mersenneforum.org  

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

Reply
 
Thread Tools
Old 2021-09-06, 15:32   #12
slandrum
 
Jan 2021
California

3·71 Posts
Default

It is only worth attracting people that will actually contribute to the work (and that actually want to) and not attempt to find ways to cheat the system in order to get the coin. If the only reason they are joining is to make money, then that is what they will attempt to optimize for, and to maximize return on that I would expect a lot of effort spent at attempting to exploit the system (it's easy to make up NF results for instance).
slandrum is online now   Reply With Quote
Old 2021-09-07, 05:39   #13
unk
 
Sep 2021

218 Posts
Default

I do not deny that the client's program code should be completely rewritten. It is in order to exclude deception.
After all, we are already sending messages to the server. Is it currently possible to fake these messages, to attribute additional work to yourself? If you can't, then everything is already fine.
Mersenne prime coin will require labor to implement. Here the developers will decide whether it is worth doing this. You can start a new project without stopping the old one, and see how the new project will show itself. I believe that Mersenne prime coin will increase the computing power of the project by 10-20 times. It's good. The project last found a prime number on December 7, 2018. It's been quite a while. It would be great to attract additional capacity.
You can run Mersenne prime coin in addition to the main project. See how the new approach will show itself. Only you need to be prepared for the fact that many newcomers will come, and many tasks will be required.

(machine translation)
unk is offline   Reply With Quote
Old 2021-09-07, 06:42   #14
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

29·199 Posts
Default

Quote:
Originally Posted by slandrum View Post
(it's easy to make up NF results for instance).
Definitely, could create problems for the project. Fake TF NF, Fake P-1 NF, fake GPU or any non-prime95/mprime LL composite results. And the backlog on LL DC is still several years.
There's ample work to be done. Without distracting the developers or end users with crypto.
A wait of a few years between discoveries is still fast compared to sometimes over a century in the past. Since the effort per primality test scales as ~p2.1, it's conjectured that on average the spacing is at a ratio of ~1.47:1, and at high exponent many more exponents must be tried between discoveries, a factor of ten increase in computing power fairly quickly gets absorbed and the progress rate and discovery rate declines. See https://www.mersenneforum.org/showpo...4&postcount=16

Last fiddled with by kriesel on 2021-09-07 at 06:53
kriesel is offline   Reply With Quote
Old 2021-09-07, 10:27   #15
unk
 
Sep 2021

17 Posts
Default

Quote:
Originally Posted by kriesel View Post
Fake TF NF, Fake P-1 NF, fake GPU or any non-prime95/mprime LL composite results. And the backlog on LL DC is still several years.
If the client currently used allows you to send statistics to the server with distortions, then the client needs to be rewritten. And if you can't send fake data now, then there's nothing to talk about.
You don't want to, as you want. Users do not need to do anything except register at all. Here the developers will have more work to do. Adapt the open source code of the cryptocurrency, configure mining (on the server!), configure the distribution of coins between users, write a new client for users.
Maybe someone will be interested.
Quote:
Originally Posted by kriesel View Post
a factor of ten increase in computing power fairly quickly gets absorbed
Calculating something in 40 years or in 4 years is a big difference. Well, maybe when the project hits the wall, they will implement cryptocurrency. In 40 years. Or someone implements it for themselves. And it will start the calculation from scratch, without using the GIMPS data.


(machine translation)
unk is offline   Reply With Quote
Old 2021-09-07, 10:34   #16
unk
 
Sep 2021

1116 Posts
Default

If the client allows you to send fake data to the server, it must be rewritten in any case.
Recently, the developers asked to upgrade to version 30.3. Well, they will release another update.

(machine translation)
unk is offline   Reply With Quote
Old 2021-09-07, 11:24   #17
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

24·13·47 Posts
Default

Quote:
Originally Posted by unk View Post
The server will calculate...
Fail...
The main idea of a criptocoin is decentralization, getting rid of the bank. Banks can be robbed and servers can be hacked. If I work hard for few days I can DDoS PrimeNet and stuck the "market" for a while, then bye-bye coins, their value will be zero. You have no freaking idea how a cc works...
(still reading the thread, this was about the first post, maybe somebody addressed it already, sorry if so)


edit: about "using calculations", a cc uses calculations to verify the transactions. If you want to make a mersenne cc, then you need to change the calculation in such a way to be useful for gimps, not the other way around. I mean, you must use gimps-related calculations to verify transactions. How do you do that?

Last fiddled with by LaurV on 2021-09-07 at 11:43
LaurV is offline   Reply With Quote
Old 2021-09-07, 11:40   #18
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

24×13×47 Posts
Default

Quote:
Originally Posted by unk View Post
Whenever a participant sends a completed task, the server administration transfers the Mersenne prime coin to the participant.
This we have already. It is called "Giga-Hertz-Days", and it has value zero, and as long as you can not securely "trade" it (i.e. transactions I was talking about) its value will remain zero. If I can not trade it, I can not buy pizza with it, so it has no value. Indeed, you really fail to understand how a currency (and especially a decentralized one) works. Of course, the server could "sell" and "buy" GHzD for real money, but then you get kicked in the butt from the anti-money-laundering guys... - real case: Puzzle Pirates game, which had an in-game currency which could be sold and bought for real money, but then later they introduced some poker and other card games in PP world, where you could win/loss a lot of "money", so they have been hit by AML and had to separate the internal currency in two, one you could buy with real money, and allow you access to features of the game, other you could play poker with it in the game, and there was no way to change from one to the other - otherwise AML guys will all be on you, imagine I buy such coins for a million dollars and "lose" them in game to some "random" person I play against, who is the lead of some terrorist organization, or of some child trafficking network (with or without my knowledge)... grrr...

