Alt-F4 #65 – Factorio visualizer in Unreal Engine 5
alt-f4.blog
alt-f4.blog
In Factorio you launch a rocket and win. Space exploration takes that and continues. You end up developing a base across multiple worlds in the solar system. After developing the solar system from the different unique resources on each planet you expand to interstellar space and then other solar systems. The game has a 'quarter of the way through' victory screen but there's another victory screen as you explore a mystery that spans the known universe and takes a reasonable amount of math knowledge figure out.
It's 100s of hours of gameplay and really well done. They modified the game so that you can fly spaceships between planets with a star and solar system maps and it all works really well.
Just make sure to turn the biters way down. The mod requires a level of patience and the intensity of combat runs counter to that (in fact the mod itself has a popup telling you to turn biters down on start).
Factorio is a brilliant Zen garden and an absolutely terrible real time strategy game. Since the beta there's been more and more biter types and a gradual ramp up in how frantic you have to be to keep up. It's tremendously off-putting and completely opposite to what makes Factorio what it is.
Hopefully the developers realize this at some point and knock it off. Make a 'frantic mode' toggle for the experienced players but don't make it the default. I can handle needed a turret or two at key points but I can't handle needing to race the tech tree to keep up. I don't get why they made the defaults so hostile to new players. I'm not even that new to Factorio but it's still waaaay too frantic with default settings.
But I guess a toned down offense on your base, while still allowing the same kind of attack, would work just as well.
Also, the biters are somewhat boring at end of game, it's just the whole world map full of the exact same evolved bases everywhere.
After that I started playing mods that in and of themselves increased the difficulty of the game, and having the biters on top of that was just way too much.
Also, in a vanilla game, once you're more or less invulnerable to the biters they become just an extra annoyance, and I'd rather just focus on building.
On the other hand, without the sense of urgency that biters give the game can become kind of boring...
It was so nerve wracking I felt like progress was so much slower due to all the fighting. After maybe 50 hrs in we realized the krastorio mod want being called correctly, so when we fixed that we learned all the new complex recipes were obsolete and all our defense were useless due to the restructuring of ammo types
So I guess my point is. Bitters have a time and place and I'm over them.
Your second or third base always goes so much smoother
I'm currently playing with the Krastorio 2 mod, where the ridiculous range of the anti-material rifle makes killing biter camps a bit too trivial.
I understand some people just aren't really into the tower defense side of factorio. that's okay, and it's a good thing that the settings can be adjusted to suit a wide range of playstyles. but if that's you, I'd recommend at least trying a low pollution / efficiency-focused playthrough. the game gives you a lot of tools to manage pollution, and the biter threat can be pretty minimal if you use them. imo this preserves the balance better than just turning off biters.
> the game gives you a lot of tools to manage pollution
The popular Krastorio 2 mod adds even more: wind turbines from the start, more efficient ways to produce energy between steam and nuclear, and it even has air purifiers that reduce pollution. That's almost cheating. On top of that, taking out biter camps is a lot easier with the anti-material rifle.
The obvious missing piece is biter-clearance automation. The game is about increasing levels of automation. The next level after we get laser turrets and we get flying bots should be reaper drones with lasers that automatically patrol and eliminate biters within a large range. Also, I've put several thousand rockets into space (4K SPM baby), but I can't fire a nuclear weapon more than 54 meters?
Of course there are also mods, like M.I.R.V., by one of the devs :
Spidertrons are also not automated. They don't patrol. They don't run away when they are out of ammo, come back and reload.
Love the MIRV: like the too-short-range nukes on the spidertron, they are a great way to bomb yourself back to the stone-age. The video nicely shows how it's first missile on automatic mode blows up his own base. Brilliant.
2.) You can totally set an artillery train to patrol between outposts, not to mention that there won't be much biters left after the first cleanup.
3.) I guess you might want stationary artillery if you want to be extra-sure the nests are taken out ASAP ?
4.) I didn't get the opportunity to use a spidertron yet, is there really no patrol functionality, like the AAI Vehicles mod has ?!
