The Samsung 870 EVO (1TB & 4TB) Review: Does the World Need Premium SATA SSDs?

Samsung has dominated the SSD market for a very long time, and every new consumer SSD model the company introduces is noteworthy. The latest product to reach the shelves is the new range of 870 EVO SSDs, building on a decade-long legacy of SATA SSDs. We have tested the 1 TB and 4 TB models from this new product family.

Samsung is one of the few companies still putting significant effort into SATA SSDs and releasing new consumer SATA models. As PC OEMs have overwhelmingly switched to using NVMe SSDs in new systems, even on the smaller capacities, the client/consumer SATA SSD market now exists almost entirely for the sake of DIY system builders and aftermarket upgrades on older systems. Most major consumer SSD brands have either stopped updating their SATA models, or decided to quietly update components without the fanfare of a new model release. Then there are companies creating odd-ball models, such as a 15.36 TB design. Either way, we don’t get many new consumer SATA SSDs in for review these days.

(It is worth noting that the enterprise SATA SSD market has seen somewhat more in the way of visible activity, because the longer product lifetimes in that market and the higher profit margins give SATA SSDs more of a long tail of commercial relevance.)

Despite this market shift, Samsung is still focused on a SATA product line, although even Samsung is showing reduced interest in SATA. It took over six months from the release of the QLC-based 870 QVO for its TLC-based 870 EVO sibling to arrive, and there’s still no sign of an MLC-based 870 PRO model (or any other MLC SSDs from Samsung based on anything newer than their 64-layer 3D NAND). Furthermore, the new 870 EVO no longer caters to either mSATA and M.2 SATA form factors, focusing solely on 2.5-inch drives. The mSATA version of the 860 EVO disappeared from the market last year, but the M.2 SATA version is still readily available and has tracked market price trends as much as Samsung’s drives ever do. It’s a bit of a surprise that they’re dropping M.2 SATA this soon given how long Samsung supported mSATA.

Given the roughly three-year cadence Samsung has set for their SATA SSD updates, the 870 generation may also be the last.

Samsung 870 EVO Specifications
Capacity 250 GB 500 GB 1 TB 2 TB 4 TB
Form Factor 2.5″ 7mm SATA
Controller Samsung MKX
NAND Flash Samsung 512Gbit 128L 3D TLC
LPDDR4 DRAM 512MB 1 GB 2 GB 4 GB
Sequential Read 560 MB/s
Sequential
Write
SLC 530 MB/s
TLC 300 MB/s 530 MB/s
Random
Read
QD32 98k IOPS
QD1 13k IOPS
Random
Write
QD32 88k IOPS
QD1 36k IOPS
Power
Draw
Burst 3.5 W 4.0 W 4.5 W 5.0 W
Average 2.2 W 2.5 W
Idle 30 mW 35 mW
Warranty 5 years
Write Endurance 150 TB
0.3 DWPD
300 TB
0.3 DWPD
600 TB
0.3 DWPD
1200 TB
0.3 DWPD
2400 TB
0.3 DWPD
Launch MSRP $39.99
(16¢/GB)
$69.99
(14¢/GB)
$129.99
(13¢/GB)
$249.99
(12¢/GB)
$479.99
(12¢/GB)

We are seeing the usual upgrades in the spec table for the 870 EVO compared to the 860 EVO from 2018. The controller is now Samsung’s MKX, a minor iterative update to the 860’s MJX. The NAND flash memory is a more significant jump: the 870 EVO is using Samsung’s 128-layer 3D TLC, first seen on the 980 PRO. More specifically, our 1TB and 4TB 870 EVO samples are using the 512Gbit dies that were introduced on the 2TB 980 PRO, a model that arrived much later than the smaller capacities built around 256Gbit dies. The preceding Samsung 860 EVO launched with Samsung’s 64L 3D NAND, and they appear to have skipped over their 92L generation with their TLC SATA drives.

