Looking at the AI's core during roundstart

Reimoo · Post by **Reimoo** » Sat Aug 01, 2015 1:23 am #108761

I just came from a malf round where I was killed in less than 10 minutes because someone was looking at my core cams very early in the round. During the subsequent deadchat tears I was told that looking at the core cams at roundstart is pretty ordinary behavior and the rule is to not hack your core APC first thing in the round for this reason. I think it's silly I have to avoid hacking the most seculded APC in the station for fear of someone looking at my core "for no reason"

Now the question here is, is it metagaming to look at the AI's core through the cams early in the round even if you genuinely do have really no reason for doing so? (doing it out of boredom, etc.)

Saegrimr · Post by **Saegrimr** » Sat Aug 01, 2015 1:26 am #108763

Solution: Don't hack your core APC because the RD will check it if he's not stupid.

Scott · Post by **Scott** » Sat Aug 01, 2015 1:29 am #108766

If you can do it why shouldn't you? The problem is someone having access to those cameras.

Scones · Post by **Scones** » Sat Aug 01, 2015 1:40 am #108768

Don't hack your core APCs

Science Officer/RD have a lot of free time and you damn well better believe they will have their faces glued to those cameras

Xhagi · Post by **Xhagi** » Sat Aug 01, 2015 1:44 am #108770

Looking at them roundstart does seem really power gamey if you ask me. Most definitely would be a round ruiner for the malf AI.

A better way to handle it is 'hey where are the borgs I have not seen them at all,' then looking at the borg camera to see an engie it autism forting on the sat next to a blue APC. I did this once as Captain, then DC'd after only getting one line out to sec and came back in dead.

Doing it right at the start ruins the round, catches the AI way early and is valid hunting to the max.

Scott · Post by **Scott** » Sat Aug 01, 2015 2:18 am #108776

How is it power gamey? It's one in a short list of Science network cameras. It's there to be used.

Konork · Post by **Konork** » Sat Aug 01, 2015 2:35 am #108779

Scott wrote:How is it power gamey? It's one in a short list of Science network cameras. It's there to be used.

Just looking at it isn't power gamey. Looking at it at round start, purely in case the AI is Malf and with nothing else to indicate whether it's Malf, is, but it's also something that won't always catch a Malf AI, so it's not really power gamey on the same level as, say, preparing for Nuke Ops because there's 5 less people on the manifest than there were players ready.

Scones · Post by **Scones** » Sat Aug 01, 2015 2:39 am #108780

Aliannera wrote:Looking at them roundstart does seem really power gamey if you ask me. Most definitely would be a round ruiner for the malf AI.

A better way to handle it is 'hey where are the borgs I have not seen them at all,' then looking at the borg camera to see an engie it autism forting on the sat next to a blue APC. I did this once as Captain, then DC'd after only getting one line out to sec and came back in dead.

Doing it right at the start ruins the round, catches the AI way early and is valid hunting to the max.

You are Security Officer Redshirt McScienceguard. You are sitting in your post which has two devices - A records console and a camera console. R&D, toxins, robotics and xenobiology do not require your supervision and you don't have access.

So you sit in your security office.

Why would you NOT watch the AI cameras? It's surveying a risky serious business piece of equipment, there is pefect IC reasoning to do so.

Seriously just don't hack your core APCs

Falamazeer · Post by **Falamazeer** » Sat Aug 01, 2015 3:03 am #108784

Wouldn't it be neat for malfs to be able to loop a normal looking round start camera image onto those consoles?
It'd be harder to check, unless you had a good reason to think something should be in the image that isn't.

that'd solve that for some of the secluded stuff. You'd look and everything would seem fine.

invisty · Post by **invisty** » Sat Aug 01, 2015 5:15 am #108798

Give cameras AI control. They can already control the light, why not the power?

Xhagi · Post by **Xhagi** » Sat Aug 01, 2015 8:54 am #108808

Because to me it's the same as performing random searches in code blue. If the AI has done nothing to indicate that it's malf yet, why would you need to look at it's cameras so early other than to go 'GOTTEM AI IS MALF RUSH IT'? It's no different than batoning an assistant doing nothing in the halls worthy of being baton'd, just walking around and such, and searching to find and emag and going 'GOTTEM OFF TO PERMA/HERE COMES THE VALID BATON.'

