6. Summary
The main contribution of this paper is to point out that
in our quest for scientific innovation we sometimes misuse
our tools, this misuse can lead to speculative and misleading
results. The SPEC CINT2000 benchmark suite is widely
used in computer architecture conferences, unfortunately it
isn’t used to its full potential. Two years after its inception
only 20% of the published papers that reference the suite in
the ISCA, Micro, and HPCA conferences fully use it. This
behavior mimics the use of its predecessor, CINT95. By
the year it was retired only 50% of papers that used it did so
fully. Projecting the use of CINT2000 into the future shows
that by 2005 we can expect 80% of the papers to fully use
the suite. However, this is a year after researchers should be