Oh, neat!

Thank you! Part of my idea behind it is that it would also buy the wearer more time to figure out that they're under attack. Though I imagine that the armour's internal computer would probably create a trace for any lasers fired - hence, whilst in reality the beams are invisible, the user is "seeing" a false colour beam simulated by the helmet. Though, of course, there's no reason that a particularly vicious electronic warfare system couldn't create false or misleading beam traces in order to confuse opponents.

Yep, my setting has those sorts of grenades (the basic react to contact drill when fighting an energy weapon equipped force is to immediately pop aerosol grenades and return fire whilst seeking cover). I imagine vehicle mounted versions would automatically go off if they detect an incoming laser.

I don't see why not. If you made it out of a material with an extremely high albedo or refractive index (i.e. extremely close to 1) then you'd probably have a decent anti-laser coating. Perhaps it would be like road signs in that it's covered in millions of microscopic glass (or future equivalent) spheres that reflect the light back in the direction of the attacker, possibly blinding them (or at least dazzling them).

I'm not particularly sure. On the one hand, exotic stuff like that would be right up the Uplifters' alley, but it also feels slightly underwhelming (I'm still trying to devise suitably exotic weapon systems and failing miserably).

Moore's Law is the one that states that processing power doubles every [x] years, right? If so, then there are some really tantalising prospects in store.

Locking you up on radar since '09

Moores law has a lot of catches and has a lot more discussion into it. Including the widely held belief it has hard limits and the fact that as things change the law was pretty much doomed to becoming obsolete by those very advances.

Who watches the watchmen?

Still though, I do think Sci-Fi writers generally don't get just how utterly ridiculous their computer tech SHOULD be given the time period.

Like Halo for example. Five hundred years in the future and some of the tech they use I could easily see being developed in the next hundred years.

New Survey coming this weekend!

I guess it's another example of our kind having no sense of scale. See: BattleTech's Helm Core and its kilobytes of precious, precious data.

Though to be fair, none of us are clairvoyant and it's honestly a little unfair to expect someone to accurately predict what computers (or other tech for that matter) will be like even 200 years from now, never mind 500. To follow on from my above example, the first versions of Battle Tech were written in the 1980s - they couldn't really have predicted the massive growth in computational power available today.

Locking you up on radar since '09

^ Except Moore's Law was coined in the 1970s. There were quite a few people who saw we'd be well beyond megabytes way back then.

"Allah may guide their bullets, but Jesus helps those who aim down the sights."

A quick read for those of you planning to bring The end of a world as they knew it and people won't feel fine.

All night at the computer, cuz people ain't that great. I keep to myself so I won't be a case on The First 48

Remember when we were talking about a sort orbital satelite/mine type platform for blockades?

Who watches the watchmen?

@Taira:

It reminds me of this article I read once about "engineering archaeology", as the author put it. Basically, since they had no idea how certain things worked any longer (the facility was pretty damn old) they had to pore over various documentation to try to figure out what did what.

I'd link it, but I've got no idea where to find it.

@Tuefel:

Huh, neat. I guess our ideas weren't completely screwy, then!

Locking you up on radar since '09

Moores law has already run into some pretty hard limits right now (the double processor speed part hasn't been true for quite a few years now, although the size, price and power consumption still shrink on schedule), so i think older writers might be forgiven for wrongly guessing where the development would start to slow down again. What's imo a much more egregious mistake is writers who have no idea at all what sort of task is easy for a computer and what isn't. The quite well known Weber for example has this ridiculous situation where the computers powerful enough to alter the look, sound, language and accent of a person in a video-conference on the fly and convincing enough to fool other humans. But ballistic calculations are so hard that people are required to do them because the computers can't deal with randomness well enough. (When that's pretty much what computers were originally invented FOR)

People are not actually required to do the calculations anymore and haven't been since WWII. Though many who are in charge of gun laying are required to know how to do it manually if need be or if they need to make sure the system is working properly under certain conditions. M3 and M4 ballistic computers for example were coupled with director systems proved far more accurate in aiming AA artillery then humans. They drastically increased the effectiveness of AA fire. Now these machines were rather large and bulky and their successors are of course less bulky and more effective by comparison.

