mersenneforum.org New Mersenne Software For Test Mersenne Prime Numbers On Android
 Register FAQ Search Today's Posts Mark Forums Read

 View Poll Results: What about This app you like it? Yes 4 36.36% No 3 27.27% Regular 1 9.09% Very bad 3 27.27% Voters: 11. You may not vote on this poll

2019-01-05, 03:09   #56
Mysticial

Sep 2016

2·181 Posts

Quote:
 Originally Posted by Batalov ...and of course, you should have first googled just for a minute or so. You cannot easily compete with the likes of y-cruncher and Mini-Pi for Pi and e. You can definitely try - but for starters, you need to know what the state of the art is before beginning.
Calling Mini-Pi a "state of the art" is probably being a little too generous.

I quite literally slapped that thing together over a weekend with no effort. No real optimizations. Even GMP Pi is faster.

The reason why Mini-Pi gets any attention is probably because it's a single self-contained file.

2019-01-05, 03:14   #57
thorken

Jan 2019

43 Posts

Quote:
 Originally Posted by Batalov ...and of course, you should have first googled just for a minute or so. You cannot easily compete with the likes of y-cruncher and Mini-Pi for Pi and e. You can definitely try - but for starters, you need to know what the state of the art is before beginning.
I know y cruncher and I repeat That this Apps are for mobile phones and With the app could calculate more decimals than 20000, this apks are for curiosity and for up temps of my Hands In winter,by the other hand the Apps run correctly and its fast not more fast than Apps use agm.

HERE a video of Mersenne app in Android and comparison With apk Lucas lehmer in Google play.

https://youtu.be/UR1jGwvUcQc

2019-01-05, 03:23   #58
GP2

Sep 2003

258710 Posts

Quote:
 Originally Posted by Batalov Now if you compute 100,000 digits of Pi today, it will not even get you laid.
Did that ever get anyone laid, at any point in human history?

Empirically [I like using that word], that sort of thing has the precisely opposite effect.

Or, uh... so I'm told.

2019-01-05, 16:13   #59
kriesel

"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

23×857 Posts

Quote:
 Originally Posted by GP2 Did that ever get anyone laid, at any point in human history? Empirically [I like using that word], that sort of thing has the precisely opposite effect. Or, uh... so I'm told.

2019-01-05, 16:50   #60
kriesel

"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

23×857 Posts

Quote:
 Originally Posted by thorken Let's see colleague who is very good at insulting you without even being able to execute the application given your extensive experience. You say my program reads numbers from somewhere? Finally apprentice magician is very good insult without having much idea of ​​programming so it seems, make your an app for Movil that calculates numbers of Mersenne, here you are making comparisons with i7 and the cpu of a mobile. This program is for mobile and is the fastest on Android, the only one I've seen is a Google Play apk and it's very slow, this one is much faster and I'm talking about Android and mobile cpus, prime95 in AtoM n270 It is slower than this app in a GT-I9195. Greetings.
Sheesh, instead of steadfastly dodging it, just answer the question: "which algorithm of LL testing do you use ?" Or, what method is used to do the squaring? Here, I'll try to make it easy, multiple choice:

a) grammar school conventional long multiplication (using what "digit size") Twice the exponent, four times as many partial products to compute and sum, for each of twice as many iterations.

b) Karatsuba divide and conquer. Twice the exponent, THREE times as many partial products, plus some additional overhead. (applied recursively, or not?)

c) Toom-Cook (for extra credit, which?)

d) modular multiplication (using which relative primes?)

e) fft (in which base?)

f) a mix depending on operand or exponent size (which, and at what breakpoints)

g) can't say, prepublication (what journal, and publication expected about when?)

h) super secret new faster than linear, proprietary, classified, NSA and KGB and Spain's National Intelligence Agency are after thorken for releasing it in executable form

i) Other (specify)

What programming language did you use?
Or put up some source code so we can look for ourselves.

(Are you by any chance acquainted with Billy Rubin?)

 2019-01-05, 20:10 #61 ewmayer ∂2ω=0     Sep 2002 República de California 25×367 Posts Does anyone reading this have an Android dev-board with which to try an Mlucas build with? It would be best if it were one with an Arm CPU supporting the 128-bit SIMD instructions (a.k.a. Arm v8+), but even failing that one could try a generic C-code build using a scalar-doubles-based FFT. (Of course the OP is welcome to try same, but at present OP seems to be busy fending off rude questions about his bigint multiply algorithm. ;)
2019-01-05, 20:18   #62
science_man_88

"Forget I exist"
Jul 2009
Dumbassville

100000110100002 Posts

Quote:
 Originally Posted by thorken Let's see colleague who is very good at insulting you without even being able to execute the application given your extensive experience. You say my program reads numbers from somewhere? Finally apprentice magician is very good insult without having much idea of ​​programming so it seems, make your an app for Movil that calculates numbers of Mersenne, here you are making comparisons with i7 and the cpu of a mobile. This program is for mobile and is the fastest on Android, the only one I've seen is a Google Play apk and it's very slow, this one is much faster and I'm talking about Android and mobile cpus, prime95 in AtoM n270 It is slower than this app in a GT-I9195. Greetings.
They were just saying that's one possible way some program without source given could be both fast and accurate to what's already known. As shown a interpretted script can run faster. Also I got that lucas Lehmer app you talked about, when it takes 27-28 seconds for 12711 then yes many programs could run faster. that's why coders use asymptotics.

