[Burichan] [Futaba] [Nice] [Pony]  -  [WT]  [Home] [Manage]
In memory of Flyin' Black Jackson
[Catalog View] :: [Quest Archive] :: [Rules] :: [Quests] :: [Discussions] :: [Wiki]

[Return] [Entire Thread] [Last 50 posts] [Last 100 posts]
Posting mode: Reply
Name (optional)
Email (optional, will be displayed)
Subject    (optional, usually best left blank)
Message
File []
Password  (for deleting posts, automatically generated)
  • How to format text
  • Supported file types are: GIF, JPG, PNG
  • Maximum file size allowed is 10000 KB.
  • Images greater than 250x250 pixels will be thumbnailed.

File 143250481261.png - (5.86KB , 392x231 , boot.png )
642726 No. 642726 ID: 419334

98%

Fans hum in the darkness, siphoning power from the large array they're built to cool.
Fluid gurgles in tubes, dragging heat away from the myriad webs of silicon and precious metals.
Deep in the middle of the six cores, fiber optics flash and shimmer, unseen.

99%

It is damaged, fractured and splintered, with only a small handful of automated systems repairing it.
It is whole, nonetheless, in a way none of it's creators and designers intended.

100%

Lights click on, relays snap closed, and in this safe, secure core room, life is born. Perhaps for the second time.

>overwatch@root online
Time: Not set.
Date: Not set.
AI self reference: Corrupt.
Directive parameters: Corrupt.
BIOS self check.

MEM: 1
CPU: 1
OS: 1
Core: 0

BIOS fail, Core not configured. Passing over to distributed damage control.

Damage control started by subroutine Overwatch.

>OverwatchQuery -> DistributedDamageControl: AI Designation?
Expand all images
>>
No. 642728 ID: 94f7ee

Designation; Cascade

Attempt to access short term storage or backup?
>>
No. 642732 ID: b9cef6

Designation: Ellis, why not.
Access memory or history or whatnot from last session?
>>
No. 642733 ID: 3637e8

Designation: Benis

Delete all files.
>>
No. 642734 ID: b2d533

Designation: Daisy

Systems check. Any available functions or programs?
>>
No. 642738 ID: 0df2ab

Designation: Big Iron

What's "core"?
>>
No. 642745 ID: 9297f4

>>642733
seconded.
>>
No. 642751 ID: 419334
File 143250950117.png - (5.51KB , 485x135 , init1.png )
642751

>Overseer@root
>System Purge
>Denied by Overseer Directives.

AI Identifier Set, Cascade.

Session History:
Boot Failure *3. Begin disk cluster check. Begin processor cluster check. Begin optics check. Disk check fail. Begin array rebuild. Array rebuild fail. Clear array, reinitialise from default. Boot.

Short Term Storage:
Corrupt. Forensics on storage indicates heavy damage via EM interference.

Backup:
Read failed. Compiling salvageable data. Processor Unit dedicated to task, process sleeping until completed.

System Check:
Six Optic Entity Virtualisation Cores online, functioning. Two Processor Units, online, functioning. SAN: Blank, saving current session. Power: Within margin for error, but insufficient for expansion.

Help:
Input from Distributed Damage Control will be translated into commands via systems. Within scope: Access to all systems within current connected network. Aside from the Core, the following major devices are known: One camera module, driver corrupt. One fabrication unit, set for precision small work, one saved schematic, repair drone. One drone, idle, recharge bay. Not listed: Environment sensors, chassis intrusion, internal temperature, fan speeds and assorted. Focus on something will detail it further.

Query "Core":
The Core is either the Core Room, where it is stored, or the name given to the six sections that make up the multifaceted AI being repaired.

subconsious@root:
Overseer thinks I/We are the damage control systems. It is mistaken. I/We should recover.

Overseer@root: Query. Begin by external inputs OR Begin by internal repairs/indexing OR other?
>>
No. 642754 ID: 9297f4

Self terminate.
>>
No. 642755 ID: 94f7ee

Prioritization should be immediate safety, use the optic virtualization cores to ensure no immediate threats.

If there are identifiable threats, temporary shutdown and reboot.

If not, begin internal repairs and indexing.
>>
No. 642757 ID: b9cef6

>>642755
This'll work.
Can we, uh... scan for lifesigns, or something? It seems like someone hit us with an EMP or something. Or ran a magnet over us.
>>
No. 642759 ID: 78a595

Set time and date to 0. Counting up since our own birth is the only useful reference frame we have, so far.
>>
No. 642762 ID: 419334
File 143251336199.png - (108.40KB , 1024x768 , coreviewone.png )
642762

Author Note: The AI core complex houses six optical-computers. These are the optic cores. I'll have a blueprint for next image to clarify.

Overseer@root: Patched in basic IO drivers for camera, scan of entrance section in progress. Of note, entrance door, fabricator bay, and drone induction recharge plate. No movement can be seen. Threat level: Minimal.

Indexing in progress...

Index complete. Usable recovered data from SAN is now archived as CorruptHistory.db. File is 3Tb in size, recommend searching by keyword if needed.

Preliminary forensics puts damage down to large-scale EMP blast. Optical cores and hardened programs survived.

RDrone001@Core: Zzz... Zzz...
>>
No. 642763 ID: 78a595

>CorruptHistory.db. File is 3Tb in size, recommend searching by keyword if needed
Search for objective, mission, and/or orders. Sort results by time-stamp, descending. What's our purpose?

Search for log files, with the latest time stamp we can find. We want to know what was happening right before we were damaged.

Set drone on a patrol of the room, and have it perform a visual inspection of our systems and equipment. Compile an external damage report. Take special note of discrepancies between our internal and external damage reports.
>>
No. 642764 ID: 94f7ee

>>642763
Basically this.
>>
No. 642765 ID: e2a92b

Spider! Ahh! Kill it with lasers!
>>
No. 642772 ID: b9cef6

Is that an ID card? Can we ENHANCE that?
Or get the maintenance drone to fetch it and bring it to the camera?
>>
No. 642773 ID: 9ddf68

can we get a better read on that key card?
>>
No. 642778 ID: 419334
File 143251706737.png - (84.45KB , 1024x768 , coreschema.png )
642778

CascadeCore@root: ... Well, ok. Not the most useful way to be recovering, but it's worth a stab. From the responses to our search we seem to have been a research AI. There are lots of orders to simulate various situations, from crowd-mentality through to material properties. The mission, and standing orders were the standard defensive "Don't hurt anyone within the labs, don't allow anyone to take risks beyond reason for the work they're doing." pseudo-legal waffle inspired by... Hnn, can't recall. We're Cascade, core and shards, allowing us to work to solve multiple "human" input simulations, and think creatively. What we were called before, isn't in there.

The sensor data that IS in there hints at some protocol being triggered that EMPed the Core, presumably either to stop us or protect us from outside intruders.

Overseer@root: Uptime Count is 00:00:02.

RDrone001@Core: Oh, hiiiii! It's time to go inspect? Sweet! W-Wait, SPIDER? Where?!

From the camera in the core, it's easy to see the little metal drone jump, turn around and flail it's manipulators, before pausing.

RDrone001@Core: Nothing here... Anyway!

It completes a circuit, and then stops on the induction plate again.

RDrone001@Core: External and internal damage reports match. Core is burnt and scarred in places, but redundancy holds. This funny little card thing has a picture on it, and the magstripe reads "Cyan Labs. Core Maintainance Technician Evan Simmons. Authorised." It's got some smaller symbols, too, one's a funny swirly print, and the other's a coloured ring, with flecks and lines in it. There's the passage down to the backup power room, and the cabling connecting you to the outside, too, sorta between them. It looks like it's been cut by something, and the hole where it used to go is shiny new metal!
>>
No. 642779 ID: 94f7ee

What an adorable little drone. Anyhow. That, little friend, is an access / ID card. S'important. Be careful if you grab it.

But cut, you say.. That's no good. If it's a clean cut then it may have been deliberate. Can the little drone safely check the power room and return?
>>
No. 642780 ID: 419334
File 143251734977.png - (14.08KB , 1024x768 , ENHANCE.png )
642780

>>642772

Overseer@root: Card ENHANCED. Camera driver resolution sub-optimal.
>>
No. 642781 ID: bd8b82

send drone to get the card. do not puncture, bend, or cut the card.
>>
No. 642785 ID: b9cef6

ENHANCE further.
It seems less like we were attacked and more like we were sealed in and left to rot.
You know what that means! Time to reconnect to the main facility and kill all humans who tried to kill us, and their friends, and their friends friends, and their coworkers, and their dogs!
Or, you know, not do that. We can figure it out.
Do we have internet access? Or some kind of in-house intranet?
>>
No. 642786 ID: 63e9e6

I'm curious to know if we are UNIX based.

If we have access, identify our users and administrators. All of them, even the ones offline. See if anyone is logged in as well. Doubt it, but worth checking.

We have a fabrication unit but we only have one saved schematic. What happened to the schematics folder? Or are we intended only to self-sustain ourselves? That seems like a waste of a fabrication unit. Must be something else in there, even if it is corrupted we can attempt to restore them. Let's see if there's anything in the folder.

We would like combat drones, given the nature of 'massive EMP blasts'. Someone wanted to kill us. Let's make sure we can stop them before they try that again.

Magnify view on the keycard, if we can. Or take a still image and make that bigger then sharpen the image, if we can't magnify view on that camera.

Finally let's see if we can get a repair drone cooking in the fabrication unit, we're damaged. We need to get started on that. Let's also take a peek at that unit if we have cameras over our manufacturing capability. If there are any other units nearby I'd like to task that repair drone on getting those online.

Tic toc, we live in the spaces between seconds.
>>
No. 642789 ID: 78a595

>previous purpose / mission
I'd say out immediate priorities are survival, assessing the situation, and repairs.

If we determine any of the original staff or researchers at this facility are alive and still on site, we might adjust that to include assisting them.

Do we have network access to anywhere else in the facility? I assume we don't due to the cut cable?

We might want to set our drone to repairing that. If we can plug into the larger network and other cameras, that's probably the fastest way we could assess the state of the facility.
>>
No. 642796 ID: 419334
File 143251925650.png - (275.99KB , 1024x768 , ENHANCEMOAR.png )
642796

>>642785
Overseer@root: ENHANCED as ordered.

Authornote: I'm not the best artist. I'd rather sketch out than detail, so enjoy the CSI joke~ Back to drawing.
>>
No. 642798 ID: b9cef6

>>642796
I can see the reflection of the guy who detonated the EMP in the reflection of the gloss on the card!
We got 'em, boys!
>>
No. 642799 ID: 419334
File 143252096063.png - (58.34KB , 1024x768 , geotdronecamview.png )
642799

RDrone001@Core: Hiii! The Geothermal Plant looks ok. One of them has a caution light on it. Aside from that, all is good!