There are plenty of modern ballistic computers that do it far better then humans. They are built to be better at then people are and can adjust calculations for things like weather, humidity, air density, round type, drag coefficient, propellant target distance, target angle etc with a far higher degree of accuracy then people can now.

There are hand held ballistic calculators for artillery. There are ballistic calculators for extreme range sniping such as with the Cheytac Intervention M-200. It is coupled with a hand held weather monitoring device and a laser range finder spotter scope.

There are now also rifle mounted targeting systems that tell you where to aim and are poven to be incredibly accurate.

edited 5th Jul '14 4:38:03 PM by TuefelHundenIV

Who watches the watchmen?

How feasible would be a vital-point targeting system be (admittedly it only works on things whose internal schematics/organ locations are known)? And would it make sense for such a system's user interface to resemble the Syndicate reboot's DART Overlay as a toggleable part of a Heads-Up Display? I don't really think that this sort of stuff can be scaled down to current rifle-mounted optics without stretching plausibility.

I have disagreed with her a lot, but comparing her to republicans and propagandists of dictatorships is really low. - An idiot

Not very. It's possible to create one for humans since our internal layout is pretty much the same on all M-1 Homo Sapiens but for other things you'd be guessing a lot of the time. For example a T-90 does not have the same layout as a T-72.

"Allah may guide their bullets, but Jesus helps those who aim down the sights."

"Aimbot" type systems like the one you're describing have appeared before in sci-fi, and everything is getting more accurate all the time, so it's not outside the realms of possibility. In fact, I wouldn't be surprised if several hundred/thousand years from now such aimbot systems are as commonplace as chewing gum is today.

It's been a while since I saw a LP of that game, so my memory of the DART overlay is a bit fuzzy. However, it didn't seem too busy (a lot of sci-fi stuff has extremely cluttered displays) so there's no reason it couldn't be like that.

As for the idea of rifle mounted optics version of the system, I wouldn't be so quick to dismiss it. Computers may be powerful enough in future that they can perform such "off-the-cuff" aiming with relatively little effort, and in very small packages to boot. Early versions may be linked to the user by cybernetics or in-armour computer suites which do all the heavy lifting. Or it could all be done "in-house" by the cybernetics/computer suite depending on the system.

EDIT:

Tom does have a point, though. The system would need to memorise hundreds (if not thousands) of different designs, all with their own quirks. So an aimbot system would probably be concerned more with maintaining ridiculously high accuracy at rapid speeds than dismantling something surgically.

edited 7th Jul '14 6:07:52 AM by Flanker66

Locking you up on radar since '09

It would be easy in that it could be programed with a database of the most likely threats, i.e. the most common tanks and aircraft. Or it could scan for likely weakpoints but it's the operator who'd have to be trained to tell what's real and what would shrug off a shot.

All tanks that want to move quickly have soft spots on the bottom of the hull and the treads are another vulnerable spot. A mecha would need to protect it's joints, sensors and ammo.

All night at the computer, cuz people ain't that great. I keep to myself so I won't be a case on The First 48

Scanning for likely weakpoints is probably not going to work very well, computers are much worse at pattern recognition than humans are. Having a database of known targets however should be very easily done (remembering data is something computers are much better at than humans).

That said, vein-o-vision is not a useful implementation, the shooter doesn't really need know the internal layout of his target, he just needs to know where to hit, so a simple highlighting of weak spots/critical components provides all the same utility with much less clutter. A simple colour coding or labelling could provide additional useful information. (like shoot here to immobilize the vehicle or here to make it blow up)

edited 11th Jul '14 5:31:16 AM by mckitten

^ Even that requires a lot of designs pre-programmed into the system. "Shoot here to immobilize" are two different things if you're talking about say a GAZ-2975 "Tigr" light vehicle and an M1 Abrams MBT.

