• 0 Posts
  • 99 Comments
Joined 2 years ago
cake
Cake day: June 25th, 2023

help-circle

  • Outer Wilds might not work well if you play sporadically, I think a big part of the joy is piecing together everything you’ve seen and grasping the connections and bigger picture… But who knows.

    Mindustry might work well though, it’s much simpler on the factory mechanics, but ties them into tower defense and RTS, needing to supply towers with ammo, and later supply factories with the right materials to create units. It’s FOSS, available for free from some official sources. And importantly, sectors are mostly isolated, meaning you can take it one mission at a time, bringing a few resources to kickstart things and building a new setup every time.

    Also, hard read on the tycoon games, I love playing OpenTTD with friends, though I lament the lack of something better than cargo distribution to require us to provide supply to the actual demand (as opposed to being paid to shunt passengers/cargo to the most convenient location). That said, I never really did any calculations in that, especially since supply and demand can change rather dynamically.


  • I haven’t properly tried satisfactory, I tried the demo back when that first came out, was asked to run around collecting leaves to put into a power generator for half an hour, and bricked my game trying to put it into borderless or something… And then I switched to Linux, the game was epic exclusive despite promises otherwise, and I passed.

    I got the impression it’s got a tedious early game, having a prebuilt map might make replaying less fun, and it sadly seems to have a very point-to-point, purpose-specific-device approach to logistics. I also like the performance of Factorio, it’s really lightweight on the GPU, and well optimized for CPU (though with the entire map and tons of individual entities loaded at all times there’s only so much you can do), which I imagine isn’t as great for the modern 3D game Satisfactory is.

    I don’t want to rant too much about it, but I think the splitter taking in and outputting two belts in Factorio is brilliant. There’s only a few types of logistics, but they are versatile and nuanced. Being able to belt items onto the side of an underground belt lets you filter out belts by side, the mechanics of belt sides and how they interact with inserters let you create compact designs or maximize throughput if you spend time on it. There’s no dedicated buffer machine, no separate splitters and mergers, all the neat things you can build come together out of component parts in an organic way.

    I will also mention that I like to try to plan ahead specifically to avoid starting over, but when rebuilding is necessary (and when laying a rail network) robots are a must-have.

    On the topic of the DLC… If you’re not drawn into the base game, might be best to pass on it, but they did a good job giving each planet some interesting unique challenges, including organic items that spoil after a certain amount of time. There’s plenty of straight content expansion mods, big and popular ones, but they mixed up the gameplay quite a bit in Space Age.

    All in all… Yeah, different people, different tastes. I’m currently doing a second playthrough of Space Age with friends, but one of them might’ve been felled by Gleba. If you want some more unsolicited gaming takes, I can recommend Mindustry and Outer Wilds ;D


  • it’s also very shallow

    You take that back!

    In all seriousness, if you’re talking about something like the fact that all machines are functionally doing the same thing, that’s kinda fair, but there’s a lot of complexity in all the options available, made even greater with DLC and mods. Just the logistics of getting items to the right places have many different approaches with various upsides and downsides, and I love all the emergent mechanics that come from belts having two sides and splitters handling two belts.

    It’s not a game for everyone, but calling Factorio shallow seems really odd. If anything, I feel like it allows you to explore its mechanics deeply, instead of having a breadth of shallow mechanics that don’t leave anything to be discovered.










  • If you license your project under GPL, and somebody submits some code (like through a pull request) that ends up in the library you use, you are now also bound by the GPL license, meaning you also have to publish the source of any derivatives.

    The way to avoid it is to use something like a CLA, requiring every contributor to sign an agreement giving you special rights to their code, so you can ignore the GPL license in relation to the code they wrote. This works, but is obviously exploitative, taking rights to contributions while giving out less.

    It also means if somebody forks the project, you can’t pull in their changes (if you can’t meet GPL terms, of course), unlike with MIT, where by default everybody can make their own versions, public or private, for any purpose.

    Though it’s worth noting, if you license your code under MIT, a fork can still add the GPL license on top, which means if you wanted to pull in their changes you’d be bound to both licenses and thus GPL terms. I believe this is also by design in the GPL license, to give open-source an edge, though that can be a bit of a dick move when done to a good project, since it lets the GPL fork pull in changes from MIT versions without giving back to them.






  • Dual booting is problematic, as mentioned you’re messing with your partitions and could mess up your windows partition, but also windows can, unprompted, mess up your Linux bootloader. As long as you’re careful with partitions and know how to fix your bootloader from a live image, there’s no real issue, but it’s worth keeping in mind.

    By the way, I recommend rEFInd for the bootloader when dual booting, it doesn’t require configuration and will detect bootable systems automatically.

    A VM sounds like a good idea to try a few things out, but do keep in mind performance can suffer, and you might especially run into issues with things like GPU virtualization. If you want to properly verify if things work and work well enough, you’ll want to test them from a live system.

    As a final note, you can give your VM access to your SSD/HDD - if you set that up properly, you can install and boot your Linux install inside a VM, and later switch to booting it natively. You still have the risk of messing up your partitions in that case, but it can be nice so you can look things up on your host system while setting up Linux in a VM.