/sci/ - Science & Math

The main problem isn't the powersouce, it's the technology which supposedly allows the suit to sustain very strong impacts (bombs, etc).
It's not physically impossible, but it's difficult as we would have to create a technology which absorbs the force of the impact and redistributes it to the rest of the suit/evacuates it. Maybe nanotech can do that, I don't really know

I'm curious to know if the other minor technologies (notably the JARVIS AI and HUD, but also stuff like the build in water filter, the mechanisms that allow the suit to fold into a case, and all of that) can be replicated?

>What fields of study should one be proficient with in order to be able to work on those types of exoskeletons

Mechanical and electrical engineering.
Basic physics, particle physics.
AeroE may be necessary.
Also Law to convince the government to let you keep it

>>5605890
>aerogel
What? How is it relevant?

Actually, if we had a power source, we could totally build that suit (well maybe the impacts could still cause problems).
Thing is the Arc reactor is sci-fi.
>inb4 muh thorium
There are lots of different theories as of how we could produce that kind of energy (beta particles, etc) but none seem viable.

I think in the comics it's said that the suit is made of nanobots, in which case it might be much easier (in theory) to build.

Also, see
http://www.wired.com/gadgets/miscellaneous/news/2008/04/ironman_physics

>What fields of study should one be proficient with in order to be able to work on those types of exoskeletons?

This interests me too.
From what I've seen, lots of different fields seemt to be necessary when building such a piece of advanced technology: from nuclear engineering and physics to biomedical engineering for the interfacing between the HUD and the user... And much more engineering fields.

Could anyone make it clearer?

>>5605994
Muh fusion

Problems:

A sufficiently energy dense power source (this would change the fucking world)

Fine motor control of a prosthesis (progress is being made here)

All other issues are solved or being solved.

Well except for absorbing huge blows like it does in the movies. I remember reading something about a proposed Nano-material that would almost immediately convert impact energy to heat. The problem then would be radiating all that heat, which could be done. I can’t seem to find anything about this material again though and it was only theoretical at the time, there was no proposal on how to actually make the stuff.

Electrodynamic induction, your iron man would have to drop a powerplant and emitter. Perhaps an emitter at the nearest pylon.

>>5606014
Would that work miniaturized?

>>5606024
>energy dense power source
I'd like to study this, actually. I've already finished studying but my job gives me loads of free time, so what degree should I get to become proficient with the study of energy?

>motor control of a prosthesis
Yeah.
The problem here is essentially the JARVIS system, can such advanced AIs be created? Things that can discuss with you with humor and sarcasm, and elaborate complex analysis of another system?

>All other issues are solved or being solved
Jet propulsion boots?
Repulsor beams?
Storing that much weaponry inside a little suit?

As for impacts, I know that nanotechnology can do that (evacuate heat upon impact).

>>5605953
Aerogels have unparalleled insulative properties and weigh very little, I'm guessing this has something to do with having rockets to move with.

>>5606035
But the propulsion itself is impossible

>>5606072
No it's not

>>5606034
I was thinking about the same exact thing the other day OP. Jarvis may be on it's way, simply look at Glass from google. I'm not saying it's the solution, but it's a step in the right direction. Years ago no one would've thought you'd be able to have some kind of real life hud.

Then look at all these videos with robots running and jumping on irregular surfaces. This is definitly a growing field and I'm sure we'll see a lot of incredible applications in the next decades.

Now for energy well, let's hope someone finds the holy grail that would save the earth from most of it's ecological problems or at least greatly reduce the impact of the problems we're facing now.
.
We're getting close brother. We're getting very close.

>>5606112
>real life HUD
>Google Glass
Yes, indeed I made a correlation.
Integrating some kind of similar technology inside the helmet, giving it voice control and all, and you're done.
Thing is, JARVIS is too intelligent. Such an AI would require immense ressources, I presume.

Yeah, there are lots of solutions for energy, none of which are applicable for now.
That's why i'd like to conduct research on energy.

>We're getting close
Yes. I hope I'll be able to see such a machine in my lifetime.

http://www.forbes.com/sites/alexknapp/2013/02/23/prototype-suit-gives-you-real-life-spider-sense/?ss=future-tech

Fear the spoderman

>>5606128
>implying

>>5606118
>JARVIS is too intelligent
The only problem is storage
Algorithms will eventually be implemented

>>5605910
This, and even worse, the physics of the suit don't even make sense. He flies through the air at super-sonic speeds, and then lands on his feet, immediately coming to zero velocity. The guy inside the suit, no matter how strong the metal is, would be reduced to liquid instantaneously.

Imagine riding in a car at 800 miles per hour. The car hits a brick wall and immediately stops. Now, you can dream up some super strong titanium/diamond/admantium bullshit car that survives that impact, but you can't come up with a way for the driver to survive the impact. F = ma just doesn't work that way.

>>5606118
>>5606422
Just look at the kinect and all these new and powerful algorithms that are popping in numerous fields. I'm sure some of them will be able to analyze situations and the composition of what you're looking at while briefing you on the subject. With everything being able to be made smaller nowadays, I'm sure there's place for a lot of features in such a machine that would cover your whole body.

>>5606445
This might be incredibly stupid, but could the suit be conceived in such a way that it detects changes in force applied to it's outer shell and then compensates, with the use of some kind of pistons/plates, by applying a similar level of force in the opposite direction? Let's assume that there is enough space in the suit for the man inside to be affected by this system.

>>5606445
Yep. This guy is right.

Thing is, building the suit out of some super hard matter wouldn't work: you'd have to directly transfer and evacuate the energy created by the sudden change in velocity.
But I don't know how you could do that

You know what I think will be the next big thing? Augmentations like in Deus Ex. It started with pacemakers and stuff like that decades ago, why couldn't we do it to simply enhance our capabilities and push our bodies and minds even further.

>>5606473
Perhaps a charged membrane that maintains elasticity via the charge, thus repelling projectiles? I don't know anything about this, just speculating. I suppose the heat would be a problem though.

>>5606467
Not really, because any way you look at it, the human body has a certain mass (let's say 70kg) and you're decelerating it over a certain time period. You could put something like airbags or the equivalent to make the deceleration a little less sudden, but it wouldn't be nearly enough.

