Gambit-list Décembre 2018

gambit-list@iro.umontreal.ca

5 participants
6 discussions

Gambit versus Python+GMP: Timing for Chudnovskys' algorithm for pi
by Bradley Lucier 07 Avr '22

07 Avr '22

Bakul Shah wrote a particularly elegant Scheme program for Chudnovskys' algorithm for pi based on the Common Lisp program here: https://bitbucket.org/tarballs_are_good/numericl/src/5fe8fe7089f48ab1c8a388… Nick Craig-Wood wrote a Python program using the GMP multiprecision library that appears to use exactly the same algorithm here: http://www.craig-wood.com/nick/articles/pi-chudnovsky/ I modified both programs a bit and include them here. They time the calculation of $10^n$ digits of pi for $n=1,2,3,4,5,6,7$. The results are heine:~/programs/gambiteer/gambit> !py python pi_chudnovsky_bs_gmpy.py 31415926535897932384626433832795028841971693993751058209749445923078164062862089986280348253421170679 ('chudnovsky_gmpy_mpz_bs: digits', 10, 'time', 1.0967254638671875e-05) ('chudnovsky_gmpy_mpz_bs: digits', 100, 'time', 3.0040740966796875e-05) Last 5 digits 70679 OK ('chudnovsky_gmpy_mpz_bs: digits', 1000, 'time', 0.00025582313537597656) Last 5 digits 01989 OK ('chudnovsky_gmpy_mpz_bs: digits', 10000, 'time', 0.00386810302734375) Last 5 digits 75678 OK ('chudnovsky_gmpy_mpz_bs: digits', 100000, 'time', 0.0834801197052002) Last 5 digits 24646 OK ('chudnovsky_gmpy_mpz_bs: digits', 1000000, 'time', 1.655979871749878) Last 5 digits 58151 OK ('chudnovsky_gmpy_mpz_bs: digits', 10000000, 'time', 30.67442488670349) Last 5 digits 55897 OK heine:~/programs/gambiteer/gambit> gsi chud1.scm Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 10, CPU time: 0.. Last 5 digits 26535. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 100, CPU time: 0.. Last 5 digits 70679. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 1000, CPU time: .004. Last 5 digits 1989. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 10000, CPU time: .028. Last 5 digits 75678. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 100000, CPU time: .472. Last 5 digits 24646. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 1000000, CPU time: 6.448. Last 5 digits 58151. Chudnovsky's algorithm using binary splitting in Gambit Scheme: digits 10000000, CPU time: 98.612. Last 5 digits 55897. So it appears that for this algorithm applied to large integers, GMP's bignum routines are about 3-4 times as fast as Gambit's bignum routines. Not so bad. For smaller bignums, GMP has a bigger advantage. The C program gmp-chudnovsky.c includes certain optimizations to this basic algorithm: http://gmplib.org/pi-with-gmp.html ftp://ftp.gmplib.org/pub/misc/gmp-chudnovsky.c On my machine, compiled with gcc -O3 -march=native -o gmp-chudnovsky gmp-chudnovsky.c -lgmp -lm the CPU times for 1,000,000 and 10,000,000 digits are 1.064 and 18.200 seconds, respectively. This is with a somewhat older machine model name : Intel(R) Core(TM)2 Quad CPU Q8200 @ 2.33GHz running Ubuntu 13.04 with heine:~/programs/gambiteer/gambit> gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/4.7/lto-wrapper Target: x86_64-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Ubuntu/Linaro 4.7.3-1ubuntu1' --with-bugurl=file:///usr/share/doc/gcc-4.7/README.Bugs --enable-languages=c,c++,go,fortran,objc,obj-c++ --prefix=/usr --program-suffix=-4.7 --enable-shared --enable-linker-build-id --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.7 --libdir=/usr/lib --enable-nls --with-sysroot=/ --enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-gnu-unique-object --enable-plugin --with-system-zlib --enable-objc-gc --with-cloog --enable-cloog-backend=ppl --disable-cloog-version-check --disable-ppl-version-check --enable-multiarch --disable-werror --with-arch-32=i686 --with-abi=m64 --with-multilib-list=m32,m64,mx32 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu Thread model: posix gcc version 4.7.3 (Ubuntu/Linaro 4.7.3-1ubuntu1) heine:~/programs/gambiteer/gambit> gsi -v v4.6.9 20130607151908 x86_64-unknown-linux-gnu "./configure '--enable-single-host' '--enable-multiple-versions' '--enable-shared'" and the Ubuntu-provided GMP 5.0.5. (I'm sure the GMP folks have a better way to build GMP on my machine than the "generic" 64-bit version provided by Ubuntu.) Brad