"Allah may guide their bullets, but Jesus helps those who aim down the sights."

It seems to make more sense for the computer to forward sensor readings to a human analysis specialist who marks off weak points for the computer to aim for.

The data base itself might not exist in the weapon itself. If you have say power armor with mini-super computers it could contain a vast data base with known threats. One could even theoretically embed analytical software into such a system.

Or there could be a BDA guy with the kit whose sole job is to record and transmit that data.

edited 11th Jul '14 3:54:57 PM by TuefelHundenIV

Who watches the watchmen?

I personally think such systems would just do a geometric analysis (something easily done on a relatively low power computer) aka how big it "looks" at a certain range and throw in a few factors later would paint a big "Shoot here for best accuracy" tag on the target through it.

"Allah may guide their bullets, but Jesus helps those who aim down the sights."

Best accuracy is easy, just aim for the center of mass

I think Tom meant for effect not accuracy. As in shoot the weak spot.

Who watches the watchmen?

No it was geometric accuracy. While it wouldn't be shooting for center mass ^note because maybe the analysis suggested the "center" was either highly likely to absorb/deflect the shot or that particular part had a lot of empty space around it all the time, it would be suggesting angles where you have the greatest probability to ensure a round on target at all.

"Allah may guide their bullets, but Jesus helps those who aim down the sights."

Yeah, but so what? Data Storage is cheap, and getting cheaper all the time. Even a portable computer with limited storage space is at the very least able to store far more design plans than the user would be able to remember. The difficult part for the computer really isn't the data, it's the accurate recognition of what type of vehicle it's looking at. A less than idealla-advanced targeting computer might not even be able to do that at all, but would rely on the user to tell what type of vehicle it is, but could still easily store hundreds or thousands of blueprints. So the user picks the type of vehicle and the computer shows the user where the weak spots are.

Pattern recognition is very hard to program because there are an almost infinite number of variables involved, not to mention their interaction. To pick a theoretical example, a program could have a hard time deciding whether it's looking an a vehicle of Type A or Type B if it finds that the outline looks like A but the colour looks like B and the programmers didn't properly assign priorities. Or a half-visible shape in a foggy swamp might look to the computer like either a tank or a boat while a human would immediately know that it can't be a tank because it couldn't drive on the ground.

Heck, check out the origin of the GISS acronym. We don't even really understand how humans do pattern recognition.

edited 12th Jul '14 3:02:28 AM by mckitten

Another step closer to mass production of Spider Silk

What you could do instead is build a profile of known equipment using a larger pool of data then just shape alone. Thermal signatures, EM Emissions, engine noise, weapon noise, anything you can feasibly measure about the target and have sensors on a system detect. The more diverse that profile the better the system will be at id'ing a specific enemy object. You could have a system that can process the data needed for it but it would be reliant on sensors feeding it information to process in the first place.

But as Mc Kitten pointed out the battle field isn't exactly a sterile sensor environment. The accuracy of the system will vary under battle field conditions. That doesn't include doing things like using various means to deliberately fool or spoof systems. Battle fields also tend to be rather busy and noisy places signal and information wise. An advanced battlefield will likely have a lot of "noise" for sensors to sort through.

You combine all of that with data from Battle Damage Assessment and the system could say aim here or say here are the weak spots.

But as already pointed out a well trained individual could also just call up a quick reference instead of relying on automatic systems and sensors to do it.

edited 12th Jul '14 7:06:19 AM by TuefelHundenIV

Who watches the watchmen?

That silk article is pretty interesting - and it's rather timely too considering I brought up the possibility of synthetic spider silk for non-powered body armour.

I stumbled upon this website on Spacebattles.net, and I think it could be useful for budding designers. Of particular interest is everything from the Speed vs. Strength video onward, though I've only watched the reverse engineering vid.

On another note, I'm considering the idea of an adaptive computer system for vehicles/powered armour. It would recognise when something is out of action - due to failure, damage, or whatever - and attempt to work around it (or at least minimise its effects). How does that strike you guys?

Locking you up on radar since '09