How does that make the round fun in anyway for crew or the AI?

A camera loop idea does not sound like a bad one, it would make an interesting power.

Scones · Post by **Scones** » Sat Aug 01, 2015 9:06 am #108809

Aliannera wrote:Because to me it's the same as performing random searches in code blue.

How is it in any way comparable to actually overstepping the lines of the alert code set?

Also I literally just gave you IC reasoning to look at the cameras, if that's not good enough I don't know what to tell you

Sat Aug 01, 2015 10:48 am

People don't look at the AI core just in case the AI is malf, they do it partly out of boredom (may as well flip through the cameras if you got 'em), and partly because the AI is really, really important, and you want to make damn sure no one is fucking with it.

Honestly there are much much much better APC' s to hack at roundstart than your core, or you could alternatively wait a little bit and let people settle in before starting to make moves, either way works

Cheimon · Post by **Cheimon** » Sat Aug 01, 2015 3:26 pm #108834

It's fine that they're looking at the core cameras, you can always just not hack the APC or have a borg disable that particular cam. Most malf AIs don't hack the core APC immediately for this exact reason: it is secluded, but it's also a camera that's checked. It's very visible: you're not forced to have this APC hacked so it's fine if people check it.

The alternative is some weird kind of "I am not allowed to check the cameras in case it would be useful" scenario which is stupid given how we work on knowledge of all other antags.

Lovecraft · Post by **Lovecraft** » Sat Aug 01, 2015 3:44 pm #108839

When I would Captain on the regular and see this I'd turn a blind eye.
Wish more players would, there's nothing fun about charging an unprepared enemy.

Steelpoint · Post by **Steelpoint** » Sat Aug 01, 2015 4:09 pm #108841

Hacking your core APC is akin to a traitor buying their equipment in a public hallway, your just begging for someone to glance at you and catch you before your ready.

ChangelingRain · Sat Aug 01, 2015 4:57 pm

It seems kind of unintuitive that the AI hacking its own APC, in a core that's meant to be extremely secure for it, is as bad as a traitor buying his powersink in the middle of the hallway in front of the captain.
Heck, it's probably more secure to hack an apc that's actually in the main halls, because you could claim it got emagged, and that's honestly really weird.

Hibbles · Post by **Hibbles** » Sat Aug 01, 2015 5:02 pm #108850

I love the camera loop idea. No clue how practical it would be to do, but it makes total sense and would be a cool way for an AI to handle this issue...

,,,that wouldn't require looking at my stopwatch to see if I should ban the Warden for checking a select few cameras yet, or if it's deemed acceptable.

Honestly, AIs are under a bit less scrutiny than in past times. Due to the fact that subversion happens less frequently, rules and policy changed, people who shit on the AI for no reason had to often back off... yeah, people are still constantly paranoid, but AIs feel like they have it better now and get a baseline of trust from most people, most of the time.

Shad0vvs · Post by **Shad0vvs** » Sat Aug 01, 2015 5:52 pm #108859

I can see how a camera loop for the sat would be worth some malf points if you're planning on fortifying it, but you should be distracting the crew with other things probably.

Malkevin · Post by **Malkevin** » Sat Aug 01, 2015 8:17 pm #108866

Camera loop was an idea that was looked at in the past but deemed impossible to implement

Amelius · Post by **Amelius** » Sat Aug 01, 2015 8:29 pm #108867

If the AI is dumb enough to hack the #1 area to look at through a camera if the borgs are acting even vaguely strangely, and someone just so happens to look at that camera, then they really deserve the loss.

You may as well be complaining that no one is handing you your victory. Just hack the goddamn vault, R&D server, arrivals maint, solars, etc. instead.

Edit:

Also, the warden and detective both have a half-dedicated job to look through cameras for suspicious activity in conjunct. A random search would be going to card the AI 'just because' and not examining a room on the station housing expensive, delicate equipment especially after a code blue alert that indicates that hey, the AI may actually be malfunctioning, so it's in NT's best interest for sec to keep tabs on it in a dual sense.

Falamazeer · Post by **Falamazeer** » Sat Aug 01, 2015 8:31 pm #108868

Well, why? I'm not saying an actual loop of some kind, but it's not like you can free eye the damn things, so why wouldn't it be able to pull up a chosen still image of the proper location instead of a live feed?

