sci.crypt
  Home FAQ Contact Sign in
sci.crypt only
 
Advanced search
November 2006
motuwethfrsasuw
  12345 44
6789101112 45
13141516171819 46
20212223242526 47
27282930    48
2006
 Jan   Feb   Mar   Apr 
 May   Jun   Jul   Aug 
 Sep   Oct   Nov   Dec 
2008 2007 2006  
total
sci.crypt Profile…
RELATED GROUPS

POPULAR GROUPS

 Up
  Re: Comaprison between MD5 and SHA         


Author: Luc The Perverse
Date: Nov 26, 2006 01:45

"Tom St Denis" gmail.com> wrote in message
news:1164491026.704166.143020@45g2000cws.googlegroups.com...
> Felix Rawlings wrote:
>>> Um? No. You can't easily multi-thread MD5 and expect performance
>>> [especially since it's basically 64 rounds of very serial operations].
>>
>> So it would yield the same performance on a single CPU machine?
>
> I don't see why not. Discounting any other processes running that is.
>
> What makes you think you can multi-thread MD5?

There is probably someway to take advantage of another core - but I doubt it
would be trivial. Perhaps I am completely wrong - I don't know enough about
MD5

Look at the big money Intel is putting into R&D and competitions to make
seemingly iterative processes run faster on multiple cores.

--
LTP

:)
no comments
  Re: L15 prize increased to         


Author: Greg Rose
Date: Nov 25, 2006 12:28

In article news1.ucsd.edu>,
Greg Rose network.ucsd.edu> wrote:
>(I'm waiting for the NIST test on a single 125MB
>sample to complete. I'll post that result, no
>matter what it says, as a follow-up to this
>message.)

I had a lot of trouble running the NIST tests on
L15, as NIST no longer supports it for Linux (only
Windows, and with a clunky user interface). But
anyway.

NIST normally recommends running the test on 125MB
of data, treated as 1000 samples of 1000000 bits
each. When run in that mode against a single
125mB stream from L15, it reported a failure of
the overlapping-templates test. Unfortunately I
clobbered this data set output and it takes so
long to re-run the test that I won't bother.
Show full article (22.06Kb)
no comments
  Re: Interesting paper: On the Power of Simple Branch Prediction Analysis         


Author: Jan Panteltje
Date: Nov 25, 2006 06:48

On a sunny day (25 Nov 2006 04:58:25 GMT) it happened Unruh
wrote in :
>It is a bit obscure. I think that the reason you missed the first
>discussion was that it was posted by Tom.

Yes I think so too.
>>But now I was sort of disappointed the discussion was not about that
>>paper but again about Tom.
>
>Sorry about that.

No need to apologise, I enjoy reading the contribution by you and many others.

Tom often replied in the style of:
'That is crap', 'You have no clue', 'This subject has been discussed to death',
the a bit later some other poster points out he is wrong with good arguments,
then a fight happens.... and he often loses.
So that makes ignoring his initial replies / attacks a good policy.
Sorry if it made me drop interesting replies by any of you., it goes then something like this:
Show full article (1.52Kb)
no comments
  sorry guys         


Author: mantinin
Date: Nov 25, 2006 01:39

"The Spieshttp://sonic.net/~doretk/Issues/99-09%%20FALL/theexpert.html
Who Shagged Us" is a book written this creep who thinks he knows
something about the CIA-perhaps his own drug failures are haunting him,
and I know them WELL

Leo Anthony Sgouros
no comments
  GLOBAL STRATEGIC SYSTEMS NEWSLETTER, January, 1997, by Markku J. Saarelainen, SOME CONCEPTS: Organizational Improvement Network (OIN) for Improving Competitiveness, Environmental Performance and Profitability         


Author: The Brigadier
Date: Nov 25, 2006 00:33

GLOBAL STRATEGIC SYSTEMS NEWSLETTER, January, 1997, by Markku J.
Saarelainen, SOME CONCEPTS: Organizational Improvement Network (OIN)
for Improving Competitiveness, Environmental Performance and
Profitability

Copyright 1997 Markku J. Saarelainen

GLOBAL STRATEGIC SYSTEMS NEWSLETTER

January, 1997

by

Markku J. Saarelainen

SOME CONCEPTS: Organizational Improvement Network (OIN) for Improving
Competitiveness, Environmental Performance and Profitability

