HDTune tests a similar level of features that HDTach does, but with a slightly different access pattern and thus can provide us with an additional set of benchmark numbers to compare between storage configurations. Here we can get the minimum, maximum and average transfer rates as well as the burst rates and access times. CPU utilization has proven negligible with modern processing horsepower, and is no longer included.
Bursts are similar, but HDTune favors the VelociRaptor.
HDTune shows minimum speeds falling just behind the VelociRaptor (but still very good).
The 3.5″ models have larger swept platter area, meaning a sharper speed drop at the end.
HDTune shows the 2 newcomers closing in on the VRaptor – or are they???
It is becoming clear that HDTune is getting long in the tooth. HDTune’s random access test appears to use 32 bit signed integers, meaning it only accesses the first 2^31 sectors when performing its test. This means it treats a 2 TB drive as a 1024 GB drive. Since the head pack only has to move roughly half the distance, access times are artificially inflated for drives greater than 2 TB. The effect is similar to what happened when TechwareLabs artificially shrunk a large drive to VelociRaptor size to boost its effective agility, except in this case the benchmark does not reflect actual drive capacity.
HDTune’s random access pattern reveals its cutoff at 1024 GB.
I submitted this as a potential bug to the HDTune authors after my RE4-GP review (4 months ago), but it appears no action has been taken, and they never responded to my submission. Forget 64 bit, HDTune is not even truly 32 bit.