# I'm a superhero

• ...and I'm going to prove it. I have inhuman speed, strength and resilience. To prove this, I present this screen shot of a java applet that tests typing speed and accuracy. Sadly, the applet was in finish, but here is a translation of the relevant parts:

Overview of the test Typing speed 2147483647 characters per minute Superb! Accuracy 98% Excellent Number of mistakes 6 characters

No, i didn't cheat. Unless there's a bug in the applet, we can trust these numbers. 2147483647 cpm is a huge number, and I'm going to try make some sense into it. Let's assume I had zero reaction time and that my finger tips move 1.5 cm on average to press a key. We need to double that because the finger has to also depress the key and return to its home row. 21474836473cm in one minute means that the tips of my fingers move at least 1073.74182 km/s on average. How much is that? Supersonic speeds are more than about 0.34 km/s. Apparently the sound barrier is broken about 35 000 000 times per second by my rapidly accelerating and decelerating finger tips. However, the intense turbulence probably causes some complications in the theory. Clearly, we need bigger speeds to have a meaning full comparison. The escape velocity on Earth, with respect to Earth's gravity is 11.2 km/s and from the earth/moon system, with respect to the sun's gravity, it is 42.1km/s. It seems that if the tip of my finger broke of with the right timing, it would clear both of those, easily, and launch in to the interstellar space. Actually, the escape velocity from the Milky Way, with respect to the solar system, is roughly 1000km/s so it might eventually even leave the galaxy. Traveling such a distance would still take pretty much an eternity.

We can calculate much more from these numbers. For example, because of the roughly 35791394 characters per second, we can estimate that my reaction time has to be less than about 28ns. That is about 1/800th of the sampling interval in cd audio. Actually, the light emanating form my display takes about 1.6ns to travel the 50cm distance to my eyes. Taking that time into account, we now estimate that my reaction time is less than about 26ns. That is including the time it takes for me to actually move my fingers.

I think we've established that I'm quite fast, but there's more. To produce these speeds, I must be quite strong. For simplicity, I'm going to assume constant acceleration, the actual peak acceleration is probably much bigger. To accelerate from rest to top speed, then decelerate equal time to press the key down, I have half of the time it takes me to press one key and return to home row. During that time, the finger tip will travel 1.5cm first accelerating and then decelerating. So, the first 0.75cm of the movement is completed at about 7ns. This means that I can produce accelerations of at least 3.0710^14 m/s2. Estimating that the mass that is accelerated is at least 2g, the force needed is at least 614 GN. That's about 340 000 times more than the space shuttle's main engine and 18 000 times more than the Saturn V at lift off. Keep in mind that this is but one finger we are talking about. My body is apparently capable of withstanding these immense forces, since my fingers feel just fine. One can say that I'm pretty much invulnerable. For example, a high end wrist watch can survive shocks of up to 5000 G, but that's nothing compared to the at least 31 000 000 000 000 Gs my fingers survive while typing. The tensile strength of the tendons and muscles in my hand is also amazing. Experiments so far give us around 60 GPa for certain carbon nanotubes. For about 0.2cm2 of cross sectional area, one tendon in my hand survives more than 30 000 000 GPa with no apparent damage.

15 points for anyone who thinks 2147483647 is a strangely familiar number.
10 points if you noticed that I ignored the effects of drag in the escape velocity scenario.
20 points more, if you noticed that I did it again when calculating the force.
100 points more if you did some calculations about how much of an effect it has.
100 points if you noticed that neither ps/2 nor usb has enough data rate for my fingers.

Yes, those are SI units, it's physics, after all.

• @AnonymousCoward said:

100 points if you noticed that neither ps/2 nor usb has enough data rate for my fingers.

USB3 does (though not in HID mode)

• @AnonymousCoward said:

15 points for anyone who thinks 2147483647 is a strangely familiar number.

Looks like int.MaxValue to me.

• @mott555 said:

@AnonymousCoward said:

15 points for anyone who thinks 2147483647 is a strangely familiar number.

Looks like int.MaxValue to me.

-100 points for explaining something that most people got immediately after that hint and the rest after a bit of thinking.

• @AnonymousCoward said:

java applet

Java? Buggy? Unpossible!

