SK Hynix Gold P31 1TB Performance Testing
We are moving towards using larger test sizes on our benchmarks, but on several tests, we also used the smaller default test sizes. This allows us to see the difference between lighter and heavier workloads. Unfortunately, our entire backlog of SSD reviews did not include this testing methodology, so we are doing the best we can with generating comparison charts.
BlackMagic Disk Speed Test
This is a popular speed test in the video community. It shows whether a storage device is suitable for throughput required at a given video format.
The Gold P31 1TB easily qualifies for all the formats on this test.
CrystalDiskMark 7.0.0 x64
CrystalDiskMark is used as a basic starting point for benchmarks as it is something commonly run by end-users as a sanity check.
Using the 1GB test set, the SK Hynix Gold P31 1TB is exceeding its advertised specs of 3500 MB/s read and 3200 MB/s write. 3600 MB/s is the fastest PCIe 3.0 SSD we have seen and is bumping up against the limit for a PCIe 3.0 drive. 3200 MB/s write speeds are also impressive, easily besting our other PCIe 3.0 drives and only losing to our larger and faster PCIe 4.0 drives in our test set.
The Gold P31 1TB is almost completely unaffected by the larger 8 GB test set, to the point where I had to double-check the screenshots to make sure I had not mislabeled the results.
ATTO Disk Benchmark
The ATTO Disk Benchmark has been a staple of drive sequential performance testing for years. ATTO was tested at both 256MB and 8GB file sizes.
The Gold P31 1TB continues leading the PCIe 3.0 pack in the first ATTO test, achieving the leadership position for both read and write speeds at our 32 MB datapoint for the graphs. Overall in the test, there is a minor bit of performance variation, but nothing like the swings we will see in our upcoming ADATA XPG SX8200 Pro 1TB review.
Much like in CrystalDiskMark, the larger 8GB ATTO test set barely affects the Gold P31 1TB. This allows the Gold P31 to maintain its position at the top of the PCIe 3.0 charts overall.
Anvil’s Storage Utilities
Anvil’s Storage Utilities is a comprehensive benchmark that gives us a very in-depth look at the performance of drives tested. This benchmark was run with both a 1GB and 8GB test size.
With Anvil running a 1GB test size, the Gold P31 1TB again turns in an impressive performance, especially in terms of read performance. Write performance stumbles a bit though; while still good a number of other PCIe 3.0 SSDs manage to chalk up some of their first wins over the Gold P31 drive in Anvil’s write score.
With 8GB test size, Gold P31 1TB drive manages to maintain its class-leading read performance, while gaining back a good bit of ground with regards to write performance.
AS SSD Benchmark
AS SSD Benchmark is another good benchmark for testing SSDs. We run all three tests for our series. Like other utilities, it was run with both the default 1GB as well as a larger 10GB test set.
In the 1GB ASSSD test set, overall performance continues to be excellent.
Thanks to the move to the larger 10GB AS SSD test set, the Gold P31 1TB drive jumps almost to the top of our result set for read performance, while also demonstrating impressive write performance. The Gold P31 achieved this jump not by performing better at the larger test than on the smaller, but instead simply by being unaffected by the larger test set while almost every other drive we have tested suffers a significant drop.
SPECworkstation, thermals, and our conclusion are up next.
I’ve been waiting for this drive to be released. :)
A quick request for SSD/NVMe drive articles, could the presence or absence of power loss protection be mentioned? Maybe in a little feature matrix? PLP, DRAM cache size, SLC cache size, IOPs, TBW, etc.
I’m assuming as a consumer drive this doesn’t have PLP, but explicitly calling it out is nice.
RQ,
Normally I put that in there – the absence (or presence) of PLP. I must have accidentally cut that from my article at some point and failed to notice. Apologies!
would be good to see the point where the SLC cache is exhausted.
My impression with flash media is that individual units of the same make perform differently depending on usage history and possibly manufacturing variations. Is there any chance in testing three to seven individual units from the same manufacturer of the same model but different production runs to see how much variation there is?
It would be great if a more current/larger Samsung SDD could be reviewed in the near future. It’s quite difficult to compare the performance of a 250 GB drive to other higher capacity drives.
Fred – the Samsung 980 Pro is coming.
Did you have a heatsink on the drive when testing temperatures? I get similar temps in my laptop with heatsinks attached.