Yes I know I'm replying to what is arguably spam.

"Joe Hanston" gmail.com> wrote in message
news:el0f8j$kqe$20@xstation.4096.net...

[evidence the gene pool has a shallow end]

You need to get out more. You also need to learn how to spell "Hilarious"
how to use an apostrophe.

And please don't drink and park, accidents creates more morons. Let
evolution take it's course and eliminate your kind.
Joe

Peter Fairbrother wrote:

That's not always true. For the same size as an Opteron you can
totally blast it out of the water for performance. The problem [as you
mentioned] is the cost. Nobody is making volume [at least general
purpose] chips that do modexp, if they made them in the volumes GPUs
were being made they'd cost ~$300.

That said, on the lower end things are better. You can totally defeat
a PPC, ARM, or MIPS with a relatively small [re: fit in a decent mid to
high range FPGA] design. A good solution if you can't afford tapeout
:-)

BRG wrote:

I have been looking at the various suggestions made (thanks!), and I am not
impressed by any crypto hardware devices, the general-purpose microprocessor
seems to give much better value. On all scales it seem that throwing an
extra processor or twenty in is much better value than adding
specifically-crypto hardware.

I suppose that's not too surprising, as the general microprocessor is one of
the most highly-developed devices ever, and it is produced in numbers which
dwarf any crypto chip.

So it's a Tyan Typhoon or Nexxus 4000 (or a SiCortex SC5832, 5,832 64-bit
MIPS cores running on 18 kW, yum :) for me. I wish.

zzz@att.net writes:

Call for Papers
and
Call for Session Proposals

The 2007 World Congress in Computer Science,
Computer Engineering, and Applied Computing
WORLDCOMP'07
(composed of 24 Joint Conferences)
June 25-28, 2007, Las Vegas, USA

Dear Colleagues:

You are invited to submit a draft paper and/or a proposal to organize
a session/workshop. All accepted papers will be published in the
respective conference proceedings. The Academic Co-sponsors of
WORLDCOMP'07 include: MIT Media Lab, MIT; Harvard University's
Statistical Genomics and Computational Biology Lab; Texas Advanced
Computing Center, The University of Texas at Austin; Statistical and
Computational Intelligence Lab of Purdue University; and University of
Iowa's Medical Imaging HPC Lab. A more complete list of sponsors can
be found below.

Hello out there

can anyone tell me if there exists something similar to
RSACryptoServiceProvider in C# that i can use to encode/decode messages
but
just the contrary way?
I mean having a private/public key on some app and use only the public
key on another app to decrypt the messages and the private+public key
on the "mastermachine".

All i want to do is shipping some data to some customers (via email
eg.) but have the data crypted.

I use c# in VS2005 with DotNet 2.0

thanks in advance

I was just wondering how Cryptography stands in the UK with relation to
current laws or anything...

More specifically, lately i've been working on my own algorithm from
scratch and i've been enjoying it immensely. Would it be legal for me
to post about it, how it works, maybe offer up some free C++ source
code for people to check it out?

Or would it be a particularly bad idea?

To the troll who is joe-jobbing me,

I'm not going to stop working on my projects, or posting interesting
results here just because you feel the need to post your pornographic
filth everywhere. I work on my projects because *I WANT TO* work on my
projects. I'm intellectually curious and improving my projects is fun.
Even if one day, I had zero users because of all of the trolling in
the world, I'd still work on the projects.

Not only does a google search for libtomcrypt go to my real domain

http://www.google.ca/search?hl=en&q=libtomcrypt&btnG=Google+Search&meta=

But nobody in the cryptographic community is stupid enough to believe
that your joe-jobs are really associated with me. I can easily filter
out the replies I get from random usenet groups. So all you're doing
is annoying OTHER PEOPLE.

Give it a rest already.

I just added Shamir's trick to the LTC library. Gets ~1.34x to ~1.4x
faster signature verifications than before with a simple 16 element
table of points.

Some numbers

---NOW---
ECC-112 verify_hash took 807645 cycles (1.34x faster)
ECC-128 verify_hash took 969613 cycles (1.40x faster)
ECC-160 verify_hash took 1391150 cycles
ECC-192 verify_hash...