It's extremely optimistic to say that you could spread that deceleration (from 800mph to 0) over a full second. Even with a deceleration that slow, relatively speaking, the force applied to the body would be

<span class="math">F = ma[/spoiler]
<span class="math">F = 70kg \frac{358\frac{m}{s}}{1s}[/spoiler]

Which is about 25000N. Enough to kill you many times over.

>>5606024
>>5606491
Couldn't it be powered by such heat? Let's assume the suit can tolerate and absorb every impact. You could just jump down a platform, thus generating power for the suit, fly away, rince and repeat.

>>5606498
Um, that doesn't work. if you jump down, you can only possible generate enough energy to go back up to where you were, and to regain even 85% of it is so hard no one would want to try to do it.

>>5606034
>Would that work miniaturized?
No. However, practical fusion technology might enable something that would work.

In theory, advanced nuclear batteries could do the job. However, due to the current cost of transmutation, the fuel for them is infeasibly costly and there has been no serious attempt at making high-power-density versions. In theory, power densities of kilowatts per gram are possible, with no separate fuel source.

Despite theoretical efficiencies nearing 100%, you'd want something like wings on the suit, to serve as radiators for the excess power, since there's no way to throttle down nuclear decay. And if the power source was ever damaged, it would be at least a short term and local radiation hazard.

>>5606024
That clip always makes me sympathetic for Loki.

If I could take a thrashing like that, and was still regarded as the wimpy little brother, I'd harbor some serious resentment of Thor too.

>>5606497
What about a speed of 90 mph? This is well above the speed limit in most countries and would allow air travel to get to places without cars.

And let's assume suits like that actually are available to the public after a shitload of courses and tests, no one would just keep going at their full speed to land. They'd probably just put their extremities in front of them and reduce their speed like you in any kind of flight transport.

>>5606518
Ah... I missed my point. Fusion might make this feasible by making transmutation much cheaper, since D-D fusion produces copious neutrons.

Pulsed D-D fusion would also produce lots of tritium and helium-3. It could be an extremely valuable reaction, even if it wasn't energy-positive by itself.

>>5606536
I think it would be better to do some kind of complex roll to remove speed. Using your hand wold probably break them and your face :^{

http://en.wikipedia.org/wiki/Powered_exoskeleton

If we had the technolgy somehow, could the suit be shaped like in the movie? With no extra radiators or anything?
I thought heat and all could be evacuated via nanotech (or nanotubes)?

>>5606538
For that matter, I suppose it's possible, at least in theory, that a compact fusion reactor could be made with helium-3 fuel.

It makes a lot of sense to interpret the Iron Man technology this way. The large arc reactor at Stark does D-D fusion, producing tritium and helium-3, but was excruciatingly expensive (only made with generous Cold War funding) and failed to produce net energy. The Jericho missile is a tactical nuclear weapon which gets a pass on nuclear weapons bans because of its small yield and lack of radioactive products, which helps make Stark such a controversial figure. The fuel in the physics package is helium-3, held at extreme densities in the pores of palladium metal, also used (in much smaller amounts) in a Stark nuclear rocket motor called the "repulsor".

Tony has had the idea for a while of making a miniaturized hybrid between the arc reactor and the palladium implosion helium-3 warhead, making electrical power rather than thrust or an explosion, but due to the cost of producing helium-3 it would have been priced out of any market. Only weaponry was an economically feasible application.

But with his life on the line, he is motivated to ignore economic feasibility and make a new kind of weapon.

>>5606655
Triggered decay is another possible explanation.

The energy source is physically impossible to be reproduced.
And that "new element" thing is silly.

>>5606863
>physically impossible to be reproduced
Prove it. The arc reactor is nonsense, but a miniaturized energy source? Not quite.

Whenever I watch an Iron Man movie, I tell myself "shit this is awesome, let's start a company and do some research and become a famous genius billionaire"

But then I realize it's fiction, and that weapon manufacturers will never be famous, that Stark's life is impossible to live, and that the suit is impossible to build.

Hold me, /sci/

>you will never be a billionaire with two PhD's, living in a multi-million dollar Mansion in Malibu, building awe-inspiring wonders of technology in your garage and fighting evil with class and sarcasm
>You will never be a famous superhero with a genius mind, friends with a secret agent, a Nordic God and a lunatic scientist, and have people wonder at your ridiculous knowledge and awesomeness
>You will never be Tony Stark
>You will never be the Iron Man.

You know it to be true, /sci/.

>>5607026
Wasn't there something about really small graphene batteries found, so that you can now hold MUCH more power in a battery soon? I'll look for it, one sec.

>>5607069
"Much more energy" is, in this case, still way less than, say, gasoline.

Not that batteries with 10 times higher energy density wouldn't be awesome, but they're a long way from being Iron Man tier.

>>5607076
plus you know, its graphene. Its very much in the scientific vogue right now, but its not something for which scaling has even close to been dealt with.

>>5607076
>>5607079
http://physicsworld.com/cws/article/news/2012/mar/08/graphene-in-new-battery-breakthrough

Found it. I can provide the paper too, if I find it (looking atm)

>>5607090
In this house, we obey the laws of thermodynamics.

>>5607104
Are they breaking them? I've only skimmed it so far

>>5607090
Well, its only in pre-print

http://arxiv.org/ftp/arxiv/papers/1203/1203.0161.pdf

>>5607118
Here is the graphene battery thing that sane people are excited about: http://www.extremetech.com/computing/105343-graphene-improves-lithium-ion-battery-capacity-and-recharge-rate-by-10x

What you linked to is a perpetual motion machine of the second kind.

>>5607123
well poop. Thanks for the link though B^)

>>5607123
Will it be soon applicable in modern electronics?
>>5607065
lik dis if u cry everytiem

>>5607161
>Will it be soon applicable in modern electronics?
We can hope. It'll take at least a year or two for them to work out the kinks and figure out economical production.

Could anyone answer the OP's question about the fields of study?
I'd really like to learn about that

>>5607207
I think if you're proficient enough in physics and mechE you can go a long way
obv you need specialisations in robotics, IT stuff and all

>>5607065
Hold me anon

>>5606445
So make something that creates a "void". Ie, like, if you sit in a car that is going 800 mph, and stops, the inside is completely "frozen". Maybe not frozen but the outside would be the only thing interacting with the outside world, leaving the inside to be moving freely. If that makes any sense.

