Lore question about telecom technologies and Codsworth

[Halstrom]

"I'm trying to explain what are the things that I find problematic because  I hope someone will point out that I'm wrong and will help restore the order :-) If I'm going into details sometimes it is not because I want to persuade everyone that I'm right. It is because I hope I give more options to somebody else to tell me that I'm wrong. This is the "selfish" part of my posts - I hope someone will give me the reason to play the game longer than 20 - 30 hours. And when I play it again in one year with various LL mods installed to not simply turn god mode on."  - Prinyo

 

Your right, it is wrong and yes there is not reason to play more than 32.7hrs.   No Need for you to make mods for it either.

 

It's not our job to give you a reason to play the game, your the one that purchased it, if you are having issue with lore, then make your own decision.  As for myself, and lore in FO 4 ...  yes  my robot flies, why , how, because I wanted one that flies and I had the money so they build it, I don't have the internet...  nope sure don't   why , how...  didn't want it, didn't need it, it was never built.    My sarcastic point here is  in the real world if there is a demand great enough, someone or some company (if possible) will fill that want or need...   what technology exist at the time of FO 4 is whatever Bethesda  says it is.    Your statement  "The lore is a balanced system of facts and relations that govern the reality of the virtual world."     Who do you think dictates facts and relations in fall out 4 , just because they have black and white TV does not  mean they could not have flying robots.... why would it?  We in real life have flown to the Moon but we still cut people open to do surgery, one advance does not mean everything advances.   We have talked to people on the moon but yet our internet speed is still limited.

 

Your statement  "

A civilization is able to create advanced AI and to experiment with cryogenic technologies while been completely stuck on the ways of communication.

There are terminals but there is no internet ( or whatever it might be called). 

All communication is analog and unreliable.

Black and white tv... umm, no. Just no."

 

Makes no sense to me...  advance AI robot but Black and white TV  umm NO !   Why not, maybe the people were not in to color TV, why do you assume if you have Robot's you must have color TV?  And internet   ... neither of which is required to build my flying robot.

 

Just my thoughts, please continue with your thoughts and enjoy the game  (or not).

 

p.s  @ Mogie56  love your sig...  posting it now on Gopher's web site.

You might want to note many vision impaired people like me can't read red on black.

[panthercom]

 Conventional vacuum tubes, as we know them, wouldn't be at all feasible for creating an electronic brain

small enough to fit in a Mr. Handy. So they would have had to have developed another technology, or miniaturized

vacuum tubes enough to make it work. I did read an article years ago about someone who had made miniature

triodes using silicon-based technology. The triode is the vacuum tube equivalent of a transistor. Interestingly, some

of the multi-element tubes or special purpose tubes, such as thyratrons, could do some pretty amazing stuff with

a much lower parts count than a solid state equivalent. If the construction of vacuum tubes changed in such a way

that rows of ladies with delicate hands and tiny tweezers were no longer needed to assemble them, who knows

what advances in miniaturization could have been made; particularly if thermionic emission was possible at lower

temperatures and higher efficiency than anything ever produced in our world. Vacuum tubes with graphene elements?

Who knows, I'm not a physicist

 

 Here's a picture of ENIAC, just to show how impractical conventional vacuum tube computers are...

[DocClox]

Here's a picture of ENIAC, just to show how impractical conventional vacuum tube computers are...

Yeah, that's why I'm thinking positronic brains or some sort of era-appropriate SciFi concept.

 

Of course, by 2077 they could have minaturised valves a lot if they'd had to do without transistors ... but I don't believe it either. The robot brains have to work on some other basis.

[DexesTTP]

Conventional vacuum tubes, as we know them, wouldn't be at all feasible for creating an electronic brain

small enough to fit in a Mr. Handy.

Here's a picture of ENIAC, just to show how impractical conventional vacuum tube computers are...

That's relevant, but it is only one side of the coin. The ENIAC computer was based on Vacuum tubes about 4-6cm long and 2-3cm wide. It used just above 40,000 of them. However, the ENIAC was both multipurpose and turing-complete, meaning it was created for no specific purpose.

