Thread Rating:
  • 1 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Nukeman Program
So let's talk about the nukeman program, and the corresponding nuclear charge. It's a bomb with the power of a maxcap canister bomb, and you can make and set one off in under ten minutes. If you'd like proof, I'll make a video of me doing it. Guide deleted on an admin's instructions. The only thing it requires is spoofing superuser access which if you ask anyone who knows anything about packets, it's literally the first thing you learn. I'm not joking, that link is to the wiki guide. 

Right now, the reason the nuke isn't set off every round is because of the nicheness of packets, and the fact that nothing on the wiki says where the nuke is, though as far as I know it's not secret content (correct me if I'm wrong). I've been trying to get more people interested in packets, and when I get around to it I'll write my own beginner's guide to everything packets. More people will learn about this, and eventually this will become abused.

Okay, I've been rambling a bit. Essentially, there is a very easy way to make something with the power of a canbomb that is only unknown because it's a subject not a lot of people are interested in. My suggestion to fix it is to require a netpass. Either require the netpass head for the actual activation (lorewise it works, three security members/heads need to approve it, then the captain has the activation code) OR require the security netpass for each sec authorization. I like the second one less since the armory is authed with the head netpass, and the security netpass is rather hard to get, especially since the armory ping reply is broken right now. Also it doesn't make as much sense lorewise. 

Anyway, thoughts?
I think the headpass should definitely be required, yeah. It's needed anyways for the way I prefer to set off the nuke, and it just adds another item to the nuclear scavenger hunt (which is good- currently it's a bit too easy to speedrun detonation).
We should probably also go ahead and fix the nuclear charges that spawn on station to actually scan for newly constructed terminals rather than just checking exactly once when the charge is first created and then never again.


Assuming we even want those charges to ever actually be detonated rather than just being window dressing

Double edit:
We literally have a nuclear authentication disk, it should probably be involved somehow in the detonation of the nuclear charge (headpass or however)
Yes, big agree with Camryn there, it makes no sense and is really unintuitive.
Also, it taking the headpass just makes sense. (I think the actual packets that are sent use the headpass too, if one does not use nukeman and instead does it directly by connecting to the nukecharge.)
Headpass makes sense from a lore perspective, but I’m not sure it really solves your problem of “this could be too easy to do given packet knowledge becoming more widespread”?

Getting headpass is currently pretty easy with a small amount of packet knowledge and the steps are detailed on the wiki, and given that the basis of this is “what if people knew a bit more about packets” all this does is gate the nuke behind a packet sniffer or equivalent setup and add one extra command you need to run first.

Requiring the auth disk could be interesting alternative and make sense.

Alternatively, I always thought it lame you could just disconnect the nuke from the wired circuit after authing it to prevent counterplay. If packet knowledge saturated to the point arming the nuke becomes a problem then surely the knowledge to disarm would become similarly well-known: require the nuke to always be connected to the mainframe and maybe the problem should self-balance?
Apologies, I meant the secpass, of course, though the nuclear authentication disk to authorize a nuclear charge sounds good too (and it does have the secpass on it as well). The disk description will become even more inaccurate though.^^
[Image: YL2FeMf.png]
Before I forget, Penne Carbonara/Penny said they were coding a new system for this after they finished the ruck re-work! You can ask them about it, but it'll require the headpass, and then return a secret nuke encryption code. That encryption code can then be used to activate it. Apparently you can also crack the encryption code manually somehow, but they didn't say much about it. 

This will make activating loads of nukes at once much harder, since it's won't just be a matter of a couple of mechcomp components AND it requires the headpass which is a lot more difficult than just superuser access!
The sec-pass would work balance-wise, maybe with also some more diversified material requirements (I assume you mean sec-pass amazingdragons: sec-pass needs a disk or armoury whilst head-pass only needs a packet sniffer).

However, I’ve thought about it a bit more and I don’t think I’d agree this would be a step in the right direction. The game already has a lot of find-all-the-pieces-of-exodia in 30-60 minutes and end the round with little counter-play: can-bombs, certain secret chems, certain cores, certain engine setups, a glove, mass-distribution of TTVs, mass-distribution of certain non-secret chems, buying kudzu and draining the smes, scanning certain things and I’m sure many more*.

Instead of focusing on making nuclear charges harder to set-off and cementing another exodia for a job, would it not be more interesting to push for hacker vs hacker play and counterplay?

The current counterplay to nukes is get off z-level or hide in a locker, with trying nukeman maybe a thing if it hasn’t been deleted/disconnected**. What if counterplay was to preemptively set-up WiFi sniffers that alerted you when a nuke began authenticating, and perhaps spam the terminal with spoof packets? What if the counter to THAT was to go on the run, switching terminals to authenticate incognito. What if the counter to THAT was to detect spikes in power usage on the power terminal and shutdown the relevant apc before it was too late? And what if the counter to THAT was to bridge to offsite mainframes, allowing that some crew member may stumble across mechcomp at the robot factory and raise an alert on the radio?

