mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Data

Reply
 
Thread Tools
Old 2015-09-17, 19:34   #89
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

100110001110102 Posts
Default

Quote:
Originally Posted by danaj View Post
I believe these and a few more ideas are shown in the simple example on RosettaCode. I would expect software doing this sort of work to be more sophisticated than that in filtering.
And they are. But to tell that to OP I had to start from somewhere to make him understand why his code is crap.
LaurV is offline   Reply With Quote
Old 2015-09-17, 19:39   #90
Gordon
 
Gordon's Avatar
 
Nov 2008

3·167 Posts
Default

Quote:
Originally Posted by Madpoo View Post

On the face of it all, it seems so unusual that someone wrote a program that is targeted so specifically to a very small group of people and then eschews all transparency in the attempt to get them to run it. Is it any wonder people are s(c|k)eptical?
Indeed, which is why I suggested that the real intent is to get people running MGCP..

BTW @ OP

1. Explain why it must be run as administator
2. Explain why virus checking must be disabled
3. Explain why wddm whatever must be turned off. - especially when Nvidia recommend you leave it turned on.
4. What web sites the software will try to connect to, where it is hosted and who the registrant is[
5. SOURCE code - because at the moment you simply come across as trying to get us to install your bitcoin mining software or even who knows what.

STILL WAITING - TIME TO PUT UP OR PUSH OFF
Gordon is offline   Reply With Quote
Old 2015-09-17, 19:42   #91
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

26×131 Posts
Default

Quote:
Originally Posted by pepi37 View Post
In short: dont like, dont use. And dont spend time on countless , endless correspondence. It is useless at start.
Just ignore him.
if I ignored everything I didn't like or things that were useless I wouldn't read my own thoughts.
science_man_88 is offline   Reply With Quote
Old 2015-09-17, 19:54   #92
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

23×433 Posts
Default

Quote:
Originally Posted by Gordon View Post
STILL WAITING - TIME TO PUT UP OR PUSH OFF
It might be worth noting that we can be a little intense around here.

This might be because:

1. We're mostly trained in the mathematical / engineering / scientific domains, where it is perfectly OK to call someone out if their wrong.

1.1. Most humans don't take kindly to being told they are wrong.

2. This is a community, which not only watches each other's back, but have their "bullshit" detectors set to *way* higher levels than most humans.

Anyone who enters this space should be comfortable with that.

Most aren't.
chalsall is online now   Reply With Quote
Old 2015-09-17, 20:19   #93
pepi37
 
pepi37's Avatar
 
Dec 2011
After milion nines:)

148110 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
if I ignored everything I didn't like or things that were useless I wouldn't read my own thoughts.
Then you have problem :)
pepi37 is offline   Reply With Quote
Old 2015-09-17, 20:26   #94
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

26·131 Posts
Default

Quote:
Originally Posted by pepi37 View Post
Then you have problem :)
I know they've been trying to diagnose me with Asperger's since grade eight all these referrals haven't really lead to much except a few Psychology appointments a psychological evaluation in which my only major downfall that got told to anyone to my knowledge was a slow processing speed. Four ( maybe more) psychology/psychiatry/neurology personnel later and nothing seems to have come of it for the most part.

Last fiddled with by science_man_88 on 2015-09-17 at 20:29
science_man_88 is offline   Reply With Quote
Old 2015-09-17, 20:43   #95
tha
 
tha's Avatar
 
Dec 2002

827 Posts
Default

Quote:
Originally Posted by pepi37 View Post
In short: dont like, dont use. And dont spend time on countless , endless correspondence. It is useless at start.
Just ignore him.
If someone appears to try to woo some good willing people to hand over their systems to him under false pretenses that may backfire to our community, each and every virtual equivalent of those fine medieval punishments is acceptable and desirable. And indeed, it is entertaining.
tha is offline   Reply With Quote
Old 2015-09-17, 21:25   #96
pepi37
 
pepi37's Avatar
 
Dec 2011
After milion nines:)

5C916 Posts
Default

Quote:
Originally Posted by tha View Post
If someone appears to try to woo some good willing people to hand over their systems to him under false pretenses that may backfire to our community, each and every virtual equivalent of those fine medieval punishments is acceptable and desirable. And indeed, it is entertaining.
Does that "good wiling people" have your own thoughts , or they need someone else to told him what they should/need to do?
What you think?

Last fiddled with by pepi37 on 2015-09-17 at 21:26
pepi37 is offline   Reply With Quote
Old 2015-09-17, 21:26   #97
philmoore
 
philmoore's Avatar
 
"Phil"
Sep 2002
Tracktown, U.S.A.