>>5607481
Would that even be possible?

>>5607481
>>5607513
I think that's what >>5606467 was talking about.

>>5606118
JARVIS is run on Stark's personal server. Presumably, he has designed his own supercomputer to handle all that processing.

>>5607065

>>5607619
Still, Jarvis would be impossible to replicate unless we manage to design learning algorithms for the machine to evolve by itself
>>5607631
that feel bro

>>5607637
>Still, Jarvis would be impossible to replicate unless we manage to design learning algorithms for the machine to evolve by itself
I wouldn't say so, but obviously we haven't figured out the kinks of Strong AI or we wouldn't be having this discussion right now

I honestly can't wait for the future. That Glass thing got me thinking about what life's gonna look like in 2040 and after.

>tfw technology

>>5607675
>yfw singularity

>>5607065
i only want his girlfriend

>>5607690
I only want his money
Then I can have everything else
>logic

>>5607701
>even with money you'll never have his charm, charisma and intelligence

I think that making the Iron Man armor is almost impossible for many reasons.
The only suit that we could be able to do is the Crysis nanosuit, based on reproduce muscles with nano fibers sensible to electricity.

>>5607710
>that fictional suit is impossible
>this fictional suit is totally possible
riveting exposition my compatriot

>>5607712

The crysis suit is much easier to replicate. You only need fibers that can extend and contract when you give them an electric impulse.

>>5607710
No

>>5607709
>charm
Practice socially. Honestly.
When you're rich you don't give a shit, I know I wouldn't. So I'd be much more outgoing.
>charisma
Yeah... You can become charismatic though
>knowledge
Educate yourself

>>5607716
What about the power source?

>>5607724
>falsely conflating knowledge and intelligence
point proven

>>5607726

Yes, the nano suit is just another layer of electric muscles that make you stronger

>>5607716
>implying you wouldn't just get crushed to death on the inside of a gigantic artificial muscle

>>5607730
intelligence implies knowledge
In Stark's case, at least

>>5607732

I'm not saying it's possible to replicate now but it has less problems than an armor like the iron man one and it's easier to project

>>5607637
>Jarvis would be impossible to replicate unless we manage to design learning algorithms for the machine to evolve by itself
We have learning algorithms. What we don't have are computers with the raw memory-searching, pattern-matching horsepower of the human brain.

>>5607741
It's sad you don't realize how retarded you are

>>5607742
Well, you have to account for a certain amount of cartoonishness in comic book movies.

Like in Iron Man 2, where he has the pocket blood tester that shows in big letters, "BLOOD TOXICITY 72%" or whatever. That's not how it would be, it's something that's done in a certain way on the screen so the audience can understand immediately.

And where he gets in robot fistfights, and falls out of the sky without being hurt, that's basically slapstick. His talking computer mostly provides exposition. Floaty glowing computer interface is more interesting to look at than typing on a computer.

People pick up on that stuff pretty quickly. Nobody talking about building an Iron Man suit is talking about bringing cartoon physics into the real world. They're usually talking about some level of impact protection and strength augmentation but especially flight, all without a big gas can on it.

>>5607763
Do we have the technology for jetpacks right now? Let's say it was connected to a source of power. I've seen the water jetpacks but shooting water to make yourself float is obviously not an option.

>>5607774
I am suddenly tempted to construct a stowable helicopter backpack, and make my supervillain debut as Ghetto Iron Man.

>>5607776
You'd become inspector gadget. I'm not sure this is something you want.

>>5607781
Hm... looking into this, I think it might be easier to make an electric scooter that folds out of my butt.

Even if it's possible it would cost more than 200 million
You won't have an iron man suit, ever, the government might but not you

>>5607763
>robot fistfights
Why wouldn't that be possible
>talking computer
That's actually useful
>glowing computer interface
HUDs are cool. And practical.

>>5605953
>>aerogel
>What? How is it relevant?

Seriously -- this is a material made to help keep open structures rigid without adding much weight.

An exoskeleton doesn't need that.

The Iron Man exoskeleton needs a massively powerful computer,
motors to assist in basic locomotion at all joints,
and deals with problems of heat dissipation, insulation, air circulation and air provision, waste management, etc.

I would suggest we cannot make a full exoskeleton that deals with ANY of those issues and is still roughly man-sized.

>>5608381
>we cannot
Decades ago you would've been told that "we cannot make a device that can hold more data than today's computers"
Smartphones exist, though.
It's the same for the suit. We see it as impossible because we haven't solved all the problems.

>>5608386
100 years ago you would've been told that "a heat engine cannot be more efficient than the theoretical limit of the Carnot cycle", and you'd still be right today, and in 100 years, and so on and so on.

>>5606112
>Jarvis may be on it's way,
Plain-language interpreters and contextually ambiguous command is pitiful, and that is the ONLY thing that distinguishes Jarvis.

>simply look at Glass from google.
What? Why are you giving credit to Google for something they haven't shown you?
Jeez, they make a couple vague remarks about what they hope to eventually produce and most people have given them credit for already getting it done!

>Years ago no one would've thought you'd be able to have some kind of real life hud.
Correct -- if by 'years ago' you mean before technology. Since technology, this has been an extremely common fantasy, and in the recent couple of decades many hobbyists have built their own without much trouble.
That's how bizarre and futuristic headset viewers are: it's been a hobbyist project for many years.

>>5608394
The suit doesn't directly break the laws of physics.
Even the impact/acceleration/decceleration problem may be solved.

Are HUD devices actually on sale right now?
I love this keyboard, and having things such as a phone w/ HUD, screens and all seems great to me, even if unpractical.

>>5606422
>>JARVIS is too intelligent
>The only problem is storage
Storage is the ONLY problem that we can assume will be suitably (!) addressed.

>Algorithms will eventually be implemented
That is a vastly unsupported vague assumption.

The only distinctive feature of Jarvis is plain-language interpretation, which is today absolutely horrible and which is improving almost not at all.

>>5608403
>improving almost not at all
Why is that?

>>5606498
>Let's assume the suit can tolerate and absorb every impact. You could just jump down a platform, thus generating power for the suit, fly away, rince and repeat.