I'm no coder though, but it doesn't seem like it would be that hard to have the list check var "if looped then NormalAIcore.jpeg" or whatever
And have it just bring up a picture of what the core is supposed to look like if nothing were done since round-start.

CPTANT · Post by **CPTANT** » Sat Aug 01, 2015 9:52 pm #108873

I frequently look at camera's as warden/security officer, including borg camera's.

Often not even because the borgs are suspicious but because they move around the station and let me see lots of things while staying in the office. But I often enough see borgs getting emagged/ killing people/ walling of the AI core next to the deep blue shiny apc. I also often just loop through the camera's and I frequently pass through important station equipment like telecomms/ smess room/ gravity generator. And yes sometimes they are hacked.

I fail to see how looking through camera's at roundstart is powergaming. It's just part of the job to look through camera's as Warden since you usually stay in your office.

But my personal opinion is that the Malf blue apc's are a shit gameplay element anyway. In my opinion it adds nothing to the gamemode.

The AI trying to get into a position as beneficial as possible before starting a fixed timer would be way more interesting.

Incomptinence · Sat Aug 01, 2015 11:41 pm

APC hacking is a really stupid mechanic. One glance at a blue APC by a non idiot and the shit is about to hit the fan. The supposed red herring of traitors hacking APC hardly ever happens so they will know consistently.

You have what vacant office, vault, AI sat, telecoms, gateway and maybe perma early on as places you maybe won't be instantly detected at random by dudes walking by? All those are on camera anyway so a random guy peeking about might see a shade of blue he doesn't like and go code red? It is like if summoning any traitor item turned your skin oompaloompa orange the mechanic is a huge tell from the get go and only the lucky get by with this crap going on.

Arete · Post by **Arete** » Sun Aug 02, 2015 12:04 am #108892

APCs turning blue when hacked is a way for an AI to gamble with a shortening of the delta timer in exchange for greater risk of detection by the crew. The issue is that there's really not much else a malf AI needs to do to distinguish it from a normal AI until the delta timer begins. If the mechanic of hacking APCs to reduce delta timer is kept but hacked APCs turning blue is not, then every malf round starts with an hour of Asimov behavior while the AI hacks every APC on its cameras before starting traitorous shit and keeping a finger poised to activate delta as soon as the call of "AI ROGUE!" goes up over the radio.

CPTANT · Post by **CPTANT** » Sun Aug 02, 2015 12:13 am #108896

Arete wrote:APCs turning blue when hacked is a way for an AI to gamble with a shortening of the delta timer in exchange for greater risk of detection by the crew. The issue is that there's really not much else a malf AI needs to do to distinguish it from a normal AI until the delta timer begins. If the mechanic of hacking APCs to reduce delta timer is kept but hacked APCs turning blue is not, then every malf round starts with an hour of Asimov behavior while the AI hacks every APC on its cameras before starting traitorous shit and keeping a finger poised to activate delta as soon as the call of "AI ROGUE!" goes up over the radio.

Thats why the hacking should be removed all together, malf AI should prepare itself as it sees fit and then activate a self destruct with a fixed time (perhaps with an ultimatum on the time for the self destruct to activate on its own).

abillities should be rebalanced around that.

callanrockslol · Sun Aug 02, 2015 9:53 am

Malkevin wrote:Camera loop was an idea that was looked at in the past but deemed impossible to implement

AI can take photo's, why not just make it a stationary photo of the core?

Fuck it add taping photo's over cameras.

callanrockslol · Sun Aug 02, 2015 9:56 am

CPTANT wrote:
Arete wrote:APCs turning blue when hacked is a way for an AI to gamble with a shortening of the delta timer in exchange for greater risk of detection by the crew. The issue is that there's really not much else a malf AI needs to do to distinguish it from a normal AI until the delta timer begins. If the mechanic of hacking APCs to reduce delta timer is kept but hacked APCs turning blue is not, then every malf round starts with an hour of Asimov behavior while the AI hacks every APC on its cameras before starting traitorous shit and keeping a finger poised to activate delta as soon as the call of "AI ROGUE!" goes up over the radio.
Thats why the hacking should be removed all together, malf AI should prepare itself as it sees fit and then activate a self destruct with a fixed time (perhaps with an ultimatum on the time for the self destruct to activate on its own).

abillities should be rebalanced around that.

