The Samsung 960 EVO (1TB) Review
by Billy Tallis on November 15, 2016 10:00 AM ESTAnandTech Storage Bench - Heavy
Our Heavy storage benchmark is proportionally more write-heavy than The Destroyer, but much shorter overall. The total writes in the Heavy test aren't enough to fill the drive, so performance never drops down to steady state. This test is far more representative of a power user's day to day usage, and is heavily influenced by the drive's peak performance. The Heavy workload test details can be found here.
The 960 EVO's average data rates on the Heavy test are slower than the 950 Pro and 960 Pro, but on par with the OCZ RD400 and faster than the Intel 750.
The 960 EVO takes third place for average service times, providing lower latency than the smallest 950 Pro despite slower overall data rates. In comparison to SATA SSDs, the latency differences are all pretty minor.
Like the 960 Pro, the 960 EVO oddly has slightly fewer high-latency outliers when this test is run on a full drive instead of a freshly-erased drive. In spite of this quirk of the drive's garbage collection routines, both drives have well-controlled latency.
The 960 EVO's power efficiency on the Heavy test is virtually the same as the 960 Pro and the 950 Pro, and not significantly worse than the fastest SATA drives.
Facebook
Twitter
RSS
87 Comments
View All Comments
ex_User - Tuesday, November 15, 2016 - link
Vaporware. 'Nuff said.Magichands8 - Tuesday, November 15, 2016 - link
This is magnificent! Not only has Samsung produced an SSD that under performs its own previous generation product but one that manages to do so while using even MORE power at an even HIGHER price per GB! They even put it on a form factor that makes the drive almost entirely irrelevant! The only thing missing is a feature that makes the modules randomly explode upon contact with the users computer.Dug - Tuesday, November 15, 2016 - link
What are you talking about?BrokenCrayons - Tuesday, November 15, 2016 - link
You have a unique perspective. :)Daggoth - Tuesday, November 15, 2016 - link
I have a question, isn't the z97 chipset capped at 2GB per second due to DMI 2.0? Isn't this a problem for the max sequential reads?Billy Tallis - Tuesday, November 15, 2016 - link
I test PCIe SSDs in the primary PCIe 3.0 x16 slot, because the riser card used for power measurement is a 16-lane low-profile card.Bullwinkle J Moose - Tuesday, November 15, 2016 - link
The 960 Pro is much better price over the duration of WarrantyI can generally kill 1 out of 4 SSD's within the warranty period
so if I buy 4 960 Pro's and 4-960 EVO's, here is the breakdown @ 500GB
4-960 Pro's = $330 X 4 or $1320 divided by 5yr warranty = $264 per year for 5 years
or
4-960 EVO's = $250 X 4 or $1000 divided by 3 years = $333.33 per year for 3 years
per year cost under warranty is WAY better for the PRO!
3 year warranty with TLC just doesn't do it for me
Bullwinkle J Moose - Tuesday, November 15, 2016 - link
I would REALLY need to be trying to kill at least 1 out of 4 but I could prolly do itSo tell me more about the internal speed Billy.....
How many seconds does it take to copy and paste 100GB to and from the same 960EVO?
and from the 960PRO?
shabby - Tuesday, November 15, 2016 - link
Why wasn't the 256gb version tested? Tom reviewed it and it was kinda meh compared to the rest of the mlc drives, it was as bad as the 600p in some cases.Billy Tallis - Tuesday, November 15, 2016 - link
The 250GB was tested. It died. See page 1 for details.