You are assuming that a mere jump could provide just ANY amount of energy, and that is ridiculous.
A jump is limited to a very specific, and very very small, amount of energy.
You know how far it would project the man jumping, at total efficiency?
Exactly as high as he jumped. (a couple inches? a foot?) and no further.

>>5606536
>What about a speed of 90 mph? This is well above the speed limit in most countries and would allow air travel to get to places without cars.
>And let's assume suits like that actually are available to the public after a shitload of courses and tests, no one would just keep going at their full speed to land. They'd probably just put their extremities in front of them and reduce their speed like you in any kind of flight transport.

Trying to make sense of this post, but...
no one is asking for a tech that gets us moving just barely over our current speed limits.

I see no reason for you to assume people wouldn't be flying at full speed often.

People do not put their extremities in front of them to slow a vehicle.
No planes land and slow to taxi because the passengers put their arms against the seat in front.

>>5606551
>I think it would be better to do some kind of complex roll to remove speed.

No aircraft reduces speed this way to land.
Doing so adds the additional problems of forcing the body to endure pressure against the sides, balance, and huge perception issues.

or, did you think that suddenly spinning a car made it easier for the driver to park?

>>5606625
>I thought heat and all could be evacuated via nanotech (or nanotubes)?

Do you understand the thread is trying to answer this using real technology?

We don't have nanotech that moves heat; we don't have nano machines, we don't have nanotubes that know how to transfer heat in controllable fashion.

You just answered a question about real technology with the word 'magic.'

>>5607774
>water jetpacks
>gayer than elton john

choose two

>>5607079
>plus you know, its graphene. Its very much in the scientific vogue right now, but its not something for which scaling has even close to been dealt with.

Quite right;
why is it that people in /sci/ often see any new exploration in tech as 'already developed, and known to have incredible properties'?

graphene is being explored -- it is definitely NOT a known technology that we can exploit to produce this suit.
Same with nano tech machines, inertial dampers, and repulsor beams.

This thread is interesting

Where can I learn about electronics/mechanics as they're used in robotics and weaponry/aeronautics?

Got any ressources?

>>5607481
>and stops, the inside is completely "frozen". Maybe not frozen but the outside would be the only thing interacting with the outside world, leaving the inside to be moving freely. If that makes any sense.

it does; you are talking about something that suggests either time control or control over inertia.
Both are ENTIRELY fantasy.
Exactly what we are NOT including in the thread.

>>5608418
>>5608420
This is why I posted >>5608394
Technology is too unpredictable and the discoveries we'll make in a close future might make us able to build this suit regardless of the contraints we have right now ("physically impossible",etc)

Protip:
If something isn't directly disproved by the laws of physics (entropy, F = ma...), it's technologically possible, in theory, to make it.

>>5607710
>The only suit that we could be able to do is the Crysis nanosuit, based on reproduce muscles with nano fibers sensible to electricity.

That is an interesting alternative, but it's still silly.
Muscle twitching doesn't happen on the epidermis, so there really wouldn't be any control, and if it is a passive sensor it would be stimulated by many other muscles.

Also, moving the force away from the joints, and thus usable leverage, reduces its force and application.

>>5607763
>Nobody talking about building an Iron Man suit is talking about bringing cartoon physics into the real world.

The people actually building exoskeletons don't think that way, but I suspect most of the people ASKING about what is possible mean EXACTLY that kind of ridiculous super-powered mayhem.

Worse, they don't really want anyone to tell them it isn't possible; they want to hear only that they can have them, soon, without having to work for it.

>>5607774
>Do we have the technology for jetpacks right now?

We did over 40 years ago -- AND they made them, AND flew them, AND still do.

How are you not aware of these?

No matter what kind of suit you build it will be limited by the g-force tolerance of the person inside it, just like current fighter planes.

>>5607774
One famous usage of a jet pack was the opening ceremony of 1984 Olympics.
http://en.wikipedia.org/wiki/1984_Summer_Olympics#General

ZPE could be the power source

>>5608386
>>we cannot
>Decades ago you would've been told that "we cannot make a device that can hold more data than today's computers"
>Smartphones exist, though.
>It's the same for the suit. We see it as impossible because we haven't solved all the problems.

Not knowing what we can do because of ignorance does not suggest what we might be wrong about after we are no longer ignorant.

We're not talking about people with no understanding of technology having a wide-open field to make anything you can imagine.
We're talking about people who really do have a grasp on what can be done (and this thread is about today and near-future, not distant future tech).

We see the Iron Man suit as impossible not because we do not know how to solve some problems, but because what Iron Man does is ridiculous in the extreme.
Please note the issues that have been brought up:
power supplies inferior by hundreds of magnitudes
inertial issues carrying a human pilot
heat dissipation
structural impossibilities
nearly every valid suggestion has to be built outside the suit

>>5608386
>Decades ago you would've been told that "we cannot make a device that can hold more data than today's computers"
>Smartphones exist, though.

That was people accurately talking about near-term technical possibilities.
It wasn't people making long-term predictions about anything that MIGHT be invented.

I really don't know why people think the ignorance of long ago means almost anything is actually possible today -- it's like they don't believe anything we have learned can be true.

>>5608399
>The suit doesn't directly break the laws of physics.
>Even the impact/acceleration/decceleration problem may be solved.

yes, it does.
There is no reason to perceive ANY of the motions might be possible.

More to the point: we do not know any way they might be solved, which is the question.

>>5608406
>>improving almost not at all
>Why is that?

Normal human language is not full of patterns and formal syntax.
people change topics, address different grammatical subjects, use indirect and non-specific pronouns, drop nouns, assume continuity of subjects, add variant instances, add interjections, use code references, change referents, use non-verbal cues, add emphasis with facial cues, and much more.
As a listener, we evaluate context, assume subjects, revert assumptions when there are failure cues, ask for clarifications, use non-verbal cues to get corrections or express problems, interpret vocal stresses and cues, carry previous subjects over, reload emotional contexts as needed, and much more.

In movies, we excuse all of the missing things because we don't have to consider the many other things we do.
But you can experience it if you watch a scene where a robotic voice with no visuals tries to mimic a person. It's pitiful, it's sad, and it always misses the major purpose and point of the conversation.

