Series Index

Previously: Meat and Plastic

This is going to be heavy on text and light on screenshots (okay it’s all text actually), since this is a break between chapters and gameplay will resume in Part 11 next week. Also going to try to break this into logical groups of thoughts, but no guarantees on that.

On game performance

I didn’t entirely know what I was getting into with this. I’ve done versions of the 100 dupe challenge before, and made it to 40+ dupes a few times, but never had the kind of performance issues I’ve been experiencing this time around. One of the reasons was outside of the game (basically a memory leak in steam causing the game to use the page file a lot) but I think the other cause was the amount of map I had uncovered.

Exploring this much was necessary because of the need to find resources on a map that didn’t have any readily available water sources – mainly to find salt to keep the rust deoxidizers running (until I got the chlorine vent connected to the saltvine farm) but also to find all the teleporters and any other vents/geysers that might be useful. Relying a lot on mushrooms also meant I needed to find slime. A contributing factor was probably the number of ranches I had set up as well. At one point I had over 400 domesticated critters on the map, and even though most of them were confined to small spaces, the combined performance impact added up.

So I don’t think it was one single thing that caused all the performance problems, but a combination of all the less-than-optimal factors that went into keeping the colony running while staying within the rules of the challenge. The early parts of chapter 2 are largely focused on optimizing the base to improve performance, which was somewhat successful – by cycle 290, I’ve managed to get the game running well enough to not have to pause between dupe actions to give them time to think about what to do next, and I can even run the game at 2x speed at times.

On tech choices

If I was on any other map, I likely would have gone straight for electrolyzer tech and not gone through everything I did with rust. This would have made taking on dupes sooner more of an option, and I might even have been able to get away without ranching entirely depending on what resources might have been available.

Getting advanced research and water sieves sooner would have helped a lot regardless of the map. The manual bathrooms were a constant leech on dupe labor for over 120 cycles, and although I had 23 dupes at the time, that was still an issue. Plus, I had nothing I could do with the polluted water that came out of the wash basins, and the polluted dirt from the outhouses just accumulated in storage bins as well.

I also did not make the best use of the resources available; I could’ve had steel made much sooner, with the massive amounts of fossil and already-refined iron around the map. This also could have gone into melting the ice room faster, providing more water to keep a SPOM running instead of spending dozens of cycles debugging the rust deoxidizer setup.

There are some techs I might have been able to take sooner if we had different resources, but there was not much point in researching atmo suits before we had reed fiber, or steam turbines before we had plastic.

On dupe choices, skills, and schedules

I think my choice of starting dupes was fine; particularly, not starting with a researcher never felt like a mistake.

Despite thinking I was getting a lot of ranching dupes, I never got skill points assigned for many of them until they had spent dozens of cycles doing things that weren’t ranching. I also needed to stay more on top of priorities.

I was also not paying close attention to traits, so I took a bunch of dupes that seemed like good mechatronics engineers…with unconstructive. I have something like 3 or 4 dupes that have mechatronics engineering but can’t build anything. I guess if I ever get around to building a power plant, they could tune up the generators.

Schedules were another nightmare to manage, and one thing I’d probably do differently is just have more/bigger bathrooms and stick to only having 4 or 6 shifts. This will probably get shuffled around anyway when we get to building microbases for all the dupes (I will go over some of these plans in Intermission 3).

On base design/layout

This is a trap I get myself in every single game I play: I spend too long trying to have the “perfect” design, and take forever to build things that need to be built instead of just building something temporary and fixing it later. Is overplanning better or worse than the “there’s nothing more permanent than a temporary solution” method?

I’m not entirely satisfied with the layout I have now either, but I’m stuck with it until I get transit tubes. It will likely have a tremendous impact on performance when I can get dupes contained to microbases and only able to access other areas via tube.

There are improvements I could have done sooner had I thought about them a bit more – centering the ranches with the living levels instead of having them all above; putting the main mushroom farm closer to the kitchen; not putting the main trash dump all the way on the one side of the base. But in a game like ONI you also have to work around the map you’re given. I couldn’t really dig more to the right without cooking the base from the oil biome, and I actually started pretty close to the top of the map, so space put a bit of a lid on things in that direction.

Also, given the (entirely necessary) span of the base, I should have set up local storages sooner. I have a bunch of dupes dedicated to just moving things around, and instead of having them spend all their time doing long runs from one end of the base to the other, they could have been doing shorter runs, and at the same time depositing materials closer to where they would be needed later.

On time management and scheduling

Not for my dupes, but for me. I will concede that being laid up for a few days didn’t help, but I could have planned the series better from a post scheduling perspective. I learned that I really need to be at least a week, and more likely two weeks, ahead of schedule when it comes to both gameplay and writing. As mentioned in the Programming Notes I’ve adjusted the posting schedule to both give me time to catch up and to build up a backlog. Performance issues meaning I could only get 20-30 cycles played in a day also didn’t help, but when combined with a 20-cycles-per-post and 5-posts-per-week plan, and a day job, it was just too much. So I think I’ve figured out how to plan better for the rest of this series and for future series.

One thing I’ve found while building up the backlog is that it’s really, really refreshing to just take a diversion off and think about something that isn’t ONI for a while. I am by no means tired of nor burnt out on the game right now, but switching things up has been a big help. And knowing I have room in the schedule for non-ONI posts and a backlog means I don’t have to stress out about cramming through the next post.

Conclusions / Lessons Learned

  • When attempting this challenge, you can probably take more dupes earlier than I did. You will get enough dupes in the long term that taking a few suboptimal ones early won’t hurt you.
  • Getting things working now is more important than making sure they won’t be in the way 300 cycles from now.
  • The game is going to run slow with 40+ dupes no matter what you do, so taking a longer pause around 20-25 dupes and getting more digging and building done at that stage (when you can still run the game at 2x/3x speed) is probably a good idea.

Next: More microbase planning

You may also like