Hello,
FWIW, there was a proposal before J3 at 167
to enhance the implicit statement so that it could
indicate a default kind for intrinsic types
not otherwise given a kind parameter value.
(I con't recall the paper number OTTOMH,
but it's the paper on the intrinsic statement
listed in the papers167 list.)
IMO, literal constants are more of a porting issue
than declarations.
YMMV, I suppose.
BTW, there'll soon be a spreadsheet (.xls)
on the J3 server listing the proposal papers,
and the initial, tentative (read: "first reaction") disposition.
--
Cheers!
Dan Nagle
Purple Sage Computing Solutions, Inc.
On Sat, 6 Mar 2004 09:34:14 -0500, Peter Shenkin
<[log in to unmask]> wrote:
>Hi,
>
>On Sat, 6 Mar 2004, 4:06pm +0900, [log in to unmask] wrote:
>> On Fri, 5 Mar 2004 22:27:46 -0500, Peter Shenkin <[log in to unmask]>
>> wrote:
>
>> > The default size is generally the size that makes
>> > the most sense on the current hardware/OS. On a 32-bit
>> > architecture, this is, well, 32 bits. Note that breaking
>> > the equivalence between default real and default integer
>> > sizes would wreak real havoc, and you really don't want
>> > 64-bit default ints in a 32-bit executable.
>>
>> The default size for real is chosen to be the *integer* size
>> that makes the most sense.
>
>Yes, good point.
>
>> This is precisely because of what
>> you say. Shouldn't Fortran, being a number crunching language,
>> chose default size for real as the size that makes the most
>> sense as a real number on the current hardware/OS?
>
>There is no size that makes the most sense. 4-byte real
>makes the most sense for me, most of the time.
>
>Why does 8-byte make more sense as the default than 4,
>16 or 32?
>
>Also, FWIW, most compilers have a commandline option to
>treat all REALs as DOUBLE PRECISION; but admittedly that's
>not part of the standard.
>
>-P.
|