You would take all of the tension and challenge out of it if you did that, currently its a game of not being caught while slowly taking over the station, if the crew finds out they have the chance to pretend they don't know and try to outplay you that way, much better than build an autismfort and then siege it out every time.

Saegrimr · Post by **Saegrimr** » Sun Aug 02, 2015 11:17 am #109001

callanrockslol wrote:
Malkevin wrote:Camera loop was an idea that was looked at in the past but deemed impossible to implement
AI can take photo's, why not just make it a stationary photo of the core?

Fuck it add taping photo's over cameras.

Now this is an idea I can get behind, as long as the sprite changes somewhat for people walking by it.

Not-Dorsidarf · Sun Aug 02, 2015 1:53 pm

People need to remember that this isn't about

"Bored science depsec officer fifteen minutes in, nothing happening, lets look around the AIsat"
or
"Warden checking the solar cameras, oh look a bloo wun"

This is about the meta of "Check the AI core camera every round as soon as the time it would take to hack is up" as the RD/scisec

Post by **Jacquerel** » Sun Aug 02, 2015 2:14 pm #109022

"You can't look in this camera until x minutes" is a very silly policy to have, and just moves the goalposts from "looked at camera at round start, meta gaming" to "looked at camera the instant it is not against rules, meta gaming".
Better to just tell ai players "don't hack apcs that are clearly visible to cameras until you don't care that they are visible"

Scott · Post by **Scott** » Sun Aug 02, 2015 2:14 pm #109023

callanrockslol wrote:
Malkevin wrote:Camera loop was an idea that was looked at in the past but deemed impossible to implement
AI can take photo's, why not just make it a stationary photo of the core?

Fuck it add taping photo's over cameras.

A photo is static, camera feeds aren't.

Not-Dorsidarf · Sun Aug 02, 2015 2:26 pm

Jacquerel wrote:"You can't look in this camera until x minutes" is a very silly policy to have, and just moves the goalposts from "looked at camera at round start, meta gaming" to "looked at camera the instant it is not against rules, meta gaming".
Better to just tell ai players "don't hack apcs that are clearly visible to cameras until you don't care that they are visible"

All APCs in the game are clearly visible to cameras, because that's how the AI sees them to hack

Scott · Post by **Scott** » Sun Aug 02, 2015 3:14 pm #109041

Not all of them. See APCs in maintenance, including the Medbay APC.

There are two or three APCs in Cargo maintenance that have no cameras pointing at them and they are visible to the AI somehow.

Post by **Vekter** » Sun Aug 02, 2015 5:15 pm #109063

Trying to enforce this is fucking ridiculous.

Solution: Stop hacking your AI Core APC five seconds into the round.

callanrockslol · Tue Aug 04, 2015 6:10 am

Scott wrote:
callanrockslol wrote:
Malkevin wrote:Camera loop was an idea that was looked at in the past but deemed impossible to implement
AI can take photo's, why not just make it a stationary photo of the core?

Fuck it add taping photo's over cameras.
A photo is static, camera feeds aren't.

That's the joke, it would be like all of those dumb robbery movies and a lot of people wouldn't think twice if they were just checking to make sure the APC was the right colour.

peoplearestrange · Wed Aug 05, 2015 10:11 am

Or you could make a "disable camera" function. Where the AI can hack a camera so ONLY it can see through it, everyone else its just static.
As various cameras are broken at round start engineers/RD might assume its just borked and go fix it (making it into perfect bait) plus it also means an AI can hide it self a little more.

How bout that? Possible in the code?

leibniz · Post by **leibniz** » Wed Aug 05, 2015 11:42 am #109971

I think we could just put the core camera on a network exclusive to the AI.
Borgs could reinforce the AI core more stealthily this way. (though they have cameras too, which they probably shouldn't)
And when there is something going on at the AI core, you couldn't just look through the cameras to check, you'd have to go to the scene. I think this would provide more suspense and interesting gameplay.

Saegrimr · Post by **Saegrimr** » Wed Aug 05, 2015 11:53 am #109973

leibniz wrote:I think we could just put the core camera on a network exclusive to the AI.
Borgs could reinforce the AI core more stealthily this way. (though they have cameras too, which they probably shouldn't)
And when there is something going on at the AI core, you couldn't just look through the cameras to check, you'd have to go to the scene. I think this would provide more suspense and interesting gameplay.