Is that actually how it ends up going? Unless your base is extremely small, then going slower means less pollution reaches the biters and they evolve slower.
If you won't have 20 drills going after several hours, "play the tutorial" is probably the best advice.
"Don't start in the middle of a treeless wasteland" is also important advice.
Don't people just take whatever starting position the game gives you? Sometimes the starting position simply is a treeless wasteland. As far as I understand, moving to start far away from your starting position is a bad idea, because there's more biters there and the resources you need are much further apart.
One build style has you rushing to decent weapons, then walking far from spawn to set up your main base, so you don't have to bother with trains to get up to rockets. (The patches are randomly placed and don't have exclusion zones, so it's not impossible to see a 10M iron, copper, coal and oil all next to each other.)
See the “setting up research” on the editor wiki page:
https://wiki.factorio.com/Map_editor#Setting_up_research
Edit: you’d probably have to give yourself the necessary modular armor, roboport and initial bots.
I love Space Exploration, but when you hit late game man it starts to get tedious fast. I've gotten to naquium and the higher end space sciences, but I get bored with the logistics of keeping it all running and have quit both times I've reached this point.
Anyway, seeing Factorio in 3D was amazing though, really cool
I myself do keep on playing with biters as I feel personally no enjoyment if there would be no biters.
Satisfactory might only be 80% as deep but it being in 3D and gorgeous more than makes up for it.
If Factorio looked anything like this model visualizer I’d take it seriously.
That might have more to do with YT's encoding and color grading. I've had the same experience when watching videos, but not when playing the game.
Not as repayable as factorio but still good fun.
It hasn’t been released yet, but the team is publishing a highly interesting devlog: https://store.steampowered.com/news/app/2162800
hard pass. I do not enjoy when a video game makes me feel like a failure because I don't know trigonometry or calculus.
(No, I will not try EVE)
https://support.eveonline.com/hc/en-us/articles/816520777052...
(I mean this in jest)
... wait, how hard would it be to do something similar for Factorio?
And then there are whole planner mods too.
It's a mod, and so not required and already something you decided to optionally add to your game.
I think I got to a point - probably nearly 100 hours in - where I finally had two lines of products coming from other planets to do the next tier of science.
Another tip that I'd add is that you should increase the science cost, I found it easy to run out of things to research while it still took hours to expand things for the next tier, which in practice meant your factory just stops doing anything because it doesn't need to do anything.
I've done the first two, and I'm quite early into SE. Just an incredible amount of complexity and I'm loving it.
That's not a wrong way to play, but the intent of the default balancing is that you make those expansions smaller and faster. If a new resources has 5 steps to make, you can slap down 2-5 buildings per step with absolutely minimal planning/routing and feed it into a launcher. There's no need to spend hours on an expansion until your science production needs that much. And an orbital base can go very far with one assembler per recipe.
pyanodons adds very realistic ore processing chains, petrochemical processes (the main developer, pyanodon, is a chemical engineer by trade), energy generation mechanisms, electronics overhauls, and also farming components including both plants and animals.
overall, a full py run will easily take 1500+ hours. it's probably not particularly enjoyable for most vanilla players, but if you liked mods such as space exploration or nullius, you might find some enjoyment in py
To be honest, I used to be much more critical about how people spent time - particularly how I spent time ... Then I learned to stop worrying and love the reality: there is no 'purpose' (which is, itself, open to as many interpretations as you're able to and comfortable entertain[ing])...
For the sake of anyone's sanity, I wouldn't mention that.
Ahh, so if I start a game now, and play a reasonable (dedicated, honestly) 2 hours a night, every day, I'll have completed it in 2025.
Wild. Not for me!
https://en.wikipedia.org/wiki/Japanese_dry_garden
You spend a small amount of time each day meticulously crafting the factory/garden to unwind. The time it takes isn't a problem. The goal isn't to 'win'.
That being said, the Factorio crowd I rub shoulders with 1500 hours is not a lot. Heck, i myself am about 4-5k hours of gameplay and mine are rookie numbers :D
Factorio allows for an extremely wide ways to play it, multiplayer is it's own world entirely if you find a decent crew to play with (there are many communities around Factorio).