>>5608423
>Technology is too unpredictable and the discoveries we'll make in a close future might make us able to build this suit


You are denying ALL of our knowledge to say that.
Of course we MIGHT eventually produce a tech that overcomes one of the major issues named.
But the point is, RIGHT NOW, we do not see a way for that to be possible.
Therefore, we actually have to say it is likely not possible.

We really have learned about some limits, and we can definitely assume that most of those limits are real.
We do not assume all limits go away just to be hopeful.

>>5608440
>ZPE could be the power source

once more, for those that still don't get it:
thread is about what we CAN DO.

Not about vaguely assuming we'll be able to do everything, and write it all off assuming everyone is totally ignorant today.

>>5608460
Even if we don't manage to make it efficient, ZPE can be harnessed with little to no efficiency and still be very powerful

>>5608456
So that takes us to the problem of storage, right
Because the algorithms and patterns of recognition we need to implement for an artificial intelligence to deduce context, irony, subtlety of human language is possible right?

>>5608442
>>5608446
>>5608447
What I mean, is that we can't immediately say something is literally impossible because we don't have the technology right now.

Instead of doing this, we should search for ways of working towards the problems mentioned.
The suit's physics are impossible to replicate in our current state of knowledge about technology and engineering, but we don't need to project ourselves in the far future. A discovery about viable power sources could make us advance a lot, and the only remaining issues will be solely about problems like heat evacuation and impact absorbing, which need research and development (or compromises) to be solved.

(II'm interested in ressources though, see >>5608421)

>>5608461
>So that takes us to the problem of storage, right
No; not storage, interpretation.
It would be a storage problem if all human communication could be stored and pulled from a databank for it's literal meaning -- we don't talk that way.

>Because the algorithms and patterns of recognition we need to implement for an artificial intelligence to deduce context, irony, subtlety of human language is possible right?

We assume so, but no one has come remotely close.
So they reduced the expectations -- in the field, they don't require emotional context or expressions (they remove them), and they restrict verbs and adjective usage.

The last huge problem I was reading about was trying to get a computer to retain relevance when people started using pronouns and dropped specific referents. None of the systems was very good at the time.
The assumption was that the listener does a lot to juggle the potential referents, take in emotional and non-verbal cues, correct them from a field of possible variables, and error-correct them against experience of how people lead conversational structure.
That last may be surmountable; it assumes most of a person's conversational structure (think of 'plot') is mimicked from experience. If so, then billions of recorded conversations in many media might be encoded to produce similar structures to help guide interpretations.

Everybody, think back to middle-school exercises in reading comprehension. Those were necessary practice for figuring out language decoding; they were also context and examples. But we have no way to get a computer to learn the same things we did from those, or from all of our talks with friends and coworkers, etc.

>>5608467
>What I mean, is that we can't immediately say something is literally impossible because we don't have the technology right now.

But we can do exactly that, and we must.
Look at it this way: if we know about inertia, and the forces in moving objects, mass and momentum, how a body fits into such a suit, and what kinds of materials could pad him, etc.

Put that into a situation where he is moving at perhaps 400kph and stops, within a quarter second or so, on the ground in a crouch...

Knowing all that, it is completely wrong to say 'yeah, we might be able to build such a suit.'

To say we could do that denies all the things we know and elevates the assumption of ignorance (the assumption we will overcome all limits regardless of our knowledge).

That is exactly the opposite of trying to be correct, accurate, real, knowledgable, or practical.

>>5608473
>interpretation
I'm not really good at anything AI related, do you mean we should teach the AI to analyze the context (so it should compare it to the rest of the conversation), tone (so variations in the voice), and implications/expressions/figures of speech to be able to hold a conversation?

From what I've seen, right now, an "AI" (things like Apple siri, I don't know if you can call that an AI) just analyze the sentence, pull out keywords and answer accordingly.
Does it depend solely on algorithms and analysis/machine learning?

>retain relevance
For the computer to do that it should compare the ongoing conversations with ones he previously had and deduce what the human means?
>emotional and non-verbal cues
How could we even do that with algorithms? You can't implement empathy
>experience
Yes that's the easiest, but the others...?

>>5608478
>I'm not really good at anything AI related, do you mean we should teach the AI to analyze the context (so it should compare it to the rest of the conversation), tone (so variations in the voice), and implications/expressions/figures of speech to be able to hold a conversation?
I am saying that plain-language conversation (the kind of talking everyone reverts to and considers most natural) would require all of those and more, yes.
But it's worse than that -- because human conversation isn't just a set of specific cues and contexts, it is also variant and inconsistent, in nearly every respect. People just don't have structured or formal speaking habits.

>From what I've seen, right now, an "AI" (things like Apple siri, I don't know if you can call that an AI) just analyze the sentence, pull out keywords and answer accordingly.
Siri is a good example of much of this: we know we are giving a command, so we change the way we talk. We know we have to make it short, not transgress or take tangential courses. We also deliberately restrict our range of referents and verbs to what we already use on the phone and in PDAs.
And Siri still, of course, gets much of it wrong -- far more than we would accept in Jarvis or the Enterprise computer.

>Does it depend solely on algorithms and analysis/machine learning?
Actually, my knowledge regards the human communication part; that's why I see this problem as so huge.
Modern grammar checkers and vocal tools like Siri are already better than I expected today, and most computer experts are hoping for a paradigm shift in software design, so there is considerable hope.

Also, there is a lot of room for people to become better in communicating. Perhaps those that become better and communicate well with computers will be the guide for others to improve communication habits. People learn most of it from hearing others, after all.

>>5608477
I'm not saying we should deny everything we know, just assume that it's eventually possible to do such things despite out lack of knowledge in the subject.

Now it may not be relevant, but we should still consider that what isn't directly disproved by physics is theoretically possible. That's all I'm saying.

Telling ourselves "well, we don't have the technology, so it's impossible, let's move on, period" isn't constructive either.
We might actually reach a point where our technology enables us to do such things in a close future, and not necessarily in several centuries.

>>5608478
>>emotional and non-verbal cues
>How could we even do that with algorithms? You can't implement empathy

We don't need to induce emotional understanding, only recognize the cues of emotions so that the proper shifts can be made in interpretation.
Human emotional communication uses fairly limited cues, but most are quite subtle, guided by our hardwired understanding of the human face and a huge amount of training.