 

The furthest we managed to reduce the size of a vacuum tube was I think about 8 mm for a centimeter and a half. Then, we switched to germanium then silicon-based transistors. We didn't really try to reduce them further as we already had better stuff.

 

In addition to that, a hardware dedicated AI would need far less components. I can't give any realistic figures as I don't have any expertise in AI-specific hardware, but I know a NAND gate takes about 4 transistors while a low/high-pass filter, analog multiplicator, analog summing ans such components used in AI takes none. If the technology is focused on using as little transistors as possible, I believe it is possible to maintain the vacuum tube's count as little as possible and that would result into small cards, small enough to fit into a Mr Handy or similar.

 

My point is, we don't actually know if it would be possible to make a viable AI with few transistors. We never even tried to do a real AI-specific hardware implementation since the late 50s (excluding things like FPGAs or Microsoft's CNN accelerator), and nowadays we mostly use graphic cards and processors as our "AI hardware", and add more as needed.

 

A vacuum-tube based technology would also use analogical computing a lot more that what we do. And, in that field we don't even know what we don't know. The last 60 years of improvement in the AI field has only yielded results intended for digital programming.

 

So, just saying it "wouldn't be feasible" is a little small-minded, as all we know for sure is that it wouldn't be feasible assuming they try to make it heavily transistor-focused.

 

edit : there were in fact attempts to make dedicated AI hardware, including analogical. However, all the data I could find about them is based on digital inputs/outputs and includes in some way CMOS chips and similar components as a way to help the digital analysis.

[panthercom]

 What I said was "Conventional vacuum tubes, as we know them" wouldn't be feasible. It takes a lot of power

to light all those filaments. Even the subminiature tubes pull a fair amount of current when you have a bunch

of them hooked up; I have some of those in my vacuum tube collection; they had relatively short lifespans too,

compared to typical 7 or 9 pin based triodes. There is also the issue of heat management; yes, tubes get hot

under normal conditions, and if they get too hot they will fail prematurely.

 

 I've been using tube audio equipment since the late 70's, and my old pair of 60 watt Dynaco MkIII's did a fine

job of heating my bedroom in the wintertime

[Arhon]

"there is no lore"

 

there actually is... just a little. have you found the arcane dagger yet?

 

That was the first unique dagger I found in Fallout 4.

 

That whole place was.... scary... very scary... and at the end when I found it ... I was totally shocked so much that I closed the game instantly....

[Guest Mogie56]

LOL Oh they definitely get hot, enough so to blister fingers.

[panthercom]

 The Mr. Handy would also need power amplifiers to drive the motor control servos for all of it's

appendages, so they would be a nice source of output tubes for all those post-apocalyptic

audiophiles and metal bands...

[Benmc20]

 

Here's a picture of ENIAC, just to show how impractical conventional vacuum tube computers are...

Yeah, that's why I'm thinking positronic brains or some sort of era-appropriate SciFi concept.

 

Of course, by 2077 they could have minaturised valves a lot if they'd had to do without transistors ... but I don't believe it either. The robot brains have to work on some other basis.

 

Well ... Do you remember the robots in Fallout 3 called "robobrain" ? Those ones had real brains in theirs chassis. Maybe it'z an explanation for :

- self awareness

- learning AI

But, if it's the case, how to interconnet such living brain to mechanical parts ? I remember some years ago that some scientits had tried with little success to grow some meurones inside of a silicon chip.

[Benmc20]

Why use and search advanced micro electronics if you have such system (real brain used as a computer) : http://fallout.wikia.com/wiki/Robobrain

 

Programming :

 

Robobrains, due to the nature of their organic processor, can be programmed with much more complex and expansive instructions than typical robots. This also means they are superior to typical combat robots in the Fallout universe, at least in terms of their tactics and strategy. Military maintenance bots can be controlled via a WLAN matrix network. In this case, a central AI operates the robots. In the Mariposa Military Base, the programs used to run robobrains were Facility Maintenance v2.1, Movement v4.8b, Repair v0.8a, Pest Control v1.4, Hearts v1.0, and Sensors '76.

 

...

