PNY Technologies is well known in the computing industry for its NVIDIA-based graphics cards, but the company also participates in the DRAM and flash-based storage products markets. In the latter, PNY markets a range of USB flash drives, SD cards, and portable SSDs under variations of the 'Elite' tag.

The company launched two new palm-sized portable SSDs earlier this year - the Pro Elite V2 USB 3.2 Gen 2 in the 10 Gbps class, and the EliteX-PRO USB 3.2 Gen 2x2 in the 20 Gbps class. Both products are based on Phison's native USB flash drive (UFD) controllers. Both products come with a Type-C interface. While the Pro Elite V2 is available only in the 500 GB and 1 TB capacity points, the EliteX-PRO has four SKUs with capacities ranging from 500 GB to 4 TB.

The Pro Elite V2 and the EliteX-PRO have identical industrial designs and dimensions, with only the product name printed on the aluminum body serving as a differentiator. The design is reminiscent of the Samsung Portable SSD T5, but manages to retain a more compact and sleek form-factor. This review takes a detailed look at the performance and value proposition of the 1 TB Pro Elite V2 and the 4 TB EliteX-PRO, with a particular focus on how they stack up against the existing players in the segment.

Introduction and Product Impressions

Rapid advancements in flash technology (including the development of 3D NAND and improvements in TLC reliability) have been accompanied by the appearance of faster host interfaces for external devices in the last decade. Together, they have contributed to bus-powered direct-attached storage devices growing in both storage capacity and speeds. The Type-C standard has also achieved widespread acceptance in the consumer market. Protocols such as USB 3.2 Gen 2x2 / USB4 and Thunderbolt riding on top of the Type-C connector have enabled the introduction of palm-sized flash-based storage devices capable of delivering 2GBps+ speeds.

One of the primary challenges with high-speed storage devices is the thermal aspect. Bridge-based solutions with multiple protocol conversion chips tend to dissipate more power due to the additional components. High-performance portable SSDs in the past have had no option but to use them - first, with SATA bridges, and then with NVMe bridges. The introduction of native UFD controllers capable of hitting 10 Gbps and 20 Gbps from Phison and Silicon Motion has opened up yet another option in this category. The Crucial X6, equipped with the Phison U17, was reviewed in August 2021 and was one of the first retail products to surpass the SATA speeds barrier by hitting 800 MBps speeds without using a NVMe bridge. Around the same time, Silicon Motion's SM2320 powered the Kingston XS2000 to 20 Gbps speeds without a bridge in the middle.

Products based on Silicon Motion's SM2320 have gained a lot of consumer mindshare because they have typically been able to hit the interface speed limits for sequential accesses in both the 10 Gbps and 20 Gbps categories. Phison's U17 and U18 were introduced with slightly lower peak performance numbers, and despite vendors like Crucial picking them up for the X6, the products didn't make any waves. The best implementation of Phison's U17, in our opinion, was seen in the OWC Envoy Pro Mini. The introduction of faster flash has since allowed portable SSDs (PSSDs) based on the Phison's native UFD controllers to hit higher speeds. As a result, we have seen an uptick in their adoption. Products such as the Sabrent Rocket Nano v2 and the Corsair EX100U are based on the Phison U18 controller. PNY has also hitched its wagon to the Phison camp for its play in the PSSD market

PNY's 2023 PSSD lineup includes the Pro Elite V2 in the 10 Gbps category, and the EliteX-PRO in the 20 Gbps one. The galleries below present the packaging and teardown process for the two PSSDs.

Both products are similar in terms of packaging (down to the length of the supplied USB Type-C to Type-C and Type-C to Type-A cables) as well as industrial design. The main difference is in the internal PCB.

PNY Pro Elite V2 Controller - Phison U17

PNY Pro Elite V2 Flash Package - HA7HG64AWA

The 10 Gbps Pro Elite V2 is a Phison U17 design. There is space for four flash packages on the PCB, but PNY has opted only for two packages in the product's top capacity. The HA7HG64AWA is a NAND flash package with 128L 3D TLC from SK hynix. It has 8 x 512 Gbit dies for a total of 512 GB per package. There is one on each side of the PCB.

PNY EliteX-PRO Controller - Phison U18

PNY EliteX-PRO Flash Package - HA8IG64AWA

The 20 Gbps EliteX-PRO is a Phison U18 design. There is space for four flash packages on the PCB, and the 4 TB capacity version has all four slots filled up. The HA8IG64AWA is a NAND flash package with 128L 3D TLC from SK hynix. The company's 128L NAND had initially appeared in the form of 512 Gbit dies, but in the second half of 2021, 1 Tbit dies also started appearing (for example, in the 2 TB version of the SK hynix Gold P31). The HA8IG64AWA is likely equipped with 8 x 1 Tbit dies for a total of 1 TB per package. There are two on each side of the PCB in the 4 TB version.

