From: jim bell <jimbell@pacifier.com>
Date: Thu, 26 Sep 1996 15:45:25 +0800
To: cypherpunks@toad.com
Subject: Re: Mitsubishi MISTY LSI
Message-ID: <199609260534.WAA00788@mail.pacifier.com>
MIME-Version: 1.0
Content-Type: text/plain


At 08:55 PM 9/25/96 -0700, Bill Frantz wrote:
>>   Nikkei English News, 24 September 1996. 
>> 
>>   [Mitsubishi] fabricated the chip as a gate array using a 
>>   0.5 micron CMOS process and its own proprietary 
>>   encryption algorithm MISTY. Operating at a maximum 
>>   input/output speed of 40 megahertz, the chip can handle 
>>   32 bits per clock cycle. 
>
>Does anyone have a reason not to consider this algorithm snake oil?  e.g.:
>Was it developed by a well known cryptographer?
>Has it been vetted by someone/some organization with reputation?
>I other words, why should we trust it?


There's not nearly enough information provided to know for sure.  
However, I'd think that any company which went to the trouble to build 
such a chip (custom, 0.5 micron process, etc) would go to the little extra 
effort to verify the algorithm is secure.

Generally, hardware is held to a higher standard because the difficulty of 
repair is higher:  Usually, you have to replace a chip.

Jim Bell
jimbell@pacifier.com