mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > No Prime Left Behind > Raiders of the Lost Primes

Reply
 
Thread Tools
Old 2010-03-10, 22:42   #243
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

37·79 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
I'm sorry but I just cannot figure out what this means at all. If someone's machine finds a prime, it's a prime, and it's sent to the server as a prime.
i'm just inserting the 'small' primes from the 12th drive into rieselprime.de for n=50k-250k.
i found many k's done by T.Ritschel some years ago with the note, the range from n=200k-350k were done by him. but i found many missing pairs in this range.

so if we don't test any non-Top5000 prime as prime, we are not sure about it! those are first-time testings without any old results to compare.
kar_bon is offline   Reply With Quote
Old 2010-03-10, 22:45   #244
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

37·79 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
The other main option would be to put in the "News" thread with a link to a new thread about the new client. That would make it much more highly visible.
i prefer this option most:
- post in news with link to separate thread
- own thread for these script, to collect suggestions/issues/notes/meanings/...
kar_bon is offline   Reply With Quote
Old 2010-03-10, 23:12   #245
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3·2,083 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
I'm sorry but I just cannot figure out what this means at all. If someone's machine finds a prime, it's a prime, and it's sent to the server as a prime. We don't need to double check anything right away. We do that several years down the road.

Why would we ever edit -2 to 0 in the tosend.txt file to make the server think that a composite is a prime?

Max, why we would need to "confirm" all non-top 5000 primes? Think about it. The only primes that "might" be PRPs are n<1000. We never check n-values that low. Anything from n=1K to ~490K is definitely a non-top-5000 prime if someone's machine says it is a prime. It won't be "confirmed" until it is eventually double checked.

Are you guys referring to very small primes/PRPs such as Frobenius PRPs? We certainly don't need any more code in there for checking those in some fashion.

Max, you may have to explain this one to me. :-)


Gary
I think what Karsten was meaning to say was that it would be slightly easier for someone with malicious intent to game the system with this new client: only a small change to the script would be required to have it send fake primes to the server. Since we've had malicious attacks on our servers in the past (read: Carlos) I think it would be wise to verify non-top-5000 primes to make sure they're good. (This could be done, say, at the end of a n=100K range when you're matching up master results files to the corresponding sieve files.)
Quote:
Originally Posted by gd_barnes View Post
If you guys disagree, I understand. Let's let the majority rule here so I'll place the 1st vote to keep it low key in the LLRnet servers thread. The other main option would be to put in the "News" thread with a link to a new thread about the new client. That would make it much more highly visible.
I'm inclined to agree with Karsten here--it would probably be best to give it the visibility of its own thread and an announcement in the News thread. It's kind of tempting to keep this as NPLB's "ace in the hole" for a while to give us an edge over other top-5000 projects but in reality it won't really affect us much (since our main competitors, RPS and PrimeGrid, don't use LLRnet in any form or fashion anyway).
mdettweiler is offline   Reply With Quote
Old 2010-03-12, 21:34   #246
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

23×5×263 Posts
Default

OK, thanks Max. That makes sense to me now about verifying all non-top-5000 primes. Actually, I think I did that in the past for k<2000; at least for n<200K. But I didn't do it for k=2000-3000 for n=50K-250K. Good idea Karsten.

I had thought about what you guys said here a little late last night and today. I agree now. It makes sense to give it good visibility. Good thinking there. I like the idea of posting it in the news with a link to its own thread. There's enough in the "LLRnet servers for NPLB thread" already. The new "Updated/improved (or whatever) LLRnet client" thread can be used as both a question/answer thread as well as for large-scale beta testing with associated issues ironed out.

Assuming the Linux README has been tweaked as per the previous issues and Max has reviewed Karsten's documentation for grammar/English and made any appropriate changes, do it!

Exciting times lie ahead!

Karsten, once again, thank you for a great idea! Your original design really only needed a small amount of modifications from the time that you conceived of it. Max, once again, thank you for quickly converting Karsten's Windows logic into a very workable Linux client that also only needed small modifications to make it work completely correctly. The original concept and design on both sides was excellent to start with and when that happens, it usually makes for relatively easy testing and good software at the end. As many issues as we found, they were really very small ones.

This was a smashing success in just a month and now we have something that we'll be able to use for years to come!

It even deserves a few of my favorite dancing Georges:




Let's rock!


Gary

Last fiddled with by gd_barnes on 2010-03-12 at 21:36
gd_barnes is online now   Reply With Quote
Old 2010-03-12, 21:40   #247
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

55538 Posts
Default

ok then.

which title for the script-thread?

"(NPLB's version of) LLRnet with LLR V3.8"?

"The Raiders Ark is here: LLRnet supports LLR V3.8!"

"A new Dimension in Prime-Hunting: LLRnet combined with LLR V3.8"

any others?
kar_bon is offline   Reply With Quote
Old 2010-03-12, 21:41   #248
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

101001000110002 Posts
Default

Just a note for you guys:

Weekends on business trips are insanely busy for me. I'll likely only make it on to answer pressing questions in the forums and in Email and to report primes; perhaps for a half-hour at most.