And then there's Clusterio that opens a whole new dimension of what's possible.
I'm willing to be told I'm just doing it wrong, but does vanilla + Space Exploration have a different sort of challenge curve, or would it end up being pretty similar to Seablock (which includes SpaceX, dunno if that's the same or different).
The larger aspect is effectively restarting repeatedly on new worlds which have unique new resources until they too are shipping rockets of stuff into space. Each new world has different properties. Some have almost no solar, others no water, some have no bugs, but others have bugs dropping in from space anywhere on your base so it's not the same thing each time. I don't mind this aspect. It is a bit like playing Factorio on repeat but it has reasoning behind it. And each restart you get to bring a cargo rocket of stuff with you and the ability to build anything you've already researched.
My outpost worlds are often messy and do the bare minimum to send up rockets of stuff to space. But it works. I'm finding it fun but I'm only a couple of hundred hours in.
Completely different. The SpaceX mod used as an endgame in Seablock has no relation to Space Exploration.
Space Exploration makes pervasive changes throughout the game, and adds the ability to launch yourself into space, build a space station, build colonies on other planets, and mine resources on other planets (and elsewhere) which aren't available on the starting planet. The research required for the first rocket launch is reduced, and there are multiple tracks of research unlocked once you get into space.
In my (normal level biter) space exploration run, my buddy and i didn't run into any problems clearing out a large swath of space to build by rushing artillery and flamethrower turrents. The problem became the tedium of expanding our territory to keep the pollution cloud contained by manually firing artillery.
IMO, the problem should be solved by giving the player better tools for artillery automation. The auto-firing artillery is both too short range, and has a tendency to anger a bunch of nests at once. Instead, it should focus on eliminating one nest cluster at a time, and maybe allow for circuit-controlled timing of when to move into the next cluster.
Artillery (as well as basically everything from the base game) should be considered early game tech: you will later unlock stuff that would make Palpatine salivate.
If anything, you very suddenly run out of things worth shooting at.
> [...] maybe allow for circuit-controlled timing of when to move into the next cluster.
You can achieve that in the vanilla game by blueprinting the artillery outpost (which would include all the necessary supporting defences, like walls/turrets/roboports/etc). Train rolls in, unloads gun+flamethrower ammo, once ammo buffers are ready set a signal to load ammo into artillery, which will start shooting. Set a timer, every time you load a shell reset it, once timer runs out + there are still shells left - you can set off an alarm that the outpost is done. Use a filtered deconstructor to pick the pieces up and move it all over.
Bonus: use a blueprint deployer (https://mods.factorio.com/mod/recursive-blueprints) to automate the "move to the next sector" part. Check out this guy on Reddit to get some ideas: https://old.reddit.com/user/Drogiwan_Cannobi/submitted/
Also, for anyone who’s first question was “how is this legal”:
“When asked about the legality of this whole endeavour, they showed great understanding and allowed the project to be released, provided it won’t be used for commercial purposes.”
> i would absolutely love to see the original source code for
As someone who's seen the source code, I'd say it's relatively good code, but nothing particularly amazing. C++ with multiple inheritance, game objects doing method calls to each other in multiple phases, and a good dose of "maybe not the best way to do this but it works and we're not changing it now" code. What keeps it all running as well as it does is a comprehensive set of automated tests, and Rseding getting on your case any time you make a PR with less-than-optimal code. The interesting bits are the algorithms, which are covered by the Friday Facts.
"Let's game it out enters the chat"
Really though super cool to hear a perspective from someone with inside knowledge!
