At 05:48 18-02-2000 -0800, Markus Kuhn wrote:
>"G. Adam Stanislav" wrote on 2000-02-18 12:31 UTC:
>> Because such conversions take time, no matter how short,
>No, it does not. The Intel BSWAP command is executed in *zero* time on
>all Pentiums. It sets just a flag for the ALU and then drops out of the
>integer pipeline. Try it before you claim otherwise.
That's not the point, Markus. The point is that standards are not to be
rigid. If someone wishes to design and use a protocol that is explicitly
little-endian, he should be free to do so. Standards are here to make our
lives easier, not to fit everyone into a rigid mold.
BTW, I wish they had added the BSWAP to the 386, but it only appeared with
the 486, which limits its usefulness. Personally, I always write software
to run on the 386 because the 486 and Pentium did not make any
revolutionary changes that would justify cutting 386 users off.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:59 EDT