Then what of the times where the AI is calling for help because someone is breaking in, or someone calls for nuke ops?

leibniz · Post by **leibniz** » Wed Aug 05, 2015 11:59 am #109975

Saegrimr wrote:
leibniz wrote:I think we could just put the core camera on a network exclusive to the AI.
Borgs could reinforce the AI core more stealthily this way. (though they have cameras too, which they probably shouldn't)
And when there is something going on at the AI core, you couldn't just look through the cameras to check, you'd have to go to the scene. I think this would provide more suspense and interesting gameplay.
Then what of the times where the AI is calling for help because someone is breaking in, or someone calls for nuke ops?

Well, as I said, people could just go to the AI core.
>but access
CE, captain, any borg, anyone that can open the teleporter

firecage · Post by **firecage** » Wed Aug 05, 2015 12:03 pm #109977

Leibniz. THat is honestly just a bad snowflake idea. "And here we have the latest model of Security Camera Consoles. It can check ANY camera on station to make security duties easier and more effective. Except for the AI core. We wish to give AI's privacy in the event they go rogue. So please do not disturb or harass them."

Steelpoint · Post by **Steelpoint** » Wed Aug 05, 2015 12:15 pm #109983

This is a non-issue, as I said prior just don't hack your fracking core APC until your ready to go loud.

CPTANT · Post by **CPTANT** » Wed Aug 05, 2015 2:56 pm #110003

Its a non issue caused by a bad gameplay element of the mode.

Not-Dorsidarf · Wed Aug 05, 2015 4:03 pm

firecage wrote:Leibniz. THat is honestly just a bad snowflake idea. "And here we have the latest model of Security Camera Consoles. It can check ANY camera on station to make security duties easier and more effective. Except for the AI core. We wish to give AI's privacy in the event they go rogue. So please do not disturb or harass them."

Security cameras can't see the AI core.

The only console onboard which can is the telescreen in the science office

Cheimon · Post by **Cheimon** » Wed Aug 05, 2015 11:09 pm #110085

peoplearestrange wrote:Or you could make a "disable camera" function. Where the AI can hack a camera so ONLY it can see through it, everyone else its just static.
As various cameras are broken at round start engineers/RD might assume its just borked and go fix it (making it into perfect bait) plus it also means an AI can hide it self a little more.

How bout that? Possible in the code?

That's called "tell your borg to break the camera in your core" and it's already something that works because the core itself can see out.

Incomptinence · Thu Aug 06, 2015 12:26 am

Maybe hacked APCs could flash blue or start turning blue as the number of hacks increased because as is an unguarded absolute tell plops onto the station somewhere as the very first steps of a malf AI. First hack last hack all equally obvious.

Gun Hog · Post by **Gun Hog** » Thu Aug 13, 2015 1:21 pm #112079

Cheimon wrote: That's called "tell your borg to break the camera in your core" and it's already something that works because the core itself can see out.

With the AI's sight nerf (Built in sight replaced by internal camera), this only partially works due to the AI being placed into an enclosure which blocks most of its view. Having broken chamber cameras will cause the AI to lose sight of the entrance and surrounding areas.

Post by **lumipharon** » Thu Aug 13, 2015 11:09 pm #112193

Just break the lights in your sat - AI has camera lights, but humans don't.

Also I was malf for the first time yesterday, and even though i had no idea what I was doing, I still managed to hack like 16 apc's without hacking the sat or getting caught.

invisty · Post by **invisty** » Thu Aug 13, 2015 11:39 pm #112202

Incomptinence wrote:Maybe hacked APCs could flash blue or start turning blue as the number of hacks increased because as is an unguarded absolute tell plops onto the station somewhere as the very first steps of a malf AI. First hack last hack all equally obvious.

Now this is clever.

You could have the APC blue for the duration of the hack, encouraging the AI to divert its attention to concealing the APC for a minute or so by preventing the crew accessing the area, being unable to fulfil other requests, etc. Once the hack completes, have the APC permanently unlocked, and the confirming meta-flag that you can't relock the APC.

Balance the time reduction against the modified risk-reward as necessary.