7 18

equal? and equal?-hash
by Marc Feeley 28 Déc '18

28 Déc '18

This is just a heads up that the implementation of equal? and equal?-hash has been completely rewritten. These functions can now handle cycles and sharing, as shown in this example: > (define x (list 1 2 3)) > (define y (list 1 2 3)) > (set-cdr! (cddr x) x) > (set-cdr! (cddr y) x) > (equal? x y) #t > (equal?-hash x) 29507628 > (equal?-hash y) 29507628 > (list? x) #f > (list? y) #f Note that equal?-hash is the default hash function that is used for tables created with make-table. Due to the fundamental nature of equal?, I would be interested in feedback from users making frequent use of equal? and tables in their code. Please report any performance degradation you notice (in particular because the new hash function may cause a change in the frequency of hash table collisions). Marc

1 0

gambit wiki
by Sonny To 21 Déc '18

21 Déc '18

is the wiki down? http://gambitscheme.org/

2 1

Voting on the Tangerine Edition of R7RS-large begins!
by John Cowan 16 Déc '18

16 Déc '18

Apologies to those who receive multiple copies of this posting. There are two ballots to vote on this time, the Tangerine Edition ballot, which decides what goes into R7RS-large, and the Orange Edition straw poll, which helps to decide what will go on the ballot for the future Orange Edition. Voting on both is open until the end of January. Please vote! Links to the ballots, which are Google Forms, will be sent to scheme-reports-wg2 ONLY. However, you can also reach the forms at at < https://docs.google.com/forms/d/1XQNvIzijmCxHgnaRq31u9MogloG0YWDaWKBs-J9CCTA> (Tangerine) and < https://docs.google.com/forms/d/1qn1Ut7tR5bzXyWOrxpbLu5O7rJJ5_m2HhAeG8YqfgG…> (Orange). You don't need a Google account. Let me know if anything goes wrong. Here are the detailed instructions for the Tangerine Edition ballot: This is a ballot to decide which SRFIs are to be included in the Tangerine (data structures and numerics) Edition of R7RS-large. This is the second of about 8 editions, so only certain topics are being voted on now. If you are seeing this ballot, you are a member of Working Group 2 provided you actually vote. However, if you have not voted on a Scheme ballot or ratification before, please send an email to scheme-reports-wg2(a)groups.google.com giving your name and a short explanation of your interest in Scheme. Choose "No vote" if you wish to abstain, which means that your vote on this question will not be counted. Note that if one person votes for alternative A, and two people for alternative B, and everyone else abstains, alternative B wins. That is because we are going by a majority of the legal votes cast for each ballot question. If no alternative achieves a majority, the question will be reballoted at a later date. Otherwise, choose "None" if you wish not to include a library of the type described in R7RS-large, or choose one of the SRFIs or R6RS according to the choices given. You can also choose "Other" for a write-in vote. There are also some yes/no questions on the ballot, for which the answers are "No vote", "No", and "Yes". If you want to revote, just vote again using the same name. Only the last vote is counted. The ballot closes at the last moment of Friday, February 1, 2019, in any time zone, which is equivalent to noon February 2 UTC. All ballots will be made public. If you object to using Google Forms (you do not need a Google account), post your ballot in an email to scheme-reports-wg2(a)groups.google.com. As an absolute fallback, send an email to cowan(a)ccil.org and I will post your vote for you. Ballots posted to other fora will be used if I see them, but are not formally supported. And here are the instructions for the Orange Edition straw poll: Please vote on the proposals you think should be considered for R7RS-large. This vote is not binding, but provides guidance to the editor about which pre-SRFI proposals should be converted into SRFIs and which should not. Ones that are already SRFIs or which have simple or existing implementations are excluded. There are four options in all cases: "No vote", "No", "Yes", and "Volunteer"; the last means that you are volunteering to write the implementation. -- John Cowan http://vrici.lojban.org/~cowan cowan(a)ccil.org weirdo: When is R7RS coming out? Riastradh: As soon as the top is a beautiful golden brown and if you stick a toothpick in it, the toothpick comes out dry.

1 1

gambitscheme.org problem
by Dan Thedens 10 Déc '18

10 Déc '18

There seems to be a redirect loop of some kind between gambitscheme.org and dynamo.iro.umontreal.ca/~gambit that doesn't resolve. When I attempt to connect to http://gambitscheme.org I get no page but a deep nesting of frames with source http://dynamo.iro.umontreal.ca/~gambit . Going to http://dynamo.iro.umontreal.ca/~gambit directly gives the same thing. I can still get to the wiki by using the full URL http://dynamo.iro.umontreal.ca/wiki/index.php/Main_Page

1 0

Updated fork of SRFI 122, Nonempty Intervals and Generalized Arrays
by Bradley Lucier 04 Déc '18

04 Déc '18

I maintain a fork of SRFI 122 https://github.com/gambiteer/srfi-122/ where I've added a few things that were not in the finalized SRFI, specifically the routines interval-cartesian-product array-outer-product array-assign! I continue to think about arrays and how to use them. I'd like to get various implementations and applications of SRFI-122 available (beyond what I use it for) and when it looks like I don't have anything more to add I might propose an updated SRFI with the new routines. In that vein, if anyone would like assistance to modify SRFI-122 to run in their favorite scheme, or to use SRFI-122 in support of their application, I'd be glad to help. Brad

3 6

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gambit-list Décembre 2018