(I have no idea if anyone is working well either with vocal emotional cues or facial emotional cues).

>>5608523
>we should still consider that what isn't directly disproved by physics is theoretically possible.

I can accept that; you're allowing for quite a lot in that,
and it still doesn't lock out the possibilities that some of our fundamental science will shift.

>>5608523
>We might actually reach a point where our technology enables us to do such things in a close future, and not necessarily in several centuries.

Agreed, and accepted --
but what we know NOW is my point.
We must never learn something, but then deny it because we know we once knew less and were wrong.

ALWAYS stand on the grounds of the facts we know best -- and not the possibility that we might be wrong about something.

>>5608522
>variant and inconsistent
I guess that if you don't speak clearly enough, the computer could still understand based on the context. So one of the main problems should be giving him the ability to analyze a sentence and deduce its context based on keywords or something, while being more sophisticated than Siri.

Isn't this the main problem?
If a computer can understand the context of a sentence, he can answer with accuracy assuming he can recognize irony, patterns and implications.

If the speech isn't structured, then the computer could still understand it if he can analyze a few words, parse the rest of the conversation and decide of a relevant answer. Would it be very difficult to implement?

>we know we are giving a command, so we change the way we talk
See above.
>far more than we would accept in Jarvis or the Enterprise
Well, those two would have much more ressources than Siri anyway. You wouldn't compare an iPhone and a supercomputer, though it is down to algorithms, storage is important.

>paradigm shift in software design
Could you elaborate on that

>recognize the cues of emotions
You would need to program a pretty complex facial analysis and recognition ability, voice/tone interpretation and analyze that based on the *meaning* of the previous sentences.
I don't think you can implement something that deals with abstract stuff like emotions without getting into psychological analysis. Face recognition isn't enough

>>5608530
Indeed, some laws of physics may be disproved in the future.
Everyone shat their pants with the neutrino experiment, and some thought Einstein was wrong. That shattered our understanding of physics, and while some people did think the experiment was flawed, it left an opportunity for our theories to be completely renewed.
>>5608537
>what we know NOW is my point
I know. I never said we should deny what we know right now, just assume that it has chances to not be correct, but also that it has chances to EVOLVE (without being disproved) in the future, allowing things to exist that we didn't think could exist.

>>5608538
>>variant and inconsistent
>So one of the main problems should be giving him the ability to analyze a sentence and deduce its context based on keywords or something, while being more sophisticated than Siri.
>Isn't this the main problem?
Yes, that's the vocal (direct) part.

>If a computer can understand the context of a sentence, he can answer with accuracy assuming he can recognize irony, patterns and implications.
Irony and sarcasm may forever be a problem; implications probably are not, since interpretation by computer presumably has to review those anyway.

>If the speech isn't structured, then the computer could still understand it if he can analyze a few words, parse the rest of the conversation and decide of a relevant answer. Would it be very difficult to implement?
It's what we haven't managed yet. Even assuming the speaker kept to just one topic deliberately, used few indirect pronouns, changed referents rarely, used short-clause sentence structure, and few adjectives, the point or goal of the speech can be easily missed.

>>recognize the cues of emotions
>You would need to program a pretty complex facial analysis and recognition ability, voice/tone interpretation and analyze that based on the *meaning* of the previous sentences.
>I don't think you can implement something that deals with abstract stuff like emotions without getting into psychological analysis. Face recognition isn't enough
It isn't for resolving the emotional issues, but it might be for proper interpretation.
Consider that most people learn the habits of expressing emotional content; turning our eyes down, pouting, rolling eyes, tilting our heads -- these are overt, and not natural or instinctive. Some people try very hard to put those cues together, even if they aren't particularly aware of doing it.

>>5608538
>>paradigm shift in software design
>Could you elaborate on that

(Again, this is from reading topical overviews, not personal knowledge of the leading edge.)
Software design hasn't changed very much. Certainly much less than hardware.
We've had major additions; Java, Python, SQL, XML -- but it mostly works within the same or smaller structures that we already use.
Programs become hugely bloated rather than elegant or efficient; we used to call it 'tight.'
Folks in the industry hope for a paradigm shift, so that software design can offer something more impressive than those bloated hulks do today.

>>5608558
I see. So "enterprise code" would become elegant.

>>5608556
>Irony and sarcasm may forever be a problem
Not sure. I'm no expert, again, but if a computer managed to recognize:
- Facial expressions commonly used when giving an ironic speech
- Voice tone that often shows irony/sarcasm
- Use of exaggerations/euphemisms or other stylistic figures that are easily recognizable
- Comparing all those to the context of the speech
Might be possible.

Well, say the speaker said "Dunno, gimme my contacts list" instead of saying "I don't know; display my contacts".
The computer could analyze the "dunno" and see it's often used as a replacement for "I don't know" (that would require a database where he can compare what the user tells him with common speech examples). Then, same thing for "gimme". As for "contacts list", he knows what it is: to decide of the operation to do (Delete? Display? Print? etc), he just compares the previous words (in this case, "gimme", so "give me") and evaluates the action that has the most chances to correspond to what the user wants. So you'd have to link "give me" to several actions (printing, displaying); the computer would then either choose one of those based on the previous sentences (so by parsing the rest of the conversation and deciding of an action) or by asking the user if he wants to print or display.

cont

>>5608588
cont

Either way, emotions (be they unnatural expressions or natural tones of voice) would need an extremely accurate facial/vocal recognition program.
For instance, those habits (rolling eyes, tilting head...) would have to be assigned to a specific "state of mind" the computer would have to choose (anger, sorrow...). If it's ambiguous (some things can mean sarcasm or anger for example), the computer relies, once again, on the previous sentences and expressions his user has had...

I find that to be pretty interesting, would you know of any ressources where I can learn basics of AI?

What should I study to be proficient in robotics and all?

>>5606012
>>5608730
Mechanical engineering

>>5605890
"Repulsor" techonolgy does not exist, neither is there such a thing as an "arc reactor", and there isn't enough room in his suit for servo motors that are as strong as his suit is, so we can't get closer than a movie.

>>5608847
did you even read the thread faggot

>>5608836
Actually no, electrical engineering seems to be more logical.
Also physics.

