The Mainstream Phoenix Rises: Samsung's 970 EVO (500GB & 1TB) SSDs Reviewed
by Billy Tallis on April 24, 2018 10:00 AM ESTAnandTech Storage Bench - The Destroyer
The Destroyer is an extremely long test replicating the access patterns of very IO-intensive desktop usage. A detailed breakdown can be found in this article. Like real-world usage, the drives do get the occasional break that allows for some background garbage collection and flushing caches, but those idle times are limited to 25ms so that it doesn't take all week to run the test. These AnandTech Storage Bench (ATSB) tests do not involve running the actual applications that generated the workloads, so the scores are relatively insensitive to changes in CPU performance and RAM from our new testbed, but the jump to a newer version of Windows and the newer storage drivers can have an impact.
We quantify performance on this test by reporting the drive's average data throughput, the average latency of the I/O operations, and the total energy used by the drive over the course of the test.
The Blue lines indicate the PM981, the OEM version of the 970 EVO.
The Orange lines are the 970 EVO.
The average data rates from the Samsung 970 EVO on The Destroyer are a slight step backwards from the Samsung PM981 OEM drive and from the 960 EVO. All of the TLC-based drives are still performing below even Samsung's older MLC-based NVMe drives, and of course the Intel Optane SSD. This year's Western Digital WD Black offers about the same performance as the 970 EVO.
Average and 99th percentile latencies for the 970 EVO are again very slightly worse than the PM981, but on these metrics the 960 EVO doesn't beat its replacement. The WD Black has notably better 99th percentile latency than the other flash-based SSDs.
There is a clear range of average read latency scores that make up the high-end NVMe market segment. The 970 EVO doesn't stand out from the other drives in that category. For average write latency, scores vary a lot more, and the 970 EVO outperforms its predecessor slightly but fails to match the very good score the PM981 obtained.
The 99th percentile read and write latency scores from the 970 EVO don't break new ground and mostly fail to match the PM981, though the differences aren't large enough to be a serious concern. The WD Black's notable QoS advantage is on the read side, where it is the only flash-based SSD to almost always keep read latency below 1ms.
We didn't have the opportunity to measure power usage of the Samsung PM981 on The Destroyer, so this is our first look at the power draw of the Samsung Phoenix controller on this test. The situation isn't good. The 970 EVO uses twice the energy as the WD Black to does despite both drives offering about the same level of performance on The Destroyer. The power efficiency of the 970 EVO seems to be a big step backwards from the previous generation and is not at all competitive.
Facebook
Twitter
RSS
68 Comments
View All Comments
bji - Tuesday, April 24, 2018 - link
You're kind of arguing against benchmarking in general here. Almost no benchmarks are directly relevant to any one person's intended use of the product. Benchmarks are not useful in that they tell me exactly how much performance to expect when running one specific program on one specifically configured hardware setup. They are useful because they allow extrapolation from measured results to expected results on workloads that actually matter to the reader.So I don't agree with your sentiment that Meltdown/Spectre are not worth consideration for their effect on system performance.
However, I am not sure that I would include Meltdown/Spectre considerations in a specific SSD review. I think these considerations deserve to be in a CPU review.
bji - Tuesday, April 24, 2018 - link
Also, may I say that users generally will not notice a 5% slowdown in any particular task; however, we've already established that readers care about minimum differences in benchmark results, because they routinely call a 5% difference clear indication of a "winner" and a "loser" for that benchmark, so for the purposes of performance reviews, the 5% difference contributed by Meltdown/Spectre definitely matters.Flying Aardvark - Tuesday, April 24, 2018 - link
It's up to 50% reduction in storage performance not 5%. You'll feel 50% loss when it happens to you.cmdrdredd - Tuesday, April 24, 2018 - link
What you are saying is misleading. SATA performance is nearly identical (within 2% difference for me). It's NVMe drives that take the hit, but even still they are faster than everything else. Processor speed is unaffected for me as well. Tested multiple times with various benchmarks both ways and it was within margin of error. I don't see the problem to be honest.LurkingSince97 - Wednesday, April 25, 2018 - link
Tell that to my I/O intensive servers that suddenly have 30% less throughput.modeonoff - Tuesday, April 24, 2018 - link
Yes but I am not an average customer. Performance is important for me.Ryun - Tuesday, April 24, 2018 - link
For everyday tasks do you guys notice an improvement in responsiveness of NVMe SSDs versus SATA SSDs?The transfer rates are definitely impressive, I've just never seen a review where I've wanted to upgrade my 500GB SATA SSD for development/gaming/maintenance tasks on my machine. Seems like boot times and opening programs are within a couple seconds of another between NVMe and SATA. Nothing like the jump between HDDs vs SSDs.
HollyDOL - Tuesday, April 24, 2018 - link
I wonder myself, got Vertex 3 (240GB) and while not permanently watching perf counters I don't see much cases of 100% load. Wonder if I would be able to see a difference if I moved to some "best enthusiast m.2/pcie ssd available". (Rest of the machine is fully capable)eek2121 - Tuesday, April 24, 2018 - link
I notice it in certain tasks. My system can get from cold boot to the login screen in about 3 seconds for instance. Editing video is much faster as well.imaheadcase - Tuesday, April 24, 2018 - link
I wouldnt say a huge performance, it really depends on certain tasks that you work with. If you work with file manager a lot with big files sure. But most people no. It makes sense if just upgrading though.