bug-gnubg
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: R?f.. . . : Re: [Bug-gnubg] Missed doubles problem ?


From: Jim Segrave
Subject: Re: R?f.. . . : Re: [Bug-gnubg] Missed doubles problem ?
Date: Sat, 30 Aug 2003 00:19:06 +0200
User-agent: Mutt/1.4.1i

On Fri 29 Aug 2003 (22:43 +0200), Holger wrote:
> At 19:37 29.08.2003 +0000, Joern Thyssen wrote:
> 
> >Even though the positiion certainly looks lige a "too good" one gnubg
> >thinks it's around the DP due to the crude logic.
> 
> Hmm, obviously, but why? The no double equity was below 0.95.
> 
>   3. No double            +0,922  ( -0,078)
> 
> Another hmm - I may have found the culprit:
> 
>         if( arDouble[ OUTPUT_NODOUBLE ] >= 0.95 ) {
> 
> in the MOVE_NORMAL case is missing the f. So every missed double that has a 
> move stored should've been around TG.

Sorry to disappoint, but numeric values with a decimal point are
treated as doubles, adding the 'f' forces the compiler to convert to
float (lower precision). For most cases, the only real need for the %f
is to hightlight when a non-fractional value is a real rather than an
integer - 0f, 10f, etc. It's just as good to use 0.0, 10.0,
particularly as floats are promoted to doubles during arithmetic
expression evaluation.

-- 
Jim Segrave           address@hidden





reply via email to

[Prev in Thread] Current Thread [Next in Thread]