Great for itty bitty files.
Seagate’s new line of hybrid drives, known as the FireCuda, promises to solve all the wrongs of past hybrid drives. These mish-mashes of mechanical and SSD drives are meant to deliver SSD performance at HDD prices. The problem is that in the past they’ve failed, which is why we rarely see them. However, the FireCuda apparently uses new tech to get around these issues.
We received a 2TB version of the FireCuda, which comes in the 3.5-inch form factor most mechanical hard drives use. It connects to your PC via a standard SATA 6Gb/s interface, so there’s no need for a fancy M.2 or PCIe connection to get the most out of it. A 5-year warranty is pretty standard these days, especially for SSDs which apparently suffer fewer failures, so it’s not much to write home about.
However, the fact that this is a two Terabyte drive for just $160 is something to be impressed with, if it does indeed offer performance levels similar to that of an SSD. For example, to buy a full 2TB SSD, you’re looking at around $750 or more, which makes the price of the FireCuda extremely competitive. According to Seagate, the mechanical portion of this drive spins at 7,200RPM, as to be expected for a performance HDD. The 2TB version of the drive is also equipped with 8GB of flash memory, which acts as a cache.
The way it’s supposed to work is like so: over time, the drive learns which applications you run most frequently, and stores them in the 8GB cache. It’s only when you’re loading something new that it reverts back to the 2TB of mechanical storage, and if you use it enough, it’ll then load that into the flash memory. It uses MLC memory in the format of a small SSD, but any modern user will probably realise that 8GB isn’t a whole lot of storage space. It might be enough to load up Windows nice and promptly, but what about all of those 20GB applications hogging your hard drive space?
Seagate calls this “OptiCache” technology, and it’s meant to boost performance over prior hybrid drives by up to 45%. And the more you use the drive, the better it’ll perform with your most used applications. So just how do you test a drive like this?
Our first set of benchmarks was to load up a bunch of games, and manually time how long it took from hitting the play button until the first logo appeared on screen. Apparently though this barely loads any data, as our comparison results against a Corsair Neutron XT showed that both drives were roughly the same speed, even after running the tests five times each.
After doing a bit of research, we realised that PCMark 8’s SSD/HDD test would be perfect. According to the supporting documents that come with the software, “Using traces recorded from Adobe Creative Suite, Microsoft Office and a selection of popular games, PCMark 8 Storage highlights real-world performance differences between storage devices.” This is perfect, as it loads around nine different real world applications onto the drive, and keeps them in the same place, then tests the trace performance (basically disk performance) when loading and using each application. This is opposed to tests like CrystalDiskMark, which test random sectors, something the FireCuda’s predictive algorithms can’t work with.
We ran the PCMark 8 Home SDD/HDD test four times, each run taking around three hours each time. Performance was not as expected, with loading times for the apps contained within the PC Mark testing process varying wildly with each run. In fact, the last run was the slowest of the lot, when it should have been the quickest. This shows the main issue with the FireCuda. With only 8GB of SSD flash memory, it can only handle a very small amount of application data before it stops doing what it’s meant to do. Our benchmark, which loaded nine different applications (typically as many of you would at home) showed that the FireCuda simply can’t handle so many different programs and still perform consistently. It’ll be fine if you just want to run Windows and Word, or your most-used app, but load it up with much else and you’ll start seeing some very strange results indeed. We saw up to 25% variance on some tests, which is then balanced out by other tests - it just can’t make up its mind which app to optimise for. While it can be very quick, we can’t give it top marks because it makes for a PC that might be having a good day, or a bad day, and you never know which to expect.