I guess I feel that we have a lot of exodias right now, and adding a step where you need to steal something from another department doesn’t make it more unique or interesting it just makes it trickier.

Alternate suggestions:
  • Make the nuke require constant mainframe access, pause the countdown if connectivity is lost.
  • Make authing the nuke take a while with some very obvious packet traffic.
  • Make authing the nuke be interruptible somehow - send dummy packets that close the session? Detect people spoofing SU and interrupt THAT somehow?
  • Make it possible to recover the nuke-authing process, perhaps from milestones.
  • Make it possible to trace terminals physical location somehow? CSI style tracing the signal?
Disclaimer: my opinions are worth exactly as much as the next guy.

*Several if these may not currently be possible (or at least I don’t know how atm, maybe others do).

**I have never seen anyone actually stop a nuke like this on main, though I have seen several claim they “probably could have disarmed it, eh” in dead chat. My experiences may not align with yours!

(06-12-2021, 01:41 PM)AmazingDragons Wrote: This will make activating loads of nukes at once much harder

I’ve never seen a round where a single nuke on the stations z-level hasn’t resulted in an immediate shuttle call - have you??? People are lazy when it comes to mass repairs (on main at least)
Should definitely use the authentication disk in some way.

Why does the bomb exclusively have to be the domain of people who know packets, which is a very small % of players and the rest of the crew are never paying attention to?

I don't think it should require more computer use than robuddies. The bomb itself should be a canbomb with extra steps, like three head IDs (genuine ones, not clones) and the auth disk. Not someone spending 30 minutes using packets without engaging with the rest of the station at all.
I'd just love to see more packet set-ups that target each other beyond just "disable mainframe or delete program" and thought this was a good place to suggest it given the nuke's current most common activation method, but maybe people feel otherwise (and maybe it would fall pray to the same issues pathology had). 

The nuke is currently just a can-bomb that uses azone & packet secrets instead of toxin secrets. However it gets changed it would be nice to have more ways to stop it after the silent set-up phase - either after it's been activated or during a long-ish auth process. I feel that even if you add a laundry list of things to scavenger hunt before activating, unless those things come with very obvious warnings to the crew it would still remain another "A can-bomb has been armed in space x3".

A proposal in the other direction:
Cement it as a a potential high-value option for traitors who manage to get the auth disk by fixing the pre-built nukes, stripping out packet functionality and forcing use of the auth disk. Either have it only activateable from the bridge or add to the self-destruct message the rough location of the terminal to get people to try and take the auth disk back, and raise the min timer to a reasonably period (ie: discount nuke ops).
As a note I would prefer for the AI to be able to set off the nuclear charge on its own, but definitely not as quietly as just packet hacking it. Noticeable enough to the point where it's risky, but not so loud as to make it impossible or too difficult to hide.

IMO, good game design rewards the player for doing difficult things, and hiding potentially obvious things is not easy. And, conversely, noticing things that someone is trying to hide can be similarly difficult, and stopping a nuke from going off is pretty damn rewarding. 'Things' is very broad, so as an example of one thing an AI may have to do to initiate a nuclear charge timer: Select several APCs to turn a very obvious color, wherein the AI's core will start trying to send packets to the APC (as it is actually network-enabled!), and as long as the packets are retrieved, progress will be made.
This presents 2 vulnerabilities for the AI to get ratted out: Anyone walking by those APCs/viewing them on camera, and anyone doing packet sniffing (the packets should be EXTREMELY obvious and loud and glaringly caps-locked) will notice that the AI is trying to set a nuke off.

This is all assuming manual charge detonation will require primarily human assistance/work.
a few thoughts:
the nuke has counters, does a station-wide message, is pretty funny, and doesn't actually get set off much?

maxcap canbombs also don't exist rn, there's no agent-b, so this is the only way to actually do this level of destruction off a single explosion.

the stuff preventing it from happening constantly is: knowledge of a certain azone item(or spawning as the right job) and mild packet knowledge - removing the azone item would make it limited to that job (and not really prevent much else interesting). i'm personally pretty fine with that. hell, isn't it an unwritten rule that you shouldn't abuse azone items or they'll be removed?

the nuclear charge just isn't that big of a deal - the only time i've ever seen one set off on station is when i personally did it, and even that involved people successfully disarming it and me rearming it with backups and such(not going into details)

Forum Jump:

Users browsing this thread: 1 Guest(s)