Home > Asimov, Larry Niven, Valve > SciFi Round Nine: Portal

SciFi Round Nine: Portal

Portal and Portal 2 are a couple of the best computer games I’ve ever come across.  I heartily recommend them, especially if you like puzzles and snark.  The basic idea in both is that you’re playing as Chell, a woman trying to escape from a laboratory controlled by a homicidal AI.  To aid you in your escape… or scientific “testing”… you get a device that shoots portals onto walls in pairs.  If you go in one portal, you come out the other one.  Instant wormhole, just add portal gun.

As ever… here, there be spoilers.

This is a portal.  In the words of GLaDOS, speedy thing goes in, speedy thing comes out.

This is a portal. In the words of GLaDOS, speedy thing goes in, speedy thing comes out.  The setup above is nice for getting the person pictured to fall at terminal velocity and also break conservation of energy.

Portal Guns Break The Universe

The main physics-breaking part is the Portal Gun itself.  It turns out, Larry Niven wrote a very good essay a while ago, The Theory and Practice of Teleportation, which covers a lot of different ideas in great detail.  I’ll just hit a couple of the main points about Portal’s portals, and then move on.

The first issue is conversation of energy and momentum.  Put a portal on the ceiling, and one on the floor, so that when you go into the one of the floor, you fall out of the ceiling… into the portal on the floor… and this keeps going on indefinitely until something stops you, somehow.  The problem here is that you can now accumulate energy until you hit terminal velocity — and then you keep dumping more energy into all the noise and heat you’re making while going that fast.  You essentially get to move “up” in Earth’s potential for “free” when you go from floor to ceiling.  But all that energy has to go somewhere, and a quick estimate suggests that this would raise the temperature of a room by a few degrees Celsius per second… which would rapidly cook Chell.

The other problem is momentum.  Put both portals on the same wall.  Throw a ball into one portal, and it comes out of the other with momentum in the opposite direction, without transmitting that momentum into any other object.  Oops.

While portals you shoot are limited to light speed travel times, this does suggest a nice way of colonizing other planets.  Or sending stuff to other places in our own solar system.  I’ll leave the exact details to Niven’s essay, but it includes sending fuel through portals to break the rocket equation… but that’s if only if you have to have an artificial surface to shoot onto.  Otherwise… fire, carefully, and wait.

At least there’s an explanation for why all this high-speed portal-hopping doesn’t kill Chell.  In fact, this aspect bothered initial testers of the game so much that Valve added Long-Fall Boots.  These are specifically designed to perfectly kill the wearer’s inertia so that they don’t go splat on the landing, and also ensure that you land feet down.  Nifty, right?  Totally doesn’t violate physics…

All AI is a Crapshoot

This is GLaDOS.  She promises cake and grief counseling.  What you get is a portal gun and deadly neurotoxin...

This is GLaDOS. She promises cake and grief counseling. What you get is a portal gun and deadly neurotoxin…

Your main antagonist in the first game and the first segment of the second is GLaDOS (Genetic Lifeform and Disk Operating System).  As you play, it becomes increasingly clear that something is wrong with the AI.  This is even more clear after you get shunted from regular testing to the android live-fire test range.  And then it gets worse.

How hard is it to program a non-evil AI?  With a strong negative weight on harm to human lives?  And then, rather than simply cutting off its access to the deadly neurotoxin, they just stick a module on GLaDOS to keep it/her from releasing the deadly neurotoxin.  Oy.  Why can’t we downweight killing humans?  Or apply the Laws of Robotics?  That last link is interesting — turns out, there are some real-world guidelines for robot construction.  Obviously a robot?  Check.  Not designed to kill humans [by running them through deadly testing chambers for “science”]?  Fail.

The other antagonist you meet is Wheatley.  He’s friendly.  He’s helpful.  He’s also deliberately constructed to be a total moron (intended to keep GLaDOS in check) and undergoes an epic and hostile personality change when you replace GLaDOS with him.  Oops.  Nice job breaking it, hero.

Cave Johnson Failed Business Planning

As a matter of fast, OSHA came in in 1970, so this matches up with the timeline.  Otherwise, this poster is an epic lampshade.

As a matter of fast, OSHA came in in 1970, so this matches up with the timeline. Yes, this poster is an epic lampshade.

Once Wheatley turns on you in the second game, you get to explore the deeply buried history of Aperture Science.  Ignoring the problem of the absurdly deep mineshaft that’s storing everything, there’s a more serious problem:  Where is Cave Johnson, CEO of Aperture, getting all the money for this?

Millions for moon rocks to make surfaces for portals… which were originally intended to be better shower curtains.  Various other nasty things are tested by and on humans, with no regard for anybody’s safety.  Cave’s idea of science is building random crap and seeing what happens, which ranges from bouncy repulsion gel (intended as a diet product — the food bounces right out!  And does bad things to your stomach…) and things like turning people’s blood into gasoline.  Or making them into an army of mantis-men.  Among other problems too numerous to mention.  WHY HASN’T THE GOVERNMENT SHUT THESE PEOPLE DOWN ALREADY???  Then again, this is all delightfully lampshaded by various signs (such as the one above), and we never see what the exterior used to look like, back in the day.  It’s quite possible that there were protestors all the time outside the facility.  Maybe even a demonstration dedicated to the missing astronauts…

Meanwhile, the game does demonstrate the increasingly dire financial straights of the company.  Since they have trouble marketing their various deadly products.  For instance:  An ad for the long-fall boots.  If you’re bored, you can even try looking for the ad about using turrets for guarding babies.  It’s even worse.  This is not a good business plan… which explains the lack of funding, but I still wonder how they managed to get any funds in the first place.