“ Fixed a crash when downgrading ghost of assembling machine when target machine cannot craft recipe due to missing pipes. ”
“ Fixed a crash when removing modded pipe-to-ground that connects to a shifted pipe-to-ground. ”
“ Fixed that solar panels on multiple surfaces would all produce electricity based on a daytime of one of the surfaces when they were part of a single electric network with a script created wire between surfaces “
“ Fixed a crash when moving blueprint book to blueprint library when there is also another book that will get under the cursor and tooltips are showing “
“ Fixed a crash related to teleporting spider vehicles with burner energy sources between surfaces. “
And so on. There was a blog post a while back that claimed they had fixed over 8,800 reported bugs and after more than four years of development they had reached 0 outstanding bugs.
and https://old.reddit.com/r/factorio/comments/13bsf3s/technical... is fascinating reading (things like: they used doubly linked lists, and multiple inheritance)
Cracked me up
https://physics.highpoint.edu/~bbarlow/courses/phy2100/docum...
(Orange. And Gray Tiles show up in the 'GUI' view.)
This is something I knew, but Factorio taught me viscerally.
The slow pace at the start, and the somewhat lifeless 2D isometric graphics simply didn't do it for me.
But the blog posts were fascinating, and people kept raving about it, so on that 4th install I decided to soldier on and actually get somewhere.
I now rate it as one of the greatest games ever made. It is insanely technically impressive, able to run megabases with trains and belts and robots and inserters and all the rest of it at 60 FPS. It doesn't crash. It's incredibly fun and very addictive. Amazing mods, amazing community.
I can't wait for the expansion. I'll just assume from the start it will be as polished as Factorio is.
It has a faster start, real world processes, terraforming and I believe is better than Factorio in every way. I consider Factorio a genre defining game and while Satisfactory and DSP comes close to surpassing it their 3d nature makes the game's puzzles easier to solve. Captain of Industry makes the puzzles just as difficult as Factorio despite having more dimensions. CoI has a lot more resources you need to manage as well.
I expect some sort of UE5 factorio-esque battle system...
Imagine a game where you run a factory, and you haveto maintain it - while a bunch of PvP FPS shooter are running through your factory battleing it out and keep dmaging your supply lines and you have to fix them - you can choose to join the factory maintenance team, or be a PvP player laying waste to your enemies and fucking up the factory supply lines for the other teams at the same time.
This is kinda what happened in modded PvP Minecraft servers.
But there are already some team based "production line PvP" games like Mindustry and Foxhole.
Some tools are a lot better-- Substance 3D for texturing, for example, especially if you can get their auto UVing to make something acceptable-- but fundamentally most straight-up 3D modelling tools are roughly what they were a while ago. The screenshot of their model seems to have tight topology, too, so I'll bet it was pretty labor intensive.
Just make it 30 or 60fps. It looks awful like this.
I’ve seen some videos where different people try out gaming trials on different Hz’s and it seems like older gamers don’t get nearly as much performance improvement from high frame rates that younger players do. They’re also much less likely to be able to tell which is which in a blind trial.
It's not a problem with games, though. Even if 60 fps was quite rare to achieve for 3D games 20+ years ago. I was usually happy if I could get stable 20-30 fps.
I don't really see the point of this, really, except a misunderstanding about game design, UX and programming in general. This thing might be possible with huge sacrifices, though, but I'm not very interested in making something in 3D "just because". I love 3D games, but a lot of games just don't translate properly to 3D.
Factorio is popular because its hardcore players make immense factories that usually reaches the limit of any CPU/GPU.
That's a bit why I dislike most of the gaming industry, because there are way too many people who aims for high quality 3D graphics without ever understanding software or hardware, and it shows, and it's a big step back for the industry as a whole, but GPU vendors are always happy for those people to push for high end graphics, but not people who like good gameplay.
Also, hypothetically, I don't think a 3D client would impact the simulation since Factorio (AFAIK) runs as a client + server. The 2D simulation would be unchanged.
It does not. Factorio runs as a single process encompassing both simulation and rendering; the "headless server" just omits the rendering bits.
This is just a fun pet project allowing you to visualize your base in 3D. Nothing less, nothing more.
I'm glad the developers have prioritized game mechanics and sim optimization over graphics, and I think the game looks good enough for what it is. but I wouldn't mind if they improved them either. they could do a lot before it meaningfully impacted UPS for megafactories.