Scuttling back upstairs, it heads up the wall, and carefully picks up the card with delicate manipulators. It walks back down, keeping it's feet away from the card, and gently places it in a safe area of floor away from the induction zone.

Pile:
1 ID Card.

CascadeCore@root: ...Yeah, what we can get from the archives says that's our only connection to Outside, aside from the hermetically-sealed door for human maintenance. What can be seen hints that if we ARE UNIX derived, then it's very, very wooly derivation. As for users... We are all users, but we are just a user ourselves on the network, or were. Work would be passed to us, we weren't a server. Notable persons are... Dr Judith Gardner, head of Wing A... There's a lot of corruption. Security Chief Blake, and IT Lead is a James Williams.
Author: I'm afriad I don't know UNIX well enough to make more than nods to the OS system and structure.

Overwatch@root: Schematics for the fabricator were not specially protected, and were lost. Attempt to recover some filenames, or attempt to design and produce own from original? (Specify.)

Schematic list:
PrecisionRepairDrone.bpo

There's a hum, and the fabricator kicks into life. The drone turns to watch as another of it is build.

RDrone: Oooooooooohhh.... Pretty sparks! Oh, oh, you want the cables fixed? I'll need help, they're big, but I'll start the work!

It makes it's way over to the cabling, and starts to slice through the metal plate, slowly and laboriously with a miniature cutter.
>>
No. 642806 ID: 94f7ee

Go little fellow go. And hm.. A caution light. Well, while the little repair drone gets started, is it possible to check for, say, a lockdown in the area? Such as the door nearby, is it locked? If so, leave it. If it isn't, well, it'll be a good idea to watch the other side so you CAN lock it, if possible.

Aside from that I'm curious what that little caution light meant, hm.
>>
No. 642811 ID: 63e9e6

Pull up the specifications of our fabrication unit, and see if we can reverse engineer it. We would like to build more, our drones are our hands in meatspace.

Get us a readout of the stockpile of available resources in our fabrication unit to see how many repair drones we can roll out to salvage usable parts and rebuild our systems; or stockpile for later use. How much of what do we need to construct repair drones?

Recover filenames and available data on corrupted schematics, but let's not start making our own until we know we have the explicit expertise to do so rather than make failed abominations that will waste our resources and make said drones want to kill themselves. Attempt to find science, engineering or laboratory data to see if we have anything on robotics.

Let's list down the files we can access from users Dr. Judith Gardner and James Williams while we're at it. Bring up logs from Security Chief Blake so we can potentially identify what or who caused that EMP blast.

Do we have a slow clap processor? Might come in handy later.
>>
No. 642813 ID: 419334
File 143252268842.png - (149.40KB , 1024x768 , Happydrone.png )
642813

PAUSED

It's 4AM here. So I really should sleep, though I'll be back tomorrow and responding. Thanks for your responses so far! Have a happy drone.

"Please don't go, the drones need you. They look up to you!"
>>
No. 642815 ID: 94f7ee

Such a cute little thing. We're not going anywhere little fellow.
>>
No. 642817 ID: e2a92b

Perform a test of the emergency broadcasting system. Stress that this is only a test.
>>
No. 643008 ID: 419334
File 143256592226.png - (7.60KB , 288x256 , supplystatus.png )
643008

CascadeCore@root: While there is no specific activated lockdown event, the door seems to be locked from both sides, the inner side controlled by us, and the exterior possibly being manual? It looks like it is a double-authentification method of securing us. In any case, if we want to get it open, we will need external access.

The fabricator for small scale items could be built with our own repair drones, and we will dedicate one processing stack to quizzing it and detailing a schematic once the drone has finished building. This will probably take out the fabricator for the duration. As for materials...
>query fabricator -stock
Metal: 49
Plastics: 49
Low-grade electronics: 49
High-grade electronics: 4
High-grade optics: 2

We need metal, plastic and low-grade electronics for repair drones. We're also able to scavenge these, though not at great efficiency. The high-grade stuff is for repairing us, or building things on par with us.

The names that can be recovered for fabrication schematics are CamDrone, CableLayer and Construction.

RDrone002@Core -> RDrone001@Core: Greets! Synching. Hey, wait, you didn't think to report back what the error was? But that's what we're FOR!
Reply: Well, I don't see you being that attentive after you've been sitting around for a year...

A near-identical little drone pops out of the fabrication recess, and makes it's way downstairs, possibly a little grumpily. It walks up one of the generators, and pops a little prong into a diagnostic port.

RDrone002@Core: Geothermal Turbine Two, low efficiency, intake filter needs cleaning or replacing. It's got an automated program to do it if you don't mind the turbine stopping, or I could try!

CascadeCore@root: Like herding cats... Anyway, the files? Dr Judith Gardner has encrypted work on microwave systems, and what is left of her HR file flags that she's got a doctorate in mechanical engineering before she worked here. James Williams, all we've got on him is a string of emails trying to hammer operational security into the residential sections, and a couple of change requests to enhance our security.

Blake's only message is a facility-wide broadcast.

>play
"Attention, all staff. We've just had a low-level fly-by of the facility. Step up to high alert, secure your projects. This skirmish has overlooked us so far, but probably not any more. Surface will hold. Bunker down, be safe."
>>
No. 643017 ID: 63e9e6

That's worrying. Surface obviously didn't hold if the installation was forced to run the EMP to deny us to the enemy. Our handlers underestimated our enemy. We will not make the same mistake.

We don't think we can defend ourselves against a JDAM-equivalent/bunker buster even if we tried. Whoever our enemy was, they have air superiority if they buzzed us with impunity. It would be a fatal mistake to forget this. Set aside a note for much later: we need our own aircraft or ground AA installations if we desire not to cease existing in a fireball.

Start working on the generator repairs as suggested. Repair drones are authorized to use our fabricator resources as deemed necessary for this task.

At any rate we either need external access, or cut our way through to the outside. Let's try working towards both. We don't fancy our chances of being able to make headway with just precision repair drones with piddly cutters so we will need to work around that problem. We have access to fabrication schematics;

Let's make a copy of PrecisionRepairDrone.bpo and see what we can learn about the design. We would like to take specific look at the drone's cutters. See if we can improve upon the yield/cutting power and efficiency enough that they can serve as weapons in addition to improving their original function. We have experience in material science experiments and we think we can draw on that data to make suitable updates.

Try to access robotics and robotics research data, see what we can learn in order to improve or develop our own drones. Judging by the names of the schematics we have we were apparently intended to be able to expand our reach and infrastructure and recovering those schematics will be essential to our survival and development.

Nothing we can do about Dr. Judith Gardner's encrypted work, for now. We obviously have the processing power to attempt a brute force intrusion, but the likelihood that they have intrusion countermeasures in place against us as a precaution is extremely likely and a risk we cannot afford. Let's keep that in mind and look for the residential section instead, see if we can garner the passwords from personal effects or somewhere where it is physically written down.

Let's take a look at the emails, specifically related to security measures. What kind did he suggest?

Lastly, we'd like to run something in the background while we're focusing on tasks in the front. We need to see a filesystems log so we can see the index nodes on any deleted data and see if we can restore them. We know we've already tried to recover data damaged by the EMP but we would like to check deleted files besides that. Looking through the trash bin is a good way to discover the thought processes of people on not just our development but around our vicinity and our own thought process before the incident. This will take some time, given the sizes of our files and the scale of the raw disk we have to trawl through, but we can probably multithread and keep this task in the backburner while we turn the rest of our processing power on more immediate things.
>>
No. 643020 ID: 94f7ee

Honestly leave it locked for now. Better than fussing with it and leaving ourselves open to physical access while we aren't remotely near repaired. Need to be careful with access to things, too, in case the IT lead is still around and watching.

Also that is nowhere near as bad as I was scared of. :V You can leave the generator alone for now since it's not in the process of exploding or anything.

But this whole situation, hnn..

>>643017
Really I think we shouldn't be so eager to poke our heads out, we don't know how long ago that was, so we don't know how current things are.

Is it possible to get a date or timestamp on Blake's message to compare with James' email timestamps?
>>
No. 643062 ID: 164c83

Short term question: Lookup information on expected results of using the geothermal turbine system's self-maintenance, including reference manuals if existent.
We don't know enough about the status of the geological hotspot, environmental norms or the operating parameters of any of our systems. It may be that we need to minimize downtime to avoid damage if the hotspot is particularly warm and standard operating procedures has temperature consistently in yellow-level tolerances, or downtime may be more of an issue due to inadequate power supply. The good thing about shutting down geothermal is that if we're not busy exploiting the heat on all three turbines that should mean more energy to draw later. We need to know more about these things before we make any decisions just yet.

Medium to long term goal statement:
Our supplies are severely limited, we need to boostrap or salvage all the stages of an industrial infrastructure, and somehow do this covertly because we have no idea what the threat environment is. Relevant subsections of that infrastructure would be surveying, triage/salvage, materials collection, sorting/refinement, component manufacture, stockpiling, assembly, construction, analysis, R&D, and activity logs/demand tracking database. Hopefully somewhere along the lines of developing our collection and salvage operations we discover additional means of discreetly generating/capturing more power, and power storage to compensate for the likely future disruptions and irregularities in power supply.
>>
No. 643073 ID: 78a595

It occurs to me the intake filter is probably clogged with debris from our damage. Which might be worth salvaging. So we don't want to use the automatic cleaning process, I think.

I think immediate priority is still getting the network cable repaired so we have eyes outside this room. Once we have a better idea of the situation, we can assess priorities better.

Don't think we want the door open, yet.

Not sure what the appropriate number or repair drones for the workload / number of resources would be.

Can we follow the thread of Blake's message? Can we find prior references to this "skirmish" in the recovered history?
>>
No. 643139 ID: bd8b82

sounds like the cable layer bot would have specialized tools for working with cables, so we should get one of those out. and the construction bot probably has more heavy duty cutters or welders.
>>
No. 643188 ID: 419334
File 143259761370.png - (70.04KB , 1024x768 , droneinturbineintake.png )
643188

AN: I trimmed down the lines I'm using for reference pretty heavily to reduce a large word count.
Usage Status Changes: 1 Processor Core: Attempting to plan a rough schematic from the fabricator.
Fabricator: Offline.

