Wrongulator: a gag calculator that gives the wrong outcome

The Wrongulator is a gag calculator that gives incorrect answers to calculations. I'm fascinated by the idea -- I wonder if they've just got a standard calculator controller in there, and then a secondary system that scrambles the results, or whether it's a pseudorandom number generator, or what. You'd want it to produce plausible outcomes (5 x 5 = 30; not 5 x 5 = 324527) but who knows if the manufacturer paid attention to this.

The Wrongulator is no ordinary calculator, its actually the world’s worst calculator as it never gives the right answer, ever! If your calculator has been exchanged for this one then every single calculation you’ve entered in it has been wrong. It is perhaps the cruelest practical joke you could inflict on your office colleague and the chances are, without being told, they’ll probably never guess….well not before it’s too late anyway! Mwhahahaha!

Wrongulator (via Red Ferret)

• [Bucket]

  I’ve been doing this for years.

  It’s what happens every time anyone asks to borrow a calculator and I hand them my HP 48GX.

  • [Brad Gilbert]

    Same! But I programmed my TI83 to give the correct answer if the problem was something you could do in your head. As soon as you use logarithms, Pi, or a large enough exponent, though, it makes subtle changes to the answers.

• [Brian Cain]

  The mind boggles at the real world disasters waiting to be created with this product.  Badly designed bridges, leaning skyscrapers, wrongly calculated space telescopes…er, wait a minute.

• [relawson]

  well wait a minute… is it “consistently” wrong? 

  if this is the only calculator used for a job, they may never figure it out because its consistent! ;)

• [betatron]

  please keep this funny joke out of my control room!  At least it’s not RPN, oh the horror… 

• [Pasketti]

  If I were evil, I’d whip up a Windows Calculator lookalike that did this.

  • [MrEricSir]

    You wouldn’t have to if you were using Windows 3.1 and its notoriously inaccurate calculator:
    http://support.microsoft.com/kb/72540

• [Jason Sutor]

  It should work correctly most of the time and only give the wrong answer the rest of the time. That’s truly sinister.

• [nosehat]

  This could be used in a basic intelligence test:

  “How much money is two twenty dollar bills and a five dollar bill.  Here’s a calculator you can use if you want.”

• [fxq]

  It’s like “Stupid Mac Tricks” all over again.

• [solitarycow]

  This is how Enron went down. True story.

• [Patrick Delaney]

  the really fiendish thing to do would be to make the answers CORRECT for single operations, say 5+5=10. When the user started doing three or more operations, like 5+5 x 3 / 2 and so on, start ratcheting up the wrongness with each operation. That way, not only would easy-to-check problems be obviously correct, every time the user was suspicious and tried to test the calculator to see if it was a gag, it would give them the correct answer and drive them a little bit closer to the edge of insanity.

• [Takashi Omoto]

  My TI-86 had a minor problem where after a few months of reliable work it would get a corrupted floating point stack. So every now and then I’d get 12/3=4.00000008. My pious amount of tinkering with it obviously didn’t help.
  Why yes, exams were pretty thrilling.

  • [semiotix]

    My TI-86 had a minor problem where after a few months of reliable work it would get a corrupted floating point stack. So every now and then I’d get 12/3=4.00000008. 

     

    Actually, 12/3 does equal 4.000000079771263 (±.000000000000005) at sea level. Most calculators ignore relativistic effects, but you got a real nice precision calculator.

• [jimkirk]

  When I was in college a friend had a new HP45 calculator that everyone kept borrowing.  He finally took the cover off, swapped the SIN and COS chicklet keys and reassembled, and only told the few people he didn’t mind using it.  Over the next week the borrowing dwindled.

• [wrybread]

  I’m thinking you could generate believable wrongness by adding or subtracting some random
  number that correlates to the size of the result. So a sum of 2 would get an adjustment of -1 or 1, while a sum of 2000 would get an adjustment between -100 and 100. Or you could go the coefficient route and multiply everything by 1.005, adjusting for the size of the number, and make sure you get some minimum amount of fucked.Hats off if this calculator generates believable results. But in my experience with these sorts of things I’d bet its of the “5 x 5 = 324527″ variety.And who makes this stuff? Amazing that someone not only had this idea, but then prototyped it, developed it, and, given that it costs under $10, mass produced it.

  • [Mark Lamb]

    Perhaps someone was faced with large pile of failures from a run of cheap calculators and cleverly avoided losing money on them.

