| Author |
Message |
|
squalyl
Advanced Newbie
Joined: 04 Aug 2009
Posts: 57
|
 Posted: 26 Aug 2009 12:09:13 pm Post subject: |
|
|
I think you have to read the whole topic to understand how to use this project :D
BOINC only intervenes in the 2nd step of the 3-steps process.
You have to run the first pass first, to find a good polynomial.
BTW how much bits are there in 149 decimal digits?
We have some calculators related keys to try before managing requests such this one :)
But why not, in the near future  |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 26 Aug 2009 01:22:06 pm Post subject: |
|
|
Lionel: the 0104 key was released by TI some years ago.
yoyo: You have to be careful what you mean by "any number." For a lot of numbers, the number field sieve is not the best idea - for small numbers, you will find that other algorithms are faster. For really huge numbers, the number field sieve becomes impractical (but so do all other general-purpose factoring algorithms.)
The largest numbers that have been factored, I believe, have been done using the elliptic curve method (which is mostly dependent on the size of the factors, rather than on the size of the number you're factoring.)
Factoring a 149-digit number is clearly doable with these tools. I don't know how much larger you can go with GGNFS. |
|
| Back to top |
|
|
RvW
Newbie
Joined: 26 Aug 2009
Posts: 9
|
| Posted: 26 Aug 2009 03:17:22 pm Post subject: |
|
|
yoyo wrote: It would mean, that you are able to factor e.g. this 149 digit number?
23373956054212729369314132440470881224763236407256033079439078784258944564500359
718028342226892427079860170972262736474225014034561266391630623618447
Can we test it?
yoyo
Do you mind if I ask where this number comes from? |
|
| Back to top |
|
|
yoyo
Newbie
Joined: 24 Aug 2009
Posts: 3
|
| Posted: 26 Aug 2009 03:25:42 pm Post subject: |
|
|
RvW wrote: yoyo wrote: It would mean, that you are able to factor e.g. this 149 digit number?
23373956054212729369314132440470881224763236407256033079439078784258944564500359
718028342226892427079860170972262736474225014034561266391630623618447
Can we test it?
yoyo
Do you mind if I ask where this number comes from?
It was just an example, which I took from the XYYXF project and for which I found a 42 digit factor with the elliptic curve method. But there remains a composite cofactor which needs to be factored.
yoyo |
|
| Back to top |
|
|
joeyadams
Newbie
Joined: 06 Aug 2009
Posts: 4
|
| Posted: 26 Aug 2009 04:10:58 pm Post subject: |
|
|
Where can I find the source for the modified GGNFS binary used with BOINC? It'd be nice to have it posted at http://boinc.unsads.com/rsals/ (not to mention GPL compliant).
By the way, great job orchestrating all this! |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
|
| Back to top |
|
|
joeyadams
Newbie
Joined: 06 Aug 2009
Posts: 4
|
| Posted: 26 Aug 2009 09:52:00 pm Post subject: |
|
|
Thanks. I compiled it for Fedora 10 ppc_7450 after a few tweaks to the build files. However, I'll need to test it and compile it correctly before we can support it. I am no stranger to endianness issues.
I imagine I could test this build by running it on a job, running the Intel build on my Intel machine, and comparing the results to make sure they're the same. How would I do this? |
|
| Back to top |
|
|
Lionel Debroux
Member
Joined: 01 Aug 2009
Posts: 170
|
| Posted: 27 Aug 2009 01:37:18 am Post subject: |
|
|
FloppusMaximus: OK for the state of the 0104 key.
(TI making the community's task easier, that's a first...)
Someone asked to censor the first post of this topic, it seems... that's not good ! |
|
| Back to top |
|
|
squalyl
Advanced Newbie
Joined: 04 Aug 2009
Posts: 57
|
| Posted: 27 Aug 2009 02:20:00 am Post subject: |
|
|
if any action has to be done privately my email address is on the project home page.
new keys added. |
|
| Back to top |
|
|
squalyl
Advanced Newbie
Joined: 04 Aug 2009
Posts: 57
|
| Posted: 27 Aug 2009 04:50:42 am Post subject: |
|
|
Do anyone still need the ftp access?
if no one screams, I'll shut it down friday ~14H GMT +1/2 (don't known, with DST)
even with vsftpd, that's a potential security hole I don't like to keep for long.
Joey: I claim no rights on the modifications made to the ggnfs lattice siever, they are so simple and yet incomplete. consider them public domain or whatever you need to do what you want.
To get your test done, run the test program in an empty folder containing a file named "input_data" with a job inside.
There are some of them in the svn repo.
just replace this:
q0: __START__
qintsize: __COUNT__
by this
q0: 10000000
qintsize: 1000
1000 values will take a few minutes to compute.
run the same job on ppc and x86 (./theprogram, it will run fine in standalone mode), then diff the ouput_data file. They should be the same.
When done, send me the ppc binary, or give me a link to it. i need it to be statically linked so it will work everywhere. The makefile should take care of this.
Last edited by Guest on 27 Aug 2009 04:59:37 am; edited 1 time in total |
|
| Back to top |
|
|
RvW
Newbie
Joined: 26 Aug 2009
Posts: 9
|
| Posted: 27 Aug 2009 07:41:34 am Post subject: |
|
|
FloppusMaximus wrote: Dear community,
I have been politely asked to remove the former contents of this post.