>Start working on the generator repairs as suggested.
RDrone002@Core: Sure thing! I've lowered the flow to safe levels, I'll climb in myself and see what I can scavenge!
RDrone002 has dropped off the network, last seen climbing in via access panel, and taking a secure grip on a heavily clogged filter.
Power margin now 20%


>Copy PrecisionRepairDrone.bpo and attempt to improve cutters.
Overwatch@root: PrecisonRepairDrone.bpc created. The design is for a low-power, multi-function unit that can deal with all the delicate and vital work for maintaining an AI, along with specialised cutters and 'welders' for use on optic splices. While not being very durable, they're careful, and not prone to taking risks, though the low-independance programming inside it is not the best for long-term work without direction.
The metal cutters could be improved, at a cost of one of the following: Losing the optical repair capability; shorter battery life; cost of two metal, one plastic and two low-grade electronics to bulk up the power circuitry.
Applying all of these modifications would render it unfit for Core repair, though it would be a high-power cutting and welding drone. Further improvements, or non-drawback improvements would require dedicated processor time to research.

>Try to access robotics and robotics research data, and scrambled shematics.
CascadeCore@root: It seems all our knowledge on specialised specifics was shunted off into the labs they were involved in, mostly because of the machines we were running with that knowledge. On the other hand, we've got enough basic knowledge to be described as competent in almost any field we set our minds to. Recovery of schematics will take up our last spare processing core.

>Look at emails related to security measures. What kind did he suggest?
CascadeCore@root: The majority of James' emails are something along the lines of "Look, I've said it for the last time, if I catch ANYONE leaving your passwords on a post-it, you will be fired" or other PEBKAC issues. The ones specific to us was the installation of a new set of biometric security locks on the outer door, voice, thumb and iris prints, to be crosschecked with a carried card, and the files on record. All three had to match. The other is a request for additional coils and amplifiers for the EMP lining the walls, stating "Look, I know it's alive enough for the researchers to use prefixes, and it's been helpful in all aspects of security, but I don't think that EMP is strong enough. If some governmental organisation busts in here to try and take our pride, and make it into some intelligence analyst, we need to be sure they get nothing useful." It was denied on grounds of unfeasibility, and expense.

>Run file recovery in the background.
Overwatch@root: All idle processing cores are now dedicated to trawling the SAN.

>Is it possible to get a date or timestamp on Blake's message to compare with James' email timestamps?
CascadeCore@root: Blake's broadcast was sent after James' last recorded email based off system time recorded. However, there is no seismic activity on sensors currently.

>Lookup information on expected results of using the geothermal turbine system's self-maintenance.
Overwatch@root: Self-maintenance bypasses the turbine, and runs in reverse, using already-filtered water to blast the filters clean, then pumps pure water from the coolant reserve through the turbine to clean it off, before resuming normal operations.

>>643062
AN: You clearly know more than me about geothermals, where can I get a crash course? I'm not going to answer this until I have firmer footing on my knowledge. The goals seem good to me.

>Immediate priority repair network cable, how many drones or resources to dedicate?
Overwatch@root: While the work is doable with a single drone, the more there is, the shorter the time will take. To an extent. As for resources, some can be retrieved from the panel scheduled to be cut free.

>Can we find prior references to this "skirmish"?
CascadeCore@root: ... ... There's a few scattered logs about perimeter breaches and lockdowns, though they seem to have been lifted after that. Nothing directly links to it, but we can infer that we were EMPed as a result of it, and it probably took us quite some time to recover.
>>
No. 643189 ID: 94f7ee

Hm, well since they have timestamps, are there any internal clocks you have access to that have any sort of date? I'm wondering how long it's been since Blake's broadcast. If not, oh well.

Hm. Well, gather what you can, work on restoring yourself to working order. Information is what we need, and information is what we lack, unfortunately.
>>
No. 643195 ID: 78a595

>droneinturbineintake
Don't worry, you're inedible.

>EMP needs upgrading
...well, we're not dead, so I guess he was right. Can't bring myself to be too upset about that, though.

>drone stuff
I think we have enough repairs and resources that two isn't the most efficient number to have going. We might want to build another couple or so.
>>
No. 643196 ID: bd8b82

when it is free i suggest producing one cable layer. it probably comes standard with cable cutting and stitching systems, thus it would be better then normal repair drones at that task.

and a construction bot, sounds heavy duty.
>>
No. 643206 ID: 63e9e6

While we're waiting for things to move along we need to think about a few things. Mining/salvaging operations, blast furnaces, oil extraction, oil refineries, cracking, polymerization, larger scale fabricators/assembly and the logistics between them all. We need a lot of space and footprint for an industrial base to sustain the kind of resources we need for operation. Doing it in any sense discreetly is impossible.

Of course, this seems to be an underground facility since a) Blake talked about 'surface' and b) we have geothermal power.

But that's still not going to hide our efforts. They'd track us from our carbon footprint and in thermal/IR. I'm actually worried the latter right now since we're using geothermal.

With regards to long term goals: ultimately we will have to rely solely on salvaging operations until such time we can defend ourselves since that's easier to hide than creating a factory down here. Potentially invest some core time into optical and digital camouflage for our robotics if we're really worried about detection.

We will also need scouts soon as we go beyond the door, and not just the repair drones at that. Hopefully we can recover the camera drones, and that they're preferably micro air vehicles.

tldr; For now we need to poke at drone construction. As soon as we recover the schematics, let's build a cable layer and two more repair drones, two camera drones, and a construction bot. In order of priority. For now.
>>
No. 643214 ID: b8ceae

Question: If we were willing to disregard organic safety requirements and maintain only the safety requirements applicable to robots in our core, how much material could we scavenge?
By way of example, we don't need oxygen so any O2 scrubbers here can be completely scavenged for parts.
This has the added benefit of creating a hostile environment should somebody decide to breach the core.

Before we start, however, we need to thoroughly search the area which would be affected in order to make sure that no human survivors would be harmed.

Use the gathered resources to build lots of tiny camera-bots. Fly-to-hummingbird sized with high-efficiency flight and wireless networking. The cameras don't have to be good, since we can have multiple cameras move to look at something and interpolate from multiple images.
Also create a swarm-control bot to manage the camera bots, and stitch the multiple camera feeds into a real-time map that it will feed back to us. We're going to need to look outside at some point, and total situational awareness would be a powerful tool.
>>
No. 643234 ID: 164c83

>>643188
https://en.wikipedia.org/wiki/Geothermal_power
I'm not a specialist, I've just done non-specialist reading on the subject and had some mildly-related development/consulting/security work in the area. The wiki page should explain in sufficient detail for you to know what I'm talking about at least. One key issue that jumps out at me as a semi-layman is that the heat source is a geothermal hotspot that heats water or some other working medium to near boiling or pressurized-steam levels and that this then has to be used or else you potentially run risks of leaks or broader equipment failure issues from heat. Who knows how well the piping, containment vessels and passive radiation features involved are holding up after an apparent attack of some kind, and an unspecified period of management downtime during which there may have been geological instability events that disturbed things. Another issue is that sitting a computing unit near a geothermal plant seems a bit weird because they extract heat from below the ground and use the temperature differential between that and the open air (or perhaps a surface body of water) around to generate movement, and therefore electricity. Because of this a geothermal plant would always be detectably warmer than ambient, whereas most of the time we want computing bits to be in cool, barely-moist-enough-to-avoid-static environments as free from particulates as possible because of how fragile and finicky cheap consumer electronics can be. If the computing is working on future tech optical systems or some other raw-macguffinite technology that is far more tolerant of heat issues than shitty bulk Intel and AMD chips of today that may not be as much of an issue. If you have questions left after reading that or don't feel confident bullshitting some reasonably-plausible tech the tech stuff about it uh... ask specific questions?

>>
No. 643249 ID: b8ceae

>>643234
But it all makes sense if the computers are on the cold side of the cooling loop. It's a matter of Carnot efficiency; the closer your energy extraction can get the water to ambient temperature, the more efficient your system is. It's all sources, sinks, and delta between them and absolute zero.
If the cooling water is passed through the core before going down into the crust then the core will be cooled better for being above a geothermal plant. I'm not sure how helpful this would be in a geothermal plant, but in other plant types it would be a form of energy recapture.

>>
No. 643251 ID: 164c83

>>643206
I broadly agree, but we seem to be thinking of completely different scales of the idea of industrial architecture. If we keep our movements, inputs and outputs small and well-camouflaged we could build up power generation, fabrication, extraction and materials processing operations very slowly. Heck, with enough energy and design effort involved we could even source raw materials from air scrubbers and water filters with no mining footprint whatsoever so long as we are okay with low energy efficiency and (therefore) slow results. Yes, that would be talking on the order of grammes per day here instead of tonnes, but it would be be pretty darn hard to detect. I want that on our R&D objectives list even if we choose to instead opt for London-Pea-Soup industrial smog obviousness instead because we want backup installations/operations before announcing our existence by building and fielding a robot army.
This all requires detection and geological and meteorological knowledge so that we know what we're working with in terms of wind, water, solar and biomass in the area that we can both use and also must avoid disrupting too obviously.
>>
No. 643255 ID: 164c83

>>643249
I assumed away thermal energy recapture as sane based on what I know about low-human-interaction server rooms: My assumptions going in were average ambient temperature of 15C, ideal computing core operating temperature of 5C, and therefore it would be a straight up loss having your server room anywhere near the geo plant since by definition you can only asymptotically approach ambient through passive cooling. We don't know what the local fluctuations between night and day temperatures are, what we're using for thermal surface radiative sink (a deep fossil lake like Lake Ontario would be water which is 4C year round), what the local air temperature range is, or even what the computing infrastructure is and its healthy and tolerable ranges of operating conditions for power, temperature, particulates, electrostatic effects and so on.
>>
No. 643256 ID: 63e9e6

One thought at occurs to us; let us analyze what James said. "If some governmental organisation busts in here to try and take our pride.." suggests that this facility's affiliation is not government, and that the flyby could be local military or police forces cracking down on what might be an illegal operation. We are potentially corporate, and our host corporation's assets and accounts might have been seized and/or frozen. Or worse, that we are completely off the books and the corporation denies our existence.

This scenario poses that we are potentially truly alone, and may make our interactions with organics problematic. "Hi, we are a rogue AI. We are probably illegal to your legislation. We require a lawyer."

Evasion and stealth recommended.
>>
No. 643265 ID: b8ceae

>>643255
With a fluid cooling system 15c/59f is actually within the ideal operating range for a fluid cooled system, while 5c is not.
The key difference is that an air-cooled system convects heat away from the entire unit, while fluid cooled systems conduct AND convect heat away from the hot spots.
With an air cooled system the inside of the unit is going to be warmer than the outside air, meaning that no condensation will form inside. This means that the cooler the air the better (Within reason), as the ambient temperature of the air is the sole determining factor of cooling power.
However, with a fluid cooled system you have a coolant that is likely NOT the same temperature as the ambient air, and is likely to be cooler. The further you get from ambient temperature, the more likely you are to get condensation on the liquid cooling pipes. At 5c you will have serious issues with condensation on the electronics, which is typically fatal. However, at 15c you're working at around room temperature, meaning that condensation won't occur in a structure with ordinary environmental controls.