>>5608847
>arc reactor
Of course not.
But miniaturized energy sources can be created

I've read the thread and you guys are contradictory as fuck

Basically, what's the list of the problems we have to deal with, that make the suit "impossible" to build for now?
What fields do these problems belong to?

Why is there such reluctance to use small gas fueled generators.

>>5609246
Wouldn't werk

>>5609268
Why?

What's more useful when trying to build such a suit:
Electrical engineering, Mechanical engineering, or something else?

>>5609516
Electrical engineering for integrated circuits and mechanical engineering for the joints and stuff. You also need aeronautics/aerospace engineering knowledge for the aerodynamics of the suit, and a good knowledge of physics

>>5609537
Thanks.
Do you know of any good textbooks for that?

Just wait a few years.

http://www.highexistence.com/10-ways-the-next-10-years-are-going-to-be-mind-blowing/

Technology is advancing very quickly.

>>5609742
>implying this has any thing to do with a high tech exoskeleton

Could that machine and the mechanisms be replicable in real life?

>>5608401
This please

>>5608422
Inertial dampeners.

>>5609913
Sure but what would you use it for?
>>5609915
HUDs are extremely unpractical and ridiculously expensive.

>>5611230
>Sure but what would you use it for?
Cleaning up old people who have soiled their beds and moving them to their powered wheelchairs.

>>5609960
>Inertial dampeners.
For when you need to get some inertia wet.

("dampers" do damping, "dampeners" do making things damp)

>>5611348
Speaking of which, Roujin Z is ripe for a Hollywood remake.

I'm not interested in creating exoskeletons but weaponry seems really passionating.

Does anyone know what ressources I should pick up to understand how missiles, satellites, bombs and all work?
Do I need chemistry, engineering?

>>5611388
What you want to do is call the Department of Homeland Security and tell them that you're interested in learning about how to make bombs and missiles.

This will get you started on the road to exciting involvement with the defense industry.

>>5611391
>implying

No, I want serious answers.
I don't even live in the US, no DHS here

MechEng, ElectEng...
These fields will help you develop the articulated joints to mimic a human skeleton, as well as actuate them with servos etc.
CompSci
In order to program said servos holistically to simulate walk/run etc.
Theology
To feel like God when you create life from metal and animate it mwuahahahahaha

>>5611398
>I don't even live in the US, no DHS here
Drones en route.

>>5611403
Wat

>>5611400
Don't forget physics to actually figure out the power source and the constraints

>>5611400
I am become Brahma, creator of men

>>5611400
>CompSci
>In order to program said servos holistically to simulate walk/run etc.


No, all compScier know is how to make calender apps on Iphones. Those servos will be programmed by the ElectEngs

>>5611896

I read that as 'Bahama' at first...
and was much confused.

>>5611917
lel

>>5611905
Uh no
You don't learn AI in EE

>>5612040
There can be a lot of programming in EE.

Anyway, programming can be done by anyone. It depends far more on individual talent and self-education than formal training.

What about using metamaterials to redirect force around the suit to the other side and then use either an elastic gel or some sort of liquid metal to expand and rapidly contract, damping the force.

Plus, ac/deceleration isn't as much of a problem as you think it is. Humans can withstand 3 to 5 gs if they're in the right position and wearing pressure suits.
At that kind of acceleration you can reach mach one in 7 to 12 seconds and get back to a dead stop just as fast.

>>5612806
That's a good idea, it seems.

>>5612806
>What about using metamaterials to redirect force around the suit to the other side and then use either an elastic gel or some sort of liquid metal to expand and rapidly contract, damping the force.
This is just comic book technobabble. You're not describing something that can actually be done.

>ac/deceleration isn't as much of a problem as you think it is.
Acceleration is the problem with getting slugged by the Hulk, crashing into the ground, or being hit with heavy weaponry.

Even if you've got a magically indestructible armor shell, you're going to get concussed to death by the sudden change of your head's velocity.

You'd need some kind of evenly distributed force like artificial gravity to accelerate the brain at the same speed as the skull, so it doesn't slosh up against it and get fucked up. Maybe you can say this is part of "repulsor" tech, but that doesn't suggest a plausible real-world mechanism.

>>5613012
>This is just comic book technobabble. You're not describing something that can actually be done.
If you can redirect light, sound and earthquakes, you can redirect force.

Think of it like a faraday cage, but for physical impact.
Make the material ripple like water, absorbing force and converting it into heat, sound, and movement.

It's sort of like a whipple shield too.

>>5613046
You know what "redirects force to the other side of the suit"? Rigidity of the suit.

Know what happens if there's nothing on the other side of the suit to redirect the force into? The suit, and everything inside it, gets pushed by the force regardless of how rigid it is.

If you want to reduce the acceleration caused by an impact, you need thickness and breakdown of rigidity, like a car's crumple zones.

>>5613094
I mean doing something like this

>>5613128
It couldn't be a tight suit. It would have to be very thick, again, like a car's crumple zones.

Electromyography can be used to detect motor signals and move at the speed you require. Internal (needles, perhaps implants) sensors are the most accurate, but recently external (pads) sensor based systems with quite good accuracy have been appearing on the market.

Do aerospace/medfags know the kind of forces a body can take? speeds would have to be based on those values.

>>5613128
Even if you could do that on a scale as small as the Iron Man suit, you forget that our organs would still move inside our body e.g. concussions. Unless we could fill our body up with some substance that completely stopped everything from moving only when we wanted it to, such a rapid stop couldn't happen

>implying anyone here could figure out what technologies to use to build an Iron Man suit
You need a fucking PhD for that stuff

>>5605890

I'm building an electrically powered exo-skeleton. I'm going to use a couple of Prius batteries to power it. It probably won't fly, thouhg.

To what extent could nanotechnologies, assuming we could use them, help in the design of the suit?

>>5614342

Those things are prohibitively heavy.

Energy density laws prevent any sort of iron man suit. Let's not even discuss how thermodynamics would melt the thing and the occupant if you tried to fly with it.

It's just a fictional concept. Like FTL and space elevators.

>>5614315
>You need a fucking PhD for that stuff
How does your mind work?

>>5614342

>need batteries

>batteries are too heavy for servos

>need more powerful servos

>servos take too much juice

>need more batteries

>servos can't handle the load

>need better servos

And so on and so forth. It's like rocket science, you can never break even, you can just barely make it work.

