop point, that felt nice, I always wanted to build a useful smaller train setup.
The "heating pipe puzzle" is essentially trivial. Besides a relatively short learning phase, initial setup of the recipes and understanding and getting over that I have to ship in basically everything else, it's a matter of finding a repeatable pattern, just like the other planets. I read some stuff about heat and energy loss on the wiki, which was a trap because the solution is to just scale out of that being a concern at all. I wanted to avoid underground belts and pipes to save fuel, but that's abundant anyway. By that point, copper, iron, chemistry are all on my space barge, so getting those materials and e.g. heat pipes wasn't a big a deal.
"Magnum Opus" and "Shenzhen I/O", have a system that checks a loop / machine once, slowly to see if it works once, then quicker 10 times to make sure it loops, then a hundred times really really quickly to make sure the solution is stable. Something like this would be have desirable for the "skill check" that is aquilo. E.g. the difference, between making 1 or 10 science packs and making 1000. If I can make 1000, I can make 1000.000, it's just a question of scaling. But scaling on aquilo is boring and painful, because construction bots are slow and if I'm short on building material I have to wait for the import. In other words, it's exactly what feels nice about unlocking construction bots, in reverse.
The heating mechanic itself is a good starting point, but introducing it on aquilo poses more questions than it answers:
- isn't space cold too?
- why aren't my foundries and ovens producing heat?
- there is a cooling process in the science production chain, but the cooling process also doesn't produce heat?
- hot fluoride pipes can freeze?
The conceptual ideas around overheating basically write themselves, so I won't bore you with that.
Entering actual complaint territory: the devs decided that stones are the insulating material that has to be used on aquilo. Not carbon. Not plastic. Not low density structures. Stones are required for heating towers too. Somehow asteroids don't contain any stones.
I can't build a sled on an ice planet. Or a ship, on the island I'm on. That doesn't mean either should be put into the game, rather picking the ice/snow/island theme was a bad idea. I built the mentioned train, but the train needed rails and the rails need stones.
Actually shipping everything in wasn't that big of a deal, it just took time. Time, during which there was nothing, else to do because I'm blocked by the lack of resources that my ship is getting for me. And then, of course, the planet / platform interaction I mentioned.
I also fell into the trap of doing math and setting up a full chemistry setup for solid fuel and then I was lacking ice until I noticed the solid fuel from ammonia recipe that would require a splitting process and that produces a ton of ice as a byproduct. So yeah, I was lacking ice on the ice planet. Partially on me, but I don't get why the solid fuel from ammonia + crude oil has to be worse than the regular splitting process?
After that it was mostly a waiting game. I could now repeat the patterns I found ad infinitum which to me means I "solved" it, I could unlock more resource clusters at absurd import costs, but I really didn't want to at this point.
The railgun is weak and pointless. It's a required unlock for the end of the game, the extremely short range (and the low view distance in space) means that the big "advantage" of penetrating targets gets outplayed by the asteroids just spawning extremely close outside of the range of the railgun and shooting one asteroid won't hit the "not yet spawned" asteroids. It's also useless since combat was solved by artillery on nauvis and gleba, and I wouldn't even import it on to gleba if combat wasn't solved since the overall design seems too slow compared to the speed of the pentas and the firing arc seems like a critical flaw against the stompers.
The overall interaction of railgun, railgun ammo and production meant that I was going stop and go for an hour or so to reach the end, which is not the kind of gameplay I play factorio for. I could have rebuilt half my platform to expand explosives and the ammo production, but I wanted to be done.
Fusion power is neat ish, but I was surprised how quickly I burned through the fuel I brought to reach the end, but maybe that was because of the railguns.
circuit networks
This playthrough, starting with throwing unneeded asteroids overboard, I really discovered and learned to love circuits. Fantastic system.Here are a few points about it:
Dear devs, please give me an egg timer for gleba.
Otherwise the one thing I would want from circuits is a one click button/switch, that I can use from the regular map by clicking on it, without opening another gui and e.g. flipping the toggle slider on the switch. Some shift, ctrl, alt, click to turn them on or off would work too.
Testing the limits a bit, it's clear that the devs made the classic fatal "mistake" of putting programming into the video game. In multiple overlapping ways, with basic logic, math and electric signals? Now we can have all the discussions about what a "good language" is, how to package logic, where, how it should be controlled, flowed, etc.. So we already know this is sort of "open ended" and the question becomes how much of a hassle using the in-game programming is and how much of a game it is to solve problems with the "bad" tools.