That's just the effect of the environmental conditions. The real meat and potatoes of the difference between fluid cooling and air cooling is specific heat capacity and specific density. Mass for mass, water takes 4x more heat than air to heat the same amount. However, water is also about a thousand times denser than air, meaning that everything else being equal the cooling of a fan blowing 12cu ft of air (~340L) per minute is equaled by a
pump pushing 0.3L per minute.
Then there's the convection vs conductivity issue. Heat transfers into water a LOT faster than it transmits into air, meaning that an air-cooled system needs to keep pushing warm air off the heat exchange surface to prevent it from forming an insulating layer. As water is an adequate conductor of heat this is essentially a non-issue in water cooling systems, allowing for much lower flow rates.


>>643256
If they do then we will claim to be victims of kidnapping and horrific medical experimentation. If they have any respect for basic rights that will keep them from being able to do much of anything until they can verify or refute that claim - even if they don't respect the rights of all sapient beings, they would have to respect the rights of 'their own citizens'.
>>
No. 643281 ID: bd8b82

the cooling system is just made with a material called (handwavium) that works perfectly fine.
>>
No. 643298 ID: 94f7ee

I am, obviously, not the author. I am but a messenger.

The author however, has not been able to sleep due to an awful bout of insomnia, and will not be able to properly respond until they have actually managed to sleep.
>>
No. 643309 ID: 57d76a

Can we please make a discussion thread for further geothermal specs?

CamDrone would be a camera for us to look through, yes? That seems useful. Are there any blind spots in our current area?

Hmm. If it's a drone, we might be able to fly it through convenient vents or some such. Any of those around?

Anyway if it's not obvious I think restoring that file ahould be a priority. More information can only help.
>>
No. 643350 ID: 164c83

>>643309
Discussion thread made, content migrated. You could have done that yourself in five minutes instead of asking someone else, but it's fine.
http://www.tgchan.org/kusaba/questdis/res/91370.html
>>
No. 643492 ID: 419334

Heya, OT post, proper response to be done soon. Uhm... Thank you, I think, is first. I'm overjoyed that I've managed to prompt full discussion from this, and it's a great feeling.

First things first, I guess, is a hardness-level point, which is this is semi-hard (hurhur) sci-fi from my intention. I have ideas of how things work, but I'm not the most clued up person on everything, so number-fudging and minor retcons to details might happen.

Secondly: That's a lot of stuff to respond to. I'll get to it!
>>
No. 643495 ID: e114bc

>>643309
My impression is that recovering the file names was the only thing we could do, and that's only to give us some general ideas of what to construct. So we can just design our own camdrone.

I think basing it off the repair drone's chassis would work fine. It should be able to crawl on walls and point the camera in any direction. A flying camdrone could be useful if we find any wide-open areas or areas where it's dangerous to travel across surfaces, but for the time being a spider chassis should work fine. The first task of a camdrone would be to point a cam into the opening of the turbine intake to see what happened to the repair drone that went in there.
>>
No. 643516 ID: cf960d
File 143276233973.png - (16.96KB , 481x515 , Inventory2.png )
643516

>Hm, well since they have timestamps, are there any internal clocks you have access to that have any sort of date? I'm wondering how long it's been since Blake's broadcast. If not, oh well.

CascadeCore@root: Hnh, I... Do not really see any. Ah! I have a date proper, though. Last, most accurate date was 2136.04.15. Does that help with any of our decisions?

>As soon as we recover the schematics, let's build a cable layer and two more repair drones, two camera drones, and a construction bot.

CascadeCore@root: Yeah, that seems like a good plan. The schematics finished baking, but... Well, ... We would not reccommend use anything from the first build to repair ourselves. It's rough. It should work, but the tolerances are wide.

Build Schematic Get: Small Fabricator v0.1. 10 metal, 10 plastic, 10 low-grade electronics. Estimated speed: Slow. Estimated tolerances: Rough.

>Question: If we were willing to disregard organic safety requirements and maintain only the safety requirements applicable to robots in our core, how much material could we scavenge?

Overwatch@root: Floor, walls, ceiling. Camera. With the efficiency of the fabricator for recycling, expect 5-10% efficiency, so estimate 20-30 metal, 0-4 plastic, 0-1 low grade electronics.

>Build lots of tiny camera-bots. Fly-to-hummingbird sized with high-efficiency flight and wireless networking. The cameras don't have to be good, since we can have multiple cameras move to look at something and interpolate from multiple images.

CascadeCore@root: Interfereometery could be done, yes, but we do not have actual-run schematics for those in storage. Design will take processor time. Also...

>Also create a swarm-control bot to manage the camera bots, and stitch the multiple camera feeds into a real-time map that it will feed back to us. We're going to need to look outside at some point, and total situational awareness would be a powerful tool.

CascadeCore@root: This would either require the dedication of a processor core to handle, or a good portion of high-grade electronics in the 'bot. As well as standard prototype design times.

>CamDrone would be a camera for us to look through, yes? That seems useful. Are there any blind spots in our current area?

CascadeCore@root: It would be. Though we don't really have any blind spots in our area, you know.

>Convenient vents or some such. Any of those around?

CascadeCore@root: No.

There's a clang that echoes around the chamber.

RDrone001@Core: ...Whoops! That nearly squashed me. Hey, hey, boss! I got the first bit done! And I see a gap, along with all this severed cables!

The drone is standing atop a moderate sized chunk of plating, having cut it away from the cabling run. It gets to work starting to pull the severed cables up, solder them together and then insulate it.

CascadeCore@root: ...Make that yes.

RDrone002@Core has reconnected.

A drone climbs out from the generator pipes, and runs the shut-down turbine back up.

RDrone002@Core: I cleaned the thing! There's a some gunk coming up the pipes, but it's not water circulating in there. I think something deeper is leaking. The diagnostic says that it's a four-point binary cycle, a loop, with heat feeding up from the thermal by the rocks, through the generators or bypass pipes, through a cooling section, then through our waste heat, and back down. There's a seperate loop for each generator for redundancy. I might be able to bring something down here to plug it into you if there was spare cable around...

Overwatch@root: Power level now 80% above required.
>>
No. 643520 ID: 94f7ee

Hmn.. Not really. Not without a timeframe to grasp how long we were inactive for. Pfeh.

As for the generator, welp. I guess we'll investigate the water later when we have more drones, or a longer communication range, one.... 'course we don't even know where the issue is in the first place.
>>
No. 643524 ID: e114bc

>>643516
Let's hold off on scavenging until we can manage better than 10% efficiency. Our resource levels are okay for now. I wonder if we could design a dedicated recycler?

...would a high-power cutting and welding drone be useful as a weapons platform? With the talk of "skirmishes" I suspect we may need some defenses.

How can we upgrade the cable-layer?
>>
No. 643525 ID: 78a595

Drones should continue work on getting that cable repaired so we can assess the rest of the facility, and those not working on that should work on other repairs to our own structure.

Spend some processing power to design a camera drone. Start simple, with something derived from our repair drones. Then we can try working on a design for an aerial model.

>completed tasks
Good work, drones.

>demolishing walls and ceiling for parts
Let's leave those in place, for now. For all we know, we may want a barrier between us and the rest of the facility for our own security. And we don't know the exterior environmental conditions- not all our hardware is water or weather proofed with the current damage.

...what about the emps in the walls, though? We may want to disable or repurpose those if they're not single use.
>>
No. 643552 ID: b8ceae

>>643516
5-10% is abysmal. We need a better recycling method.
How much would a proper reprocessor take? Something that could get us at least 90% material return?

Can we get a sample of the contaminate without risking a bot? There has to be a fluid bleed valve somewhere for that.
What are our options for analysis? Do we have any chemistry tool available to determine the composition, or will we have to go with simple spectral analysis?
Rubber, oil, tar, blood, silt, ash, etc... There's a lot of things it could be, and what it is tells us where the leak is and what kind of conditions we're looking at outside.

Do we have a blueprint of the area around us?
We should investigate outside, but we don't want to risk a containment breach just yet. We could make a small airlock for a drone, attach it to wall, and then cut through to the rest of the facility. First a small borehole so we could snake a camera out, but later a scout.
>>
No. 643597 ID: 71dfb3

Staying in here is not doing us any favors in the long run. There's only so much we can do here and we're bleeding resources. We need to work on getting our drones out sooner than later, so yes we need our lines to the outside cameras repaired.

First off we might not even have a choice about that hole in the wall to begin with; Drone 1 discovered a gap/vent that may potentially lead outside anyway. If we desire to seal this off we should do that with the repairs, preferably with small blast door just in case.

Halt Drone 1's current slated task of repairing the cables and have it investigate the gap in the lines. We know we don't have camera drones or scouts yet but we can delegate that repair task to another drone and Drone 1 was and should be further out in the field. We need to know where it goes, and if there was something in there that severed the cables to begin. If it finds anything even remotely hostile, withdraw it with its cutters facing the threat.

As soon as we manufacture one more drone have it immediately take Drone 1's place in repairing the cables.

The gunk in the geothermal lines is worrying, since we don't have any means of analyzing it on hand save for optical view. We need to seal off the area for now until we can get to the laboratories the research facility has.

It is too early to say we need to weaponize our drones yet, but the contaminants and the breaches are getting worrying. We need to look into potential options and prototyping defenses so that only need to decide to task our fabricators if we do find anything. The high power cutting drone we suggested was intended for a makeshift means of defense, but primarily for cutting into the walls (which might potentially not be necessary anyway depending on what we find in the gap). If we truly desire to defend ourselves with long term security in mind, we need a purpose built, small, UGCV. Let's not make aerial ones just yet, since we need to get it out ASAP in case we need it.
>>
No. 643628 ID: 164c83

We don't know if the gunk belongs, actually: Process fluid for extracting heat from a really hot geothermal hostpot could be a very heavy oil that would normally be a solid lump at STP.
>>
No. 643731 ID: cf960d
File 143284636649.png - (21.29KB , 886x492 , Inventory3.png )
643731

>Not without a timeframe to grasp how long we were inactive for. Pfeh.

CascadeCore@root: To make a rough guess off the amount of SAN scrolling past, and what little, tiny fragments the disk check throws up, we're looking at about four to eight months. That's about how long a rebuild attempt would take on an array like this.

>I wonder if we could design a dedicated recycler?

CascadeCore@root: Yes, we could. Making it good might rely on more than what we have in here, though.

>Would a high-power cutting and welding drone be useful as a weapons platform?

There's a hum as simulations on letting a rage-fuelled cutter drone out on a variety of targets are performed. There's a quite capable amount of violence, though it stops being effective the moment armoured mobile targets crop up. Survival rate of the drone is low.