• [Chevan]

  I can just see someone leaving a couple of these scattered around a classroom the day of a final exam in something like chemistry (where you’re often allowed to use calculators, but not advanced ones) for all the people who forgot theirs to find.

• [GIFtheory]

  I could see accounting firms buying these up by the bushel. Can you say, “plausible deniability?”

• [Gilbert Wham]

  Shit, I can do this with a normal calculator.

• [mypalmike]

  It’s the Calcucorn!

  http://video.google.com/videoplay?docid=-7064178994016272127
  (Fast forward to about 4:05)

  • [nuschu]

    That is the first thing that popped into my head :)

• [TooGoodToCheck]

  If I was going to have a useless calculator, I’d rather have one that represent results greater than four as “a suffusion of yellow”

  • [fxq]

    That’s a Dadadder. Brion Gysin used to sell ‘em.

• [awjt]

  A lotttttttta people are already using these.  Most of them work for the government.

• [Zix]

  I’m guessing that this thing *does* generate obviously wrong results (like 5+2 = -973.284), rather than subtly wrong ones. The former makes it obviously a gag/annoyance, whereas the latter could actually endanger lives, financial wellbeing, and property, and expose the users, sellers, and manufacturer of this product to liability.

• [lightning]

  Considering that most people can’t do simple arithmetic without a calculator, these could be really funny nasty.

• [BarBarSeven]

  John Boehner called.

• [Terry Mcintyre]

  I have been told that HP did a test when electronic calculators were new: they gave a large group of people brand-new calculators, and told them to “test the calculators” by solving a series of problems. This was back in the days when people actually were taught to do arithmetic without calculators.

  Nevertheless, only one of the sample, a 12 year old boy, noticed that the calculators were deliberately programmed to give wrong answers.

• [Lane Yarbrough]

  I bought a level that was not manufactured properly, I wasted so much wood. It’s strange how long I chose to believe the fault was mine, I never questioned the accuracy of the level itself. 

  Don’t people do the same with “spell check”?

  • [jackie31337]

    Don’t people do the same with “spell check”?

    Yep, that’s how I once ended up with “opera rational objectives”. ;)

    • [Teresa Nielsen Hayden]

      I have yet to see an error-free spellchecker. They also tell you that long-established words and equally well-established alternate spellings don’t exist.

• [stuck411]

  @donttrythisathome:disqus  tweeted about his calculator giving wrong answers. Was this what Adam was talking about?

• [Justin Cram]

  That explains why the debt ceiling hasn’t been resolved!

• [Antoine Tallon]

  French Police has been using this device to figure out the number of strikers in the streets for a long time now, they never coplained about it !

• [knoxblox]

  Having never been an engineer or mathematician, I never really needed a calculator, so I still have my trusty TI-36X SOLAR. It never died, so I still use it. The stamp on the back says it was manufactured on June 6, 1994.

  So…what is the equivalent of the technology I’m using? Tin-can telephone?

• [Steve Bosman]

  If I was devising a calculator like this and wanted the effect to be really subtle I would always keep the first and last digits of integer results correct that way back checking wouldn’t be as easy.

  However, I agree with everyone else who thinks there are some bad consequences lying in wait for the users of such a product and I hope one that subtle is not made.

• [frogfactory]

  Does it return “A suffusion of yellow”?

• [Mike Perkowitz]

  Regarding Cory’s question, if they actually guarantee that it’s *never* right then it can’t just be generating random answers.

• [Thad_E_Ginathom]

  I had a piece of software that did this. It was a piece of middleware which allowed staff to extract and use data from the company’s accounting system directly into spreadsheets on their PC. It was, in business terms, really wonderful, and revolutionised a lot of routine stuff. It also saved me the chore of running dozens of batch files with different parameters to extract that data. It was a hot all around.

  However, there was one very un-obvious parameter in the set up file, which, if you got wrong, would pass completely unnoticed until some bright spark in accounts noticed that it was churning out numbers, as it should — but they were the wrong numbers! 

  Ahhh… red-faced days in my career. Happy memories! (sort of!) 

