Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Comments on P = NP Proof

0 views
Skip to first unread message

reas...@gmail.com

unread,
Feb 3, 2009, 4:48:19 PM2/3/09
to
Has anyone looked at this paper?

A Polynomial Time Algorithm for 3-SAT
http://arxiv.org/abs/cs/0701023v1

I think I read a review that tore it apart and
showed several examples of 3SAT instances
where this method failed. But, I can't seem
to find the article I read.

Also, I am looking for relatively small
unsolvable 3SAT instances. About 20
clauses would be great. Does anyone
have some test instances handy?
I have looked at SATLIB, but the
smallest ones they have are over 90 clauses.


Russell
- 2 many 2 count

Gus Gassmann

unread,
Feb 3, 2009, 6:55:30 PM2/3/09
to
On Feb 3, 5:48 pm, reaste...@gmail.com wrote:
> Has anyone looked at this paper?
>
> A Polynomial Time Algorithm for 3-SAThttp://arxiv.org/abs/cs/0701023v1

>
> I think I read a review that tore it apart and
> showed several examples of 3SAT instances
> where this method failed. But, I can't seem
> to find the article I read.

I just did a quick google search on "Sergey gubin" and found numerous
hits. This one is near the top:
http://adsabs.harvard.edu/abs/2008arXiv0804.2699C

vet...@gmail.com

unread,
Feb 9, 2009, 3:36:02 PM2/9/09
to
On Feb 3, 3:55 pm, Gus Gassmann <horand.gassm...@googlemail.com>
wrote:

> On Feb 3, 5:48 pm, reaste...@gmail.com wrote:
>
> > Has anyone looked at this paper?
>
> > A Polynomial TimeAlgorithmfor 3-SAThttp://arxiv.org/abs/cs/0701023v1

>
> > I think I read a review that tore it apart and
> > showed several examples of3SATinstances
> > where this method failed. But, I can't seem
> > to find the article I read.
>
> I just did a quick google search on "Sergey gubin" and found numerous
> hits. This one is near the top:http://adsabs.harvard.edu/abs/2008arXiv0804.2699C
>
>
>
> > Also, I am looking for relatively small
> > unsolvable3SATinstances. About 20

> > clauses would be great. Does anyone
> > have some test instances handy?
> > I have looked at SATLIB, but the
> > smallest ones they have are over 90 clauses.
>
> > Russell
> > - 2 many 2 count- Hide quoted text -
>
> - Show quoted text -


At www.timescube.com, there is a demo of the algorithm (http://
arxiv.org/abs/cs/0701023v4) .

I tryed it with 30 clauses - it works. I would recomend to use fast
browsers like Chrome, Opera, Safary, or FireFox.

0 new messages