What is Digg?Sponsored by newegg
Missed out on the best electronic deals last Black Friday? view!
newegg.com - Newegg.com's Cyber Monday Promotion has you covered. No Lines, No Crowds; Just Click and Save.
90 Comments
- sbcea, on 10/29/2009, -10/+70Maths?
- InTheUnion, on 10/30/2009, -3/+37YEA, YOU HEARD IT RIGHT!
hmm so I'm guessing there are more American Diggers... - thomas, on 10/30/2009, -0/+29It is just how they say/spell it in the "Queen's English". It is just like colour vs color and centre vs center. We say math and they say maths.
- Lith25, on 10/30/2009, -1/+29 The processor does exactly what you tell it to do every single time. These are all the fault of the programmer, not the processor. The programmer needs to understand the limitations of the machine they are working on and factor that into the code.
- InTheUnion, on 10/30/2009, -2/+27I always thought maths was short for mathematics
- R0B0Ninja, on 10/30/2009, -0/+25But can it run Cry...
But can it do arithmetics? - SammyboyKIDDAH, on 10/30/2009, -3/+28centre
- dfross, on 10/30/2009, -0/+22It stands for
Mathematical Anti-Telharsic Harfatum Septomin. - snipe123, on 10/29/2009, -3/+25I've never seen a processor doing a mistake but as a programmer god i did tons of mistakes.
Everytime I did one the computer was happy to tell me that I did not do my FOR correctly causing a infinite loop :(. - cx0der, on 10/30/2009, -1/+18Colour
- executex, on 10/29/2009, -3/+18It's not the processor, blame the inventor of floating point. They came up with this idea of making floating point variables, in the form of bits that represent certain numbers that are plugged into a formula to get the correct answer (makes it easy to calculate). All due to the limitation of having 32 bits. Unfortunately there really is no way around it.
Instead programmers should think about these possibilities of floating point errors and reprogram their calculator to avoid the problem, like splitting the numbers into multiple operations instead of trying to add them all at once. - bongone, on 10/30/2009, -4/+19Mathematics / Maths
We don't use the "s", people taught proper English do. It is their language. - Wolfsoap, on 10/30/2009, -0/+14God that's an obscure reference. But thanks.
Thanks ants. Thants. - colinmhayes, on 10/30/2009, -0/+13if you can find a limitation of mathematica, stephen wolfram will come and beat your ass personally.
- jeffwmartin, on 10/30/2009, -0/+11*Sea kittens
- jodokast, on 10/30/2009, -0/+1042
- rmxz, on 10/30/2009, -0/+10Seems it keeps things in symbolic form as long as it can before converting to decimal numbers.
"
Input: (sqrt(2) + pi) * (sqrt(2) + pi) - 2 - pi*pi - 2 * pi * sqrt(2)
Exact result: -2-2 sqrt(2) pi-pi^2+(sqrt(2)+pi)^2
Alternate form: 0
"
nice. - belumaves, on 10/29/2009, -3/+12so as a user of the americanized version of english I am perplexed every time I see someone call it maths. is there a reason that it's pluralized?
- simplyintricate, on 10/30/2009, -0/+8Wolfram Alpha uses Mathematica (by the same company) which uses the GNU Multi-Precision library:
http://en.wikipedia.org/wiki/GNU_Multi-Precision_L ...
Apparently, it has "no practical limit to the precision". - plaguepony, on 10/30/2009, -0/+8Floating point numbers are wonderful. Without them there would be many problems that cannot be solved with a decent amount of time. The problem is their misapplication.
- Jektal, on 10/30/2009, -2/+10They had it first, theirs can be the proper version.
Plus, we're the ones who have created Cheez Whiz, Toys R Us, Ez Pass, and thousands of other perversions of the language in the name of capitalism & laziness. - bjs3171, on 10/30/2009, -0/+7it would have been pretty funny if Data, from TNG constantly got simple arithmetic wrong.
- kitsua, on 10/30/2009, -3/+10Aluminium
Hypothesise
And Herb has a ***** H in it.
- inactive, on 10/30/2009, -0/+7No wonder I suck at maths... Math Blaster was wrong!!!!
- FlyingSquidwolf, on 10/30/2009, -0/+7u mean just like Mathematics .....is also pluralised.
- Fozefy, on 10/30/2009, -1/+7You can say this, but in general the fact is that most "programmers" don't understand what is going on here. If you try to do this calculation in assembly code such as MIPS or even in a "mid-level" language like C you're going to get these errors. The math behind how to get around problems like this can get very advanced and there will always be holes in it, that's just the nature of it. We just have to make the holes as small and as infrequent as possible.