@AnonymousCoward said:

2147483647 cpm is a huge number, and I'm going to try make some sense into it. Let's assume I had zero reaction time and that my finger tips move 1.5 cm on average to press a key. We need to double that because the finger has to also depress the key and return to its home row. 2147483647*3cm in one minute means that the tips of my fingers move at least 1073.74182 km/s on average.

Wait a min...

@AnonymousCoward said:

because of the roughly 35791394 characters per second, we can estimate that my reaction time has to be less than about 28ns. That is about 1/800th of the sampling interval in cd audio.

What the...

@AnonymousCoward said:

So, the first 0.75cm of the movement is completed at about 7ns. This means that I can produce accelerations of at least 3.07*10^14 m/s2. Estimating that the mass that is accelerated is at least 2g, the force needed is at least 614 GN.

Am I in math class? WHAT IS THIS!?

@AnonymousCoward said:

For example, a high end wrist watch can survive shocks of up to 5000 G, but that's nothing compared to the at least 31 000 000 000 000 Gs my fingers survive while typing. The tensile strength of the tendons and muscles in my hand is also amazing. Experiments so far give us around 60 GPa for certain carbon nanotubes. For about 0.2cm2 of cross sectional area, one tendon in my hand survives more than 30 000 000 GPa with no apparent damage.

It's the dream! The dream where I'm naked in physics! aaaaaaAAAAAA!

(Seriously, though, you've spent about 3 times more time describing the bug than the guy who wrote the applet spent QAing the damned thing.)

• @b-redeker said:

@mott555 said:

Looks like int.MaxValue to me.

-100 points for explaining something that most people got immediately after that hint and the rest after a bit of thinking.

Relax, he's just trying to claim his Internet Points.

• @AnonymousCoward said:

Overview of the test Typing speed 2147483647 characters per minute Superb!

Amazing! That's about the same amount of characters per minute as minutes this train was late:

• @b-redeker said:

-100 points for explaining something that most people got immediately after that hint and the rest after a bit of thinking.

Seriously, people need a hint in order to recognise that? But then again, given how many people keep posting it as if it's something fabulous and exciting, I do have to wonder.

• @AnonymousCoward said:

For simplicity, I'm going to assume constant acceleration, the actual peak acceleration is probably much bigger. To accelerate from rest to top speed, then decelerate equal time to press the key down, I have half of the time it takes me to press one key and return to home row. During that time, the finger tip will travel 1.5cm first accelerating and then decelerating. So, the first 0.75cm of the movement is completed at about 7ns. This means that I can produce accelerations of at least 3.07*10^14 m/s2. Estimating that the mass that is accelerated is at least 2g, the force needed is at least 614 GN. That's about 340 000 times more than the space shuttle's main engine and 18 000 times more than the Saturn V at lift off. Keep in mind that this is but one finger we are talking about. My body is apparently capable of withstanding these immense forces, since my fingers feel just fine. One can say that I'm pretty much invulnerable.

You're too modest; you didn't even mention the immense heat dissipation.  Using your model, the work done in applying that force over the 3cm of each keypress is 18.42GJ, which considering the duration of each keypress is no more than 28ns means that you are going to be emitting something like 659.277479629PW.  Now, that's about 3.8 times the entire amount of energy reaching the entire surface of the planet from the sun in a given period, or to put it another way, it's about 10,464 Hiroshima bombs per second.

So although you might be ok, I'm a little concerned for your neighbours.  And anyone else within a thousand kilometers.

Or indeed on the same planet as you.  Would you mind being a little less awesome before you blow it up?

(Edit: That's ten thousand Hiroshimas per second per finger, of course.  I forgot to mention that.)

• Also, let's not forget that your computer is pretty cool too.  Let's assume an ~2GHz CPU, and you're doing 2gig keys per minute; that basically means that your processor has about 60 machine instructions in which to acknowledge the interrupt, decode the key, pass it to the JVM, run that through all the java bytecode and update the screen.  It must be a fairly specialised CISC machine, I'll bet ....

•  You're probably still o.k. leaving out relativistic effects, as you're still well below the speed of light.

• @DaveK said:

You're too modest; you
didn't even mention the immense heat dissipation.  Using your model, the
work done in applying that force over the 3cm of each keypress is
18.42GJ, which considering the duration of each keypress is no more than
28ns means that you are going to be emitting something like
659.277479629PW.  Now, that's about 3.8 times the entire amount of
energy reaching the entire surface of the planet from the sun in a given
period, or to put it another way, it's about 10,464 Hiroshima bombs per
second.

