[RC5] AMD X86-64 port

gentryj at gci.net gentryj at gci.net
Mon Apr 28 08:08:45 EDT 2003


As I understand it from having examined the RC5 algorithm...  the actual 
"crunching" is hardcoded to use 32 bit rotates.  On native 64 bit chips this is 
bad because, as someone already pointed out, it must call extra opcodes to 
handle the 32 bit operation.  However, AMD's 64 bit chip doesn't appear to be 
native 64....  (I'm still slightly confused about how they are claiming a 64 
bit CPU can actually perform better at 32 bit applications?!?).

Anyone had a chance to run one of these Hammers yet?  IMHO, it could really go 
either way.  Depends on if AMD is telling the truth.

James / Warbird
Club Overclocker

> On Sun, 2003-04-27 at 14:13, Karl wrote:
> > > One of the core guys (Acidblood) is raving about the new chip.  I'm
> > > pretty sure he thinks it's the best thing since sliced bread.  So
> > > there's definitely interest internally.
> > > 
> > 
> > I'm most interested to see what those extra registers in 64-bit mode can do
> > for RC5-72 performance.
> 
> Acidblood has said that the simple fact of being 64-bit won't help, and
> I think he even said it would harm it.  The operations of rc5 are
> particularly friendly to 32bit processors, I believe he mentioned the
> modulo operations work really well on 32bit processors.  But of course,
> if the new Opteron is overall a faster processor, it should surely go
> faster.  The fact that it runs 32bit code pretty well should be
> interesting too.
> 
> --
> Scott
> sdodson at distributed.net
> sdodson at eagle4.cc.gasou.edu
> --
> To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
> rc5-digest subscribers replace rc5 with rc5-digest
> 


--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list