Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Will most of the math be done on a computer?

  1. May 1, 2014 #1
    Just curious but how much math will I actually do say on a white board vs putting a few equations into something like excel?
  2. jcsd
  3. May 1, 2014 #2


    User Avatar
    Gold Member

    Computers are faster and more reliable.
    You can do unlimited number of equation using computers until it gets overheated..It will only take time and energy.

    You can also do unlimited number of equations using a white board until you get bored and stop.It will also take your time and energy.
  4. May 1, 2014 #3


    User Avatar
    Science Advisor

    Depends. Most of my initial analysis is done on paper or a white board. Then it is all done with Matlab, Excel, or a specialized differential equation solver.

    That said, it's tough to explore new designs with Excel... you need to be able to rapidly estimate mathematical relationships on paper or on a white board (if in a group) to be a good engineer.
  5. May 1, 2014 #4


    User Avatar
    Science Advisor

    But how do you know what equation to put into the computer?

    Garbage in, garbage out.
  6. May 1, 2014 #5


    User Avatar
    Gold Member

    You develop the algorithms on paper. You implement them and run them on the computer and then analyse the results. You might develop the algorithms on the computer, but it is just taking the place of paper.

    You might develop a spreadsheet right on the computer, but, again, it is just taking the place of paper. Then you crunch the numbers.

    The goal is to avoid "crunching numbers" on paper. Use the number cruncher.
  7. May 1, 2014 #6


    User Avatar
    Science Advisor
    Gold Member

    A spreadsheet is only as useful as the brain that programmes it. It's fine if all you want to do is 'accounting' style calculations with totals of figures and to use available Functions but that is only half the job, if you are starting from scratch on a new bit of analysis. I have to admit that I often find myself plotting a graph of a function with Excel, when I want a Max, Min or zero crossing and I could have done it all symbolically. When I get an answer the Excel way, I have not helped my own understanding of how and why that function works (the Physics behind it, in many cases).
    The Mathematica package goes further; it lets you write Analytical Functions and will do the simplifications, differentiations, integrations etc. for you - all symbolically. This is very good but someone actually programmed all those rules into Mathematica and, if you're not careful, you won't even be aware of the niceties of transformations, trig identities, Taylor Series. It will just be turning the handle and getting a result. A bit like SatNav journeys, when you get somewhere but have no idea how you got there. It really acts against your best interests, in many cases.
  8. May 1, 2014 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    I'm so awkward with spreadsheets that i do such investigations in Basic.
    It helps me keep that 'awareness' to which Sophie speaks above.

    Ahhh memories... One interpreted Basic's ATAN function blew up on me - so i wrote a Taylor series for it.
    A small step for mankind but a great one for me.
  9. May 1, 2014 #8
    Just curly but could any of the higher math have a program written around the equation you came up with in say c++ or anything that compiles quickly?
  10. May 1, 2014 #9


    User Avatar
    Science Advisor
    Gold Member

    "Curly"? Was that a typo or auto-correct? I'm not familiar with it.

    If you use any of the advanced packages, the numerical calculations are based on compiled processes and not crude, interpreted stuff. The Maths Co-pro in your computer will be accessed if there is one.

    Excel is pretty clever but it can only do things cell - by - cell unless you call and use VBA routines. VBA has a pretty decent compiler afaik. But it's not the way to approach serious number crunching because it still looks at cells and that takes ages.
  11. May 1, 2014 #10
    as a side note, prior to the math coprocessor multiply and divide operations was done by adding and subtraction. The coprocessor added a direct multiply and divide operand. The need to increase mathematical functions of a microprocessor also led to MMX (Matrix Math Extensions) which increases FPU, Floating point unit math functions. http://en.wikipedia.org/wiki/Floating-point_unit

    the MMX term is a AMD trademark name. However its used by Intel as well. the instruction set can be found here.


    the other advances in CPU processing power is the addition of more than 1 ALU (algorithmic logic unit)

    the 486 and previous CPU's could only process one math function or operand at a time, Pentium added more ALU's to their processors. The number depends on the make and model of the processor. This allowed multiple pipeline operations to be performed at the same time.

    The program capability depends on how well the software is programmed to take advantages of the CPU's increased capabilities.
    Last edited: May 1, 2014
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Will most of the math be done on a computer?
  1. Analog Computer (Replies: 2)

  2. Computations in EE? (Replies: 3)