I will be back late Monday but have some family matters to tend to. By Tuesday, I'll be back in full swing of the projects. I kind of had to "pay the piper" finally for spending 3-4 hours/day on the projects almost non-stop for the month before I left and this busy time is my payment. lol Unfortunately I'm not retired just yet but I know I'll stay busy when I am. :-)

Either one of you, if you can help answering any others questions or anything else adminstratively on the proejcts at this time, that would be greatly helpful.

Max, you might keep an eye on the public and private servers. I think you usually do anyway. It appears that port 6000 will need loading up before Monday.

Thanks again guys.


Gary

Last fiddled with by gd_barnes on 2010-03-12 at 21:42
gd_barnes is online now   Reply With Quote
Old 2010-03-12, 21:48   #249
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

23·5·263 Posts
Default

Quote:
Originally Posted by kar_bon View Post
ok then.

which title for the script-thread?

"(NPLB's version of) LLRnet with LLR V3.8"?

"The Raiders Ark is here: LLRnet supports LLR V3.8!"

"A new Dimension in Prime-Hunting: LLRnet combined with LLR V3.8"

any others?
I like the last one and can't think of any others off the top of my head. Max, any ideas?

One thing that I thought of: I wonder if the original design of LLRnet was the way it was for security reasons. In other words, was it "static" in its LLR version so that people could not easily "fake" primes? Well, they could but it wouldn't be as easy. So...for security reasons, I wonder if for a future release, there would be an easy way to make the Awk and Perl scripts into only an executable/binary for the client with no source code available for them? That way, it would take some serious effort for someone to reverse-engineer it back to the source code and change the code in order to fake results and/or primes or non-primes.


Gary
gd_barnes is online now   Reply With Quote
Old 2010-03-12, 21:53   #250
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

B6B16 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
I like the last one and can't think of any others off the top of my head. Max, any ideas?

One thing that I thought of: I wonder if the original design of LLRnet was the way it was for security reasons. In other words, was it "static" in its LLR version so that people could not easily "fake" primes? Well, they could but it wouldn't be as easy. So...for security reasons, I wonder if for a future release, there would be an easy way to make the Awk and Perl scripts into only an executable/binary for the client with no source code available for them? That way, it would take some serious effort for someone to reverse-engineer it back to the source code and change the code in order to fake results and/or primes or non-primes.

Gary

i've found a batch-script 'compiler' but not tested it yet (it makes in WIN-Dos a *.com from a batch file).
OTOH i even don't know why there's a llrnet.exe (with LLRV3.5 included/compiled) but there exists the *.lua files also! perhaps all could be compiled in one exe!?

so i will create that new thread the next hour (with note in News, too).
the name could be changed later if we find a better one!

i take the last links of the downloads: 2 from Max and the one from me.
could be changed later, too!

Last fiddled with by kar_bon on 2010-03-12 at 21:55
kar_bon is offline   Reply With Quote
Old 2010-03-12, 22:18   #251
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

23·5·263 Posts
Default

Quote:
Originally Posted by kar_bon View Post
i've found a batch-script 'compiler' but not tested it yet (it makes in WIN-Dos a *.com from a batch file).
OTOH i even don't know why there's a llrnet.exe (with LLRV3.5 included/compiled) but there exists the *.lua files also! perhaps all could be compiled in one exe!?

so i will create that new thread the next hour (with note in News, too).
the name could be changed later if we find a better one!

i take the last links of the downloads: 2 from Max and the one from me.
could be changed later, too!
I don't have time to check them. Has Max tweaked the Linux README and reviewed your documentation for grammar/English?
gd_barnes is online now   Reply With Quote
Old 2010-03-12, 22:19   #252
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

37×79 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
I don't have time to check them. Has Max tweaked the Linux README and reviewed your documentation for grammar/English?
i don't know, he has nothing said about that!
kar_bon is offline   Reply With Quote
Old 2010-03-12, 22:42   #253
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3·2,083 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
I don't have time to check them. Has Max tweaked the Linux README and reviewed your documentation for grammar/English?
No, sorry, I haven't done that yet. I'm really busy today and tomorrow; can it by chance wait until after then?

I don't think we need to worry about "statifying" the program to make it more secure. That's not really a particularly big deal and it defeats the whole purpose of making the application modular so that new LLR versions can be used. At any rate, if someone's really set on faking primes, they can do it with the old LLRnet client if they want. It might be a tad harder but not much so--I can think of a way right off the top of my head that would work and be reasonably easy.
mdettweiler is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Anti-poverty drug testing vs "high" tax deduction testing kladner Soap Box 3 2016-10-14 18:43
What am I testing? GARYP166 Information & Answers 9 2009-02-18 22:41
k=243 testing ?? gd_barnes Riesel Prime Search 20 2007-11-08 21:13
Testing grobie Marin's Mersenne-aries 1 2006-05-15 12:26
Speed of P-1 testing vs. Trial Factoring testing eepiccolo Math 6 2006-03-28 20:53

All times are UTC. The time now is 13:05.


Mon Oct 18 13:05:48 UTC 2021 up 87 days, 7:34, 0 users, load averages: 1.48, 1.51, 1.58

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.