3×373 Posts
Default

Does the author have any benchmarks for this program? How long does it take to LL test a 100 million digit candidate on a system with at least 5 GPUs? The trial factoring code is obviously not very competitive with existing programs, but the author claims to have created a new multiplication algorithm that speeds up the LL test over programs using existing FFT algorithms.

If I had a system with 5 GPUs, would the most efficient use of this system to LL test be to run five instances of CudaLucas? If so, this program would have to be 5 times as fast to be an improvement.
philmoore is offline   Reply With Quote
Old 2015-09-17, 21:48   #98
Madpoo
Serpentine Vermin Jar
 
Madpoo's Avatar
 
Jul 2014

1100111100012 Posts
Default

Quote:
Originally Posted by tha View Post
If someone appears to try to woo some good willing people to hand over their systems to him under false pretenses that may backfire to our community, each and every virtual equivalent of those fine medieval punishments is acceptable and desirable. And indeed, it is entertaining.
Besides, what if this guy is on the level and he's really on to something incredible? Like, some groundbreaking new way of doing LL sequences in a non-sequential way.

Consider his own words from 4 years ago:
Quote:
The core of the problem (excuse the pun) is the multiplication of large numbers a large number of times. In the current serial implementation, the program multiplies two 315,653 decimal-digit integers 8,386,560 times (with different integers each time, of course). It takes a long time because, to put it simply, it performs each multiplication serially - it doesn't start the next multiplication until the previous multiplication has finished.

The parallelized version, on the other hand, can perform each multiplication in parallel. Not incidentally, it does that part of the algorithm without the need for synchronization. If it takes the one core of the dual-core Intel processor "T" seconds to compute each multiplication, and assuming that each "scalar processor core" of the GPU is at least as fast at executing each multiplication, then 1,024 "scalar processor core" threads (one CTA) should get approximately 1,024 multiplications done in time "T".
I, for one, am curious to know how he's doing this when the result of one multiplication is used in the next one. It'd be an incredible feat to multiply a couple of numbers when you don't know what those numbers are. So color me intrigued.

That plus his desire to do the following:
Quote:
I have to reduce a serial process that currently takes one core of a dual-core 2.4 GHz Intel processor one full month to execute, down to exactly 1/100th of a second. Then I have to run that for a full year, non-stop.
Like I said, if he could do what it takes a dual-core 2.4 GHz CPU a month to do in only 1/100th of a second, wouldn't that be impressive?

In short, I don't think any of that is possible, and all of this was said by him 4 years ago and I'm sure by now he's realized the problems inherent in that line of thinking, but I'm still curious where he wound up.
Madpoo is offline   Reply With Quote
Old 2015-09-17, 22:00   #99
Madpoo
Serpentine Vermin Jar
 
Madpoo's Avatar
 
Jul 2014

3,313 Posts
Default

Quote:
Originally Posted by Madpoo View Post
...I, for one, am curious to know how he's doing this when the result of one multiplication is used in the next one. It'd be an incredible feat to multiply a couple of numbers when you don't know what those numbers are. So color me intrigued....
It was pointed out to me that what might be involved is breaking down a single LL iteration into parallel chunks and then combining the results.

I wonder if the author considered the effects of limited memory bandwidth. working with billion-digit Mersenne #'s, I imagine that's some hefty amounts of data flying around, and the need to run across 5 GPU's (which I think is tied to him wanting to break it down into 8,190 chunks of work) means a lot of data on PCI as well.

Oh well... those are problems I'm sure he's solved by now.

EDIT: specifically it sounds like the author was interested in this at one point:
https://en.wikipedia.org/wiki/Karatsuba_algorithm

Last fiddled with by Madpoo on 2015-09-17 at 22:08
Madpoo is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Fulsorials a1call Miscellaneous Math 46 2020-08-03 00:31
Java applet alternative a1call Programming 19 2019-11-08 22:31
Alternative to LL paulunderwood Miscellaneous Math 36 2019-08-26 08:09
Is CEMPLLA 1.5 "the only software in the world capable of discovering" something? Not really. CRGreathouse Number Theory Discussion Group 51 2018-12-16 21:55
free alternative to EasyFit? ixfd64 Software 1 2008-04-26 21:28

All times are UTC. The time now is 23:47.


Fri Oct 22 23:47:31 UTC 2021 up 91 days, 18:16, 0 users, load averages: 2.48, 2.29, 1.98

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.