Quote:
 Originally Posted by ewmayer What exponents are those timings for, and on what compute hardware?
same exponents as OP, on the only working hardware I have at last check an android phone. I think it has a snapdragon processor, and PARIdroid was 2.11.0.1.5 edit2: seems my PARIdroid thinks they aren't prime though...

Last fiddled with by science_man_88 on 2019-01-05 at 21:02

 2019-01-05, 21:09 #63 kriesel     "TF79LL86GIMPS96gpu17" Mar 2017 US midwest 23×857 Posts Ok, in fairness to thorken, in post one, he says he found the application on a phone, and makes no claims to having written it. From his third screen shot: Code: Analizando M9967 Con 3000.37 Digitos Numeros Primos De Mersenne Encontrados 22 Tiempo Del Calculo 0h 2m 56s Calculo Den Numeros Primos De Mersenne A 0.12 PrimosM/Seg Indice De Velocidad 0.125 Puntos. MAX=100 Numeros Primos De Mersenne Calculados De 2^0-1 A 2^10000-1 Ultimo Numero Analizado Tenia 3000.37 Digitos Archivo Creado En Tu Telefono Con Los Numeros Primos De Mersenne ((( ---- ||| ---- Fin Del Programa ----- ||| ----))) Programado Por Marcos Ochoa Diez ...@gmail.com http://youtube.com/... MvIiIaX Corp. 2019 serranoltexx:/sdcard/download # Which Google translate converts (amusingly translating part of the name, Diez, to Ten) to Code: Analyzing M9967 with 3000.37 Digits Mersenne Primеs Found 22 Calculation Time 0h 2m 56s I calculate Mersenne's Prime Numbers at 0.12 PrimosM / Sec Speed Index 0.125 Points. MAX = 100 Mersenne Prime Numbers Calculated from 2 ^ 0-1 to 2 ^ 10000-1 Last Number Analyzed Tenia 3000.37 Digits File Created On Your Phone With Mersenne Cousins Numbers (((---- ||| ---- End Of Program ----- ||| ----))) Programmed by Marcos Ochoa Ten ...@gmail.com http://youtube.com/... MvIiIaX Corp. 2019 serranoltexx: / sdcard / download # which, with a little human help, becomes Code: Analyzing M9967 with 3000.37 Digits Mersenne Primеs Found 22 Calculation Time 0h 2m 56s I calculate Mersenne's Prime Numbers at 0.12 PrimosM / Sec Speed Index 0.125 Points. MAX = 100 Mersenne Prime Numbers Calculated from 2^0 - 1 to 2^10000 - 1 Last Number Analyzed Had 3000.37 Digits File Created On Your Phone With Mersenne Prime Numbers (((---- ||| ---- End Of Program ----- ||| ----))) Programmed by Marcos Ochoa Diez ...@gmail.com http://youtube.com/... MvIiIaX Corp. 2019 serranoltexx: / sdcard / download # (Note though, that he did invite questions. The general rule of be careful what you ask for, particularly applies here on the forum.) Searching online for Marcos Ochoa Diez yielded (and I by no means endorse any of it) https://loteria-primitiva.uptodown.c...dows/descargar https://generador-de-numeros-primos....wn.com/windows https://www.amazon.com/Marcos-Ochoa-.../dp/B00RN9BYNS https://yepdownload.com/monitor-lunar-ip (Nothing on github or sourceforge in what I reviewed.) Last fiddled with by kriesel on 2019-01-05 at 21:35
 2019-01-05, 21:23 #64 pinhodecarlos     "Carlos Pinho" Oct 2011 Milton Keynes, UK 2·2,539 Posts “Tenia” translates into “had”. “Mersenne cousin numbers”, please replace cousin with prime. Last fiddled with by pinhodecarlos on 2019-01-05 at 21:25
2019-01-05, 23:12   #65
GP2

Sep 2003

13×199 Posts

Quote:
 Originally Posted by ewmayer So your simple Pari script running on Android is nearly 4x faster than GP2's GMP-based LL tester running on an AWS Skylake core?
I didn't write it. It's adapted from RosettaCode, mostly by ripping out the preliminary TF stuff and making it print out a residue.

GMP isn't that complicated and I could have written a simple LL test myself, however this code incorporates an optimization for the modulo that probably wouldn't have occurred to me.

2019-01-06, 09:52   #66
LaurV
Romulan Interpreter

"name field"
Jun 2011
Thailand

236068 Posts

Quote:
 Originally Posted by pinhodecarlos “Tenia” translates into “had”.
You are wrong, I am sure it was about either this or better this...

 Similar Threads Thread Thread Starter Forum Replies Last Post allasc Math 34 2022-09-11 13:03 JonathanM Information & Answers 25 2020-06-16 02:47 paulunderwood Miscellaneous Math 18 2017-01-26 20:33 primus Miscellaneous Math 1 2014-10-12 09:25 jocelynl Math 8 2006-10-20 19:36

All times are UTC. The time now is 07:55.

Mon Oct 3 07:55:24 UTC 2022 up 46 days, 5:23, 0 users, load averages: 0.73, 0.91, 0.94