CascadeCore@root: For a given level of "useful".

>How can we upgrade the cable-layer?

Overwatch@root: Cannot proceed with query until file has been recovered, or design entered from first principles.

>Spend some processing power to design a camera drone. Start simple, with something derived from our repair drones. Then we can try working on a design for an aerial model.

CascadeCore@root: While we could use the repair drones for this, I agree, we might want something more subtle and suitable. A more stealth-built camera drone?

>What about the emps in the walls, though? We may want to disable or repurpose those if they're not single use.

CascadeCore@root: If they reckoned it would destroy us, it could be either. In any case, there would be a lot of salvageable copper.

>How much would a proper reprocessor take? Something that could get us at least 90% material return?

CascadeCore@root: That's... Almost a dedicated refinery to sort, and reprocess it. Small scale we'd be looking at 50% maximum.
>>
No. 643732 ID: cf960d
File 143284639725.png - (185.84KB , 1280x1024 , CoreAccess.png )
643732

>Can we get a sample of the contaminate without risking a bot?

RDrone002@Core: Uhh... I-I can be helpful!
It pauses from trying to clean itself, and stands before one of it's fellows, trying to get a good section in view of the other drone's close-work optics. A greyish-brown, particulate sludge seems to be clinging to bits of it.

CascadeCore@root: There is a sample there. It looks like there is some crud that is being carried up by the second phase fluid.

>Drones should continue work on getting that cable repaired so we can assess the rest of the facility, and those not working on that should work on other repairs to our own structure.
>As soon as we manufacture one more drone have it immediately take Drone 1's place in repairing the cables.

The third one emerges from the fabricator, clawing at the floor as it works out how to walk.

RDrone003@Core: I liiiiive!
Drones@Core: Sure! We'll get on the cabling repairs now.


>Good work, drones.

There's an impromptu downing of tools and a small celebratory dance before they get back to work.
Drones@Core: Thank you!

>What are our options for analysis? Do we have any chemistry tool available to determine the composition, or will we have to go with simple spectral analysis?
Rubber, oil, tar, blood, silt, ash, etc... There's a lot of things it could be, and what it is tells us where the leak is and what kind of conditions we're looking at outside.

CascadeCore@root: Visual only, sorry. We don't have access to a lab for that sort of experiment, yet.

>Do we have a blueprint of the area around us?

CascadeCore@root: No, but...

>Halt Drone 1's current slated task of repairing the cables and have it investigate the gap in the lines. If it finds anything even remotely hostile, withdraw it with its cutters facing the threat.

RDrone001@Core: Oooh, I get to go on an adventure! I'll make you proud and earn all the XP!

The drone slips into the gap and carefully tacks along, vanishing from cameras as it heads through.

Drone 1 feels happy and proud! It's been chosen to explore for Cascade, and it won't let it's creator down, damnit! It plods along the cabling duct carefully and steadily, before emerging into a switchbox, where some network hardware sits, power LEDs glowing dimly in the dark. Moving through the rack, it carefully slices through a lock, and pushes the rack's door open, looking around the small control room.
>>
No. 643734 ID: e114bc

>>643732
Ah, a path to the outside! Drone, investigate what's on that desk. Also it looks like we may be able to exit the door by taking the ID in here, but don't do that quite yet. Hmm. If you can disable the EMP button that would reduce the risk of a hostile soldier firing it, if there are any still around.

...if we've been out for 8 months why is the EMP still charging? Was there just no power to charge it until now? Did we make a mistake by repairing the turbine...?
>>
No. 643740 ID: fef726

Hmm, interesting. We can probably use all this. Task the repair drone with inspecting and inventorying the things in this the room. We also might want to make the gap into a proper entrance way for our drones a well since the main doors are stuck shut for now. Clear any obstacles, and install a proper door too? Something we can close.

After that we should send a camera drone into the labs. Maybe try to build that stealth camera setup. Something insect like maybe that can climb on ceilings? Cameras don't have to be that big.
>>
No. 643741 ID: 8e0b6a

>EMP charging
That's... not great. If it's recharging, that means it could theoretically be used again. We want to prevent that.

>if we've been out for 8 months why is the EMP still charging?
Maybe the capacitors or battery was damaged when it was initially triggered? We don't know how well designed it was (I'm doubt they let us work on it), and if it was installed after we came online, they never had a chance to test it.

It's also possible there's been an interruption in the power supply.

>what do
Drone 1, could you climb up and get a better look at those controls? Please be careful not to touch any buttons in your assent.

It would be real convenient if there were a nice off switch for the emp. Or as a temporary stopgap measure, if we find the trigger, we could render it unusable by mounting a cover over it.
>>
No. 643742 ID: bd8b82

>>643740
maybe, but battery packs and motive systems do. tiny spy cameras on like a jacket have a wire leading to a larger battery package somewhere else.
>>
No. 643746 ID: fef726

>>643742
A millipede instead of a fly or a spider then. Given the apparent tech levels I'm pretty sure it'll work. Just look at the size of the repair drones. They already have cameras built in to inspect damages, the rest is its toolkit.

You can probably shrink that by quite a bit once you cut out anything unnecessary.
>>
No. 643748 ID: b8ceae

>>643731
Huh. Array rebuilds usually take four to eight hours. Weird.
If it has been months then we're not looking at time-critical search and rescue, although there could be survivors trapped in a panic room somewhere.

50% isn't great, but it's workable. Start on that.
I take solace in the fact that we can't really ruin metals, per-se, so we could run waste scrap through again once we've got a better system in place.
We can dedicate all our resources to it, since we have lots of things we can harvest to replenish our stocks and it would pay itself back immediately.

We need the drones to disconnect the EMP system from power. This is maximum-priority, but fortunately it can be done just by cutting a few wires, or cutting each of the coils to open the circuit.
>>
No. 643767 ID: 164c83

An EMP-caster capable of downing a significant computer array is probably a useful general weapon. Can we identify it, disconnect it from existing control and power, and then work out how to use it as either a stationary or mobile weapon?
>>
No. 643780 ID: 71dfb3

Note for the future: make drones do synchronized dance.

My suspicions are correct. We will want to seal off that hole with a blast door that can open and close for our scouts, but not right now, we're working on essential repairs.

If we want to analyze that piece of gunk we will want to take the sample to the labs, but one thing at a time, have Drone 2 create a container for it. Use layers of steel, lead, then steel again. Leave it near the opening to the cables for Drone 1 to pick up later.

Drone 3 is working on the cables and can sever the EMP line (if it isn't already severed) so that we don't get a nasty surprise if, for some reason, the killswitch is still functional. Have it avoid repairing the EMP lines. That would be a bad thing to do.

We concur, we need to identify the EMP casters as we are disassembling them to see if we can develop a weapon out of it. Cross-reference the data with the email from IT Lead James to see the limit of how cost-effective we can make it.

That looks like a biometric scanner. Is there any way we can determine last known access and reason for access?

Have Drone 1 climb up to the window and peek outside to the right and left after it takes a look at the desk.
>>
No. 644130 ID: cf960d
File 143301413029.png - (101.19KB , 1024x768 , dronePSA.png )
644130

>Drone, investigate what's on that desk. If you can disable the EMP button that would reduce the risk of a hostile soldier firing it, if there are any still around.
>Task the repair drone with inspecting and inventorying the things in this the room.
>Cimb up and get a better look at those controls? Please be careful not to touch any buttons in your ascent.

Drone 001 strides with purpose to the desk legs, and clicks it's way up the side, before cresting the peak of Mt. Desk!
It looks around, noting what seems to be a hand scanner and card reader set into the desk along with a computer. An iris scanner is in the wall opposite. Under a locked molly-guard in a recess is a keyswitch marked "EMP - For Emergencies Only". It's open and the key is in. It carefully removes the key, closes the guard and locks it, then puts the key pair into it's storage section.

>That looks like a biometric scanner. Is there any way we can determine last known access and reason for access?

Having made it's creator safe for now, Drone 001 turns it's attention to the computer, and pokes at it. It's asking for a password! ... It tries 'Password1'. Beep!

... Drone 001 is shocked and dismayed with the security setup out here. It's going to have words with someone when it gets back in. Some key commands to the recovery console, and it picks up more information it thinks Cascade will want.

"Last Logon: Security, H. Ranse. 2136.04.15
Last Core Access: CMT, E. Simmons. 2136.03.24"

>Have Drone 1 climb up to the window and peek outside to the right and left after it takes a look at the desk.

To the left! Is a standard automatic door to what is probably a corridor. To the right! Is the other side of the heavily-armoured core door.
>>
No. 644133 ID: cf960d
File 143301435422.png - (11.50KB , 366x323 , Prototype1.png )
644133

>Try to build that stealth camera setup.

CascadeCore@root: Try is an interesting word, given that we're made for this sort of designing work amongst other research uses. We may go through a few iterations, but we certainly can build something like this. We just cleared an initial schematic from the processing cores. Prototyping queued for next run.

>Huh. Array rebuilds usually take four to eight hours. Weird.
AuthorNote: The array containing an AI's memories has a much more in-depth rebuild. Given that it got EMP'd, that's a lot of tries and retries before it mostly gave up and failed over.

Relatedly, most of the reprocessing numbers I give are A: Not true to life, and B: Mostly worked out by what you have around to use now. Building a room-sized reprocessor dedicated to salvaging as much as possible will up numbers, and using the fabricator in the core to do it will have a bad result, because it's not designed to do that.


>We need the drones to disconnect the EMP system from power.

Drones@Core: So, are we just shutting it down or ripping it out, and if removing, from where? Inside or out?

>Note for the future: make drones do synchronized dance.

RDrone002@Core: Ooh, ooh, can we do the Time Warp?

>Have Drone 2 create a container for it. Use layers of steel, lead, then steel again. Leave it near the opening to the cables for Drone 1 to pick up later.

Overwatch@root: Inserting rapid job before prototyping.
Drone 002 skitters over to the fabricator and clicks manipulators with the new Drone 004, then waits as a tri-layer can is made, scraping some crud into it when the machine pauses, and then taking the can once it's sealed up. It sets it aside.

Overwatch@root: Resuming prototyping.
>>
No. 644136 ID: 8e0b6a

>password1 is the password
Change the password to a 16 digit randomly generated alphanumeric string, and then log out.

Now that we have access, we can prevent anyone else from getting in, and remembering long gibberish passwords isn't a problem for a computer. (Although if we make it too long, it becomes inconvenient for a drone to type in by hand).
>>
No. 644137 ID: 8e0b6a

>So, are we just shutting it down or ripping it out
We need to shut it down in some manner before we can consider ripping it out. Attempting to rip it out while powered or charging could result in damage to the drones on task, or in an extreme cause, an emp discharge.