• [Guest]

  Hee! XD

• [Brandon Sachs]

  A program I wrote a number of years ago:/*This is the MATHULATOR 9000, it tries to do math*/

  #include
  #include /*include time travel capability*/

  /*declare stuff*/
  int sel;
  int loop;
  float num1;
  float num2;
  float answer;

  /*here is the program*/
  int main()
  {
  loop = 0;

  srand(time(NULL)); /*Putting clock into the random!*/

  /*main menu*/
  while (loop == 0){
  printf(“Welcome to the MATHULATOR 9000n”);
  printf(“You are using the worlds most advanced number calculating machinen”);
  printf(“Please enter the type of calculation you would like to perform:nn”);
  printf(“1. Combine two numbers… to make one bigger numbern”);
  printf(“2. Remove one number from another numbern”);
  printf(“3. Multiplixicaton!n”);
  printf(“4. Make a number into a certain number of pilesn”);
  printf(“5. Quantum calculusn”);
  printf(“6. Exitnn”);
  printf(“Make your selection now: “);

  scanf(“%d”, &sel);

  /*addition portion*/
  if (sel == 1){
  printf(“Enter the first number you wish to use: “);
  scanf(“%f”, &num1);
  printf(“Enter the second number to combine with the first: “);
  scanf(“%f”, &num2);
  printf(“Now Addifying!n”);
  answer = num1 + num2 + (rand()%10) – (rand()%10);
  if (answer > 1000) answer = answer + (rand()%100) – (rand()%100);
  /*answer output*/
  if (answer > 0 && answer < 10000) printf("Those two numbers make something near %.0f… or something like thatn", answer);
  if (answer 9999) printf(“The answer is: a lot, like more than four digits or somethingn”);
  if (answer > 1000000) printf(“what the hell is wrong with you? trying to make me add numbers like that!n”);
  if (answer == 1337) printf(“I hate that number!n”);
  printf(“nnn”);
  }

  /*subtraction portion*/
  if (sel == 2){
  printf(“Enter the first number you wish to use: “);
  scanf(“%f”, &num1);
  printf(“Enter the second number to take away from the first: “);
  scanf(“%f”, &num2);
  printf(“Now booleaning!n”);
  answer = num1 – num2 + (rand()%10) – (rand()%10);
  if (answer > 1000) answer = answer + (rand()%100) – (rand()%100);
  /*answer output*/
  if (answer > 0 && answer < 10000) printf("That means you have %.0f apples, i mean, uh. numbers n", answer);
  if (answer 9999) printf(“The answer is: a lot, like more than four digits or somethingn”);
  printf(“nnn”);
  }

  /*multiplication portion*/
  if (sel == 3){
  printf(“Enter the first number you wish to use: “);
  scanf(“%f”, &num1);
  printf(“that number how many times?: “);
  scanf(“%f”, &num2);
  printf(“Now multiplexoring!n”);
  answer = (num1 + (rand()%3) – (rand()%3)) * (num2 + (rand()%3) – (rand()%3));
  /*answer output*/
  if (answer > 0 && answer < 10000) printf("That's like %.0f, I thinkn", answer);
  if (answer 9999) printf(“The answer is: a lot, like more than four digits or somethingn”);
  printf(“nnn”);
  }

  /*division portion*/
  if (sel == 4){
  printf(“Enter the first number you wish to use: “);
  scanf(“%f”, &num1);
  printf(“How many piles do you want to make?: “);
  scanf(“%f”, &num2);
  printf(“Now multiplexoring!n”);
  /*make sure there is no divide by zero. because that sucks*/
  if (num2 != 0){
  answer = num1 / num2 + (rand()%10) – (rand()%10);
  if (answer > 1000) answer = answer + (rand()%100) – (rand()%100);
  /*answer output*/
  if (answer > 0 && answer < 10000) printf("That's like %.0f, I thinkn", answer);
  if (answer 9999) printf(“The answer is: a lot, like more than four digits or somethingn”);
  }
  if (num2 == 0){
  printf(“ARE YOU TRYING TO KILL ME OR SOMETHING? BYE!nn”);
  break; /*the program is really pissed off*/
  }
  printf(“nnn”);
  }

  /*quantum calculus*/
  if (sel == 5){
  printf(“nnNow cal… Uh, you know, It’s been a long time since school and all.n”);
  printf(“I’m like sort of tired and, well, why don’t you just pick a simpler one.nnn”);
  }

  /*exit*/
  if (sel == 6) loop = 1;
  }
  return (0);
  }

• [J-Zen Wenzoski]

  Slap a different label on it and voila! A codebreaking game!