R E V I E W // C O R S A I R D O M I N A T O R P L A T I N U M C O N T R A S T S P E C I A L E D I T I O N
insist on not 'risking' their DRAM .
So the special edition contrast,
it seems the memory can match
just about any and all similarly
rated kits, but of course with a
minor difference. That being the
ability to go the distance and
reach the previously stated
4,500MT/s without much fuss.
Granted this is largely board
dependent, but if you have any of
the high-end boards (which is a
plausible assumption, given the
type of person that considers
spending $479 on a 32GB kit) this
should be possible especially with
the ROG or other two DIMM
motherboards.
It’s because of this massive
headroom that the testing of this
kit was done at frequencies above
4000MT/s with the XMP profile
only serving as a reference. It’s
almost criminal to operate this kit
at the rated 3,466 MT/s.
That said, if you look at the
performance figures, what you will
notice is that
24 The OverClocker Issue 45 | 2018
the gaming benchmarks for the
most part are academic. With a
sufficiently fast enough CPU the
impact of DRAM frequency and
timings is diminished and the
test configuration made this very
evident. This is not to say that
overclocking the memory isn’t
worth it, in fact I for one always do
it. However, not everyone is going
to have a 5.2GHz gaming capable
CPU which doesn't need copious
amounts of voltage, and this CPU
in particular could do the
frequency at a relatively low 1.3v. It
would have been easier or simpler
to show the impact of performance
using the default clock on the
8086K, but right now with the 9th
generation core CPUs all capable
of similar frequencies, this is a
very likely scenario (5GHz+ capable
CPUs) and the testing/results
reflect that.
In terms of synthetic
benchmarks, at least the ones
used for competitive overclocking,
the performance differences are
a lot more obvious, especially in
the memory intensive benchmarks.
This is actually relevant to those
who use productivity applications
as well which are dependant on
memory latencies, bandwidth etc.
The ability to render or process for
instance 0.3fps more with the
overclock over the stretch of
several hours adds up. For
instance, going from the X.M.P
profile to the overclocked 4266MT/
s CL17 settings boosts the
rendered frame rate by 0.35fps
(HWBot X.265 4K render), which
over the course of just one hour
translates into an additional 1,260
rendered frames or roughly 52
seconds of video (at 24fps). This is
significant and even though it’s
most likely an edge case, if that
rendering time has a cost attached
to it, then such memory, with such
overclocking leg room makes
complete sense.
Even if this isn't relevant to you,
the performance in and of itself is
simply just cool to have.