Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120503133432.GA25957@openwall.com>
Date: Thu, 3 May 2012 17:34:32 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: SM in README-CUDA

Lukas, myrice, magnum -

On Thu, May 03, 2012 at 03:31:15PM +0200, magnum wrote:
> On 05/03/2012 03:23 PM, myrice wrote:
> > I add XSHA512 format in Performance issues section in README-CUDA. I
> > follow your format. I want to know what does SM mean? I think GTX 460
> > has 7 SMs. But there is 20.
> 
> It refers to the architecture abbreviation: sm_10, sm_20 and so on.
> sm_20 is Fermi.

If myrice was confused by this, then surely our users would as well.
I think we should not be abbreviating things, and we should describe
what these are and where to edit them.

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.