So these comments are particularly tongue in cheek, because I can gleam just how much of a pain it is to build or extend a system like this.
I would really like visual feedback over which signals are passed out of deciders and selectors, the same way you show what contents are in a chest with 'alt mode'. Especially since the monitors don't really do that, for some reason.
Packaging / distribution: logic blocks and groups aren't copyable like requester settings.
I more or less built a "requester" system with trains and I noticed I can't read requests from requester chests. To use them to set a filter on an inserter and unload the train into the requester chest. I can build something like it with deciders and regular chests, but it's a shame I can't reuse/interact with the request system that way. Like, it would be cool to be able to fulfil requests with trains, if it's a request at a particular place, etc..
For more complex logic setups the amount of deciders that are needed just end up taking too much space. I really like the visual nature of the system, and there is a good justification for e.g. the power switch to take up some space. But I don't think it's "right" that a few simple deciders (if else) take more space than an assembler. That's usually where my initial positive impression of the gamey nature of the circuits can't be held and is replaced by wishing I could just write code.
Overall they do change the entire game though, I'm looking forward to another playthrough where I fully leverage them from the start.
where I stumbled
It did take me a while before it clicked for me that just because something has to be *made* on a specific planet, doesn't mean it has to *stay* there. So I was really happy about being able to export foundries and electromagnetic thingies and equally disappointed when I noticed that the same advantage didn't apply to the biolab or the cryogentics thing.A *very* fun moment happened when I didn't read the asteroid briefing on my first inter planetary trip and my platform was nearly completely destroyed and I frantically set up launch capability on fulgaro to be able to save my ship. Chaos, but good "losing is fun" chaos.
I set up a train loading system that made "mixed trains", and during loading just taking the contents from the train, feeding that into a "anything < 100" and piping that into filter settings for an inserter will NOT work, the decider also needs a "train ID !=0" which is off by default. I also tried reading logistics network content twice to automatically insert new bots and I forgot to check the checkmark to actually read the contents twice, so in one case I just produced way more logistics bots than I needed. Both of those can just be activated by default I think.
mixed malfunctions and obstacles
some stuff that was annoying and just can't be the intended experience:Stack inserters actually cause too many problems to make them really useful. If quality items are produced or things spoil while being held, or if the production recipe is changed, the programming of the stack inserter dictates to wait for more things that fit the things that are already being held, apparently indefinitely. So it will just stop if something "unexpected" happens. It can be solved with custom filters or circuits, but it's a bad default experience.
There are some issues with quality filters, I think they worked correctly in inserts and splitters and didn't work with "greater than", etc. in circuits and had to be manually specified? Maybe there is a "multiply with quality levels" arithmatic that can be done. Quality filters can definitely be built with circuits, it's just a pain and takes way more space than is reasonable.
Copy pasting material requirements from an assembler to a requester chest, seemed to vary widely in the amount of items that are being set. Sometimes dozens, sometimes hundreds and I don't need multiple hundred complex items to just sit in a requester chest.
I like new(?) factorio-pedia in game, but I often misclicked the name and the "change product" gear icon in assemblers / production buildings and was confused why I couldn't select the wiki entry as production recipe, until I noticed that I had misclicked. Also, repeatedly configuring assembler, logistics or circuit limit, requester chest amounts, took a lot of a clicking. This ties into the circuit network stuff a bit. Maybe I can make something that's easier to use with circuits, should the QoL of that be locked behind using circuits? Idk.
meta
At late game, there were some issues with large amounts of things that I had to manually set up another recycler loop to get rid of things and moving filled logistics storage chests, was... bad. It would be nice if there was a better (and different, interesting) system to handle thousands of items at once. Basically containers.It's interesting to me that there is a "glass ceiling" for planning. You start playthroughs with manually building, then copy pasted automated building, then... nothing and you'd need basically factorio not 2.0 but 2.0 ** 2 because abstracting it more needs entirely different systems. And I'm not sure if there is an audience for that, even if the devs wanted to make that. I just wanted to mention that there is "more itch that wants scratching" but I've reached the end of the tools that factorio has right now. It is still multiple levels above any of the competition I've seen, but it's also still not the pinnacle of factory automation I want.
