The ADATA Ultimate SU800 SSD Review (128GB, 256GB, 512GB)
by Billy Tallis on February 1, 2017 12:01 PM ESTPerformance Consistency
Our performance consistency test explores the extent to which a drive can reliably sustain performance during a long-duration random write test. Specifications for consumer drives typically list peak performance numbers only attainable in ideal conditions. The performance in a worst-case scenario can be drastically different as over the course of a long test drives can run out of spare area, have to start performing garbage collection, and sometimes even reach power or thermal limits.
In addition to an overall decline in performance, a long test can show patterns in how performance varies on shorter timescales. Some drives will exhibit very little variance in performance from second to second, while others will show massive drops in performance during each garbage collection cycle but otherwise maintain good performance, and others show constantly wide variance. If a drive periodically slows to hard drive levels of performance, it may feel slow to use even if its overall average performance is very high.
To maximally stress the drive's controller and force it to perform garbage collection and wear leveling, this test conducts 4kB random writes with a queue depth of 32. The drive is filled before the start of the test, and the test duration is one hour. Any spare area will be exhausted early in the test and by the end of the hour even the largest drives with the most overprovisioning will have reached a steady state. We use the last 400 seconds of the test to score the drive both on steady-state average writes per second and on its performance divided by the standard deviation.
The 512GB ADATA SU800 delivers lower steady-state random write performance than the 525GB Crucial MX300 despite the latter having less built-in overprovisioning. The 256GB SU800 isn't the slowest in its class, but the 128GB is in last place with barely half the performance of the 120GB Samsung 750 EVO. The Silicon Motion engineering sample with more overprovisioning outperformed the SU800 with twice the capacity.
The performance consistency scores from the SU800 are typical for a low-end SSD, but there's clearly room for improvement. Samsung's 750 EVO scores far higher than any of the other budget SSDs, and the Crucial MX300 has the next highest score.
 |
|||||||||
| Default | |||||||||
| 25% Over-Provisioning | |||||||||
For the first few seconds of the test, the SU800 delivers close to the advertised IOPS, then it drops into a fairly typical pattern of widely variable performance above 5k IOPS as it burns through the spare area. Once the SLC cache and spare area are full, the SU800 spends most of its time slowed down by garbage collection, with regular short bursts of higher performance.
 |
|||||||||
| Default | |||||||||
| 25% Over-Provisioning | |||||||||
The garbage collection cycles on the SU800 appear to last about a minute each, separated by several seconds of faster performance. Overall performance and variability are higher for the larger capacities.
Facebook
Twitter
RSS
35 Comments
View All Comments
Arbie - Thursday, February 2, 2017 - link
@eek2121: Anandtech has changed most of its drive tests in the past two years, but it's easy to compare the Mushkin Reactor to this Adata in IOMeter 4KB random reads & writes. In these, the Mushkin is 50% faster. So please provide your basis for saying it's "drastically slower". While it may "soon disappear", that's been true for the past 18 months and it's still there, at $240. And FYI "clickbait" is the swathe of pandering pix labeled "From the Web". Anandtech provides these to raise its status as a on-line source of repute and save us the trouble of looking for garbage on our own.Billy Tallis - Thursday, February 2, 2017 - link
The Mushkin Reactor was one of the last drives tested with our 2013 test suite. Even though we used IOmeter back then, the test protocol was different. The 4kB random access numbers reported in that review were for QD3 using an 8GB test file, while our current 2015 test suite uses a 16GB file for random writes and the whole drive for random reads, and the score reported is the average of QD1, QD2, and QD4 performance. If the numbers were directly comparable, we'd be directly comparing them.TheinsanegamerN - Thursday, February 9, 2017 - link
"drastically"A few thousand IOPS is "drastically" now eh? For this market, they would never notice the difference.
eek2121 - Wednesday, February 1, 2017 - link
What is so click-baity about this? It's a review. Did you bother reading it? They came to the same conclusion you did. It's not that great. The Mushkin Reactor has been discontinued, why would they include it?lopri - Thursday, February 2, 2017 - link
That drive is not faster than what is benched. MLC does not automatically make a good SSD.Having said that, I do sometimes think AT neglect hardware of the past in their review, which is regrettable. Consumers not only compare current products but also upgrade their old gears, and they want to see if it is a worthy undertaking.
fanofanand - Wednesday, February 1, 2017 - link
Some might say I am being greedy or unrealistic, but I think that when SSD's hit 1TB for under $100, then we will finally be able to put the axe in mechanical hard drives with the exception of the 10+ TB drives for archiving or media libraries. If I could pick up a 2TB for under $200 I would buy it today. We are so close, yet so far.Great_Scott - Wednesday, February 1, 2017 - link
1TB drives were around $200 6 months ago.Most people seem to have around 3TB of data, not including any bloated Steam libraries, $600 over the course of a year (late 2015 to late 2016) seems reasonable, it's what I did.
Of course, the recent price increase probably blindsided most people, but if it was made a priority many enthusiasts don't need to be using spinning rust pretty much regardless of non-archive non-media (bulk) data.
But I keep seeing posts like yours. Does everyone want to only use 1 SSD in thier PC for some reason?
Laststop311 - Wednesday, February 1, 2017 - link
You have to not include blu ray disc image libraries either. I have a 6 bay NAS with 6x 4TB in raid 5 giving me 20TB of space 10TB of that is gone from 200 blu ray disc images and another 5TB from FLAC audio. When I have to upgrade all my disc images to 4k blu ray disc images they double in size + i will have more fo them. I will have to rebuild my nas with 6x 12TB drives at that point. So plenty of enthusiasts will still need spinning rust. For my next PC tho I plan on using a 128-256GB optane drive for the OS and most used productivity apps and VLC media player and chrome and my favorite couple games. A 1TB nvme drive for the majority of my most played games and 2x 2TB SATA 3 2.5" SSD's that my torrents will download onto and seed for awhile before being moved to the NAS or installed on a faster drive or deleted.The_Assimilator - Thursday, February 2, 2017 - link
"Does everyone want to only use 1 SSD in thier PC for some reason?"Why use a whole bunch of drives if you can only use 1? Saves on cable clutter, physical space (important for the rise in SFF systems) and just makes things simpler overall. Personally I can't wait for 4TB SSDs to become mainstream in price so that I can replace my mechanical disk.
JimmiG - Thursday, February 2, 2017 - link
I have about 1.2 TB of SSD storage spread over 240 GB, 480 GB and 500 GB SSD's, bought in the last couple of years. 1.2 TB in a single drive would be much more convenient since you wouldn't have to juggle your game installations and other files to keep the amount of free space on each SSD reasonable and balanced.After I got a NAS, I only have an old 1 TB HDD (in addition to the SSDs) in my PC for general data storage and less frequently used applications and games. I don't think "spinning rust" is going anywhere, but I think the smaller capacities (<3TB) are going away. People want them for bulk storage in NAS devices or as external USB drives.