>>5615122
>fictional
Yes, doesn't mean impossible
>>5615123
The level of knowledge in EE and ME is far higher than anyone here knows

>>5615127

>Yes, doesn't mean impossible

Point being? The iron man suit breaks several laws of physics. It is impossible. Fiction or no.

>>5615133
>breaks several laws of physics
As it's shown, probably (impacts and all)
But making an exoskeleton with flight, weapons and all isn't impossible

>>5615127
>The level of knowledge in EE and ME is far higher than anyone here knows
So, "the level of knowledge in EE and ME is far higher than" you know?

Then why do you assume that this stuff you don't know is required?

>>5615138

Because he acknowledges the existence of people smarter than himself.

You should try it someday.

>>5615137

Are we talking about an iron man armor or something completely different like a powered exoskeleton?

Making it with flight is damn near impossible. Where do you keep the fuel? Why does it need to fly? What sort of engines propel it? How do you thermally isolate their waste heat from the pilot?

If it's going to fly, then just make a VTOL jet. It will have a greater range and payload than some man-shaped thing, and it won't have nearly the amount of failure modes as a flying exosuit.

>>5615142
We're going to end up arguing about the evolution of technology and how it's potentially possible
I'll stop

How do I become Tony Stark, /sci/?

>>5615144

Yeah, it was probably going to lead to the "technology will eventually break physics" idiot talking point.

Flexible actuating materials would go along way to make something a bit more human shaped. Currently given shape restrictions we would have to use a high number of solid actuators to get the movement required.

But there are still huge barriers in power generation, material technology and the fuel-less propulsion.

>>5615154
>implying I said that
It was going to end up taking about how today's technology is limited and tomorrow's technology might give us more opportunities, that's all

>>5615155
>But there are still huge barriers in power generation, material technology and the fuel-less propulsion.

That cannot be surmounted any easier than the conservation of energy.

I don't really know what to think about what's been said in this thread.

Does the Iron Man suit DIRECTLY violate some laws of physics? What aspects of the suit would need to be toned down for it to be, in *theory*, physically possible, regardless of the nature of the technology we should use to make it possible?

Now what are the technological limitations that make us unable to create such an exoskeleton? Can those limitations be solved in our lifetime, in your opinion?

>>5615225

Pretty much.

>>5615143
>herp derp hurrrrr

>>5615225
>What aspects of the suit would need to be toned down for it to be, in *theory*, physically possible, regardless of the nature of the technology we should use to make it possible?
The big problem is Tony not getting concussions and similar high-acceleration injuries when he gets knocked around in the suit.

The toughness of the suit itself is also pretty much comicbookland stuff. Chemical bonds just aren't that strong. Accordingly, the strength is quite absurd. He throws loads around that should crush the suit's structure.

Such compact and powerful thrusters may be possible, but we have no idea how to make them.

>>5615644
To address the pressure issue, wouldn't filling the suit with some kind of perfluorocarbon help with that? I've read somewhere that it can help lessen the effects of g-forces by around a factor of 6 or something like that.

Obviously achieving the sort of performance in the movies is out of the question, but it doesn't seem implausible to believe that a suit that brings your strength up to that of a couple of body-builder's is out of the question.

>>5615644
>concussions and similar high-acceleration injuries
Several ideas have been formulated in the thread - do none of them work?
>toughness of the suit itself
>chemical bonds aren't that strong
Okay.
I'm going to get called out for pseudoscience, but don't nanomachines (not that we actually know how to use them, but hypothetically) operate at the molecular level?
>the strength is quite absurd
Yeah.

You don't need the suit to absorb impacts exactly.

Relativistic quantum lock would do the trick. (would be fun too.. Superman's infinity punch would theoretically be stopped up to a certain sustainable point, as Stark would have all of matter to distribute the force. Bye-bye universe....)


The key isn't the component, but rather the function of the interaction of particles. Modify the phenomena that allows matter to pass through each other such that the impact source is more evenly distributed through the entire atomic structure of the suit, in lieu of a limited positional structure which affects the rest of the suit. Doing it this way, you could have an absorption rating much higher then the more traditional deflection and absorb methods. (say the first 5 layers of atoms respond to the impact interaction instead of the first 1 or two...)

>>5616231
>Relativistic quantum lock
What is that, does it exist and where can I learn about it?

>>5616231
>Relativistic quantum lock
10/10

>>5616235
Lol
It doesn't exist

Hey guys
Hey
Guys

What if we wanted to create a drone, instead of an armor? Would that be more realistic, guys?

>>5616231
>Relativistic quantum lock
What? What's that?
>broscience

I can access the deep web with my tl.0.0 closed shell system quantum lock too

Check this out: robotics becomes so advanced that robots frame mimics humans except much stronger. Advances in neuroscience directly locate our individuality in brain; pattern of neurons that creates personality. Directly upload to robotic core. No need for suit. Profit!

>>5616481
See >>5616389

>>5617302
>drones
The impact problem would be solved.
Not the propulsion problem or the power source one.

Honestly, if I can even get that user interface/HUD/AI he has in his helmet, I'm good. I don't need a suit.

https://www.youtube.com/watch?v=ZwOxM0-byvc

>>5617519
Replicating the interface is easy. There are tutorials out there. Not sure about HUDs.
Gotta wait a few years for the AI, though, and have enough money to buy a supercomputer

Which aspects of the suit can we replicate in our current state of knowledge?

I'd like to know if those HUDs and stuff are actually possible/can exist though (the link >>5617519 posted)

>>5619523
bump for this

>>5620115
>>5619523
you taking about this?
Yes, it's possible, but you likely can't afford it

>>5620147
half of those holograms are gimmick anyway

What programming languages are commonly used for AI and such interfaces?
C?

>>5617498
Implying in the future we won't exist entirely in virtual reality where nothing will be limited by physics.
Implying that we don't already exist within a simulation.
Implying computers will be able to simulate universes and individual conscious patters in its inhabitants in a few years.

we all fuqd

>>5620464
You forgot the meme arrows.
Also, if the world is a computer system, why can't we crack reality to execute a script, bending the laws of physics?

Don't derail the thread though, create another one to discuss this (stay classy, /sci/)