Cave’s other major problem seems to be that he thinks of science as throwing together a bunch of random stuff, and then seeing what happens when some poor sucker tries to use it.  That’s not science.  That’s cruel and unusual.  And darkly hilarious for the player.

But anyway, Cave, you don’t know what science is.  And don’t get near my house with those combustible lemons.

  1. michaelbusch
    2013/01/06 at 7:28 pm

    Rachel is quite correct that Niven’s essay discusses possible uses of things like the portal device. So have many other writers – TVTropes indexes them under “Portal Door”.

    But just so that it is made clear here, if Aperture Science had developed the portal technology properly, instead of feeding Cave’s fixation with putting boxes onto buttons, things would have gone something like this:

    1. Portal the Moon, with the other end of that link in a nice big vacuum chamber. The link is either made independent and self-sustaining of whatever was used to set it up, or _that_ portal gun is locked and vaulted somewhere very well secured.

    2. Profit! Cave is vindicated as a genius, rather than as a man who should not be in charge of anything.

    3a. Attach the smallest possible portal and its surface to a rocket engine and a sufficiently solid shield. Run fuel and power lines through the little portal, and send it off. 3b. When the portal reaches its destination, slide through more surfaces and open portals on them. You can recycle the rocket by passing it back through the new portals to the ground. The situation is similar to the one I outlined here: https://clementsgame.wordpress.com/2012/09/22/fantasy-league-round-2-the-silicon-mage/#comments . Also, portals are _oval_. This means that if the size of the portals can be adjusted, the portal surface for a large portal can be slid through a smaller one, making it even easier to go anywhere in the universe.

    4. More profits.

    The cost estimates here are iffy. GLaDOS claims that the portal gun is more valuable than the internal organs of everyone in Chell’s hometown. But we don’t know how big that town was or if we’re going by illicit-organ-trade prices (~1 million USD or less per person) or insurance valuations (~10 million USD or more), and GLaDOS is an unreliable source anyway. The portal gun as designed by Aperture in the game is outrageously over-engineered, so say that one portal device costs 1 billion USD (2012 dollars) with minimal returns to scale. Then we have the following, given that the rockets accelerate at 1 g. Using the links once they are set up would cost more, of course:

    -Cost and mission time for initial Moon-shot demonstration: ~1.5 billion USD, ~1 day from device firing.

    -Mars link: ~2.5 billion USD, ~1 week from launch
    The cost includes a second device to open a larger portal after the small one has been landed; recall that the first device can be recycled for another link.

    -Europa link: ~3 billion USD, ~12 days from launch
    Here the cost includes a second device and shielding for the rocket to handle Jupiter’s magnetosphere.

    -Sedna link: ~3 billion USD, ~30 days from launch
    Here the shielding is necessary to protect the rocket from dust collisions at 10^4 km/s, the fuel and power bill is ~0.1 billion.

    -Alpha Centauri link: ~5 billion USD, ~5.5 years from launch
    The fuel and power bill is now ~2.5 billion; it can be cut by a billion or so if long-fall technology can kill relativistic velocities. Peak speed for the rocket is upwards of 0.85 c – although what happens to a portal when there are significantly different time rates on opposite sides? You probably don’t want to stand halfway through one. Peak speed could be raised to 0.97 c, but at the expense of raising the cost by a couple of billion. Additional acceleration gets harder going up the relativistic curve. Sending spacecraft to other stars takes longer than going to Alpha Cen, but costs about the same since they spend most of their time coasting.

    You can see the exploit here, especially when the small rockets can be recycled by pushing them through other portals. That saves all of the cost and time of flying them up to speed and out to that distance, cutting the cost of additional interstellar links by at least a factor of two. End result: Aperture Science becomes _the_ transport and communications company, and provides links to and between a thousand stars within 50 years of when they invented the portal device (around 1950). If normal patent rules were followed, they would have had twenty years of exclusive rights to the technology; which is enough to give them enough of a head-start over any competition that they would dominate the portal market for at least that much longer. More likely, Aperture would have been compelled to license the technology for appropriately astronomical compensation, with considerable government and independent oversight.

    Above I’ve just considered the possibilities for interstellar and interplanetary transport and telecom, but if one link cost 1 billion USD, it would be worthwhile to install portal links between hundreds of cities on Earth. Even if the portals were too small for high-speed heavy rail links, they would be plenty wide for low-latency fiberoptics and for fifty thousand passengers per hour on a narrow-gauge line. Replace a large portion of the current telecommunications backbone and every international airport – or rather, those would not have been built, given the c. 1950 invention of the portal device (this is most similar to Niven’s society #4 for anyone who looked at the teleportation essay). Diversifying from transport and telecomm, put a few portals in solar orbit much closer in than Earth and maybe power satellites become appealing.

    Lower the cost of a portal link and things get even more awesome. Consider using the infinite fall for electricity generation: 1 portal pair near Earth’s surface gives ~1 GW of clean power (~1 m radius portals, dropping a 100 ton magnet at 1 km/s through a linear alternator. Higher speed means more Coriolis forces and problems with the magnet going sideways). Very valuable at 1 billion/portal (1 dollar/watt).

    But such a story would not have made anywhere near such a compelling puzzle game, although there would have been interesting possibilities for time-travel paradoxes with the universe as broken as it would be. And the cake would probably not have been a lie.

  1. 2013/04/12 at 7:11 pm
  2. 2013/07/01 at 1:59 am
  3. 2013/11/07 at 7:57 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: