Index of /WallStreet/6351/txt
Name Last modified Size Description
![[DIR]](/icons/back.gif)
Parent Directory 20-Mar-2010 15:21 -
![[TXT]](/icons/text.gif)
0001.html 27-Apr-2009 09:28 20k
0002.html 27-Apr-2009 09:28 9k
0003.html 27-Apr-2009 09:28 9k
0004.html 27-Apr-2009 09:28 9k
0005.html 27-Apr-2009 09:28 10k
0006.html 27-Apr-2009 09:28 13k
0007.html 27-Apr-2009 09:28 14k
0008.html 27-Apr-2009 09:28 20k
0009.html 27-Apr-2009 09:28 11k
0010.html 27-Apr-2009 09:28 3k
0011.html 27-Apr-2009 09:28 21k
0012.html 27-Apr-2009 09:28 15k
0013.html 27-Apr-2009 09:28 33k
0014.html 27-Apr-2009 09:28 19k
0015.html 27-Apr-2009 09:29 32k
0016.html 27-Apr-2009 09:29 15k
0017.html 27-Apr-2009 09:29 63k
0018.html 27-Apr-2009 09:28 59k
0019.html 27-Apr-2009 09:29 41k
0020.html 27-Apr-2009 09:28 5k
0021.html 27-Apr-2009 09:28 49k
0022.html 27-Apr-2009 09:28 36k
0023.html 27-Apr-2009 09:29 30k
0024.html 27-Apr-2009 09:28 82k
0025.html 27-Apr-2009 09:28 75k
0026.html 27-Apr-2009 09:29 33k
0027.html 27-Apr-2009 09:29 55k
0028.html 27-Apr-2009 09:29 31k
0029.html 27-Apr-2009 09:29 3k
0030.html 28-May-2009 19:54 15k
0031.html 28-May-2009 19:54 197k
0032.html 28-May-2009 19:54 39k
0033.html 28-May-2009 19:54 27k
0034.html 27-May-2009 22:54 64k
0035.html 21-Jul-2009 19:54 83k
0036.html 20-Dec-2009 20:54 100k
56_game_hitting_stre..> 27-Apr-2009 09:29 5k
bit.html 08-Jun-2009 18:37 3k
bnews.html 27-Apr-2009 09:29 20k
busy_beaver.html 27-Apr-2009 09:29 1k
distributed_computin..> 27-Apr-2009 09:29 3k
fermat.html 11-Jan-2010 16:54 6k
livestation.html 27-Apr-2009 09:29 2k
mile_per_gallon.html 27-Apr-2009 09:29 1k
okinawa2004.html 27-Apr-2009 09:29 1k
okinawa2008.html 27-Apr-2009 09:29 1k
scifi.html 29-Jun-2009 08:54 25k
tc-16as.html 08-Jun-2009 18:37 6k
uucp.html 27-Apr-2009 09:29 20k
video.html 27-Apr-2009 09:29 33k
www.americafree.tv.html 27-Apr-2009 09:29 1k
Post-it
[Post-it]
---20100316/memo
http://www.perfsci.com/free-software.asp#giantint
k = 19;
while ((unsigned int)k<B)
{
if (isprime(k))
{
ell_mul(xr, zr, k, An, Ad, N);
if (k<500) 小さい素数は２回含む
ell_mul(xr, zr, k, An, Ad, N);
if (cnt++ %1000==0)
dot();
}
k += 2;
}
第一ステージ
---20100316/memo
http://homepage2.nifty.com/m_kamada/math/ecm_ja.htm
未知の素因数の桁数とB1の関係（READMEより）
知の素因数の桁数 B1 繰り返す回数と発見できる確率
63% 86% 95%
20 11000 74 148 222
25 50000 214 428 642
30 250000 430 860 1290
35 1000000 904 1808 2712
40 3000000 2350 4700 7050
45 11000000 4480 8960 13440
50 43000000 7553 15106 22659
55 110000000 17769 35538 53307
60 260000000 42017 84034 126051
65 850000000 69408 138816 208224
http://mersenneforum.org/showthread.php?t=13146
73 digit ECM factor
we (Joppe Bos, Thorsten Kleinjung, Arjen Lenstra, Peter Montgomery)
found the following 73-digit prime factor
1808422353177349564546512035512530001279481259854248860454348989451026887
of 2^1181-1 by ECM, completing the factorisation of this number.
Stage 1: we implemented arithmetic functions for Playstation3s for
Mersenne numbers. Stage 1 for 24 curves in parallel and for B1=3*10^9
took less than 23 hours on one PS3, i.e., less than one hour per curve
per PS3.
Stage 2: we parallelised some functions, this stage with the default value
of B2 of about 10^14 took about 15 minutes on 4 cores (per curve).
We ran more than 30000 stage 1 and 8800 stage 2 computations.
Using B1=3000000000-3000000000, B2=103971375307818, polynomial x^1, sigma=4000027779
Step 1 took 0ms
http://mersenneforum.org/showthread.php?t=13051
GIMPS' second Fermat factor!
F14 has a factor
116928085873074369829035993834596371340386703423373313
I found this after approximately 750 curves ran on F14 with B1=110M.
http://mersenneforum.org/showthread.php?t=12168
GIMPS' first Fermat factor!
User Buckle turned in this gem recently!
F19 has a factor: 37590055514133754286524446080499713
ECM found a factor in curve #3, stage #2
Sigma=7121198363696307, B1=3000000, B2=300000000.
UID: Buckle/G810, F19 has a factor: 37590055514133754286524446080499713
ECM on Fermat numbers
Digits in factor 25 30 35 40 45 50 55 60
Bound #1 50,000 250,000 1,000,000 3,000,000 11,000,000 44,000,000 110,000,000 260,000,000
Curves to test 280 640 1,580 4,700 9,700 17,100 46,500 112,000
4096 Done Done Done Done Done Done Done 23,152
To attack F_n = 2^(2^n) + 1 with recommended internal paremeters:
% fermat n
To test performance/integrity on your machine:
% fermat n -t
(which runs known elliptic curves, with certain expected results).
To force a stage limit of B = 1000000, say:
% fermat n -b 1000000
---20100313/memo
http://www.perfsci.com/free-software.asp#giantint
giantint/numberth$ cc -O factor.c giants.c -lm -o factorfactor.c: In function ‘main’:
factor.c:688: warning: format ‘%d’ expects type ‘int’, but argument 4 has type ‘long int’
giantint/numberth$ time ./factor -1 137
Sieving...
Commencing Pollard rho...
...
Commencing second stage, curve 15...
..............................................................................................................................................................................................................................................................................................................................................
32032215596496435569
* 5439042183600204290159
real 0m20.900s
user 0m20.845s
sys 0m0.028s
giantint/numberth$ time ./factor +1 137
...
Commencing second stage, curve 18...
..............................................................................................................................................................................................................................................................................................................................................
32127963626435681
* 105498212027592977
real 0m27.445s
user 0m27.394s
sys 0m0.040s
giantint/numberth$ time ./fermat 7
Attacking 2^128 + 1
Selecting elliptic curve 1 seed s = 2006972216; B = 10000000, C = 500000000:
.............................................................................................................
---20100313/memo
http://www.rkmath.rikkyo.ac.jp/~kida/kima.htm
木田祐司先生作成のUBASICで書かれた楕円曲線法のプログラム GMP化が目標
10 ’ecm2
20 input "n=";N
30 gosub *Ecm2
40 print N;"=";GD;"*";N\GD
50 end
60 ’
70 *Ecm2
80 dim F(48)
90 P=isqrt(N)
100 L1=min(50000,int(1400*exp((log(P)-31.0)*0.14))+100)
110 L2=40*L1
120 LG=log(L1):LS=isqrt(L1)
130 print "L1,L2=";L1;L2
140 randomize
150 ’
160 *MainLoop
170 EC%=irnd:U=EC%+1
UBASIC では % の付いた変数は −32767 から 32767 までの整数
180 A0=2*U*modinv(3*U^2-1,N)@N
modinv ある数を法とする逆数計算
190 if A0*(A0^2-1)*(9*A0^2-1)@N=0 then 170
200 A=(-3*A0^4-6*A0^2+1)*modinv(4*A0^3,N)@N
210 AA=(A+2)*modinv(4,N)@N
220 X=(3*A0^2+1)@N:Z=4*A0@N
230 print "A=";A;" X,Z=";X;Z
240 ’* 1st step
250 ’ * for power of 2
260 for I%=1 to int(LG/log(2))
270 W1=(X+Z)^2@N:W2=(X-Z)^2@N
280 X=W1*W2@N:Z=(W1-W2)*(W2+AA*(W1-W2)@N)@N
290 next
300 GD=gcd(Z,N):if GD>1 then *Atta
310 ’ * for powers of odd primes
320 IP%=2
330 repeat
340 P=prm(IP%):IP%+=1
350 for I%=1 to int(LG/log(P))
360 gosub *ECmulti(&X,&Z,P)
370 next
380 GD=gcd(Z,N):if GD>1 then *Atta
390 until prm(IP%)>L1
400 ’* 2nd STEP
410 ’* make polynomial
420 UX=X:UZ=Z:’ =R
430 Deg%=1:F(1)=1:F(0)=(-X*modinv(Z,N))@N
440 W1=(X+Z)^2@N:W2=(X-Z)^2@N
450 TX=W1*W2@N:TZ=(W1-W2)*(W2+AA*(W1-W2)@N)@N:’=2R
460 W1=(X-Z)*(TX+TZ)@N:W2=(X+Z)*(TX-TZ)@N
470 X=(W1+W2)^2@N*UZ@N:Z=(W1-W2)^2@N*UX@N:’ =3R
480 for I%=5 to 210 step 2
490 WX=X:WZ=Z
500 W1=(X-Z)*(TX+TZ)@N:W2=(X+Z)*(TX-TZ)@N
510 X=(W1+W2)^2@N*UZ@N:Z=(W1-W2)^2@N*UX@N:’ =7R,9R,...
520 if I%=105 then DX=X:DZ=Z:’ =105R
530 if gcd(I%,210)=1 then
540 :Deg%+=1:Xw=(-X*modinv(Z,N))@N
550 :F(Deg%)=1
560 :for J%=Deg%-1 to 1 step -1
570 :F(J%)=(F(J%)*Xw+F(J%-1))@N
580 :next
590 :F(0)=F(0)*Xw@N
600 UX=WX:UZ=WZ
610 next
620 ’
630 W1=(DX+DZ)^2@N:W2=(DX-DZ)^2@N
640 X=W1*W2@N:Z=(W1-W2)*(W2+AA*(W1-W2)@N)@N:’ =210R
650 UX=X:UZ=Z
660 W1=(X+Z)^2@N:W2=(X-Z)^2@N
670 TX=W1*W2@N:TZ=(W1-W2)*(W2+AA*(W1-W2)@N)@N:’=420R
680 W1=(X-Z)*(TX+TZ)@N:W2=(X+Z)*(TX-TZ)@N
690 X=(W1+W2)^2@N*UZ@N:Z=(W1-W2)^2@N*UX@N:’ =630R
700 ’
710 for Q%=2 to L2\420
720 X1=X*modinv(Z,N)@N:X2=X1^2@N:X3=X1*X2@N:X4=X2^2@N
730 W=1
740 for J%=47 to 0 step -4
750 W=(W*X4+F(J%)*X3+F(J%-1)*X2+F(J%-2)*X1+F(J%-3))@N
760 next
770 GD=gcd(W,N):if GD>1 then cancel for:goto *Atta
780 WX=X:WZ=Z
790 W1=(X-Z)*(TX+TZ)@N:W2=(X+Z)*(TX-TZ)@N
800 X=(W1+W2)^2@N*UZ@N:Z=(W1-W2)^2@N*UX@N
810 UX=WX:UZ=WZ
820 next:Q%=0
830 goto *MainLoop:’try next curve
840 ’
850 *Atta
860 print:return:’GD is a divisor
870 ’
880 ’* main subroutine
890 ’ M must be odd
900 ’ N,AA : global
910 *ECmulti(&X,&Z,M)
920 local I%,TX,TZ,UX,UZ,W1,W2,W3,W4
930 TX=X:TZ=Z:UX=X:UZ=Z
940 W3=TX+TZ:W4=TX-TZ
950 for I%=1 to len(M)-1
960 W1=W3^2@N:W2=W4^2@N
970 TX=W1*W2@N:TZ=(W1-W2)*(W2+AA*(W1-W2)@N)@N
980 W3=TX+TZ:W4=TX-TZ
990 if bit(I%,M)
1000 :then W1=W4*(X+Z)@N:W2=W3*(X-Z)@N
1010 :X=(W1+W2)^2@N*UZ@N:Z=(W1-W2)^2@N*UX@N
1020 :else W1=W4*(UX+UZ)@N:W2=W3*(UX-UZ)@N
1030 :UX=(W1+W2)^2@N*Z@N:UZ=(W1-W2)^2@N*X@N
1040 next
1050 return
run
n=? 18^16+1
L1,L2= 564 22560
A= 90032034639039456184
X,Z= 65814232592604926128 45923549317295498171
A= 77216027138829881607
X,Z= 94698138176341449409 52282093912201079366
A= 10384364739415668074
X,Z= 27901642140712720029 35238368532640931443
121439531096594251777 = 30894471809 * 3930785153
http://www.perfsci.com/free-software.asp#giantint
には、CソースのECM法
---20100201/CUDA
http://www.nvidia.com/object/cuda_develop.html
"CUDA 2.2 Quick Start Guide" is Install Guide.
---20100131/memo
$ time ./a.out f24.param |more
GeneFer 1.3 (standard) Copyright (C) 2001-2002, Yves GALLOT
A program for finding large probable generalized Fermat primes.
Usage: GeneFer -b run bench
GeneFer -t run test
GeneFer <filename> test <filename>
GeneFer use interactive mode
Start test of file 'f24.param'.
3.96e+00 536^1058076+1...
0.00e+00
$ time ./a.out f26.param |more
GeneFer 1.3 (standard) Copyright (C) 2001-2002, Yves GALLOT
A program for finding large probable generalized Fermat primes.
Usage: GeneFer -b run bench
GeneFer -t run test
GeneFer <filename> test <filename>
GeneFer use interactive mode
Start test of file 'f26.param'.
4.00e+00 536^4194304+1...
0.00e+00
---20100123/memo
http://www.mersenne.org/report_exponent/?exp_lo=36042257&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22018729&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22020091&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22021367&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22041823&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22052431&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22676837&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22676839&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22679021&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22679203&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22681523&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22860017&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22860527&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22873453&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22882129&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22893397&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=23277439&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=23277557&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=23704859&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=23704969&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22688521&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22833439&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22842103&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22849927&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22856573&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22876219&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22889017&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22892447&exp_hi=&B1=Get+status
---20100111/CUDA
On ubuntu9.04/32bit/GTX260
$ time ./mfaktc.exe 66362159 64 65
mfaktc v0.01 C...
...
no factor for M66362159 from 2^64 to 2^65 bits
tf(): total time spent: 273133msec
real 4m33.207s
user 4m30.925s
sys 0m2.288s
Compiletime Options
THREADS_PER_GRID 1048576
THREADS_PER_BLOCK 256
SIEVE_SIZE_LIMIT 32kiB
SIEVE_SIZE 230945bits
SIEVE_PRIMES 50000
USE_PINNED_MEMORY enabled
USE_ASYNC_COPY enabled
VERBOSE_TIMING disabled
SELFTEST disabled
MORE_CLASSES disabled
---20100109/SIGPS
69991758 2048 12289
I think it cannot be really the last candidate until 70M
$tail 60000000
69991628 2048 12289
69991640 2048 12289
69991650 2048 61441
69991676 2048 12289
69991692 2048 12289
69991694 2048 12289
69991696 2048 12289
69991730 2048 12289
69991758 2048 12289
project3
12289,40961,61441,65537,86017
f0:2000000-230000000の割算の結果
s0:候補
f1:候補の篩いの結果
s1:残ったもの
2336,414724444161
,123456789012
> 69991832 2048 12289
> 69991924 2048 40961
> 69991944 2048 12289
> 69991990 2048 12289
> 69992004 2048 61441
> 69992028 2048 12289
> 69992032 2048 86017
> 69992046 2048 12289
> 69992078 2048 12289
> 69992090 2048 12289
> 69992134 2048 40961
> 69992148 2048 12289
> 69992158 2048 12289
> 69992220 2048 40961
> 69992226 2048 40961
> 69992248 2048 61441
> 69992256 2048 12289
> 69992274 2048 86017
> 69992288 2048 61441
> 69992298 2048 12289
> 69992310 2048 12289
> 69992322 2048 12289
> 69992342 2048 40961
> 69992362 2048 40961
> 69992416 2048 40961
> 69992424 2048 12289
> 69992458 2048 12289
> 69992462 2048 40961
> 69992554 2048 12289
> 69992644 2048 12289
> 69992652 2048 86017
> 69992672 2048 12289
> 69992674 2048 40961
> 69992724 2048 12289
> 69992824 2048 40961
> 69992830 2048 12289
> 69992884 2048 12289
> 69992930 2048 12289
> 69992952 2048 12289
> 69993062 2048 86017
> 69993068 2048 61441
> 69993076 2048 12289
> 69993082 2048 12289
> 69993114 2048 12289
> 69993198 2048 12289
> 69993246 2048 12289
> 69993320 2048 12289
> 69993336 2048 12289
> 69993354 2048 40961
> 69993356 2048 12289
> 69993366 2048 12289
> 69993376 2048 12289
> 69993382 2048 40961
> 69993434 2048 12289
> 69993438 2048 12289
> 69993498 2048 12289
> 69993500 2048 40961
> 69993502 2048 12289
> 69993534 2048 86017
> 69993588 2048 12289
> 69993634 2048 40961
> 69993648 2048 12289
> 69993760 2048 12289
> 69993782 2048 61441
> 69993786 2048 61441
> 69993788 2048 86017
> 69993846 2048 61441
> 69993850 2048 86017
> 69993888 2048 12289
> 69993900 2048 12289
> 69993904 2048 61441
> 69993950 2048 61441
> 69993956 2048 12289
> 69994070 2048 12289
> 69994152 2048 12289
> 69994180 2048 12289
> 69994228 2048 40961
> 69994234 2048 61441
> 69994238 2048 86017
> 69994258 2048 12289
> 69994264 2048 12289
> 69994270 2048 12289
> 69994274 2048 61441
> 69994278 2048 12289
> 69994292 2048 40961
> 69994310 2048 86017
> 69994312 2048 12289
> 69994328 2048 86017
> 69994338 2048 61441
> 69994344 2048 12289
> 69994378 2048 86017
> 69994392 2048 12289
> 69994420 2048 12289
> 69994422 2048 40961
> 69994510 2048 12289
> 69994522 2048 40961
> 69994546 2048 12289
> 69994554 2048 86017
> 69994580 2048 12289
> 69994600 2048 12289
> 69994664 2048 86017
> 69994682 2048 12289
> 69994738 2048 12289
> 69994744 2048 61441
> 69994748 2048 12289
> 69995010 2048 12289
> 69995172 2048 12289
> 69995174 2048 12289
> 69995178 2048 86017
> 69995192 2048 12289
> 69995234 2048 40961
> 69995240 2048 40961
> 69995244 2048 40961
> 69995290 2048 12289
> 69995314 2048 12289
> 69995320 2048 40961
> 69995324 2048 12289
> 69995372 2048 12289
> 69995474 2048 12289
> 69995476 2048 61441
> 69995496 2048 12289
> 69995504 2048 12289
> 69995520 2048 12289
> 69995604 2048 12289
> 69995660 2048 12289
> 69995664 2048 12289
> 69995676 2048 61441
> 69995680 2048 12289
> 69995690 2048 12289
> 69995700 2048 12289
> 69995738 2048 12289
> 69995776 2048 61441
> 69995834 2048 12289
> 69995846 2048 12289
> 69995856 2048 12289
> 69995878 2048 12289
> 69995906 2048 12289
> 69995908 2048 12289
> 69995922 2048 12289
> 69995932 2048 61441
> 69995934 2048 40961
> 69995944 2048 12289
> 69995964 2048 40961
> 69995972 2048 12289
> 69995994 2048 12289
> 69996026 2048 40961
> 69996040 2048 12289
> 69996058 2048 86017
> 69996070 2048 61441
> 69996082 2048 12289
> 69996094 2048 12289
> 69996126 2048 12289
> 69996142 2048 12289
> 69996144 2048 61441
> 69996174 2048 61441
> 69996190 2048 86017
> 69996206 2048 12289
> 69996222 2048 40961
> 69996458 2048 12289
> 69996510 2048 86017
> 69996574 2048 12289
> 69996602 2048 12289
> 69996624 2048 12289
> 69996642 2048 40961
> 69996732 2048 12289
> 69996742 2048 12289
> 69996816 2048 40961
> 69996958 2048 12289
> 69996970 2048 12289
> 69996990 2048 12289
> 69996998 2048 86017
> 69997002 2048 12289
> 69997030 2048 12289
> 69997032 2048 61441
> 69997050 2048 40961
> 69997076 2048 12289
> 69997090 2048 12289
> 69997100 2048 40961
> 69997102 2048 61441
> 69997112 2048 12289
> 69997118 2048 61441
> 69997148 2048 12289
> 69997152 2048 12289
> 69997154 2048 12289
> 69997160 2048 12289
> 69997254 2048 12289
> 69997266 2048 40961
> 69997276 2048 12289
> 69997308 2048 40961
> 69997320 2048 40961
> 69997324 2048 12289
> 69997348 2048 12289
> 69997386 2048 12289
> 69997408 2048 12289
> 69997424 2048 40961
> 69997454 2048 12289
> 69997456 2048 40961
> 69997470 2048 12289
> 69997502 2048 12289
> 69997548 2048 12289
> 69997580 2048 40961
> 69997610 2048 61441
> 69997630 2048 12289
> 69997652 2048 61441
> 69997674 2048 61441
> 69997678 2048 61441
> 69997728 2048 86017
> 69997758 2048 40961
> 69997770 2048 12289
> 69997772 2048 12289
> 69997774 2048 12289
> 69997792 2048 12289
> 69997800 2048 12289
> 69997914 2048 12289
> 69997950 2048 12289
> 69998020 2048 12289
> 69998024 2048 40961
> 69998034 2048 12289
> 69998044 2048 61441
> 69998068 2048 12289
> 69998098 2048 86017
> 69998122 2048 61441
> 69998146 2048 40961
> 69998158 2048 86017
> 69998194 2048 40961
> 69998300 2048 61441
> 69998326 2048 40961
> 69998338 2048 12289
> 69998358 2048 12289
> 69998372 2048 12289
> 69998374 2048 12289
> 69998410 2048 40961
> 69998428 2048 61441
> 69998452 2048 12289
> 69998476 2048 12289
> 69998500 2048 12289
> 69998516 2048 12289
> 69998522 2048 40961
> 69998588 2048 12289
> 69998638 2048 12289
> 69998674 2048 61441
> 69998694 2048 40961
> 69998704 2048 40961
> 69998726 2048 12289
> 69998742 2048 12289
> 69998868 2048 12289
> 69998914 2048 12289
> 69998962 2048 12289
> 69998964 2048 12289
> 69998972 2048 12289
> 69998974 2048 12289
> 69999002 2048 12289
> 69999068 2048 12289
> 69999106 2048 12289
> 69999128 2048 12289
> 69999134 2048 12289
> 69999140 2048 12289
> 69999176 2048 12289
> 69999180 2048 12289
> 69999198 2048 12289
> 69999200 2048 12289
> 69999220 2048 61441
> 69999242 2048 12289
> 69999266 2048 12289
> 69999272 2048 12289
> 69999328 2048 12289
> 69999330 2048 12289
> 69999342 2048 40961
> 69999358 2048 61441
> 69999372 2048 86017
> 69999396 2048 61441
> 69999414 2048 61441
> 69999416 2048 86017
> 69999434 2048 40961
> 69999474 2048 12289
> 69999476 2048 12289
> 69999480 2048 40961
> 69999514 2048 12289
> 69999530 2048 40961
> 69999570 2048 61441
> 69999574 2048 40961
> 69999664 2048 12289
> 69999714 2048 12289
> 69999720 2048 12289
> 69999740 2048 12289
> 69999746 2048 40961
> 69999788 2048 12289
> 69999798 2048 40961
> 69999822 2048 40961
> 69999856 2048 12289
> 69999882 2048 12289
> 69999936 2048 61441
> 69999938 2048 12289
> 69999940 2048 40961
> 69999976 2048 12289
> 69999980 2048 86017
296行
.input 50549184:229999997 2048 2
60000000 357171:69991758 2048 12289
input 50191717:229999997 2048 2
project3/20100110 開始時点のセーブ
project3/20100110.0/input 新 input
2140003^2048+1)/2
2235100^2048+1)/12289
2140398^2048+1)/40961
2235064^2048+1)/61441
65535 は全ては含まれていない。
/project3/20100110.0/s3 setver* から作った。/2 以外。
---20100109/ubuntu 9.04/CUDA 2.3/CUFFT
http://www.mersenne.org/report_exponent/?exp_lo=36042257&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22018729&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22020091&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22021367&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22041823&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22052431&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22676837&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22676839&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22679021&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22679203&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22681523&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22860017&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22860527&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22873453&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22882129&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22893397&exp_hi=&B1=Get+status
---20100104/ubuntu 9.04/CUDA 2.3/CUFFT
http://www.mersenne.org/report_exponent/?exp_lo=36042257&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22018729&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22020091&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22021367&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22041823&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=22052431&exp_hi=&B1=Get+status
---20100103/ubuntu 9.04/CUDA 2.3/CUFFT
MacLucasFFTW.cuda.J.tar.gz
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m46.313s
user 0m1.216s
sys 0m0.348s
2048k fft sec/iter = 0.0107
Hi,
Version "J" at .0107 sec/iter for the 2048K
(This version support only 2048k) on GTX260.
$ time ./MacLucasFFTW 216091
Iteration 10000 M( 216091 )C, 0x30247786758b8792, n = 2097152, MacLucasFFTW v8.1 Ballester
...
Iteration 210000 M( 216091 )C, 0xcfe091c8f59f8a7b, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 216091 )P, n = 2097152, MacLucasFFTW v8.1 Ballester
real 41m36.711s
user 0m21.581s
sys 0m4.932s
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 8388608, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 8388608, MacLucasFFTW v8.1 Ballester
real 9m4.028s
user 0m2.080s
sys 0m0.600s
---20100103/ubuntu 9.04/CUDA 2.3/CUFFT
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m46.683s
user 0m1.252s
sys 0m0.396s
---20100102/ubuntu 9.04/CUDA 2.3/CUFFT
$ ../../bin/linux/release/transposeNew
[TransposeNew]
> Device 0: "GeForce GTX 260"
> SM Capability 1.3 detected:
> CUDA device has 27 Multi-Processors
> SM performance scaling factor = 1.00
Matrix size: 2048x2048 (64x64 tiles), tile size: 32x32, block size: 32x8
Kernel Loop over kernel Loop within kernel
------ ---------------- ------------------
simple copy 91.25 GB/s 91.48 GB/s
shared memory copy 88.07 GB/s 90.39 GB/s
naive transpose 4.56 GB/s 4.54 GB/s
coalesced transpose 63.89 GB/s 70.29 GB/s
no bank conflict trans 71.45 GB/s 70.85 GB/s
coarse-grained 71.42 GB/s 70.80 GB/s
fine-grained 87.92 GB/s 90.78 GB/s
diagonal transpose 69.93 GB/s 73.15 GB/s
Test PASSED
Press ENTER to exit...
float -> double
$ ../../bin/linux/release/transposeNew
[TransposeNew]
> Device 0: "GeForce GTX 260"
> SM Capability 1.3 detected:
> CUDA device has 27 Multi-Processors
> SM performance scaling factor = 1.00
Matrix size: 2048x2048 (64x64 tiles), tile size: 32x32, block size: 32x8
Kernel Loop over kernel Loop within kernel
------ ---------------- ------------------
simple copy 92.16 GB/s 83.95 GB/s
shared memory copy 88.38 GB/s 92.47 GB/s
naive transpose 5.27 GB/s 5.36 GB/s
coalesced transpose 73.44 GB/s 72.82 GB/s
no bank conflict trans 74.19 GB/s 72.69 GB/s
coarse-grained 74.18 GB/s 72.69 GB/s
fine-grained 88.35 GB/s 92.45 GB/s
diagonal transpose 83.29 GB/s 90.03 GB/s
Test PASSED
Press ENTER to exit...
AA=0
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m56.451s
user 0m1.148s
sys 0m0.280s
AA=8
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m56.262s
user 0m1.156s
sys 0m0.312s
AA=16
$ time ./MacLucasFFTW 33333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m56.293s
user 0m1.148s
sys 0m0.316s
$ time ./MacLucasFFTW 63333333
Iteration 10000 M( 63333333 )C, 0xa5d7a917d728239a, n = 4194304, MacLucasFFTW v8.1 Ballester
M( 63333333 )C, 0xa5d7a917d728239a, n = 4194304, MacLucasFFTW v8.1 Ballester
real 3m40.366s
user 0m1.576s
sys 0m0.332s
#define NCPU 2048
#define IDX(i) ( ( (i) >> 11 ) + (((i) & 2047) << 11) )
UL j;
const int numThreads = blockDim.x * gridDim.x;
const int threadID = blockIdx.x * blockDim.x + threadIdx.x;
g_N = 1024 * 2048
const int stride = g_N/NCPU; 1024
const int js= stride * threadID;
const int je= js + stride;
BIG_DOUBLE temp0,tempErr;
BIG_DOUBLE maxerr=0.0,err=0.0;
BIG_DOUBLE carry;
0-1023,1024-2047
---20091231/ps3
7612 ? RNl 130441:55 ./MacLucasFFTW.k.6 -c1000000 c43112609
26732 ? RN 61589:58 ./dg2-1.043.ps3.out input
Hi,
My PS3 vrified M43112609.
$ ./MacLucasFFTW.k.6 -c1000000 43112609
...
43100001 4194304
43110001 4194304
M( 43112609 )P, n = 4194304, MacLucasFFTW v8.1 Ballester
ELAPS TIME 59 days
Thank you,
---20091227/ubuntu 9.04/CUDA 2.3/CUFFT
MacLucasFFTW.cuda.H.tar.gz
$ time ./MacLucasFFTW 33333333
time ./MacLucasFFTW 63333333
Iteration 10000 M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 33333333 )C, 0xd717246f501c7d94, n = 2097152, MacLucasFFTW v8.1 Ballester
real 1m56.446s
user 0m1.296s
sys 0m0.320s
$ time ./MacLucasFFTW 63333333
Iteration 10000 M( 63333333 )C, 0xa5d7a917d728239a, n = 4194304, MacLucasFFTW v8.1 Ballester
M( 63333333 )C, 0xa5d7a917d728239a, n = 4194304, MacLucasFFTW v8.1 Ballester
real 3m40.547s
user 0m1.520s
sys 0m0.388s
$ time ./MacLucasFFTW 216091
Iteration 10000 M( 216091 )C, 0x30247786758b8792, n = 2097152, MacLucasFFTW v8.1 Ballester
...
Iteration 210000 M( 216091 )C, 0xcfe091c8f59f8a7b, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 216091 )P, n = 2097152, MacLucasFFTW v8.1 Ballester
real 43m39.499s
user 0m19.289s
sys 0m4.268s
Hi,
Version "H" at .0117 sec/iter for the 2048K FFT and .0221 sec/iter for the 4096K FFT on GTX260.
and GTX260 result.