We all know that it is important to have reliable, accurate and timely
information for decision making purposes to ensure adequate quality in
our decision making. However, one of the most excellent questions...
Show full article (6.13Kb)
1 Comment
  Re: Interesting paper: On the Power of Simple Branch Prediction Analysis         


Author: Unruh
Date: Nov 24, 2006 20:58

panteltje@yahoo.com writes:
>Unruh schreef:
>> Again, you take on the mantle of net cop with allacrity.
>> If there really were net cops you would find yourself with tickets far more
>> than most posters. Another news rule is that you remain polite with people
>> and that if someone does something you do not like, you exhibit tolerance
>> of that. Furtehrmore, the newer the poster, the more tolerance you exhibit. But that
>> is a rule you do not like, so you ignore it. With your example, why should
>> anyone follow any rules at all, since your key rule seems to be that your
>> own personal preferences take precedent over anything else.
>OK I will react.
>First I read this group every day, but usually first the 'headers'
>(subjects),
>and if I think the subject is interesting _to_me_ I will follow it.
>Because of this, you sometimes mis discussions about things.
>I just checked with google (posting from that now) and could not find a
>reference to that
>paper here.
Show full article (1.92Kb)
no comments
  DES: left "circular shift" of key bits         


Author: Mike McNally
Date: Nov 24, 2006 19:33

Hello all,

In conjunction to reading Applied Cryptography 2nd Ed., I'm also using Cryptool as a
learning aid. In learning modern encryption algorithms I figured I'd start with the
DES, my logic being that it will give me a good foundation for learning the rest of
the algorithms.

However, I do have a question about the "circular shift" of the key after it has been
reduced to 48 bits and then split into two 28 bit halves.

Cryptool has some animated illustrations of some of the classic ciphers, and also for
DES.

This is what it shows for the first 'R' and 'L' half of the 28 bits before and after
the left circular shift, on the first round of the 16, which would make it a 1 bit
shift.

(I have aligned the shift for simplicity)

For the 'R' half:

1 0 1 1 1 1 1
1 0 0 1 1 0 0
0 1 1 1 1 0 0
1 1 1 1 0 0 0
Show full article (2.28Kb)
4 Comments
  OT. Before it gets to Christmas time... My melancholic look back at the last year         


Author: David Eather
Date: Nov 24, 2006 12:10

In many ways I still see myself as an outsider so I see things perhaps
in a slightly different way.

I would like to thank sci.crypt for an educational year over a goodly
range of topics. It is an amazing gift to the public that professors,
doctors and genuine experts give their time to deal with the "small
matters" raised by non-experts.

I also thank Rodger Schlafly who puts out his crypto mini-faq every
month but rarely gets thanked for his efforts. And JS, DG, BU and JP
(who seems to be no more) - even if I disagreed with you in an OT
thread, I respected your intellect and you did change me.

Merry Christmas Sci.crypt and best wishes for the new year
1 Comment
  OT. Before it gets to Christmas time... My melancholic look back at the last year         


Author: David Eather
Date: Nov 24, 2006 12:09

In many ways I still see myself as an outsider so I see things perhaps
in a slightly different way.

I would like to thank sci.crypt for an educational year over a goodly
range of topics. It is an amazing gift to the public that professors,
doctors and genuine experts give their time to deal with the "small
matters" raised by non-experts.

I also thank Rodger Schlafly who puts out his crypto mini-faq every
monthly but rarely gets thanked for his efforts. And JS, DG, BU and JP
(who seems to be no more) - even if I disagreed with you in an OT
thread, I respected your intellect.

Merry Christmas Sci.crypt and best wishes for the new year
2 Comments
  Re: Interesting paper: On the Power of Simple Branch Prediction Analysis         


Author: Unruh
Date: Nov 24, 2006 10:27

"Tom St Denis" gmail.com> writes:
>Unruh wrote:
>>>Not *trying* to be a netcop but if you guys aren't even going to read
>>
>> No, you do it with ease, you do not have to try.
>Well it's no secret that you support re-news.
>If we let everyone repost shit on a weekly basis what would be the
>point of archiving any of this?

A lot more than the point of archiving this whole discussion, or most of
the discussions on netnews.
>It's a simple trick. Read the group for a while, and *then* start
>posting. It'd be different if they were months apart but ...
Show full article (1.23Kb)
1 Comment
1 2 3 4 5 6 7 8 9