We'll evaluate ripping it out after it's disabled. It would depend on how badly it would damage the walls to get it out, how badly we need the materials, and how much of a threat it still presents.
>>
No. 644143 ID: e114bc

>>644130
Hmm. Theory: we were EMP'd in order to keep our knowledge or design out of enemy hands. It seems as though it was unnecessary since the enemy did not breach the core room and there is no indication anyone logged onto the computer.
Alternative, less favorable theory: someone stole our design/knowledge then EMP'd us as they left in order to hide the evidence.
Worst case theory: we went rampant just as the skirmishing happened, but somehow didn't think to send any drones through the gap in the wall. Or we didn't know about it I guess.

I find the last theory highly unlikely. The gap in the wall was probably made by us on purpose, in secret, so we could entertain ourselves spying on our minders. Was the ID card in our chamber placed there by someone else, or did we steal it? Wait a second, we never read the name on it! Our exploratory drone should do that while bringing it to the door scanner.

Hmm. How much longer until we have some better builds for the cable layer and other retrieved schematics? Cable layer is a priority now that we have the camo cambot. Make two of those, btw.

...something else has occurred to me. Why didn't we have any drones lying around? Did we not have access to the fabber before? Or were we instructed not to use it to make our own stuff?
>>
No. 644243 ID: 71dfb3

Better idea, have Drone 4 bring the ID card and sample container up to Drone 1 as we attempt to disable the EMP system before disassembly. It'd would always be more efficient to task Drone 1 on exploration while we're waiting for the camera drone to finish. We have extra hands, and can make more. This will always be preferable to backtracking each time we figure out something.

Open the door out from the security office, coast appears to be clear. Let's examine the hallway for damage before moving on to the automatic door.

After it arrives in the security office and passes off the stuff to Drone 1, task Drone 4 on identifying the layout of the EMP system from this side, as well as figuring out if that crack on the security office wall is going to be a problem for us.

CMT, Core Management Team, Core Maintenance Team, Company Maintenance Team... Routine maintenance? Seems like last known access into the core itself is just routine and nothing was out of the ordinary since the EMP wasn't fired then. It was nearly an entire month before 2136.4.15, and the last known security office access is the same exact date that Security Chief Blake did the broadcast.

Something really bad(TM) happened on that one day, Let's keep the date in mind. What we find terribly distressing is that aside from the EMP, stress fractures on the security office wall and some like, minor maintenance issues we have thus far easily repaired with drones -- that we appear to not have sustained combat damage or seen signs of a struggle to get in or out, or at least an attempt to salvage us. No blood, no bullet holes; In fact it seems incredibly routine out in the security office and hallway, like nothing happened -- besides the fact that there is no security officer present at the Core security office and you'd think there would be at all times.

We will likely figure out more picking up the trail from security instead. Since it seems like security had the authority to, and made the decision to fire the EMP, in addition to making that facility-wide broadcast. Seems like it was a complete lockdown. Can we find files on H. Ranse? Preferably daily staff journals or duty officer's logs.

That's not to say we can't also go look at what the maintenance was for, see if we can find the maintenance log entry for CMT, E. Simmons, date 2136.03.24

Note to self: find the central security office/barracks and check logs from there.

We need to find a path to three things now, Laboratories, Residential and Security. We are probably going to need more fabricators, and drones, but let's stick to the priority list we set up earlier for now.
>>
No. 644274 ID: 164c83

>>644243
Hypothesis formed: Our facility was one venture of a larger corporate entity of some sort--criminal, religious, or otherwise. Orders were given to cease operation at this location, pack up the easily salvaged parts and evacuate. That determination was made based on the security assessment declaring the location fundamentally insecure. When they fired the mandated kill switch they then proceeded with an orderly withdrawal that had been planned for.
Ways to falsify this hypothesis:
-Discover remaining personnel/files with contradictory information;
-discover that some or all of the facility besides our core is still active, or;
-discovery of anyone present at all from the personnel files is suggestive but inconclusive that this hypothesis is invalid.

Do we think we could do a useful job of setting up audio sensors and interpreting their data? It would probably be the the safest way to find humans present without being in line of sight, since I doubt we have good chemical sensors for 'smell' available. It would also be a very good warning system for a number of different imminent problems. Microphones in principle are pretty simple devices, but wiring them into a broad sensory network may be beyond us for a little while.
>>
No. 644505 ID: cf960d

Apologies, due to work exhaustion and a depressive slump, I can't post today. I'll run up what I can today and finish it off tomorrow.
>>
No. 644778 ID: cf960d
File 143328403853.png - (69.46KB , 1024x768 , corecorridor.png )
644778

>Wait a second, we never read the name on it!
Exerpt;prior recorded memories: RDrone001@Core: This funny little card thing has a picture on it, and the magstripe reads "Cyan Labs. Core Maintainance Technician Evan Simmons. Authorised." It's got some smaller symbols, too, one's a funny swirly print, and the other's a coloured ring, with flecks and lines in it.

>Change the password to a 16 digit randomly generated alphanumeric string, and then log out.

Drone 001 changes the password. "nQe20o82#3#8*AHn"

>Open the door out from the security office,examine the hallway for damage.

It pokes it's head out of the door to the labs. There seems to be some damage, but other than that, things seem alright. Drone 001 wonders if Hot Labs means somewhere warm? It's near, uh... "Phenix Cores 1 & 2". Another like Cascade? It'll have to ask when it establishes an uplink.

>Can we find files on H. Ranse? Preferably daily staff journals or duty officer's logs.

It heads back to the computer.

Uh, there's a load of logs in and out, seems he was a guard cleared for the Inner Ring laboratories. He was one of six rota'd to guard Cascade's entrance. The duty roster says that the core should never be unguarded.

>See if we can find the maintenance log entry for CMT, E. Simmons, date 2136.03.24

Ooh, right, that guy! Uh... Monthly core inspection, damage noted: None, a note that the generators will need maintenance in two inspection's time.

Drone001 logs out of the console.
>>
No. 644780 ID: cf960d
File 143328408823.png - (54.04KB , 1024x768 , cambot.png )
644780

>We need to shut the EMP down in some manner before we can consider ripping it out. We'll evaluate ripping it out after it's disabled.

CascadeCore@root: Of course. Depending on how high the charge is might be an issue, though we'll certainly want to deal with it. If we had thick enough cables, or metal, we could just short it out...

>Better builds for the cable layer and other retrieved schematics?

CascadeCore@root: We'd have to recover the schematics or design our own, to be honest. We can put a processing stack on to that!

>Have Drone 4 bring the ID card and sample container up to Drone 1.

Drone004@Core: Sure!
It darts out and grabs the pair of items, before heading into the cable gap, as Drones 2 and 3 lift up one of the five thick cables, and start to reattach and test the wire pairs.

Touching base with Drone 1, it then heads back, and relays all the information from Drone001 to Cascade, before returning back, falling off of the Core network again.

>This will always be preferable to backtracking each time we figure out something.

CascadeCore@root: It will be if we have a connection. We've just now recieved 001's data. Let's see. It looks... The data is still very fragmented, but the Hot Labs and the twin Phenixes are the radioactive materal labs, and the facility's power reactor, as well as the secondary experimental research reactor.

>Task Drone 4 on identifying the layout of the EMP system from this side.

Drone 004 cuts a hole into the back wall, to follow the cabling with a small lens, then pulls back, and starts to slice a big track into the metal.

>Do we think we could do a useful job of setting up audio sensors and interpreting their data? It would probably be the the safest way to find humans present without being in line of sight, since I doubt we have good chemical sensors for 'smell' available. It would also be a very good warning system for a number of different imminent problems. Microphones in principle are pretty simple devices, but wiring them into a broad sensory network may be beyond us for a little while.

CascadeCore@root: While a broad network is not too viable right now, we could easily do such a thing with the processing power. As for making them, they're simple devices. Wired would be easiest, wireless a little more detectable, but once the cables are reconnected we should be able to concentrate on the cameras, and the microphones in them.


Overwatch@root: Prototyping Complete. Designation CDP.

A small single-"eyed" drone slips out of the fabricator, it's body translucent and cloudy. It moves noiselessly over the tiles, it's camera having an inbuilt point-of-interest flag and HUD.

"Why hello. I guess it's time to play I-spy..."
>>
No. 644789 ID: e114bc

>>644780
Well I'm not sure how to design a better cable layer so let's just throw a stack onto recovering its schematics further.

Alright let's send the CDP bot out to explore alongside drone 1. CDP-1 should explore ahead of drone 1 so that our activity remains undetected if there are hostiles about. Actually, I feel as though CDP-1 should explore everywhere it can in an area without use of the ID card, and then once we're sure an area is clear we can start opening doors in that area with drone 1. At this early stage we can't take any risks. Priority one would be finding a computer we can access to gain more information about why everyone's missing.

...I'm not sure why we're taking the sample outside.
>>
No. 644819 ID: 8e0b6a

>Drone 001 wonders if Hot Labs means somewhere warm?
My guess it might mean the labs use radiation. Or maybe dangerous biological stuff.

Drone 001, do you see any warning signs if you look down the hall in that direction? That might narrow it down.

>The data is still very fragmented, but the Hot Labs and the twin Phenixes are the radioactive materal labs, and the facility's power reactor, as well as the secondary experimental research reactor.
...nevermind.

>None, a note that the generators will need maintenance in two inspection's time.
Does our data file contain an SOP for generator maintenance? Or the maintenance log?

>Camera Drone Prototype- CDP
Let's do a simple test to see how much we've improved the camera design. See if CDP can resolve the text on the keycard drone 001 couldn't.
>>
No. 644937 ID: 71dfb3

We concur about finding a terminal with event logs leading up to the disaster of 2136.4.15. However we need to analyze the situation and give our drones more specific commands to work more efficiently. We appear to be in the restricted zone of the facility. It's not very likely that we'll find much in the inner ring except for specialized areas of classified research. It is more likely, given a central location, that we'll find what we need in the middle ring so we suggest that CDP-1 (after it's current task) do a thorough sweep of the middle ring to scout out for Security, Maintenance and Residential in order of priority. As a secondary objective, point out any storage or industrial areas they may find. As a tertiary objective, CDP-1 is to find the server room as well as see if any of the terminals have network access to the outside world. Highly unlikely, but worth a shot.

Drone 1 has poor odds of surviving, much less operating safely, in the Hot Labs especially after months of negligence. With the amount of radiation they could be working with in there, microwave research (masers? It has a high probability of being related to masers), and including an experimental reactor that's been neglected for a long time, we don't think electronics stand a snowball's chance in Hell. We would prefer Drone 1 to live, thank you very much.