There is no specific thermal solution for the controller or flash in either of the PSSDs. The SK hynix 128L 3D TLC NAND is noted for its power efficiency, and that probably played a role in PNY's decision. The interesting part of the EliteX-PRO PCB is the presence of two TI DDR4000 ICs. These are 1:2 DDR switches / multiplexers. On this PCB, they are used for high-speed signal switching to help interface four flash packages with the two-channel U18 controller.

The industrial design of the Pro Elite V2 and the EliteX-PRO is not unique, with many similarities to the Samsung Portable SSD T5. Thankfully, PNY has opted for an aluminum casing and this lends a solid look and feel to the PSSDs.

Since we are dealing with two different PSSD classes / capacity points, this review will include separate comparison graphs and tables for each. CrystalDiskInfo provides a quick overview of the capabilities of the internal storage device. Despite TRIM not being noted in the features list, we found NTFS volumes on both  PSSDs successfully processing TRIM commands from Windows. All S.M.A.R.T features such as temperature read outs worked well.

Prior to looking at the benchmark numbers, power consumption, and thermal solution effectiveness, a description of the testbed setup and evaluation methodology is provided.

Testbed Setup and Evaluation Methodology

Direct-attached storage devices (including thumb drives) are evaluated using the Quartz Canyon NUC (essentially, the Xeon / ECC version of the Ghost Canyon NUC) configured with 2x 16GB DDR4-2667 ECC SODIMMs and a PCIe 3.0 x4 NVMe SSD - the IM2P33E8 1TB from ADATA.

The most attractive aspect of the Quartz Canyon NUC is the presence of two PCIe slots (electrically, x16 and x4) for add-in cards. In the absence of a discrete GPU - for which there is no need in a DAS testbed - both slots are available. In fact, we also added a spare SanDisk Extreme PRO M.2 NVMe SSD to the CPU direct-attached M.2 22110 slot in the baseboard in order to avoid DMI bottlenecks when evaluating Thunderbolt 3 devices. This still allows for two add-in cards operating at x8 (x16 electrical) and x4 (x4 electrical). Since the Quartz Canyon NUC doesn't have a native USB 3.2 Gen 2x2 port, Silverstone's SST-ECU06 add-in card was installed in the x4 slot. All non-Thunderbolt devices are tested using the Type-C port enabled by the SST-ECU06.

The specifications of the testbed are summarized in the table below:

The testbed hardware is only one segment of the evaluation. Over the last few years, the typical direct-attached storage workloads for memory cards have also evolved. High bit-rate 4K videos at 60fps have become quite common, and 8K videos are starting to make an appearance. Game install sizes have also grown steadily even in portable game consoles, thanks to high resolution textures and artwork. Keeping these in mind, our evaluation scheme for portable SSDs and UFDs involves multiple workloads which are described in detail in the corresponding sections.

• Synthetic workloads using CrystalDiskMark and ATTO
• Real-world access traces using PCMark 10's storage benchmark
• Custom robocopy workloads reflective of typical DAS usage
• Sequential write stress test

In the next couple of sections, we have an overview of the performance of the two PNY PSSDs in these benchmarks. Prior to providing concluding remarks, we have some observations on the drives' power consumption numbers and thermal solution also.

Performance Benchmarks - PNY Pro Elite V2 1 TB

Benchmarks such as ATTO and CrystalDiskMark help provide a quick look at the performance of the direct-attached storage device. The results translate to the instantaneous performance numbers that consumers can expect for specific workloads, but do not account for changes in behavior when the unit is subject to long-term conditioning and/or thermal throttling. Yet another use of these synthetic benchmarks is the ability to gather information regarding support for specific storage device features that affect performance.

PNY claims read and write speeds of 1100 MBps and 1000 MBps respectively, and numbers close to this are hit in our ATTO benchmark configuration. ATTO benchmarking is restricted to a single configuration in terms of queue depth, and is only representative of a small sub-set of real-world workloads. It does allow the visualization of change in transfer rates as the I/O size changes, with optimal performance being reached around 16 MB for a queue depth of 4.

Full System Drive Benchmark Bandwidth (MBps)Full System Drive Benchmark Latency (us)Full System Drive Benchmark ScoreExpand All

Consistent with the observations made in most of the workload components of the PCMark 10 Storage Bench, the PNY Pro Elite V2 makes up the bottom part of the graphs in this section. Only the much smaller OWC Envoy Pro Mini in a different form-factor (thumb drive) lands below the Pro Elite V2.

