Reading and writing files to and from a hard drive is one of the most crucial choke points in a PC. Buying a new internal hard drive can make apps load faster and shorten system bootup times, while springing for a new external drive could shave minutes off your next routine backup or the transfer of your large collection of photos and videos.
A hard drive's speed can vary significantly based not only on its theoretical top transfer speeds, but also on your usage patterns. So we've designed our drive-testing procedures at PC Labs to approximate the speeds you can expect while accomplishing a comprehensive array of common tasks: reflecting both best-case scenarios, and real-world use.
We test external drives using an Intel X299-based testbed equipped with all the trimmings, including a USB 3.1 Gen 2 USB Type-C port. We also test on a 2016 MacBook Pro, using the Type-C/Thunderbolt 3 ports, for certain of our tests.
PCMark 7 Secondary Storage Test
Our most wide-ranging benchmark is the PCMark 7 Secondary Storage test. It comes from UL (formerly, FutureMark). The company's benchmarking apps are used not only by publications like PCMag, but also by drive and PC manufacturers themselves.
The Secondary Storage test is trace-based, which means it measures how quickly software commands are performed during everyday workloads to which you might subject your PC. These simulated workloads include a Windows Defender virus scan, an image-import task, a video-editing session, and application launches.
The PCMark 7 Secondary Storage test results in a proprietary score that is usually in the thousands. We mention this score in each review and compare it with the scores that other, competing drives received. The score is only meaningful relative to other scores.
BlackMagic Disk Speed Test
Our next test measures a drive's throughput in bits per second. We use the macOS-only BlackMagic Disk Speed Test app from professional media software firm DaVinci to perform this test. This utility is typically used to discern whether a given drive has enough throughput to play back specific video formats smoothly. But it also returns some useful throughput measurements.
BlackMagic offers both a read score and a write score, which we also compare with those of other, similar drives. These scores are useful in discovering the theoretical maximum speed that a drive can achieve.
Crystal DiskMark 6
We also use the Crystal DiskMark utility for a second opinion on throughput. Crystal DiskMark's sequential-read tests measure read/write activity with data written in a large contiguous block on the drive, which is similar to how manufacturers themselves test drives to advertise their performance.
We also use Crystal DiskMark's 4K tests to measure random reads/writes, which reflect data activity in which the drive is fetching and writing scattered files and pieces of files across the hard drive or solid-state drive.
Folder Transfer Test
The final test for external drives is a drag-and-drop test. It uses the Windows Explorer or macOS Finder to copy a 1.23GB test folder full of several different file types from the testbed's internal drive to the external hard drive being tested. We hand-time the scores (in seconds).
We perform each of these tests using all of the interfaces that the drive supports. For most current drives, that means we almost always perform one round of tests using USB 3.0 (either via Type-A or Type-C connectors, of the fastest kind the drive supports, say, USB 3.1 Gen 1 versus Gen 2). If the drive supports Thunderbolt 3, we run the tests using that interface, too.
Network Attached Storage & Internal Drives
The process for testing internal drives is slightly different. We mount the drive in our testbed using its native interface and set it up as a secondary drive. We never install system files onto the internal drive we're testing. Internal SSD and hard drive testing is done strictly on our Intel X299-based testbed, employing the native SATA ports or one of the M.2 PCI Express x4 slots, in the case of PCI Express M.2 SSDs.
Internal Drives: PCMark 8 Storage, Crystal DiskMark & AS-SSD
First, we run the PCMark Storage test (in this case, PCMark 8's Storage test, instead of PCMark 7's) for a high-level view of how the drive will function under various everyday workloads like word processing, videoconferencing, and the like. Like PCMark 7, it generates a proprietary score meaningful relative to other scores.
Next, we run the Crystal DiskMark utility as described above.
Finally, we use AS-SSD's copy tests to simulate moving different types of files from one location on the drive to another. This is very similar to the folder drag-and-drop test above that we use for external drives, but it's slightly more comprehensive since AS-SSD offers three different transfer tests: a sample game-install folder, a program folder, and a single big ISO file.
Testing NAS Drives
Because a wide range of network conditions outside of our control can affect the performance of network attached storage (NAS) devices, we only perform a folder transfer test on these devices. The specific hard drives installed in a NAS are also a big variable, if you're looking at a NAS drive that doesn't come pre-populated with drives.
This test uses a different folder from our internal and external drive tests, comprising 4.9GB worth of music, video, photo, and office document files. We hand-time how long it takes to transfer the folder to the NAS via a wired connection, as well as how long it takes to transfer the file from the NAS back to our testbed.