Last fiddled with by LaurV on 2021-09-07 at 11:46
LaurV is offline   Reply With Quote
Old 2021-09-07, 14:28   #19
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

577110 Posts
Default

Quote:
Originally Posted by unk View Post
If the client allows you to send fake data to the server, it must be rewritten in any case.
Recently, the developers asked to upgrade to version 30.3. Well, they will release another update.

(machine translation)
Not the client singular, the GPU client application types plural, plus perhaps Mlucas, and any other client application that some developer decides to create in the future. Only prime95/mprime developed and maintained by George Woltman has the secret security code software for secured interaction with the PrimeNet server. There's only one of him. And he has already produced prime95/mprime up to v30.6b4, and begun mentioning v30.7. I suggest you learn more about the project & personnel, try to get a current understanding.

It's unclear to me how you propose to force all the other developers to upgrade the numerous application types to implement separate security software implementations and find enough time for the server administrators & server code developer to upgrade the server to accommodate that, and fund the necessary server upgrades, and force upgrading of the installed base of client software by thousands of end users on many more machines and instances. Only CUDALucas and clLucas are from the same developer (msft). Some of those developers are no longer actively participating; msft in particular.
Refusing unsecured client apps' output has been impractical because of the heavy dependence historically for GPU applications (which are now producing the bulk of the project throughput) on manual reservations and submissions, and because of the substantial reduction in throughput that refusing them would produce.
The algorithms for proof of work in TF NF to eliminate such cheating have been demonstrated. It's implementation that is the obstacle. (IIRC something similar is applicable to P-1 also. LL production first testing should just go away, eventually.)

There is already a financial incentive to GIMPS participation, that discourages cheating, called the EFF prizes. Cryptocoin for some projected benefit to GIMPS has been proposed and rejected in the past. Here. See also the bitcoin thread. Financial gain is not what drives GIMPS participation though.

There is also established, a set of obstacles to doing what some of the crypto proposals require, doing partial PRP, and issuing, receiving, and storing the full interim residues on the PrimeNet server. Cost of sufficient data storage & backup, time by scarce developer talent, internet delays for users with slow links, etc.

Last fiddled with by kriesel on 2021-09-07 at 14:35
kriesel is offline   Reply With Quote
Old 2021-09-07, 21:50   #20
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

23×17×73 Posts
Default

Quote:
Originally Posted by kriesel View Post
It's unclear to me how you propose to force all the other developers to upgrade the numerous application types to implement...
It often amuses (and frustrates) me how often those who don't understand how things actually work throw out ideas as the ideal solution for everything.

Crypto/Blockchain. AI. Quantum Compute. Etc, etc, etc...

The devil is in the details. Sadly, few who advocate for change understand even the high-level details.

There is no panacea. Deal with it.
chalsall is offline   Reply With Quote
Old 2021-09-07, 23:01   #21
Uncwilly
6809 > 6502
 
Uncwilly's Avatar
 
"""""""""""""""""""
Aug 2003
101×103 Posts

270A16 Posts
Default

Quote:
Originally Posted by chalsall View Post
It often amuses (and frustrates) me how often those who don't understand how things actually work throw out ideas as the ideal solution for everything.
https://imgflip.com/i/5g4jcy
Uncwilly is offline   Reply With Quote
Old 2021-09-07, 23:03   #22
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

29×199 Posts
Default

Quote:
Originally Posted by kriesel View Post
Only CUDALucas and clLucas are from the same developer (msft).
Should perhaps have said primary developer there. George has made contributions to gpuowl, and IIRC there's a bit of his code in mfaktx or is it CUDAPm1 also. As well as in the server implementation, which I know little about. Also he created MMFF, and gwnum which is used in some other packages outside hunting Mersenne primes.
Quote:
Originally Posted by chalsall View Post
It often amuses (and frustrates) me how often those who don't understand how things actually work throw out ideas as the ideal solution for everything.
...
The devil is in the details. Sadly, few who advocate for change understand even the high-level details.

There is no panacea. Deal with it.
Yes, and far too many of them are in positions of power or taking to the streets violently. And their emotion-driven wrongheaded votes count equally as much as those of the clueful.

Pat Paulson nailed it totally deadpan on LaughIn. "The leading cause of forest fires is trees." We would not have certain issues if we did not have the successes that produced the issues' prerequisites.

Something I'm unsatisfied with is my ability to distinguish among:
naivete
ignorance
stupidity
trolling
illogical sales pitches with an underlying undisclosed motive
drastically excessive optimism
outright dishonesty
sociopathy and its assorted manifestations

Last fiddled with by kriesel on 2021-09-07 at 23:18
kriesel is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
New Mersenne Software For Test Mersenne Prime Numbers On Android thorken Software 66 2019-01-13 21:08
Do-it-yourself, crank, mersenne prediction thread. Uncwilly Miscellaneous Math 85 2017-12-10 16:03
Coin Paradox petrw1 Puzzles 1 2015-02-06 23:17
coin tossing davieddy Puzzles 62 2010-05-29 01:14
Coin Problem. mfgoode Puzzles 65 2007-01-26 00:25

All times are UTC. The time now is 22:46.


Sat Oct 16 22:46:05 UTC 2021 up 85 days, 17:15, 1 user, load averages: 1.38, 1.23, 1.20

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.