The performance specifications are very similar to the 860 EVO. Almost all of those performance numbers are limited either by the throughput of the SATA interface or the latency of the NAND itself, so most of the performance numbers are the same across the entire capacity range. The 250GB and 500GB models still have significantly lower sequential write speed after the SLC cache runs out, but this time we don’t have official numbers on SLC cache sizes (but we have tested that in this review). On the other hand, Samsung continues to be one of the few brands to provide performance specifications for queue depth 1 in addition to the high queue depth ratings.

Internally, the 870 EVO continues to illustrate just how ridiculously oversized the 2.5″ form factor is for modern consumer storage. Even the 4TB model’s PCB takes up just a fraction of the internal space, and the 1TB model’s even smaller PCB has only a handful of tiny components for the power regulation on its back side—this could have easily been a single-sided board if Samsung wasn’t saving space for the extra flash on the 2TB model.


1TB (Left) vs 4 TB (Right)

Gallery: Samsung 870 EVO 1TB & 4TB

The SATA Competition: Hard to Determine

Truth be told, there aren’t that many new SATA SSDs on the market. To put this into perspective, the most recent SATA competitor we reviewed, the SK hynix’s Gold S31, was launched in 2019.

Part of this is down to the lack of focus on the SATA market as mentioned above, but also because SSD vendors have taken to silently updating the hardware inside a product line without changing the product number or announcing the change. Their argument is that as long as they can still achieve the same performance listed in the specification sheet, and adhere to the warranty, the customer shouldn’t have to worry about what is actually inside the product. 

We generally disapprove when SSD vendors silently swap out major components without renaming a product, because changing the SSD controller or NAND flash can have a major impact on a drive’s performance and power efficiency. Unfortunately, this has long been a common practice for entry-level models, and a few manufacturers have tried it on more upmarket models and been deservedly excoriated when they’re caught (eg. ADATA, with the SX8200 Pro).

However, for mainstream SATA SSDs based on TLC NAND and controllers with DRAM caches, there’s a lot less potential downside for consumers when the manufacturer silently upgrades the NAND. The SATA interface imposes such a severe performance bottleneck that it’s unlikely new NAND would result in a slower drive. Switching to newer NAND with a higher per-die capacity could adversely affect the smallest models in a product line, but the tendency has been for 120GB models to simply disappear instead of being updated—and they aren’t missed. We’re disappointed that (for example) Micron’s Crucial MX500 is now shipping with at least 96-layer NAND rather than the 64-layer NAND it launched with, and likely with Silicon Motion’s SM2259 controller that features improved error correction over the SM2258.

Newer NAND in existing SATA models is far more likely to bring benefits to consumers. Maximum throughput cannot increase, but latency and performance at lower queue depths can still be improved marginally. Newer NAND also tends to bring power efficiency improvements, which means that upgrading the storage capacity in a laptop that’s a few years old shouldn’t hurt battery life even though larger drives tend to be more power-hungry.

The other factor in this is that we recently updated our SSD test suite for 2021, which means re-testing older drives for new performance numbers. In our SSD test suite 2021 article, we prioritized NVMe SSDs, as that is where the market is. We are getting around to retesting the older SATA drives, although these take the longest time to complete – 24 hours for a fast 1 TB drive, up to 100 hours or more for a 4 TB drive, not including the analysis.

So while we’re a bit lacking in SATA review samples that reflect the current state of the market, we do have a diverse selection of entry-level NVMe SSDs that have been pushing into the SATA SSD price bracket. These may not be viable alternatives for some use cases (older systems that don’t support NVMe, or data hoarders that need lots of multi-TB drives), but for any reasonably recent consumer system with ordinary storage needs, these NVMe drives are real competitors to SATA SSDs. That’s the focus on this review.

Our AnandTech Storage Bench tests are traces (recordings) of real-world IO patterns that are replayed onto the drives under test. The Destroyer is the longest and most difficult phase of our consumer SSD test suite. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

ATSB The Destroyer
Average Data Rate
Average Latency Average Read Latency Average Write Latency
99th Percentile Latency 99th Percentile Read Latency 99th Percentile Write Latency
Energy Usage

For SATA drives, the Samsung 870 EVOs turn in class-leading scores on almost all of the performance metrics. But these improvements are all marginal at best; the SATA interface bottleneck almost completely levels the playing field. The small improvements to read latency brought by the 870 EVO pale in comparison to what is achieved by even entry-level NVMe SSDs.

