Re: AltiVec

Tim Dierks (timd@consensus.com)
Fri, 15 May 1998 11:41:50 -0700

• Next message: William H. Geiger III: "Re: RSA's SecurPC not-so-"Secur""
• Previous message: Bruce Schneier: "Re: RSA's SecurPC not-so-"Secur""
• In reply to: Black Unicorn: "Re: RSA's SecurPC not-so-"Secur""
• Next in thread: Eric Young: "Re: AltiVec"

At 11:35 AM -0400 5/14/98, Dan Bailey wrote:
>Does anyone know if it includes a double-length multiply?
>
>Getting 128 x 128 bit multiplications would be very nice.

No, but it does have 4-way vectorized 16 x 16 bit multiplication (I don't
know if this is a significant win).

This would be a great processor for Shamir's vectorized DES. 32 128 bit
registers.

Tim Dierks - timd@consensus.com - www.consensus.com
  Director of Engineering - Consensus Development
  Developer of SSL Plus: SSL 3.0 Integration Suite

• Next message: William H. Geiger III: "Re: RSA's SecurPC not-so-"Secur""
• Previous message: Bruce Schneier: "Re: RSA's SecurPC not-so-"Secur""
• In reply to: Black Unicorn: "Re: RSA's SecurPC not-so-"Secur""
• Next in thread: Eric Young: "Re: AltiVec"