So although you might be ok, I'm a little concerned for your neighbours.  And anyone else within a thousand kilometers.

Or indeed on the same planet as you.  Would you mind being a little less awesome before you blow it up?

(Edit: That's ten thousand Hiroshimas per second per finger, of course.  I forgot to mention that.)

Fear not, people of Earth, I can assure you that I will use my powers to fight for cosmic justice and only occasionally show off.

@DaveK said:

Also, let's not forget that your computer is pretty cool too.  Let's assume an ~2GHz CPU, and you're doing 2gig keys per minute; that basically means that your processor has about 60 machine instructions in which to acknowledge the interrupt, decode the key, pass it to the JVM, run that through all the java bytecode and update the screen.  It must be a fairly specialised CISC machine, I'll bet ....

There's a really nice place in Titan, they sell the best hardware of this system.

@RogerWilco said:

You're probably still o.k. leaving out relativistic effects, as you're still well below the speed of light.

I was refering to drag in fluid dynamics, not frame dragging.

• @AnonymousCoward said:

@RogerWilco said:

You're probably still o.k. leaving out relativistic effects, as you're still well below the speed of light.

I was refering to drag in fluid dynamics, not frame dragging.

Y'know, that's probably going to be pretty hard to quantify.  Those 18.42GJ per keystroke are equivalent to 1.14968609x10^29 electron volts.  Now, assuming standard temperature and pressure (STP), 1 mole (Avogadro constant # of molecules, 6.02x10^23 mol^-1) occupies 22.4 liters of volume (.0224 m^3), so the mean separation between molecules is (.0224/6.02x10^23)^1/3, or ~3.34nm.  That means that your fingers are going to displace 3x10^-2m/3.34x10^-9m = 8982035.9 (call it 8982036) air molecules during the average stroke, and assuming that that's how you dump the heat you're generating (yes, I'm neglecting what will probably be quite significant black-body radiation in this; we should assume that you're bathing the surrounding countryside in hard gamma rays) then that means you're dumping approximately 12.8ZeV (zetta electron-volts) into each molecular when you collide with it.  That's as much 914285714 LHCs, which is far more than entirely sufficient to fission it, and indeed you should already have discovered the Higgs boson, and have probably created unbound quarks for the first time since the earliest stages of the big bang, but how we're going to calculate the wind resistance I have no idea; bet it's not a simple term in the square of velocity any more.  After all, each stroke of your finger summons into existence a shower of particles with mass-energy equivalent to 18.42GJ divided by the speed of light squared, approximately 205 micrograms, and doing that every 26ns means that 7.8 tonnes of energy and matter is showering from your fingers per second as you type.  But that's not all!  Grand Unification is thought to arise at 10^15 GeV, and the temperature of the universe during the Plank epoch is thought to have been maybe 10^19GeV, i.e. 10^24 and 10^28 eV respectively, so your 12.8ZeV means that somewhere between every one hundred to one million keystrokes, you should create a new universe.

I hope you have a lead-lined hoover on standby.

• He is a cosmos unto himself.

• Since this number is Int.MaxVal, you probably typed more but it got trunc'd. Try finding an app that stores the keystrokes as Long.

• 6 characters in error make up 2% of total characters typed. You only typed 300 characters in about 10.737418233 nanoseconds.

My math might be wrong but I calculated:

```2,147,483,647                                 c/minute      divided by 60 to give
35,791,394.11          + 1/150             c/second      divided by 1,000 to give
35,791.39411       + 1/150,000         c/millisecond divided by 1,000 to give
35.79139411    + 1/150,000,000     c/microsecond divided by 1,000 to give
0.03579139411 + 1/150,000,000,000 c/nanosecond  multiplied by 300 to give
10.737418233   + 3/1,500,000,000   nanoseconds
```

I wouldn't assume that you could maintain this rate for long enough to exhibit most of the phenomena described.

•   This is inefficient. After you type one letter, either the next letter is typed with the same finger or it isn’t.

If it is, your finger should immediately proceed to the next letter, without touching the home row. This would be the case for "TR", for example.

If the next letter is typed with a different finger, and especially if the third letter is not typed with the same finger, you could quite easily slow down the return trip.