In stark contrast to the performance numbers, the 870 EVOs turn out to be the most power-hungry TLC drives in this bunch: they sacrifice some of the efficiency improvements the 860 EVO provided, even though drives like the SK hynix Gold S31 have been able to deliver significant improvement on this.

The ATSB Heavy test is much shorter overall than The Destroyer, but is still fairly write-intensive. We run this test twice: first on a mostly-empty drive, and again on a completely full drive to show the worst-case performance.

ATSB Heavy
Average Data Rate
Average Latency Average Read Latency Average Write Latency
99th Percentile Latency 99th Percentile Read Latency 99th Percentile Write Latency
Energy Usage

The scores for the Heavy test paint much the same picture as for The Destroyer. The full-drive test runs additionally show that the worst-case performance of the mainstream SATA SSDs is still superior to many entry-level NVMe SSDs, even though the NVMe SSDs significantly outperform SATA for any more normal workload.

The ATSB Light test represents ordinary everyday usage that doesn’t put much strain on a SSD. Low queue depths, short bursts of IO and a short overall test duration mean this should be easy for any SSD. But running it a second time on a full drive shows how even storage-light workloads can be affected by SSD performance degradation.

ATSB Light
Average Data Rate
Average Latency Average Read Latency Average Write Latency
99th Percentile Latency 99th Percentile Read Latency 99th Percentile Write Latency
Energy Usage

On the Light test, the measurable but imperceptible performance advantages of the 870 EVOs over other SATA drives have basically disappeared. The read latency scores on the full-drive test runs may be a tiny bit better than the 860 EVO, but the only scores that have clearly shifted with this new generation are the energy consumption figures that have creeped up.

The PCMark 10 Storage benchmarks are IO trace based tests similar to our own ATSB tests. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

PCMark 10 Storage Traces
Full System Drive Overall Score Average Bandwidth Average Latency
Quick System Drive Overall Score Average Bandwidth Average Latency
Data Drive Overall Score Average Bandwidth Average Latency

The Full System Drive test from the PCMark 10 Storage suite shows a much wider spread of performance scores among SATA drives than our ATSB traces, but also a much smaller advantage for the NVMe drives. Judging by this test, the 870 EVO offers a small but real improvement to performance compared to earlier SATA drives. The 4TB 870 QVO also scores quite well since it benefits from the same controller and has enough SLC cache to almost match the performance of the 4TB 870 EVO.

The subset of tests included in the Quick System Drive and Data Drive benchmarks show a more level playing field among SATA SSDs, and a greater advantage for NVMe drives. Since we run these tests before the Full System Drive test, each drive is closer to its fresh out-of-the-box state, which helps these tests get closer to showing the theoretical peak performance of a drive.

Our burst IO tests operate at queue depth 1 and perform several short data transfers interspersed with idle time. The random read and write tests consist of 32 bursts of up to 64MB each. The sequential read and write tests use eight bursts of up to 128MB each. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

QD1 Burst IO Performance
Random Read Random Write
Sequential Read Sequential Write

Our burst IO tests show little to no performance differences between the Samsung 870 EVO and other top SATA SSDs. The 1MB sequential transfers are already hitting the SATA throughput limits even at QD1, and the 4kB random IOs are at best marginally improved over Samsung’s previous generation. Samsung’s slight improvement to random read latency is enough to catch up to Micron’s as shown by the Crucial MX500, but a 10% gain hardly matters when NVMe drives can double this performance.

Our sustained IO tests exercise a range of queue depths and transfer more data than the burst IO tests, but still have limits to keep the duration somewhat realistic. The primary scores we report are focused on the low queue depths that make up the bulk of consumer storage workloads. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Sustained IO Performance
Random Read Random Write
Sequential Read Sequential Write

On the longer synthetic tests that bring in some slightly higher queue depths, the improved random read performance of the 870 EVO is a bit more clear. In one sense it is impressive to see Samsung squeeze a bit more performance out of the same SATA bottleneck, but we’re still talking about small incremental refinements where NVMe enables drastic improvements. Aside from random reads, the 870 EVO’s performance improvements are exceedingly minute and it should be considered essentially tied with most other recent mainstream TLC SATA drives.