Performance Benchmarks - PNY EliteX-PRO 4 TB

The details of the different performance benchmarks have already been covered in the previous section. Cutting right down to the chase, PNY claims read and write speeds of 1600 MBps and 1500 MBps respectively, and these are actually exceeded in the ATTO benchmarks provided below. This is only representative of a small sub-set of real-world workloads. The change in transfer rates with I/O size can be observed, and optimal performance is reached around 8 MB for a queue depth of 4.

Full System Drive Benchmark Bandwidth (MBps)Full System Drive Benchmark Latency (us)Full System Drive Benchmark ScoreExpand All

The bandwidth numbers are excellent for the EliteX-PRO despite the use of a DRAM-less native UFD controller. However, these suffer from random access performance issues, causing latencies to be much higher than bridge-based solutions. As a result, the slight edge in the bandwidth graph is not enough to prevent the PSSD from slipping to the third place behind the two bridge-based solution in the full system drive benchmark score.

Miscellaneous Aspects and Concluding Remarks

The performance of the PNY Pro Elite V2 and the EliteX-PRO in various real-world access traces as well as synthetic workloads was brought out in the preceding sections. We also looked at the performance consistency for these cases. Power users may also be interested in performance consistency under worst-case conditions, as well as drive power consumption. The latter is also important when used with battery powered devices such as notebooks and smartphones. Pricing is also an important aspect. We analyze each of these in detail below.

Worst-Case Performance Consistency

Flash-based storage devices tend to slow down in unpredictable ways when subject to a large number of small-sized random writes. Many benchmarks use that scheme to pre-condition devices prior to the actual testing in order to get a worst-case representative number. Fortunately, such workloads are uncommon for direct-attached storage devices, where workloads are largely sequential in nature. Use of SLC caching as well as firmware caps to prevent overheating may cause drop in write speeds when a flash-based DAS device is subject to sustained sequential writes.

Our Sequential Writes Performance Consistency Test configures the device as a raw physical disk (after deleting configured volumes). A fio workload is set up to write sequential data to the raw drive with a block size of 128K and iodepth of 32 to cover 90% of the drive capacity. The internal temperature is recorded at either end of the workload, while the instantaneous write data rate and cumulative total write data amount are recorded at 1-second intervals.

CrystalDiskMark Workloads - Power Consumption [ 4 TB PSSDs ]
TOP: PNY EliteX-PRO 4TBSamsung T7 Shield 4TBSanDisk Extreme PRO v2 4TBCrucial X6 4TB	BOTTOM: Samsung T7 Shield 4TBSanDisk Extreme PRO v2 4TBCrucial X6 4TBPNY EliteX-PRO 4TB

Unlike the bridge-based solutions, the EliteX-PRO can hit a true zero deep sleep state. Peak power consumption is 4.9W, but that is way lower than the 7.23W of the SanDisk Extreme Pro V2. This is only to be expected given the presence of four flash packages as well as the high speeds (1550 MBps) to the external host. Samsung's T7 Shield possesses a comprehensive thermal solution despite its peak power consumption being only 4W. PNY needs to equip the EliteX-PRO with additional support to handle the heat output.

Final Words

The PNY Pro Elite V2 and EliteX-PRO have been available in the market for a few months now. The 1 TB version of the Pro Elite V2 has a current street price of $70, while the 4 TB version of the EliteX-PRO is available for $300. At the Pro Elite V2's current price point, the main competition is the Transcend ESD310C. However, the two products target different market segments / consumer types. There is not much to complain on that front. The EliteX-PRO's pricing analysis is not quite that straightforward. The 4 TB T7 Shield is priced at $260, but it is a Gen 2 device. The SanDisk Extreme PRO v2 is also priced at $300, and has a much better performance profile and stands up well to sustained stress. However, the EliteX-PRO consumes much less power. On balance, the EliteX-PRO can be said to provide value only if it is priced lower than the SanDisk Extreme PRO v2.

Phison's native UFD controllers have been a bit of a mixed bag so far. Not being able to hit the interface speed limits and small SLC cache sizes (that is arguably an important aspect for direct-attached storage workloads) are some of the complaints that any reasonable user might want to file. Bridge-based solutions, despite their higher power consumption numbers, offer much better value. OWC was able to work around these problems in the Envoy Pro Mini by careful firmware tuning and creating a unique product in a thumb drive form factor. Unfortunately, the same can't be said of PNY's offerings. While the Pro Elite V2 offers compelling all-round value at $70, low-power operation is the only advantage for the EliteX-PRO. Consumers looking for a 4 TB off-the-shelf high-speed PSSD are better off with the SanDisk Extreme PRO v2 (or even the T7 Shield if 10 Gbps / Gen 2 speeds are enough).