[latency vs throughput]

Exactly, I was just writing the same thing.  You have to compare both
time to complete, and how long the unit blocks before it can issue
another insn.

>In my Usenet archives, I found the following table of fixed point multiply
>latencies and throughputs, posted on comp.arch in 1996 by
John Carr <jfc@mit.edu>:

[snip]

>In my opinion, this _is_ bad.  Particularly the Alpha.    They have
>improved since then, but then so has floating point performance.

Huh?  21064s are pretty old, but. even so, the raw clockrate was
higher than many other CPus of the same vintage. Overall, FP
performance wasn't so bad.