• ilya-stromberg (5/5) Aug 29 2013 Hi.
• Craig Dillabaugh (5/10) Aug 29 2013 This isn't a D library, but GMP might fit your needs:
• ilya-stromberg (5/9) Aug 29 2013 Thanks.
• qznc (4/9) Aug 29 2013 Phobos has std.bigint, but the documentation recommends GMP for
• ilya-stromberg (4/7) Aug 29 2013 No, I need double, not integer, so I can't use std.bigint.

"ilya-stromberg" <ilya-stromberg-2009 yandex.ru> writes:

Hi.

Do you know any library with high-precision doubles? I need at 
least 20-30 decimal digits without rounding, it will be better to 
have 50-100 decimal digits.

Thanks.

"Craig Dillabaugh" <cdillaba cg.scs.carleton.ca> writes:

On Thursday, 29 August 2013 at 09:33:12 UTC, ilya-stromberg wrote:
 Hi.

 Do you know any library with high-precision doubles? I need at 
 least 20-30 decimal digits without rounding, it will be better 
 to have 50-100 decimal digits.

 Thanks.

This isn't a D library, but GMP might fit your needs:

http://gmplib.org/

It is in C, so you should be able to interface with D code.

Craig

"ilya-stromberg" <ilya-stromberg-2009 yandex.ru> writes:

On Thursday, 29 August 2013 at 13:37:47 UTC, Craig Dillabaugh 
wrote:
 This isn't a D library, but GMP might fit your needs:

 http://gmplib.org/

 It is in C, so you should be able to interface with D code.

 Craig

Thanks.

Anybody knows a D interface for the GMP library? I can't find it 
at the Deimos.

"qznc" <qznc web.de> writes:

On Thursday, 29 August 2013 at 09:33:12 UTC, ilya-stromberg wrote:
 Hi.

 Do you know any library with high-precision doubles? I need at 
 least 20-30 decimal digits without rounding, it will be better 
 to have 50-100 decimal digits.

 Thanks.

Phobos has std.bigint, but the documentation recommends GMP for 
"very large numbers".

http://dlang.org/phobos/std_bigint.html

"ilya-stromberg" <ilya-stromberg-2009 yandex.ru> writes:

On Thursday, 29 August 2013 at 13:53:42 UTC, qznc wrote:
 Phobos has std.bigint, but the documentation recommends GMP for 
 "very large numbers".

 http://dlang.org/phobos/std_bigint.html

No, I need double, not integer, so I can't use std.bigint.
I must provide at least 20-30 decimal digits without rounding 
after a decimal point.