nokko

Members
  • Content Count

    10
  • Joined

  • Last visited

Community Reputation

1 Neutral

About nokko

  • Other groups InGame Verified,
    Github Contributors,
    Patreon supporters
  • Rank
    Newbie

Personal Information

  • Byond Account
    nokko

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. PRs that change Lavaland seem to be contested a lot. (#13202, #13013, #13075) It would be a lot easier for coders to figure out what should and shouldn't be added to/removed from Lavaland, if there were a set of agreed-upon-by-maintainers answers to some basic questions about the nature of Lavaland. (And the Shaft Miner job, in general.) I think that Lavaland on Paradise feels sort of incomplete and cobbled together right now, and adding content to it / changing it to be a more unified experience would be good. I'm in the process of asking some players what they think of a fairly major proposed set of changes, and I'll post a proposal once I have something workable that I can deliver as a PR in a timely manner. Knowing more about what exactly we want Lavaland to be on Paradise would help me make a proposal that isn't a complete waste of everyone's time. Here's a list of questions, I'd really appreciate it if some devs and maintainers shared their thoughts about these: What do we want miners to do on Lavaland? • Yes, mine. How much should they mine? • What should they do when no more minerals are needed? • When should they go tendril hunting? • When should they kill megafauna? In what order? ⁃ Why? (In-universe/In-game, what rewards, what incentives?) What loot should drop on Lavaland? • If new loot is added, how should it be themed? • Is the old loot fine? Currently, a lot of items in the loot pool are only really useful to antagonists. ⁃ (Think Staff of Slipping, meat hook, ash storm staff, lava staff. They don't really help miners against fauna, and can't really help other departments do their jobs.) Secondarily, I have two gripes with Lavaland that are less big-picture. Please tell me if you think these are problems that warrant solutions, or if they are fine as-is. Lavaland megafauna can sometimes either spawn or wander very close to the Mining Outpost. In-game, this is a problem. An unlucky miner can easily kite a Colossus, Ash Drake, etc. to the mining shuttle, and cause a station emergency. In-universe, this makes no sense. NanoTrasen have set up a Base Camp on Lavaland already - I'd say that it makes no sense to send a team of miners to an outpost where evil GPS signals are picked up just 50 metres north. Solution: Move megafauna (and swarmer) spawns further from the Outpost. Make megafauna AI try to move away from Base Camp if they are too close to it, and aren't aggroed onto anything. I'll try to write a PR as soon as I figure out the AI code. Tendrils sometimes spawn within chasm distance of one another. In-game, this is very annoying to tendril-hunting miners. Solution: Make tendrils always spawn at least 5 tiles away from each other. Thanks for reading. Please do consider replying with your take on Lavaland, whether you're a dev, maintainer, or you just really like to play a Shaft Miner.
  2. Here's my piece: SciChem as it stands is just completely pointless. 60% of the rounds I play Sci, I see chemists beating up monkeys for no reason, making lube, making space drugs, making toxins, and using the lab for nothing else. Logical: there is no other thing to do, since making like 20 space cleaner grenades (more than a janitor will ever need) takes barely 10 minutes, nobody asks for anything (there's nothing non-illegal to ask for other than healing/cleaning/firefighting nades, which are all different tiers of useless), and there's no consequence if you do or don't do your job, correctly or no. The chem dispensers in that room are more useful to the Botanists (for, y'know, growing plants that Sci needs for R&D) than to the Science department. You can make the argument that it's there to train medical chemists.... I've never once seen it used that way, at least not in any structured sort of exercise. Regardless, even without SciChem, a novice chemist could just build their own chem dispenser and setup with R&D's help, and learn that way. It's... pointless, as it stands. Basically just antag bait.
  3. I'd love more job-specific items, too, but there may be some problems with this proposal. Librarian, Book of Forbidden Knowledge (15 TC): The problem I see here is that Wizard/Cult spells can't be removed from someone who uses this item. So, Sec may have to deal with a confirmed traitor... who keeps Blinking out of perma/execution, and can't be contained. Mechanic, Spacepod Combat Upgrade (15 TC): The Mechanic is a Karma job, giving Mechanic players laser guns and pod armour, (especially given that Space Pod Pilots aren't guaranteed every round) is... yipes, it's something. You could already do this if you have access to R&D as a traitor, but this just makes it trivial. Chief Engineer, The Supermatter Sliver (10 TC): Well.. easy permakills aren't great, one-use portable crematoriums aren't great either. (Even if this one requires rad protection to use.) A traitor CE could take their ID and advanced hardsuit off and replace it with a regular engineer suit, grab a stunprod, and some cable cuffs, and completely punk their target without much effort, and with no chance of them getting back. (Since most people don't do pre-scans.) Chief Medical Officer, Implantable Chemical Synthesizer (15 TC): Might be reasonable, but it all depends on what's on the list. :P
  4. It happens that the problem of figuring out the average number of steps that it takes to get a block to DAC can be modelled as an Absorbing Markov chain, and solved analytically. I'll post the math once I get the code all figured out.
