512GB Class M.2 Performance Testing
We kept the charts light today to focus on the two new drives. We did include two older M.2 SATA SSDs still sold today. The MyDigtialSSD BP5e M.2 uses a Phison S10 controller with Toshiba 15nm TLC memory. This drive represents a similar configuration to several OEM-focused SSDs shipping in notebooks sold in 2016 and 2017. The SanDisk X400 M.2 is an actual SSD that currently ships in notebooks and desktops from large system builders. The X400 M.2 is the only 1TB drive in this article.
There are some fine details to consider with this form factor. Heat can become an issue without proper airflow. The circuit board is smaller than a standard 2.5" SSD. The SATA drives don't suffer from thermal throttling as much as the NVMe drives that are capable of reading and writing four to six times faster than the products we're testing today. You should still be mindful of placement.
Not all motherboards will allow you to use every M.2 slot for SATA. We use Z270 series boards from Asrock to test SSDs and the slot between the CPU and first PCI Express slot is PCIe-based storage only. We tried a M.2 SATA drive in the system and the initial screen noted an error and refused to boot further. We had to configure one of the M.2 slots at the bottom of the board to M.2 SATA in the BIOS.
Another consideration is that some motherboards will not dedicate the maximum bandwidth to the M.2 slot for SATA. This is mainly an issue with older boards. For instance, the ASUS Z87 generation boards would only drive 1 PCIe lane's bandwidth to the M.2 slot, around 500 MB/s.
Sequential Read Performance
All of the drives exhibit very high sequential read performance. Most variation between products comes at lower queue depths, but that is where most consumer workloads fall. This test is the equivalent of reading a large block size file from the drives, like a movie. The queue depth (QD) is now many files you read at one time. A QD2 test is the same as reading two movies back from the drive, or transferring them from one drive to another.
Sequential Write Performance
The sequential write test is the same but in the reverse order. In this test, we write large block size data to the drives. This is where Samsung has historically dominated during the 3-bit per cell (TLC) era. Micron closed the gap to Samsung's write performance with 64-layer flash and not long after Toshiba/SanDisk released 64-layer memory to do the same.
In our burst test, the drives are all fairly equal as the SLC buffer does a very good job on each of these products to keep performance high for a short period of time.
Sustained Sequential Write Performance
The MX500's SLC buffer is not as fast as the 860 EVO's but it does stay at a higher speed tier for longer. Post SLC buffer the MX500 delivers a higher native TLC write speed, around 375 MB/s for the 500GB drive. The 860 EVO 500GB writes large block size sequential data at 300 MB/s after the initial SLC buffer fills.
For reference, the other two drives use older 2D TLC memory and have a lower sustained write speed after the initial SLC hit. There are many advantages to using 3D memory. For people that transfer large volumes of data this can be the most significant.
Random Read Performance
We talked about Micron catching up to Samsung's VNAND but in some areas, the Micron 64-layer memory is superior. The MX500 and some NVMe SSDs like the Adata SX8200 with Micron flash technology outperform Samsung's SSD in random read performance.
This is where the user experience comes from, what makes your computer "feel" fast. Samsung's 860 series took a small step backwards in QD1 random read performance compared to the previous generation, 850 EVO.
The MX500 has a strong advantage here and it's one of the reasons why we like this product series so much.
Random Write Performance
Micron flash doesn't win every test through. This review will show the MX500 and 860 EVO going back and forth depending on the workload. Samsung's TurboWrite, the fancy name for the SLC buffer technology, is a fixed size that writes random data faster. Crucial's Data Write Acceleration (DWA), the brands fancy name for the SLC buffer technology, is dynamic. It will change the amount of high-speed area depending on the amount of data already on the drive, and the working workload at the time. Both technologies are very good and superior to what we see from most companies today. In the performance charts, the Samsung shows an advantage but under real-world use, there is very little difference at the keyboard.
70% Read Sequential Performance
We start to see some division in performance under mixed workloads. With sequential mixed data, the Samsung outpaces the Crucial with a gap you may notice transferring large amounts of data in your system or over a high-speed network.
70% Read Random Performance
In the same vein, the Crucial delivers better random mixed workload performance thanks to the faster memory at low queue depths. The MX500 will feel faster under normal use. In two identical notebooks, we ran the MX500 and 860 EVO side-by-side during the SYSMark test. We were able to see the MX500 completing tasks faster, even booting and rebooting. The problem is that without a side-by-side comparison, you wouldn't see the improvements. Both drives feel very fast as the boot media for a PC.
PRICING: You can find the product discussed for sale below. The prices listed are valid at the time of writing, but can change at any time. Click the link below to see real-time pricing for the best deal:
- We at TweakTown openly invite the companies who provide us with review samples / who are mentioned or discussed to express their opinion of our content. If any company representative wishes to respond, we will publish the response here.