Re: SUN vs Lisp machine hardware differences?

["David A. Moon" <Moon@scrc-stony-brook.arpa>]

    From: larus%paris.Berkeley.EDU@berkeley.edu (James Larus)
    Date: Tue, 11 Aug 87 13:56:29 PDT

    If Symbolics has evaluated the tradeoffs and really believes that this
    check is worth the costs, then they should publish a paper and
    convince the rest of the world.

In this connection you might note that SUN's latest machine contains
some gestures in this direction (not for forwarding pointers, though).
Among other things this shows the essential meaninglessness of the CISC
vs RISC debate.  Of course you should put into hardware only what you
think is important, and not expend hardware optimizing things that don't
need to be optimized.  At the next level of detail, reasonable people
can disagree on what is important and what should be optimized.  A
machine that is labelled a RISC can have parallel tag checking, just as
a machine that is labelled a CISC can execute one instruction per cycle.
It's just name-calling.

It isn't necessary to reply to this message.