  5. Yep! The times I put in my post were the times to get to at least DAC, or higher. So, obviously, /E../, /F../, /D[B-F]./ all count as well as /DA[C-F]/
  6. Nokko's Speed Strats for Genetics! So, you want to hulk out, see through walls, toss spears and stunbatons at Security, and eat the bolas and flashbangs they toss back. Bad news - completing Genetics takes forever, and is *very* boring. Lots of people want a genetics rework on Paradise, which is fair: staring at a menu all shift and only getting to use your powers on the escape shuttle sucks. Fortunately, Nokko is here with a new strategy for completing Genetics research much faster! (There is some informational prelude to the actual strategy. For a list of steps, scroll to the bottom of this post.) Why is Genetics so slow? At the start of a shift, there are 1-2 Geneticists working in the Genetics Lab. There are 54 unknown DNA blocks that may contain either powers or disabilities. (This means each geneticist must test 27 genes each shift, if they divide the work evenly.) There is a cooldown of at least 2 seconds between each block you irradiate, because the minimum irradiation time is 2 seconds. There is a cooldown of 25 seconds between each DNA injector you print (per console). I timed myself getting a monkey's first 12 DNA blocks to DAC (or higher), here are the results: 50sec, 76sec, 103sec, 17sec, 78sec, 7sec, 79sec, 10sec, 23sec, 80sec, 31sec, 1sec Lowest: 1sec, Highest: 103sec, Median: 40.5sec, Mean: 46.25sec This means that, if it takes 25-103 seconds to get a block to DAC, an optimal geneticist checking 27 blocks, uninterrupted, will spend anywhere from 675-2781 seconds just getting injectors. (11-46 minutes) Plus, however long it takes to test each injector, clean any unwanted disabilities, label the mutations you discover, give Chemistry monkeys, etc. All in all, a geneticist working solo, on one console, can unlock every mutation in ~28 minutes on average, meaning flipping every block except 55 to DAC or higher, and printing injectors for each. However, a geneticist's time might be taken up by other things. SSD, handing out mutations to Security, fighting off randos who want Hulk 5 minutes into the round, cloning people when the coroner is dead, getting killed by traitors, etc. TL;DR: It's slow because of the injector cooldown, the time it takes to get a block to DAC, and the time it takes to test each injector. How can we make Genetics faster? To solve the problems listed on our earlier list, we can: 1. Get more Geneticists working at the same time, or more DNA Scanners per Geneticist, 2. Reduce the search space from 54 genes to a lower number, 3. Don't go SSD, 4. Reduce the cooldown on the time it takes to irradiate a block, 5. Reduce the cooldown of 25 seconds between each DNA injector you print (per console), 1 and 3 are possible in-game already if you have a cooperative Research Director and can coerce a few Scientists into helping you out - the speed increase is obvious. 4 and 5 seem impossible from looking at the code. This guide is about one approach to tackling option 2, reducing the search space. Genes To understand how to find what we are looking for, we must first understand what there is. In the code, there are two lists of genes: GLOB.bad_blocks is the list of all the disabilities*, and GLOB.good_blocks is the list of all the powers. There are 24 genes in the good list, and 24 in the bad. Sources of mutation There are several ways for lifeforms with DNA to get mutations. Radiation storms - Every time a radstorm mutates you, there is a ~95% chance to get a bad mutation. Rad storms are not guaranteed to occur every round - not easily accessible. Radiation (from nuclear reactors, uranium, uranium golems, etc.) - I believe that radiation in itself does not give you mutations, it only gives you radiation damage. Do correct me if I'm wrong, though, I only looked through the code briefly. Somewhat easily accessible if you get R&D to print SUPERPACMAN generator boards+components. The "Pulse Radiation" function of the DNA Modifier access console - 93.25% chance of bad mutation, cooldown 2 seconds. Greater radiation strength does not increase the amount or likelihood of good/bad mutation. Easily accessible. Unstable Mutagen - Only gives bad mutations, if ingested. Easily accessible if SciChem and MedChem aren't being total nerds. You need about 2-6 beakers to get to 24 mutations.** * The Incendiary Mitochondria mutation is technically a disability in-code, so you won't get that one with this strat alone. ** I haven't done the math on this yet, but 2-6 beakers seems fairly reliable. Unstable Mutagen seems to be the best source of bad mutation. Why do we use a chemical that generates 100% bad mutations? Because, of course, once you know the list of bad chemicals, you know the list of good chemicals too - the ones that aren't bad. Synthesis (What the hell is the strat, nokko!?) Get like 4-6 beakers of unstable mutagen. Get 3 bottles of potassium iodide, which heals radiation damage. Get 2 radiation suits, one for you and one for your geneticicist buddy. Feed a monkey all of the mutagen. (You can also use yourself as the monkey, but then you need a cooperative colleague to get you back to normal once you get epilepsy, tourettes, etc.) Count how many mutations the monkey has - if it's 24, that's all of the disabilities! Feed the monkey more mutagen as needed, until you're at 24. Scan in all the disabilities, feed the monkey potassium iodide and mutadone to cure the mutations and radiation damage. Do genetics as normal! Now you have about 45% the amount of genes to check, and all of them are powers! ((54-30)/54 = ~44.44..%)
  7. nokko

    RISSA Tips 2.0

    Rissa C, Can Thermite Grenades be made with Chemistry? How? How do grenades work, in general? Are they useful for Science? How do they generate research/tech levels?