http://www.mersenne.org/report_exponent/?exp_lo=36000127&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36000521&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36010921&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36007753&exp_hi=&B1=Get+status
---20091227/memo
ps4
7612 ? RNl 130441:55 ./MacLucasFFTW.k.6 -c1000000 c43112609
26732 ? RN 61589:58 ./dg2-1.043.ps3.out input
ELAPS TIME 59 days
ps3
1879 ? S 0:00 time ./f26.ps3.1.out f26.param
1880 ? Rl 182157:55 ./f26.ps3.1.out f26.param
32654 ? RN 146379:00 ./dg2-1.043.ps3.out input
http://www.mersenne.org/report_exponent/?exp_lo=36000127&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36000521&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36010921&exp_hi=&B1=Get+status
http://www.mersenne.org/report_exponent/?exp_lo=36007753&exp_hi=&B1=Get+status
---20091222/ubuntu 9.04/CUDA 2.3/CUFFT
4096*4096*8=128Mbyte
---20091221/ubuntu 9.04/CUDA 2.3/CUFFT
$ time ./MacLucasFFTW 11111111
Iteration 1000 M( 11111111 )C, 0x563e0cb0741ec03d, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 11111111 )C, 0x563e0cb0741ec03d, n = 2097152, MacLucasFFTW v8.1 Ballester
real 0m13.486s
user 0m0.440s
sys 0m0.216s
$ time ./MacLucasFFTW 11111111
Iteration 1000 M( 11111111 )C, 0x563e0cb0741ec03d, n = 1048576, MacLucasFFTW v8.1 Ballester
M( 11111111 )C, 0x563e0cb0741ec03d, n = 1048576, MacLucasFFTW v8.1 Ballester
real 0m7.449s
user 0m0.368s
sys 0m0.108s
$ time ./MacLucasFFTW 1111111
Iteration 1000 M( 1111111 )C, 0x59722dd388ae5361, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 1111111 )C, 0x59722dd388ae5361, n = 2097152, MacLucasFFTW v8.1 Ballester
real 0m12.770s
user 0m0.468s
sys 0m0.156s
$ time ./MacLucasFFTW 1111111
Iteration 1000 M( 1111111 )C, 0x59722dd388ae5361, n = 65536, MacLucasFFTW v8.1 Ballester
M( 1111111 )C, 0x59722dd388ae5361, n = 65536, MacLucasFFTW v8.1 Ballester
real 0m3.107s
user 0m0.208s
sys 0m0.092s
$ time ./MacLucasFFTW 2222222
Iteration 1000 M( 2222222 )C, 0xbddaec90077bf804, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 2222222 )C, 0xbddaec90077bf804, n = 2097152, MacLucasFFTW v8.1 Ballester
real 0m12.705s
user 0m0.464s
sys 0m0.160s
$ time ./MacLucasFFTW 2222222
Iteration 1000 M( 2222222 )C, 0xbddaec90077bf804, n = 131072, MacLucasFFTW v8.1 Ballester
M( 2222222 )C, 0xbddaec90077bf804, n = 131072, MacLucasFFTW v8.1 Ballester
real 0m4.029s
user 0m0.280s
sys 0m0.080s
$ time ./MacLucasFFTW 4444444
Iteration 1000 M( 4444444 )C, 0x2843bac364b588ae, n = 2097152, MacLucasFFTW v8.1 Ballester
M( 4444444 )C, 0x2843bac364b588ae, n = 2097152, MacLucasFFTW v8.1 Ballester
real 0m12.689s
user 0m0.460s
sys 0m0.188s
$ time ./MacLucasFFTW 4444444
Iteration 1000 M( 4444444 )C, 0x2843bac364b588ae, n = 262144, MacLucasFFTW v8.1 Ballester
M( 4444444 )C, 0x2843bac364b588ae, n = 262144, MacLucasFFTW v8.1 Ballester
real 0m4.528s
user 0m0.256s
sys 0m0.112s
---20091220/ubuntu 9.04/CUDA 2.3/CUFFT
2048*2048=2^22
---20091219/ubuntu 9.04/CUDA 2.3/CUFFT
http://forums.nvidia.com/index.php?showtopic=69801&hl=CUFFT
CUDA_INSTALL_PATH ?= /usr/local/cuda
INCLUDES := -I. -I$(CUDA_INSTALL_PATH)/include
float -> double
$ ./FFT
Device: GeForce GTX 260, 1404 MHz clock, 895 MB memory.
--------CUFFT------- ---This prototype---
N Batch Gflop/s GB/s error Gflop/s GB/s error
8 524288 8.7 9.3 1.7 83.1 88.7 1.6
64 65536 55.0 29.3 2.4 168.5 89.9 2.5
512 8192 199.0 70.7 2.1 205.1 72.9 2.5
Errors are supposed to be of order of 1 (ULPs).
Device: GeForce GTX 260, 1404 MHz clock, 895 MB memory.
Compiled with CUDA 2030.
--------CUFFT------- ---This prototype--- ---two way---
N Batch Gflop/s GB/s error Gflop/s GB/s error Gflop/s error
8 1048576 8.6 9.1 1.7 83.4 89.0 1.6 83.5 2.1
16 524288 19.3 15.4 2.1 108.0 86.4 1.3 108.0 1.9
64 131072 55.4 29.6 2.5 170.0 90.7 2.5 168.0 2.9
256 32768 167.3 66.9 1.7 166.0 66.4 2.0 165.9 2.9
512 16384 199.8 71.0 2.1 206.2 73.3 2.5 206.5 3.7
1024 8192 188.1 60.2 2.3 177.1 56.7 2.5 177.3 3.9
2048 4096 171.0 49.7 2.6 140.8 41.0 3.0 140.6 4.5
4096 2048 162.4 43.3 2.3 151.6 40.4 3.3 151.4 4.9
8192 1024 166.8 41.1 2.3 164.3 40.4 3.4 163.9 5.2
Errors are supposed to be of order of 1 (ULPs).