Heesob Park wrote in post #1109662:
> Hi,
>
> 2013/5/21 Rob Biedenharn <rob / agileconsultingllc.com>:
>>> For a 64-bit build numbers should be Fixnum all the way up to (2**63-1).
>> Fixnum#size returns the number of bytes in the representation. On a 32-bit
> system, you'd get 1.size==4 instead.
>>
>> -ROb
>
> Windows uses LLP64 data models, which maintains compatibility with
> 32-bit code by leaving both int and long as 32-bit.
>
> C:\Users\phasis>ruby -ve 'puts 1.size'
> ruby 2.0.0p195 (2013-05-14) [i386-mingw32]
> 4
>
> C:\Users\phasis>ruby -ve 'puts 1.size'
> ruby 2.0.0p195 (2013-05-14) [x64-mingw32]
> 4
>
> C:\Users\phasis>ruby -ve "puts ['a'].pack('P').length"
> ruby 2.0.0p195 (2013-05-14) [i386-mingw32]
> 4
>
> C:\Users\phasis>ruby -ve "puts ['a'].pack('P').length"
> ruby 2.0.0p195 (2013-05-14) [x64-mingw32]
> 8
>
> Refer to http://en.wikipedia.org/wiki/LLP64#64-bit_data_models
>
> Regards,
> Park Heesob

So what is the effective difference between what you can/can't do with 
the 64-bit version versus 32-bits? Greater memory space, etc?

I want to do math with large numbers and assumed the 64-bit version 
would be faster for large numbers, but is that assumption true?

-- 
Posted via http://www.ruby-forum.com/.