Redditor / u / Amaroko put this WD Blue drive on a blank cardboard box and held a Blue Yeti microphone right above it while it was on. Spectral analysis of the audio shows a 90 Hz baseline – 90 cycles (revolutions) per second and 60 seconds per minute equals 5,400 rpm.
This WD Elements 8TB drive is listed as "5400 RPM Class" on its datasheet, and its firmware reports 5400 RPM via SMART – but it hums at 120 Hz. = 120 Hz * 60 s / min 7,200 rpm.
This WD MyBook 8TB drive also reports spinning at 5,400 RPM via SMART, but spectral audio analysis shows it is rotating at 120 Hz (7200 RPM).
Last week the good Redditors from / r / DataHoarder were once again upset about Western Digital – this time because of incorrectly displaying the speed of their WD Red Network Attached Storage hard drives. (Although the linked post turns things upside down, members of the German-language forum hardwareluxx.de began investigating the problem more than a year ago.)
We found this controversy which is reminiscent of previous complaints that Western Digital did not properly disclose the use of Shingled Magnetic Recording technology in their NAS drives. The new complaint, however, is that Western Digital is referring to 7200 RPM drives as the "5400 RPM class" – and its proprietary firmware report 5400 RPM over the SMART interface.
Recently, redditor / u / Amaroko set out to prove or disprove the results of previous internet users. For each of several drive models, Amoroko placed a sample of this drive on an empty cardboard box with a blue Yeti microphone above it, and switched the drive on. Spectral analysis of the recorded audio with Adobe Audition resulted in a base frequency of 120 Hz for two models of a WD 8 TB drive of the "5400 rpm" class.
120 cycles / s multiplied by 60 s / min result in 7,200 cycles / min. In other words, these 5400 RPM drives were really spinning at 7,200 RPM.
Who doesn't want a faster speed?
Enlarge /. When we compare datasheets between an 8 TB 5400 RPM Barracuda and an 8 TB "5400 RPM Class" Red, we see a huge difference in power consumption.
At first glance this doesn't seem to be a problem – who wouldn't prefer a drive with a faster spindle speed? Unfortunately, faster spindles not only mean a potentially lower search latency, but also lead to a sharp increase in both noise and power consumption.
This increase in noise and performance has put many users on the trail of Western Digital's fake spindle speed of 5,400 RPM in the first place – those users bought drives that they expected to roll slowly and slowly, but they have more Noise, heat and power consumption than expected.
When comparing the data sheets between a Seagate Barracuda with 5,400 rpm and a Western Digital Red of the "5400 rpm" class, which according to the audio spectrum analysis actually rotates at 7,200 rpm, we can see a clear difference in power consumption. The 8 TB Red consumes 8.8 W for the Barracuda's 5.3 W when it is active and 800 mW for the Barracuda's 250 mW when it is in standby mode. The difference here, for the most part, isn't in brand – when spinning at the same actual speeds, Western Digital and Seagate drives consume roughly the same amount of power.
To be fair to Western Digital, you can't buy an 8 TB NAS drive that spins at 5,400 rpm at all. Both the Red and Ironwolf entry-level NAS ranges from WD Red and Seagate use 7,200 RPM spindles. To suit the consuming public, "RPM" has a very specific meaning, and there isn't even a footnote in the WD Red datasheet to explain what – if anything – "5400 RPM Performance Class" actually means.
Response from Western Digital
When we reached out to Western Digital in the course of researching this story, a representative confirmed the conclusions of various forum visitors and Redditors – that is, "5400 RPM class" does not mean a drive is spinning at 5,400 RPM.
For many years, Western Digital has published the speed for selected products within a “class” or “performance class” instead of publishing specific spindle speeds. We also optimize selected hard drive platforms and the associated hard drive properties to create different variations of such platforms to meet different market or application needs. In this way we can use our economies of scale and pass these savings on to our customers. As with any Western Digital product, our product details, including performance, acoustics, and performance (data rate), are tested to meet the specifications listed on the product data sheet and marketing materials.
In our editorial opinion, this answer, however polished, is not helpful – nor is the as-yet undefined "RPM class" itself not helpful. A consumer reading this data sheet will be more confused, not less, by the marketing fuzz. And the "typical consumers" Western Digital may be trying to protect themselves from information they don't understand are unlikely to read hard drive datasheets at all.
If you'll pardon an automotive analogy, we don't notice it as much as a window sticker on a V6 sedan declaring it "four-cylinder class" without specifying what is actually under the hood.
Listing image by Brian Wong / Flickr