Sustained IO Performance
Random Read Random Write
Sequential Read Sequential Write

Power consumption is one area where Samsung could theoretically offer more significant improvements despite still being constrained by the same SATA interface, but the 870 EVO doesn’t really deliver any meaningful improvements there. The 4TB model is consistently a bit less efficient than the 1TB model on account of having more memory to keep powered up, but when comparing the 1TB model against its predecessor and competing drives there’s nothing particularly noteworthy about the 870 EVO. SK hynix’s Gold S31 has a modest efficiency advantage for random IO while Samsung is technically the most efficient of these SATA drives for sequential IO.

Random Read
Random Write
Sequential Read
Sequential Write

The queue depth scaling behavior of the 870 EVOs is almost identical to the 860 EVOs and still quite typical for mainstream SATA drives. For random reads the 870 EVOs saturate around QD16, while for random writes QD4 suffices. On the sequential IO tests there’s only a small performance gain from QD1 to QD16, and the more interesting question is how stable performance is through the rest of the sequential tests. The 1TB 870 EVO seems to run out of SLC cache a bit earlier than the 860 EVO when the sequential write test is running on an 80% full drive, but the 4TB model has plenty of cache to finish out that test at full speed.

This test illustrates how drives with higher throughput don’t always offer better IO latency and Quality of Service (QoS), and that latency often gets much worse when a drive is pushed to its limits. This test is more intense than real-world consumer workloads and the results can be a bit noisy, but large differences that show up clearly on a log scale plot are meaningful. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Consistent with most of our other read performance tests, the Samsung 870 EVO shows slightly better average and 99th percentile random read latencies than most of its SATA competition. Even some of the entry-level NVMe drives that can deliver higher random read throughput than is possible for the 870 EVO still have clearly higher latency across most or all of the throughput range that the 870 EVO can cover. A QLC-based or DRAMless TLC NVMe SSD can potentially offer far higher throughput than any SATA SSD, but clearly beating the 870 EVO on both throughput and latency requires stepping up to a more mainstream NVMe design with DRAM and TLC NAND.

Our benchmark suite includes a variety of tests that are less about replicating any real-world IO patterns, and more about exposing the inner workings of a drive with narrowly-focused tests. Many of these tests will show exaggerated differences between drives, and for the most part that should not be taken as a sign that one drive will be drastically faster for real-world usage. These tests are about satisfying curiosity, and are not good measures of overall drive performance. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Pass 1
Pass 2

Some of our other tests have shown a few signs that the 870 EVO’s write performance can drop when the SLC cache runs out, but this straightforward sequential write pass over the entire drive doesn’t reveal any such behavior. The 870 EVO’s sequential write performance is extremely consistent, even on the second write pass.

Sustained 128kB Sequential Write (Power Efficiency)
Average Throughput for last 16 GB Overall Average Throughput

Due to the excellent performance consistency, the Samsung 870 EVOs edge out the other SATA drives with marginally higher average sequential write speeds. The entry-level NVMe drives end up much worse off than the mainstream SATA drives once their caches run out, but the more mainstream NVMe drive blows them all away.

As expected, the Samsung 870 EVO’s random read performance shows basically no variation across a range of working set sizes, and that read performance is at least a little bit faster than any of the other SATA drives or the entry-level NVMe drives.

Random Read
Random Write
Sequential Read
Sequential Write

There are no particular surprises in how the Samsung 870 EVO handles IOs of different block sizes. Unlike some drives, it has no trouble with sub-4kB IOs. It offers moderate improvements over the 860 EVO for mid-sized random reads (up to about 128kB). The one negative is that for writes we again see more inconsistency from the 870 EVO than the 860 EVO when testing an 80% full drive. The simple whole-drive sequential write test may not have been able to reveal any SLC caching troubles, but it does seem clear that the caching behavior has some performance regressions for more complicated workloads on a drive that’s more well-used—though it’s still unlikely to matter for any typical real-world consumer workload.