/3rd year Software Eng student - plaguepony, on 10/30/2009, -0/+6LOOK AROUND YOU!
- elnerdo, on 10/30/2009, -0/+5Why authors are British.
- Ascus, on 10/30/2009, -1/+6Very old news, I remember this problem from "Intro to Computer Programming" 101, in the 70's.
- Dustmuffins, on 10/30/2009, -1/+6Ye olde shoppe.
- thomas, on 10/30/2009, -0/+5It is just how they say/spell it in the "Queen's English". It is just like colour vs color and centre vs center. We say math and they say maths.
- Atario, on 10/31/2009, -0/+5Don't you mean "ariths"?
- sinembarg0, on 10/30/2009, -0/+5overclock and you'll see the processor make some mistakes, or disable all error correction. Processors aren't perfect, but they deal with their imperfections quite well.
- echris, on 10/30/2009, -0/+5http://www.youtube.com/watch?v=EN5JYg1vDAY
- baranovich, on 10/30/2009, -0/+4Imhotep is invisible.
- MovieBlog, on 10/30/2009, -0/+4Just for fun I tried the following equation on WolframAlpha.com:
599,999,999,999,999,999,999,999,999,999,999,999,999,999,999,999,999 - 599,999,999,999,999,999,999,999,999,999,999,999,999,999,999,999,998
I wonder if it has unlimited precision? - ultrafez, on 10/30/2009, -1/+5ENGLISH. DO YOU SPEAK IT *****?
- the8thbit, on 10/30/2009, -0/+3Score one for free software.
- mdizz21, on 10/31/2009, -1/+4Some possible Canadian diggers: Jim Carey, Mike Myers, Michael Cera, Pamela Anderson (minus her *****, cause those were done in LA)
- rmxz, on 10/30/2009, -0/+3Fixed point math would have the same problem with 0.9 and 0.1 that the article describes.
Though I do notice that 256-bit-fixed-point with the decimal point in the middle is extremely cool, being able to represent just about any real-world value without the need for floating point - for example, measuring the size of the universe, and measuring the smallest atom can both be done in the same scale with 256-bit fixed point. [1]
To deal with 0.9, though, you need a system that understands fractions; and even that won't help you deal with pi or sqrt(-1). Of course there are software packages that deal with all that too ("sage" is a nice one); but most of the time I don't want to know that something should be pi feet long. I'd rather the calculator just told me 3.14 or whatever I can see on a tape measure.
[1] http://graphcomp.com/info/specs/java3d/j3dguide/Vi ... - the8thbit, on 10/30/2009, -0/+3On the day that Wolfram Alpha was released I found that it did order of operations incorrectly, which was somewhat disturbing.
- r00ts, on 10/30/2009, -1/+4If only more people understood this.
It's not the computers that are bad with math, it's the programmers. Use the correct data type for the job and you won't have these sorts of problems. - ultrafez, on 10/30/2009, -0/+3Oh you're so funny! Just because you don't understand something written in English as opposed to one of its derivatives.
- sinembarg0, on 10/30/2009, -2/+5ah, but what exactly is math short for? In the US, it's short for mathematics, which is already plural. What's it short for in the UK, mathematic?
- mcprogrammer, on 10/30/2009, -0/+3The Excel example is a formatting bug, not a problem with floating point numbers. It is related to floating point accuracy because the .1 in the calculation is what causes the bug, but it's not really the same as the other examples. Essentially, a few numbers around 65535 get converted to 100000 when they are displayed. Internally, the value is correct (or as close as it can be given the limitations of floating point), so other calculations using the result aren't affected by it.
- the8thbit, on 10/30/2009, -1/+4I don't know about computers in general, but the Pentium II was pretty bad at math.
- DeskFlyer, on 10/29/2009, -0/+3Not sure, but it does seem to be a popular word:
http://www.google.com/#hl=en&safe=off&num= ... - colto, on 10/31/2009, -0/+3for(int i=1; i>0; i++){
System.out.println ("Dammit I did it again....");
} - achansen121, on 10/30/2009, -0/+2pft..
Brits speak one kind of English; Americans speak another kind.
There are many fish that live in the sea.
But there are many salmon in the river.
Yes, I am arrogant.
In truth, this only really annoys me on something like the wikipedia article for Hybridization. Something invented by an American should have an article written in American English. - Atario, on 10/31/2009, -0/+2"Unfortunately there really is no way around it."
Sure there is -- use an appropriate data type. -
Show 51 - 92 of 92 discussions
© Digg Inc. 2009
— Content created and posted by Digg users is