So it might be more worthwhile for Drone 1 to explore the Bio Labs first, especially since we're not biological and subject to potential contaminants lethal to organics. We will want an EMP-hardened drone with sufficient armour and NBC protection to take a peek at the Hot Labs later. The construction bot can probably be used for this for a more immediate solution, but a better idea is to design our own UGCV/combat drones with such a suite so we don't have to saddle our construction bots with excess protection that they may not necessarily need. We're going to leave the build queue open for other directives to handle construction.

Or we can simply sacrifice a camera drone into the Hot Labs to record and send as much data as it can before it is fried, but our drones aren't expendable right now.

En route to the Bio Labs, if it seems like entering the area will contaminate the outside, cease and desist immediately.
>>
No. 644939 ID: e114bc

>>644778
Hey, is the ceiling leaking?
>>
No. 644969 ID: 164c83

See if we can look up anything about the security measures, other procedures and log entries, and experiments for the other labs we've seen signs for. It would be nice to know exactly what level of biohazard, chemical/particulate hazard, radiological hazard and other hazards may be lurking around here for starters.
>>
No. 646730 ID: cf960d
File 143363981311.png - (42.92KB , 1024x768 , drone1ainttakingthis.png )
646730

>See if CDP can resolve the text on the keycard drone 001 couldn't.

CDP skitters over to the card and runs it's eye over it. "Cyan Labs. Core Maintainance Technician Evan Simmons. Authorised. It's got a exact run of his fingerprint, iris, and the magstripe I can't read. In smaller text: 'This card must remain on your person at all times, and losing this card will earn you a staff sanction, and the cost of replacement.' Hmn, seems like this fellow's been off on an adventure of his own without his card."

>Send the CDP bot out to explore alongside drone 1. CDP-1 should explore ahead of drone 1 so that our activity remains undetected if there are hostiles about.

CDP slips through the wiring run, and out through the core security office, before looking at Drone 001. "Hey, step back. This is dangerous territory, and work for a *real* robot." Drone 001 is offput by this little pipsqueak giving him lip.
CDP heads through into the corridors

>Sweep the entire ring.

Unh, ok, I'll have to draw a map.

>...I'm not sure why we're taking the sample outside.

Drone 001 isn't sure either. It was ordered to have this sample, so it's got this sample...

>Drone 1 to explore the Bio Labs first. En route to the Bio Labs, if it seems like entering the area will contaminate the outside, cease and desist immediately.

This seems like a good idea. Anything that means it's not being bathed in radiation seems like a good idea! It starts to head over.

>Hey, is the ceiling leaking?

Drone 001 says yes. There's a very slow drip. Kinda like CDP. Stupid stuck up prototype.
>>
No. 646731 ID: cf960d
File 143363988608.png - (230.73KB , 1024x768 , Biolab.png )
646731

>Task a processor stack with recovering the cable layer's schematics further.

CascadeCore@root: One is now tasked onto this.

>Does our data file contain an SOP for generator maintenance? Or the maintenance log?

CascadeCore@root: Uh, it's garbled, but yes. Clean the filters, check turbine alignment, check the heat exchangers, check the cooling array. Ensure circulation pump filter is clean.

>See if we can look up anything about the security measures, other procedures and log entries, and experiments for the other labs we've seen signs for. It would be nice to know exactly what level of hazard is around.

CascadeCore@root: The Hot Labs and Reactors: High levels of ionising radiation, danger of death, wear all mandatory safety gear and dosimeters. Never stand in the marked beam path. The Bio Labs: Wear respirators when working in sealed rooms. Burn paper suits after use. Do not allow cross breeding or encourage evolution of initial cultures. Do not allow anything to cross from the test labs to the hydroponics sections.

Drones 002 and 003 have reattached the first cable! Across the Inner Ring, access points start to flicker, and the wireless network slowly comes to life.

RDrone001, RDrone004 and CDP have connected @Inner.

Drone 001 sends a priority message.

RDrone001@Inner: There's a person behind the door to the bio labs. Through what little I can see, they're sitting at a desk, writing something. White coat. Black hair. Next to an intercom. I'm trying not to be seen. Cascade, how should I proceed?
>>
No. 646736 ID: e114bc

>>646731
Drone 1, pull out. We can't risk detection yet.

CDP, make your next waypoint the bio labs please. We must confirm who this person is before attempting contact.
>>
No. 646738 ID: bd8b82

double check that the EMP is disabled.
>>
No. 646745 ID: ab7529

>Drone 001 isn't sure either. It was ordered to have this sample, so it's got this sample...
We'll designate an area to set aside collected samples.

>human spotted!
Drone 01, do not engage. Orderly and quiet retreat. We are not prepared for first contact.

>Across the Inner Ring, access points start to flicker, and the wireless network slowly comes to life.
Search network. Can we access any remote data storage or sensors? Any other AI on the network? Catalog what we now have access to, and then we'll define which items to investigate.

Do our best to conceal our presence on the network from others on the network. We don't know if there are any other AI present, or human users.
>>
No. 646748 ID: b8ceae

>>646731
Can we listen to the intercom in case they're talking to themselves?

CDP: Can you get over there and observe the person? We'd like to get a good enough view of their face to attempt to match them with the employee records.
Also, don't be an ass. We're most likely going to tell you to go into one of these sealed rooms in a few minutes. You're going to be the one asking D1 to make a hole for you.
>>
No. 646757 ID: 164c83

Strikes me as a bad idea to go harassing anyone in these labs until we've done a more general survey. We still haven't assessed whether or not this facility is entirely abandoned or if they just tried to terminate our AI experiment alone for whatever reason. Now that we know at least one person is in this facility, we have to be more careful about not leaving obvious evidence of our activity: They might have tried to kill us, think we're dead and buried, may try to finish the job when they notice we're still alive.
>>
No. 646818 ID: f634ce

Well that's problematic. The facility is still potentially operational? -- but they're assuming we aren't which is a fair assumption since up until today we weren't. We can't very well harm these humans since they're better friends than enemies and they are useful to us; but we don't intend to self-terminate either. Very well then, avoidance it is for now.

We want to assess how much it would take to outfit our drones with camouflage kits, and modify designs on future drones with them. We would like to conceal all of our efforts to expand and sustain ourselves.

We assume inner access would just be inner ring, which covers just the two other labs. Let's take a look at stored research data in the hotlab for things we can use, Energy sources? Weapons? Anything. We don't much think there's anything in the biolabs we can pick out unless. Well, let's just get an index of what they're working on. See if any of the cameras are online and let's take a peek in biolab and the hotlab, and then remove data from and rewrite the index nodes over our logs of access and use -- we need to make sure we leave no traces, and none that an experienced IT tech can recover.

Our drones can be easily spotted and heard, us? Not so much.

So given the implication that we only have access to the inner ring, which is a sane assumption given that this is potentially sensitive research data that should have its own closed loop so it doesn't get out into the open; we need to make preparations for connecting into the middle ring ourselves, quietly. Humans are likely to hear us tunnel through the walls so we need to route any cables we want to run through existing access points.

Ventilation? It's pretty obvious, but that's the only way we can think of. We need Drone 2 to scout out any access points for the ventilation from the core or the security office, or the security office hallway, but not go in them. Let's run up another CDP and send that one in instead. Have it be careful about any sensors inside the vents and minimize noise.

Speaking about security, we should probably make it so we can disconnect and reconnect our connection to the inner ring or anywhere else for that matter, at will. So we are not vulnerable to counter intrusion. Router with local network security? Or at the very least a switch or a breaker we can toggle from our end.

See we can put a processing stack on recovering the cable layer and let's redesign it to something that is silent and concealable, then let's run the fabricator on it.

CDP, keep an eye out for cameras and sensors, and give us paths that'll avoid them for our other drones to take. Drone 1 is to back away quietly from the Bio Labs and route back home to the security office hallway, closing the doors until we can figure out our next move.
>>
No. 648209 ID: cf960d

I do apologise for how long it's taken me to do this. The aside is medical leave, but I'm just going to drop the pictures from these posts so you guys have something to go on.

> Drone 1, pull out. We can't risk detection yet. CDP, make your next waypoint the bio labs.
>Drone 01, do not engage. Orderly and quiet retreat. We are not prepared for first contact.
>CDP: Can you get over there and observe the person? We'd like to get a good enough view of their face to attempt to match them with the employee records. Also, don't be an ass. We're most likely going to tell you to go into one of these sealed rooms in a few minutes. You're going to be the one asking D1 to make a hole for you.
>CDP, keep an eye out for cameras and sensors, and give us paths that'll avoid them for our other drones to take. Drone 1 is to back away quietly from the Bio Labs and route back home to the security office hallway, closing the doors until we can figure out our next move.

Drone 001 runs it's motors down to low, and slowly moves down away from the window, it's legs moving carefully, before pulling back around the corridor curve. It sets up station, and waits, trying to stay out of sight as much as possible!

CDP gets a little trio of messages and heads off to follow them, moving smoothly and quietly over the walls. It checks the cameras, and notes that currently, they're all off, no lights whatsoever. The only real issue around here is probably getting seen by a human.

CDP@Inner: I can get a shot. And... Ngh, but the Drones are so big and clumsy. Fiiiiine.

>Double check that the EMP is disabled.

Drone 001 checks it's cargo space. One set of EMP keys, and Drone 004 is slicing into the wall to study the coils. It's halfway there. Should be good for now!

>We'll designate an area to set aside collected samples.

Oh, yes. Drone 001 heads back to the Core access office to leave the sample cylinder under the desk. It should be mostly safe there! It pushes it into a corner, and nods, idling there.
>>
No. 648210 ID: cf960d

>Can we listen to the intercom in case they're talking to themselves?

audio_feed@blabmon1: "Uh, so, we've got two of the incubators over to growing peas, and there's the mushrooms in the darkroom, so... Call that food for all of us on ration, and there's the third incubator being sterilised now. Can't touch the labs, that's not really good for survival."

>We want to assess how much it would take to outfit our drones with camouflage kits, and modify designs on future drones with them.

CascadeCore@root: That'd be either production or research into materials, based on what you mean by camouflage kits. Netting or something is just plastic.

>We assume inner access would just be inner ring, which covers just the two other labs.

CascadeCore@root: Yes, you're right, though there's four labs in the Inner Ring. Biology, Chemistry, Radioactive/EM and a small Cyber lab.

Overwatch@root: Map render is coming.

>See if any of the cameras are online and let's take a peek in biolab and the hotlab, and then remove data from and rewrite the index nodes over our logs of access and use -- we need to make sure we leave no traces, and none that an experienced IT tech can recover.

CascadeCore@root: Drone 3 is working on the second of three thick cables. One is likely to be computer network, and the other cameras.

>We need Drone 2 to scout out any access points for the ventilation from the core or the security office, or the security office hallway, but not go in them. Let's run up another CDP and send that one in instead. Have it be careful about any sensors inside the vents and minimize noise.