[prinyo]

There are two questions here:

1. Can the civilization produce something like Codsworth?

2. Is the civilization using the technologies in other areas that are considered important?

 

>>>   If the post seems too confusing to follow please skip directly to the tl;dr. The post is simply following my internal logic that led me to the results written in the tl;dr

 

 

And there is one more thing. From reading the overseer's terminal at the start of the game we learn that people were told to stay in the vaults for 3 months. Maybe some stayed longer than that. But because he/she stayed in cryo the PC emerges more than 200 years after the bombs. There is more than enough time for the other survivors to recover and start moving forward.

 

Codsworth is not only "Decisional" and "Learning" AI. He is self-aware, socially-aware and emotional. This makes creating him way harder.

And in a world where the warfare is considered very important - what are the technical achievements in the warfare that come close to the dish-washing  robot?

 

Positronic brain is a "black box" - the reader/user is not supposed to care how something works. What is important is that it does work. 

For example - Data in Star Trek also uses positronic brain.  This means some cool futuristic technology that the viewer is not supposed to think about. But Data fits within the Star Trek lore because  the same civilization that created him is technologically advanced enough to also create warp drive, holodecks, transporters, tractor beams and so on. 

 

But we can explain Codsworth as a positronic brain. It would require a lot of assumptions but after all the energy is the main problem our civilization has when trying to create anti-matter. In a world where small and powerful energy sources can exist, experimenting and using anti-matter becomes believable. The assumption that the FO civilization has found a way to utilize  anti-matter is enough to make Codsworth's existence believable. Yes, it is a "black box" but it makes sense. 

 

 

The question that remains is -  In a world where the civilization is in a full stagnation (as I described in detail in an earlier post) are there any other areas where such technologies are used? 

Mister Gutsy, Mister Handy and Mister Orderly. Is this enough? Depends from the point of view. If in the Star Trek world there were no warp drives and all the other things I listed would creating 4 different models of Data be enough to make his existence believable? 

 

Yes, in the case of FO4 - from a timeline point of view.

But it opens another question. About the civilization in the 200 years after the war.

 

tl;dr

 

Assuming that  Codsworth and the others use positronic brain is a perfectly good way to fit them into the lore (question 1). The existence of more powerful (and compact) power sources would allow the use of anti-matter in ways that "our" civilization can not predict.

And it also fits the FO4 timeline. The robots were created before the war and are a promise for a brand new world of technological achievements and opportunities (question 2).

The question that opens now is about the 200 years after the bombs. They were  not enough for the civilization full with so much promise to pick itself up and move forward?

 

 

[Ernest Lemmingway]

Well ... Do you remember the robots in Fallout 3 called "robobrain" ? Those ones had real brains in theirs chassis. Maybe it'z an explanation for :

- self awareness

- learning AI

But, if it's the case, how to interconnet such living brain to mechanical parts ? I remember some years ago that some scientits had tried with little success to grow some meurones inside of a silicon chip.

 

 

 

Robobrains were introduced in the first Fallout game. There was even a quest in FO2 to make your own with a choice of brains. Including a psychotic one. Get the assembly wrong and the robobrain will explode! Fun times.

 

Actually there is some viable research on using neurological structures to house and access digitized data. The interface is also less of an issue that most people think. Chips that allow quadriplegics to control computer pointers via wired connection to computers have already been created and proven to work. That's not true brains-in-a-jar, but it is an important first step. Given enough time (and fewer morals and scruples) scientists could bridge the gap within a few years.

 

Problems arise in using a brain from an already extant creature; an adult brain is literally wired to work with their body and would have to unlearn everything, not to mention the high likelihood of total insanity when cut off from normal stimuli. Children's brains would have less to unlearn, but are still mostly wired for a human body and its stimuli as well as being emotionally and intellectually immature. Theoretically cloned brains are ideal since they could be adapted to a cybernetic life from the time of self-awareness, as would wholly artificial brains. But coming to self-awareness is a traumatic experience. Imagine a newborn with the body of a Terminator.

 

I'll spare you all the issues of chemical imbalances in a brain. Suffice to say too much or too little of any neurotransmitter, which is common even in a human body, can cause all manner of problems.