Re: Pi to 10 trillion places

>> There is no money to fund it. There is no need for it. Where would
>> you
>> put it? How could you check it?
>
> No intrinsic need, indeed. But my understanding is that the driving
> force behind these extended calculations is the need to test new
> hardware. If I run the calculation on new hardware to 2 trillion
> places, and it agrees with the previous 1.2-trillion place record, then
> the chances are very good that there are no hardware bugs.
>

This would have to be the stupidest hardware test I have ever heard of.

Doing add, multiply, shift 2 trillion times tells you only very little more
than doing it a million times, and tells you nothing about 90% of your CPU.
It would be like testing a new car design by driving it in reverse for 2,000
miles and pretending that it is a comprehensive test.

If you want to test hardware, you exercise every instruction in the hardware
set, in as many different combinations and sequences as you can think of,
just as if you are testing a car you drive at different speeds on different
roads in different conditions. For a CPU this would typically include
interrupt handling, multi-threading, super-scalar computations, DMA
performance, Bitblt ops (such as MMX) - all the hard things in CPU design,
none of which would be tested in a pi calculation.

If you want to check out hardware, devise a decent test. But don't try and
bs anybody that calculating pi is any way a reasonable way of verifying CPU
operation.




.

Relevant Pages

  • Re: antennas
    ... on products that never saw the benefits of a calculation. ... Not so with hardware. ... Antenna design has other important advantages. ... RF and antennas are ...
    (sci.electronics.design)
  • Re: antennas
    ... on products that never saw the benefits of a calculation. ... hardware, software does not wear out with age. ... design, to avoid the warranty recall experience. ... RF and antennas are ...
    (sci.electronics.design)
  • Re: problems calculation very long numbers....
    ... The reason is not ... pointer that can only be 2 bytes (compiler limitation) So what I do, ... values) the final result of this calculation is 2EB54FB16 which I have to ... > Is the format the special hardware uses compatible or the same as your ...
    (alt.comp.lang.learn.c-cpp)
  • Re: Mean time between failure in Windows XP
    ... complete system the customer is not only really interested ... in how often just the hardware crashes. ... Do you know of any other measurement on ... >>>| calculation. ...
    (microsoft.public.windowsxp.perform_maintain)
  • Re: Quad core CPUs loaded at only 50% when running a CPU and mmap intensive multi-threaded task
    ... Well, I just bought this new box, so there isn't a kernel version that I know that worked on this hardware. ... However on my previous box I have never seen such a problem (that the CPU is loaded only 50% with clamd) and I've been ... # Device Drivers ...
    (Linux-Kernel)