1) TI still cares about these keys? That's unexpected... What consequences does this have for the rest of the project?
2) The cat is out of the bag; there's no way all copies can be taken offline, that's just not how the Internet works. |
|
| Back to top |
|
|
Synthesis2k2
Newbie
Joined: 18 Aug 2009
Posts: 10
|
| Posted: 27 Aug 2009 07:49:50 am Post subject: |
|
|
I started running BOINC on both my systems, yesterday. The main box is a quad-core Q9550 and a P4 3.0. Glad to contribute.
-Syn  |
|
| Back to top |
|
|
squalyl
Advanced Newbie
Joined: 04 Aug 2009
Posts: 57
|
| Posted: 27 Aug 2009 08:20:38 am Post subject: |
|
|
| RvW> that sounds a bit like 09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0, huh? |
|
| Back to top |
|
|
Synthesis2k2
Newbie
Joined: 18 Aug 2009
Posts: 10
|
| Posted: 27 Aug 2009 02:14:48 pm Post subject: |
|
|
squalyl wrote: RvW> that sounds a bit like 09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0, huh?
Ah, that brings back memories of not that long ago. I remember how big of a thing it was on digg. |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 27 Aug 2009 03:57:19 pm Post subject: |
|
|
FloppusMaximus wrote: Dear community,
I have been politely asked to remove the former contents of this post.
Dear Floppy,
If you're not too depressed can you show us what a so polite request look like?
We're all with you: you'll never be forgotten in the TI community (and not only for that - all your contributions are great!)
We all love you 
Last edited by Guest on 27 Aug 2009 03:58:28 pm; edited 1 time in total |
|
| Back to top |
|
|
Graphmastur
Advanced Member
Joined: 25 Mar 2009
Posts: 360
|
| Posted: 27 Aug 2009 04:05:06 pm Post subject: |
|
|
FloppusMaximus wrote: Dear community,
I have been politely asked to remove the former contents of this post.
Two things:
Please define "politely", and can TI really make you remove that? Did we do something ileagel? |
|
| Back to top |
|
|
RvW
Newbie
Joined: 26 Aug 2009
Posts: 9
|
| Posted: 27 Aug 2009 04:11:33 pm Post subject: |
|
|
squalyl wrote: RvW> that sounds a bit like 09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0, huh?
Is that DeCSS?
Yeah, same thing basicly; any crypto system is only as good as the secrecy of its keys. Once the keys get published on the Internet you can go to court (or politely request removal) as much as you like, the protection is gone; anyone (especially the people really interested in it) will be able to obtain the keys in one way or another.
On a related note, did TI ever make any substantial amount of money through these keys / by signing apps for third party developers? |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 27 Aug 2009 04:45:59 pm Post subject: |
|
|
Graphmastur wrote: FloppusMaximus wrote: Dear community,
I have been politely asked to remove the former contents of this post.
Two things:
Please define "politely", and can TI really make you remove that? Did we do something ileagel?
I suppose "politely" means more or less:
Please remove and stay silent - in that case we won't do anything against you.
I've allready been asked something like that 1 time.
(it was not by TI) |
|
| Back to top |
|
|
joeyadams
Newbie
Joined: 06 Aug 2009
Posts: 4
|
| Posted: 27 Aug 2009 06:10:47 pm Post subject: |
|
|
squalyl wrote: Joey: I claim no rights on the modifications made to the ggnfs lattice siever, they are so simple and yet incomplete. consider them public domain or whatever you need to do what you want.
To get your test done, run the test program in an empty folder containing a file named "input_data" with a job inside.
There are some of them in the svn repo.
just replace this:
q0: __START__
qintsize: __COUNT__
by this
q0: 10000000
qintsize: 1000
1000 values will take a few minutes to compute.
run the same job on ppc and x86 (./theprogram, it will run fine in standalone mode), then diff the ouput_data file. They should be the same.
When done, send me the ppc binary, or give me a link to it. i need it to be statically linked so it will work everywhere. The makefile should take care of this.
Thanks. This simply segfaults on the PPC after running for a little while. Looks like I have some debugging to do :)
Does the binary need to be compiled statically against boinc_api, boinc, dl, and pthread as well? Also, do I need to download gmp et al. to make static builds, or can I somehow use the .so's in /usr/lib for this? |
|
| Back to top |
|
|
Synthesis2k2
Newbie
Joined: 18 Aug 2009
Posts: 10
|
| Posted: 27 Aug 2009 10:10:09 pm Post subject: |
|
|
A big FYI for whomever is running the project server, I got the following error messages, starting about an hour and a half ago (8:35pm Eastern):
[These error messages were taken from my other machine, I can't cut and paste through remote desktop]
Code:
8/27/2009 10:01:26 PM RSA Lattice Sieving [error] Error reported by file upload server: Server is out of disk space
8/27/2009 10:01:26 PM RSA Lattice Sieving Temporarily failed upload of tita_04870000_10000_20090826134035_0_0: transient upload error
8/27/2009 10:01:26 PM RSA Lattice Sieving Backing off 1 min 0 sec on upload of tita_04870000_10000_20090826134035_0_0
-Syn |
|
| Back to top |
|
|
|
Open Topic & United-TI Talk => General Open Topic
