Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 20 May 2011 08:51:21 +0400
From: Solar Designer <solar@...nwall.com>
To: crypt-dev@...ts.openwall.com
Subject: Re: alternative approach

David, Yuri -

On Fri, May 20, 2011 at 08:34:50AM +0400, Solar Designer wrote:
> Also, it appears that you synthesized for the smallest Virtex-6 device,
> with 11,640 slices.  The largest described in that overview is 6.4 times
> larger.  If my math is right, assuming the 66x figure from my previous
> e-mail, it could give us a 420x advantage over a quad-core CPU, or as
> much as 2500x for Pico's 6-chip board.

Oh, I was wrong about the largest.  I looked at the last line in the table,
but that's not the largest device.  The largest in terms of slice count
(XC6VLX760) is 10x larger than the smallest, but it is not the largest
in terms of Block RAMs and DSPs.  Also, it lacks PCIe interface blocks,
which could be relevant to us.

David - which specific Virtex-6 devices are we going to use in production?

Alexander

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.