Optimizing further, there is nothing saying you can't look ahead and position multiple fingers at the same time.

For example, typing turtle with this look ahead would have you move your fingers to "t" and "u" at the same time (1.5 CM), depress "t" (.25CM), depress u (.25), move to "r" (1.5 CM), depress "r", move to "t" while prepositioning "l" and "e", (1.5), press"t" (.25),"l" (.25), and "e" (.25)

That totals in 5.75 CM, instead of the previous estimate of 12 CM.

• @cdosrun said:

This is inefficient. After you type one letter, either the next letter is typed with the same finger or it isn’t.

If it is, your finger should immediately proceed to the next letter, without touching the home row. This would be the case for "TR", for example.

If the next letter is typed with a different finger, and especially if the third letter is not typed with the same finger, you could quite easily slow down the return trip.

Optimizing further, there is nothing saying you can't look ahead and position multiple fingers at the same time.

For example, typing turtle with this look ahead would have you move your fingers to "t" and "u" at the same time (1.5 CM), depress "t" (.25CM), depress u (.25), move to "r" (1.5 CM), depress "r", move to "t" while prepositioning "l" and "e", (1.5), press"t" (.25),"l" (.25), and "e" (.25)

That totals in 5.75 CM, instead of the previous estimate of 12 CM.

While you make a sound argument about how fingers really move when typing, it is difficult to quantify exactly, sometimes fingers need to take (relatively) long leaps and sometimes they just stay at the same space hammering the same key. For this reason I chose 1.5cm, because is sounded plausible that it is shorter than the shortest distance the finger is going to move. That is, simply pressing the key down and then back up. The probability of that the same finger needs to write two letters one after another is even more complex, it depends at least on the keyboard layout and the language used. So I just ignored it completely, which should be ok, since we are just approximating here and not trying to get good quality numbers.

• ... bonus points for why this number is sometimes known as "Barlow's Number".

• @joe.edwards said:

6 characters in error make up 2% of total characters typed. You only typed 300 characters in about 10.737418233 nanoseconds.

I wouldn't assume that you could maintain this rate for long enough to exhibit most of the phenomena described.

I'm not sure to which phenomena you are referring to, but even assuming I only typed 300 characters(didn't check your math), approximations on speed, reaction time, acceleration, force, tensile strength and power (yes, it's over 9000) are all still valid. Obviously waste heat depends on how long I type, but even a short burst would produce a scintillating flash of em radiation.

• @DaveK said:

Also, let's not forget that your computer is pretty cool too.  Let's assume an ~2GHz CPU, and you're doing 2gig keys per minute; that basically means that your processor has about 60 machine instructions in which to acknowledge the interrupt, decode the key, pass it to the JVM, run that through all the java bytecode and update the screen.  It must be a fairly specialised CISC machine, I'll bet ....

All the comparisons in this thread were very amusing, but this really put things into perspective. Even if someone was able to type fast enough to release the equivalent of ten thousand Hiroshima bombs per finger per second, they would still be sixty times slower than a single core on a four year old laptop. That is, if the CPU had a TYPE_RANDOM_LETTER instruction. But I'm sure you can write a decent PRNG that uses less than sixty cycles.

I used to think of it as merely gigahertz. But from now on I'm going to think of it as thirty Supermans.

•  OMG the funniest WTF posted in a very long time.  And thanks everyone else for the comments; this is just so precious.

• @Faxmachinen said:

I used to think of it as merely gigahertz. But from now on I'm going to think of it as thirty Supermans.

•  @mott555 said:

@AnonymousCoward said:

15 points for anyone who thinks 2147483647 is a strangely familiar number.

Looks like int.MaxValue to me.

This is one of the reasons why I hate "high level" languages that are at too high a level. I think they must be suffering from hypoxia or something.

(By the way, I'm talking both about the bug, because "modern languages are supposed to prevent silly mistakes" and the fact that they often have such a cumbersome syntax for labeling a particular value.)

This is one of the reasons why I hate "high level" languages that are at too high a level. I think they must be suffering from hypoxia or something.

(By the way, I'm talking both about the bug, because "modern languages are supposed to prevent silly mistakes" and the fact that they often have such a cumbersome syntax for labeling a particular value.)

So you're saying the bug was somehow caused by the language? And which bit exactly is cumbersome? What is your proposal on both counts?