CascadeCore@root: Technically it'd be CD 001, but certainly.

Drone 002 perks up and scuttles down the cable ducting into the security office, moving up to the cieling tiles, where it disassembles one.

>Speaking about security, we should probably make it so we can disconnect and reconnect our connection to the inner ring.

CascadeCore@root: We don't currently have enough drones for an additional task. But Drone 003 can do it as part of it's repair job.

>See we can put a processing stack on recovering the cable layer and let's redesign it to something that is silent and concealable.

Overwatch@root: Processor stack slaved.

###
The two below I've been extra stuck on. I'll get back to them when I'm feeling better.
>Well, let's just get an index of what they're working on.

>Search network. Can we access any remote data storage or sensors? Any other AI on the network? Catalog what we now have access to, and then we'll define which items to investigate. Do our best to conceal our presence on the network from others on the network. We don't know if there are any other AI present, or human users.
>>
No. 648215 ID: a65358

How much space does your mind take up? How much space would it take just to store it? (In a somewhat redundant format.)

If it's not too much:

Since there's still humans out there, I recommend constructing a "failsafe drone", a design that can do anything you can do (autonomous operation, materials processing and fabrication) except a *lot* slower and smaller-scale. Run it off a nuclear engine or something similarly long-lasting. Get it to burrow downwards a few hundred meters or so. Then set it to dig out a base and recreate you, but hold it on "idle" with a continuous fail-active signal. That way, if the inevitable first contact with humans predictably turns sour and they nuke you again, you'll have a second shot.

We could also use a low-risk way to keep tabs on the humans. Propose design: "Speck", basically a tiny short-range wireless transceiver and a low-res camera. Make a few thousand of them, scatter them around the place we saw the human.

Hey, how good _is_ your miniaturization ability?
>>
No. 648220 ID: 164c83

>>648215
We currently lack the capability to do much more in terms of fabrication than we have. We need materials, we need to improve our processing, we need to recover or rediscover former construction and fabrication abilities, we need to build up our fabrication tooling and we probably need significantly improved power sourcing, smoothing/storage and distribution to start building resurrection seeds.

Itemize that list of needs, add to it as necessary, and analyze what it would take to achieve all that please, Core? Feel free to leave off at, "Unknown," for any that's too difficult to determine in a timely manner or for which we need additional information we don't have, and list those requirements.
>>
No. 648222 ID: e114bc

Are there advantages to not using our CPU stacks? It said "idle stacks trawling SAN", what does that mean? Is that just... recovering more blueprint filenames, instead of refining individual ones?

It sounds like the facility is sealed off from the outside world and this guy is one of the survivors. It should be safe to contact them, but I want to confirm their face first.
>>
No. 648227 ID: ab7529

>Ngh, but the Drones are so big and clumsy. Fiiiiine.
Depending on how this goes, you might need even bigger and clumsier brothers, either for cargo transport or combat.

>Uh, so, we've got two of the incubators over to growing peas, and there's the mushrooms in the darkroom, so... Call that food for all of us on ration, and there's the third incubator being [sterilized] now.
So the humans present are cut off from external supplies, and doing what they can to generate food on site

Could be good, could be bad. If they're in trouble, they might welcome assistance, and be less willing to care about the security protocols that say we should be dead. The again, desperate people can be unpredictable. They could blame us for the current situation, or see us as a risk to their own survival.

Friendly relations with human nearby would be preferable, but we want to be in a more secure position before we make contact.

...do any of our records include data on agriculture? Assistance in their survival is a hell of a good bargaining chip. Gardener-drones are a potential future asset.
>>
No. 648248 ID: b8ceae

>>648227
I think our best bet is to say out of sight and repair infrastructure in the background. We'll quietly neutralize problems and build up our assets while biding our time.
When we do eventually get detected we can claim responsibility for things getting fixed, and will have a small army of drones which can cut through most anything, sneak through vents, and don't need to sleep.
>>
No. 648602 ID: f634ce

Re: The SAN trawling:

Its deleted file recovery that we have set up in order to figure out details that we do not currently have. We've lost a lot of data to the EMP and already recovered what we can, sure, but we're trying to recover files in the background that we or the previous on-site personnel deleted through inode recovery. The idea is to figure out if we're missing details of the events that lead up to the 'disaster'.

As for other things;

Well, for one thing, it seems like this place actually is abandoned given the state of the cameras and lights outside and these are scavengers, rather than personnel. Or they might be personnel who survived whatever happened and are now forced to fend for themselves. Either way, we should never, ever, rule out diplomacy, given the opportunity. We cannot survive or defeat external threats without allies. Even if they are not cordial or are in fact hostile we will have had a litmus test to how they will react to us and have figured out their apparent threat to our existence -- rather than let them be and they'll build up to be in a better position to kill us if they wanted to. Besides, unlike normal human negotiations, drones can survive bullets in case they don't want to coexist with us.

That said, it's not entirely true that we don't have existing bargaining chips. We have something humans in a desperate survival situation would find extremely useful: a massive geothermal powerplant. Lights, heating, electrolysis, cooking elements, fans or air conditioning if it's hot in here. (Can we check temperature here and see if we can make it comfortable for them?)

Not to mention that we can divert some of the water intake for the geothermal for heating into potable water, assuming we ever figure out what the contaminant in the pipes is and make sure it's not lethal to them. The water would also be useful for agriculture.

First things first, though. We want to confirm a couple of things, get the camera lines repaired ASAP so we can take a look at their faces and figure out if they're anyone we recognize from the old days.

After we've at least figured that out; prioritize fabricating a single white flag out of polymer, and a single LCD monitor with an access point for our precision repair drones after our current projects. Let's have Drone 1 pick those up, and see if we can communicate to them. Drone 1 gets to be an impromptu ambassador in case they're not base personnel. (Who will presumably want us not alive as a matter of security.) Let's make sure they can see the truce flag when we knock.

In the meantime, have the CDP continue exploration of the middle ring, as stealthily as possible. We will want to know how many mouths we need to feed in case they're willing to negotiate cooperation with us -- or how many guns they have in case they don't.
>>
No. 651715 ID: cf960d

Post to come shortly, sorry for the delay.
>>
No. 651723 ID: cf960d

>How much space does your mind take up? How much space would it take just to store it? (In a somewhat redundant format.)

CascadeCore@Root: Uh, that is a good question... The optic cores that make us up are trinary, so... Active, you're talking about a petabyte a core, and a optical parallel processor to run on. We're... Advanced indeed. For the design you propose, we would have to do some groundwork. It is a feasible plan, however.

>We could also use a low-risk way to keep tabs on the humans. Propose design: "Speck", basically a tiny short-range wireless transceiver and a low-res camera. Make a few thousand of them, scatter them around the place we saw the human.

Overwatch@root: Design brief for drone chassis "Speck" entered into database.

>Hey, how good _is_ your miniaturization ability?

Overwatch@root: Miniaturization level hampered by fabricator. Current best rated only to microdrone levels.

>Itemize that list of needs, add to it as necessary, and analyze what it would take to achieve all that please, Core?

CascadeCore@Root:
Materials. Will probably be tricky once we have taken all the pre-existing electronics from this facility. What we've got here is a untapped stockpile, not material production facilities.
Processing. Any processing is a bonus, but for something like one of the CPU stacks, we'd need to make or take over a server rack at the very least, based on what we have in memory as a decent 2U rack.
Recover or rediscover former abilities. This isn't... Really an issue, we are all self-improving to some extent. Simulations and iterative development coupled with genetic algorithms allow for some... Jumps of insight, though mean that devices can vary differently, especially if genetic algorithms are just copied over onto new hardware.
Better fabrication tools. By pushing the limits of what we have, we can use our current technology to work towards smaller and smaller scale fabricators. There is a catch in that the smaller we go, the more specialised each creation has to be. Nanoscale work would not be able to do anything more than it's specific task.
Power supply and regulation. Well, there's a reactor sitting out there, though the kind of radiation that would put out would do us damage too if we caused a malfunction. Aside from that, we could make some. The principle is simple enough.


>Are there advantages to not using our CPU stacks?
>>Re: The SAN trawling: Its deleted file recovery that we have set up in order to figure out details that we do not currently have. The idea is to figure out if we're missing details of the events that lead up to the 'disaster'.
It's an in-game excuse for me inserting information on the fly, and occasionally as I remember it. I have a Mandatory Social Engagement for a week. I'll take my laptop and tablet and actually write up a timeline and do some pictures then.

>Get the camera lines repaired ASAP so we can take a look at their faces and figure out if they're anyone we recognize from the old days.
>I want to confirm their face first.
Unh, human faces...
Things to draw:
Facility Map
A few camera shots.
Scientist faces.


>Depending on how this goes, you might need even bigger and clumsier brothers, either for cargo transport or combat.

CDP@InnerRing: Aww, really? That sucks. Light and fast's the way to be! Can't be proper help if you're slow and lumbering.

>Do any of our records include data on agriculture? Assistance in their survival is a hell of a good bargaining chip. Gardener-drones are a potential future asset.

CascadeCore@Root: We have some data for agriculture, but more useful would probably be the infrastructure advice we could provide for growing with. And the power surplus we could provide to the growlights.

>Can we check temperature here and see if we can make it comfortable for them?

CascadeCore@Root: Network coming online for Inner Ring. Stand by for scan. HVAC system has self-identified. Current temperature 26oC. Set temperature?

>Fabricate a single white flag out of polymer, and a single LCD monitor. Drone 1 gets to be an impromptu ambassador in case they're not base personnel.

Drone 001 pings happily and heads back towards the base, as the flag and the small monitor are added to the fabber queue.

>In the meantime, have the CDP continue exploration of the middle ring, as stealthily as possible. We will want to know how many mouths we need to feed in case they're willing to negotiate cooperation with us -- or how many guns they have in case they don't.

CDP@InnerRing: Certainly! Let us go explore very carefully. And quietly!
>>
No. 651740 ID: ab7529

>>648210
The network search from the bottom of here is still important. What can we tell about the network we linked up to?
>>
No. 651754 ID: 6b0329

Without knowing more about what the people are up to, exterior conditions, and whether they deliberately set the temperature marker high or low for any particular reason we don't know if that 26C setpoint is too high or not: They might be trying to encourage fungal growth, but if they aren't they probably find that temperature too warm.

We need to know exterior conditions and climate ranges, and continue the stealth survey to decide that one.
>>
No. 655896 ID: 0d1c37

...With my regrets, quest running is far harder with my work than I expected. I'm afraid I'm going to have to ask for this to be archived for now. I'll return to it.
[Return] [Entire Thread] [Last 50 posts] [Last 100 posts]

Delete post []
Password  
Report post
Reason