🚀 go-pugleaf

Fascinating.  I am finding it very hard to locate *any* published
cross-platform benchmarks that compare 486's to Macintosh's in
meaningfully comparable applications, such as wordprocessing.  So far
I've only located two.  Does anyone know about any others?

The first one I found is in PC Magazine for May 12, 1992, which
compares a $7,000 Quadra 700 to a $9,275 Dell PowerLine 450DE
(80486-50 cpu).  The latter has a 4MB cache in its EISA controller!
(The article says the Dell is running Windows, but the article is not
at all clear whether the applications are Windows applications or DOS
applications>.

The other is in the Dec. 23/30, 1991 issue of PC Week.  (The one
mentioned in the current Mac ads in Scientific American, for
instance.) It compares a $8,500 Quadra 900 against a Compaq 486/50L of
no specified price and few other details about the configuration.
(That one at least made clear it was using MS Word for Windows.)

Again, is anyone aware of any other published benchmarks comparing
Mac's to Intel boxes?

>From the meagre bits of information in the PC Week article, I conclude
that the current mid-range Mac (the IIvx) is of roughly equal or
better price/performance than all of the current 486DX2/50's, using
the exhaustive (and very informative) benchmarks of 24 of the latter
in the Oct. PCWorld.  The price/performance ratios of the 24
platforms, not counting the PS/2, but counting the Mac, have a 2:1
range, whereas the prices have a 3:1 range. (If you plot price vs.
performance you see very little correlation.) 21 of the 486's have a
worse p-p-ratio than the Mac, and it is only 12% worse than the best.
(The only 486 in the 24 faster than the Mac was only 5% faster but 30%
more expensive; nearly half of them were more expensive than the Mac)

To make the price comparisons fair I included a "works" package
(Microsoft or Claris) and WordPerfect, since most of the PC's come
with the former & I need the latter.  But I'm not confident in the
analysis since I computed the Mac's relative performance on the
assumption that the Compaq 486/50L in PC Week has roughly the same
performance as the Compaq Deskpro 486/50M in PC World and that the
Quadra 900 & 950 have roughly the same performance.  *If* the 50L is
the same or faster than the 50M, and if the 950 is the same or faster
than the 900, then the IIvx is at least as good as I've figured;
otherwise I could be all wet. (I've seen one anecodotal report claiming
the 950 is 30% faster than the 900, and if the 486/50L is a true 50MHz
machine and not a DX2 it would very likely be faster than the 50M.  Time
for a return trip to the library.)

Ted Lee <tmplee@TIS.COM>