(By the way, I'm talking both about the bug, because "modern languages are supposed to prevent silly mistakes" and the fact that they often have such a cumbersome syntax for labeling a particular value.)

Having a "cumbersome" syntax for labeling a particular value, I suppose, is better than having no syntax at all for labeling that particular value? Like C?

Lemme tell you: back in the day it was *really* fun writing C code for a system with 16-bit ints. Will this sample code I found work? WHO KNOWS! They wrote it assuming 32-bit ints, and guess what, that was a fucking bad assumption! And now I'm in a world of hurt fixing the damned thing. (Which isn't to say int.MaxValue would have fixed the problem, but it would have been nice to have.)

• Nevermind all that --

I have some new and fairly sophisticated equipment to turn all of that into useable, marketable energy.  We'll have to keep you unconsious and feed you intravenously, but such is the price to save the world from global warming  the next ice age the end of peak oil   the energy crisis  Sarah Palin John Stamos Khaaaaan!  Climate Change.

•  Woah, I didn't realize blakey was quite so high-strung today... perhaps I should have been a bit more obvious with my mostly sarcastic remark.

In fact, C has a much more cumbersome syntax for platform-independent 2^(max)-1, which is

`((unsigned)(~0))>>1`

(Mostly my comment was based on the observation that, independent of features like an "easy to remember name" for some system value like maximum positive integer, programmers still find ways to break their programs in severe and amusing ways.)

• @DaveK said:

Y'know, that's probably going to be pretty hard to quantify.  Those 18.42GJ per keystroke are equivalent to 1.14968609x10^29 electron volts.  Now, assuming standard temperature and pressure (STP), 1 mole (Avogadro constant # of molecules, 6.02x10^23 mol^-1) occupies 22.4 liters of volume (.0224 m^3), so the mean separation between molecules is (.0224/6.02x10^23)^1/3, or ~3.34nm.  That means that your fingers are going to displace 3x10^-2m/3.34x10^-9m = 8982035.9 (call it 8982036) air molecules during the average stroke, and assuming that that's how you dump the heat you're generating (yes, I'm neglecting what will probably be quite significant black-body radiation in this; we should assume that you're bathing the surrounding countryside in hard gamma rays) then that means you're dumping approximately 12.8ZeV (zetta electron-volts) into each molecular when you collide with it.  That's as much 914285714 LHCs, which is far more than entirely sufficient to fission it, and indeed you should already have discovered the Higgs boson, and have probably created unbound quarks for the first time since the earliest stages of the big bang, but how we're going to calculate the wind resistance I have no idea; bet it's not a simple term in the square of velocity any more.  After all, each stroke of your finger summons into existence a shower of particles with mass-energy equivalent to 18.42GJ divided by the speed of light squared, approximately 205 micrograms, and doing that every 26ns means that 7.8 tonnes of energy and matter is showering from your fingers per second as you type.  But that's not all!  Grand Unification is thought to arise at 10^15 GeV, and the temperature of the universe during the Plank epoch is thought to have been maybe 10^19GeV, i.e. 10^24 and 10^28 eV respectively, so your 12.8ZeV means that somewhere between every one hundred to one million keystrokes, you should create a new universe.

I hope you have a lead-lined hoover on standby.

You are way overestimating the energy that is present in the collisions between my finger and the surrounding air. The 18.42GJ of energy that is used in one key press is at first translated into kinetic energy of the finger and then, when the finger is decelerating it should get absorbed by my body (it's not like the air can stop the finger on it's own, especially in such a short time period) which would(most likely) transformed into heat in the body. This would mean quite a bit of black body radiation being produced when I'm typing (and afterwards, just to cool down). For a better estimate of the energies involved in collisions of my fingers and the air I'm going to use the speed of the finger and the speed of a typical nitrogen gas molecule in stp conditions. Using the ideal gas law we get 0.34872 km/s for the rms speed of the nitrogen molecule. In a head on collision with a carbon atom in the finger and a nitrogen atom in the air, there should be about 71.759 keV present in the collision, which is about 0.001% of the 7 TeV in LHC.

• @Medezark said:

We'll have to keep you unconsious and feed you intravenously,

Not possible.

The required density of his frame would preclude anything penetrating his skin.

•  That is awesome and absolutely you are superhero

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.