For details on our mixed IO tests, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Mixed IO Performance
Mixed Random IO Mixed Sequential IO

The mixed random IO test provides the Samsung 870 EVO with one of its biggest performance wins yet over the rest of the SATA field and the entry-level NVMe competition. But most of that comes from the capacity advantage the 4TB model has over most of these comparison drives; the 1TB 870 EVO is only about 5% faster overall than the 860 EVO. On the mixed sequential IO test, the SATA bottleneck keeps most of the performance scores within a fairly narrow range, and the 1TB 870 EVO’s performance is actually a bit of a regression compared to its predecessor.

Mixed IO Efficiency
Mixed Random IO Mixed Sequential IO

As with our separate tests of random reads and writes, the top efficiency scores for mixed random IO go to SK hynix, with Samsung’s TLC drives turning in the next best scores and having a clear advantage over other competing brands. Over on the sequential IO side of things, the efficiency scores more closely mirror the performance scores, and the 870 EVO doesn’t have any real advantage over other mainstream SATA drives.

Mixed Random IO
Mixed Sequential IO

The 1TB 870 EVO’s performance during the mixed random IO test is more consistent than the 860 EVO’s, but still has a few unpleasant drops that aren’t present for the 4TB model. On the mixed sequential IO test, the 1TB 870 EVO’s performance is actually a bit less consistent than the 860 EVO. But aside from those occasional outliers, the general trend is for the 870 EVO to provide superior random IO performance and link-saturating sequential performance across a wide range of workload mixes.

Real-world client storage workloads leave SSDs idle most of the time, so the active power measurements presented earlier in this review only account for a small part of what determines a drive’s suitability for battery-powered use. Especially under light use, the power efficiency of a SSD is determined mostly be how well it can save power when idle.

SATA SSDs are tested with SATA link power management disabled to measure their active idle power draw, and with it enabled for the deeper idle power consumption score and the idle wake-up latency test. Our testbed, like any ordinary desktop system, cannot trigger the deepest DevSleep idle state. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Idle Power Consumption - No PMIdle Power Consumption - Desktop

The Samsung 870 EVO may feature an updated controller compared to the 860 EVO, but there’s no real difference in idle power consumption, for either active idle or the desktop (non-DevSleep) idle states. Samsung’s idle power figures are best in class, with SK hynix offering the only close competition.

Idle Wake-Up Latency

The Samsung SATA drives all take about one millisecond to wake up from using SATA link power management. This is higher than several of the other SATA drives, but not really enough to be of much concern for system responsiveness.

The Samsung 870 EVO is a new SATA SSD in a market where all the interesting action is centered around NVMe SSDs. The 870 EVO is necessarily a low-key product refresh, but Samsung deserves praise for actually making this a new model instead of simply updating the parts used in the 860 EVO.

Given the limitations imposed by the SATA interface, our expectations for any new SATA SSD are mild. Performance can only improve in the corner cases and power efficiency cannot make big leaps without getting rid of the SATA performance limits. Prices can go down, but we’ve been seeing a lot of that even without a new generation of 3D NAND and SSD controller: the 860 EVO is currently selling for about a third of what the launch MSRPs were in 2018. The Samsung 870 EVO’s newer 128L flash may be setting the stage for future price drops, but this early in Samsung’s transition to 128L flash it’s not bringing any savings to consumers.

Knowing that any changes the 870 EVO brings relative to its predecessor will be minor, the most important function of this review is simply to check whether Samsung remains at least consistent with the refresh. As far as we can tell, all seems to be well. Our testing didn’t reveal any serious performance regressions, though several signs point to the 870 EVO’s SLC caching being a bit less effective. Since this only shows up on tests that are deliberately more strenuous than any common consumer workload, we’re not concerned by these results. Otherwise, the 870 EVO continues to be just about as fast as possible for a SATA SSD, and is a fine replacement for the 860 EVO.

