19 Apr
2002
19 Apr
'02
8:09 a.m.
> His next machine, the Z4, although in size,
weight and power
> like the Z3 is already 30 times faster, barely gets around a
> Kilozuse ...
I think your formula needs some serious tinkering. I
think the lowest
number should reflect the most classic machine.
Naaa - never. Usualy one assumes 'better' with higer numbers,
and I think we agree that more classic is better.
The Zuse 1 should come out to 1 Zuse as others have
suggested. So by
this measure, the Babbage Difference Engine should come out at something a
bit less than 1 Zuse.
Naaa. The real relative value has to be determinated by the formula.
And tweeking the formula to give a 1 to a specific combination is
against the basic idea of nature defined values.
The Pentium III with 256MB and 10GB hard disk sitting
on my desk should be
something in the TerraZuse range (or much higher).
So this stupid Pentium is a 10^12 times more geeky then a Zuse ?
Come on. This is the geeknes factor. If the pentium gets a nanozuse
it's already way overrated.
We should also factor in user I/O. Machines that used
front panels and
blinkenlights should have much lower (i.e. more classic) scores than those
that have keyboards and 16MB bit-mapped displays with 24 million colors
and what not.
Machines with blikenlights are bigger and more heavy than modern
workstations, so you got this included. If we use to arbitrarily
citeria (like Display, MS ability, etc. pp. it looks less 'serious'.
Also someone suggested to include things like bus wide, data pathes
etc., but these are again measurements only true for some specific
architectures. Also, htey realy don't metter for the Boah Ey (*)
factor.
Gruss
H.
(*) Boah Ey, sound of an educated person when confrontated with
old hardware...
--
VCF Europa 3.0 am 27./28. April 2002 in Muenchen
http://www.vcfe.org/