It is a little disappointing that the 870 EVO doesn’t bring further improvements to power efficiency. Since the 860 EVO’s launch, SK hynix has raised the bar for consumer SSD efficiency in both the SATA and NVMe market segments, but Samsung is not challenging that leadership with their recent launches.

Widespread adoption of NVMe in the consumer space means the role of SATA SSDs is shifting and shrinking. There’s no longer any point in competing to offer the fastest SATA SSD, and not much reason to compete on write endurance when any workload that actually pushes the endurance limits of mainstream consumer SSDs would benefit greatly from NVMe performance. Most systems that are too old to support NVMe SSDs probably have more serious performance bottlenecks than storage performance. So the 870 EVO has to compete more in the role of secondary storage, providing extra capacity for things like an overflowing video game library. With game developers only just beginning to explore ways to make use of NVMe performance, most any mainstream SATA SSD will offer more than enough performance and endurance for this use case now and for the near future.

  250 GB 500 GB 1 TB 2 TB 4 TB
Samsung 870 EVO $39.99 (16¢/GB) $64.99 (13¢/GB) $129.99 (13¢/GB) $249.99 (12¢/GB) $479.99 (12¢/GB)
Samsung 870 QVO     $109.99 (11¢/GB) $218.00 (11¢/GB) $411.77 (10¢/GB)
Samsung 860 EVO $39.99 (16¢/GB) $59.99 (12¢/GB) $109.99 (11¢/GB) $229.99 (11¢/GB) $444.76 (11¢/GB)
Samsung 860 PRO $68.80 (27¢/GB) $99.99 (20¢/GB) $199.99 (20¢/GB) $379.99 (19¢/GB) $729.99 (18¢/GB)
WD Blue 3D NAND $40.48 (16¢/GB) $59.99 (12¢/GB) $97.99 (10¢/GB) $199.99 (10¢/GB) $442.99 (11¢/GB)
Crucial MX500 $48.99 (20¢/GB) $53.99 (11¢/GB) $104.99 (10¢/GB) $209.99 (10¢/GB)  
SK hynix Gold S31 $43.99 (18¢/GB) $56.99 (11¢/GB) $104.99 (10¢/GB)    
NVMe
Samsung 970 EVO Plus $59.99 (24¢/GB) $79.99 (16¢/GB) $164.99 (16¢/GB) $320.44 (16¢/GB)  
SK hynix Gold P31   $74.99 (15¢/GB) $134.99 (13¢/GB)    
Sabrent Rocket Q   $64.99 (13¢/GB) $109.98 (11¢/GB) $219.98 (11¢/GB) $599.98 (15¢/GB)
WD Blue SN550 $42.99 (17¢/GB) $59.99 (12¢/GB) $109.99 (11¢/GB) $224.99 (11¢/GB)  

Now that its successor is out, the Samsung 860 EVO will eventually be going away, but it’s likely to still be in stock with major retailers for at least several months, and with third-party sellers for much longer. For now, the 860 EVO is cheaper than the 870 EVO for all but the smallest capacity, and that makes the 860 the smarter buy. But as Samsung transitions more fab capacity to their 128L TLC, this situation will change. (The 860 EVO also manages to be priced quite well against the 870 QVO, which really should offer more than just $10 savings at 2TB.)

Other major brands like Western Digital, Crucial and SK hynix offer great SATA SSDs that are generally cheaper than Samsung’s 870 EVO. Samsung’s performance advantages are too slight to justify any significant price premium. I also don’t think that Samsung’s reputation for quality is so much stronger than these competitors that Samsung should be charging $25 more at 1TB and $40-50 more at 2TB compared to eg. Western Digital.

The decline of the SATA SSD market broadly will take at least a few more years. But Samsung’s niche as the premium choice within the SATA SSD market is shrinking much more quickly. If you want to spend a bit more to get a nicer than average SSD, the obvious route it to spring for a decent NVMe SSD that at least offers the possibility of being noticeably faster. But if you just need another terabyte or two of good-enough storage in a system where space is getting tight, there area a variety of cost-effective models with similar performance that fit the bill.