Final Words

The Phison E7 NVMe SSD platform has provided us with a very interesting case study of the effects SSD controller firmware can have on the performance of drives with almost identical hardware. All Phison E7 drives on the market use Toshiba 15nm MLC NAND. The firmware has evolved significantly since the first retail release in the spring of 2016, but it has not produced an across the board improvement in performance.

On our ATSB tests of real-world desktop storage workloads, the NX500's best showing was on the Heavy test, the most write-intensive of the three. Digging deeper, our ATSB tests show the NX500 is generally slower than its siblings for writes, though often fastest of the three for reads. This is not where we expected its strengths to lie, though the benefits of the large spare area do show up in the relatively small performance hit the NX500 suffers when the tests are conducted on a full drive.

The NX500 is at its best with sustained high queue depths. It's reasonable for a drive with this much overprovisioning to take other measures to optimize for heavy workloads, but clearly the NX500 overshot any sensible consumer workload target. Even the heaviest desktop workloads don't reach QD32 very often, and their overall performance is determined primarily by how the drive behaves at low queue depths. At lower queue depths, the NX500 mostly fails to deliver.

Our synthetic tests mostly mirror the ATSB tests in showing lackluster write performance compared to how the NX500 ranks on the read speed tests. The sequential write speed of the NX500 is pretty good in the grand scheme of things, but the other two Phison E7 drives are slightly faster still.

The Corsair Neutron NX500 consistently scores poorly on power efficiency. Since it is a desktop-only drive and only consumes a few Watts at most, this is insignificant. One contributing factor is that the NX500 has twice as much DRAM as is typical for its flash capacity, providing a small but constant extra power draw that apparently doesn't do much for performance.

I suspect the firmware used on the NX500 borrows some more from Phison's enterprise SSD firmware than the Patriot Hellfire's firmware does. The Hellfire's performance clearly suffers greatly when the drive is full, more so than either of the other two Phison E7 drives we've tested, and more than most MLC SSDs. The Patriot Hellfire's ranking tends to be better on our short burst I/O tests at QD1 than on the sustained tests. All of those are common characteristics to see on consumer drives that sacrifice some high-end performance for the sake of better real-world performance. The Corsair Neutron NX500 isn't a clear loser on all of the real-world and low queue depth tests, indicating that it hasn't completely sacrificed consumer performance optimization in the pursuit of higher synthetic benchmark scores.

  250-256GB 400-512GB 800-1024GB 1.6-2TB
Corsair Neutron NX500   $319.99 (80¢/GB) $649.99 (81¢/GB) TBA
Samsung 960 EVO $142.84 (51¢/GB) $234.00 (47¢/GB) $477.99 (46¢/GB)  
Samsung 960 PRO   $299.99 (59¢/GB) $579.99 (57¢/GB) $1129 (55¢/GB)
Intel SSD 600p $165.59 (65¢/GB) $212.99 (42¢/GB) $355.00 (35¢/GB)  
WD Black $109.99 (43¢/GB) $198.98 (39¢/GB)    

While the tradeoffs of different Phison E7 firmwares are interesting, they're not too relevant to the current state of the market. All consumer PCIe SSDs using planar MLC NAND flash are squeezed between cheap TLC drives like the WD Black and Intel 600p, and Samsung's 960 EVO which offers better real-world performance than pretty much everything except the 960 PRO. At the moment, the price spread is a mere $35: from 39¢/GB for the WD Black up to 47¢/GB for the 960 EVO. It's hard to argue that there's any room for a product to carve out a niche somewhere in that small range. Based on performance alone, the Corsair Neutron NX500's MSRP is about twice what its actual retail price ought to be. But even with a massive price cut, the NX500 will need to rely on aesthetics and brand loyalty to sell.

Power Management
Comments Locked

45 Comments

View All Comments

  • Kristian Vättö - Wednesday, August 16, 2017 - link

    There are numerous 8-lane enterprise SSDs already.
  • hlm - Wednesday, August 16, 2017 - link

    e.g. HGST FlashMAX III and HGST Ultrastar SN260 products are eight-lane devices.
  • The_Assimilator - Wednesday, August 16, 2017 - link

    Hey look, another SSD that has no reason whatsoever to exist!

    I don't understand why manufacturers don't, y'know, try to COMPETE with Samsung instead of re-re-releasing the same old, tired, slow controllers with slightly different but ultimately insignificant spins on them.
  • DanNeely - Wednesday, August 16, 2017 - link

    Because unless you have a billion dollars to spend and a few years to wait, you can't create your own controller. That means almost all of the other companies selling drives have to pick and choose between a handful of controllers made by Phison/etc. Until they recover from Samsung's blind siding them and design new higher performing architectures from the ground up none of them have anything in the same performance class. If what happened at the start of the market when Intel's controllers were unbeatable is any indication we should hopefully have competitive designs available in another year or so.
  • FunBunny2 - Wednesday, August 16, 2017 - link

    -- Because unless you have a billion dollars to spend and a few years to wait, you can't create your own controller.

    well, isn't a controller an implementation of physics and math? which is to say, unless something new happens with NAND chips (not just node size or xLC), may haps we've reached the one-true-answer to the controller problem? may be there's just no more there, there.
  • Samus - Wednesday, August 16, 2017 - link

    Wow. That was disappointing.
  • RaistlinZ - Wednesday, August 16, 2017 - link

    Current Newegg Prices:

    1. 500GB Samsung 960 Pro = $299.99
    2. 1TB Samsung 960 Pro = 600.82

    The NX500 has no reason to exist. The price needs to be cut in half to make it even REMOTELY attractive.
  • alpha754293 - Wednesday, August 16, 2017 - link

    I'm surprised you didn't bother comparing it against the Intel 750 Series 400 GB PCIe NVMe SSD.
  • Billy Tallis - Thursday, August 17, 2017 - link

    I had originally planned to include the 400GB 750, but some of the results from it looked funny and I decided it wasn't worth postponing the review for several days to re-test the 750. That drive's a pain to test, because I have to run each test twice in order to record the power on both the 3.3V and 12V lines, and the performance has to match between the two runs for the results to be valid.
  • alpha754293 - Friday, August 18, 2017 - link

    Depending on how you want to tackle/handle it.

    There are statistical methods available out there that even with noisy data (e.g. high standard deviations) that you can still use it to process data that might otherwise not make sense at first glance, on the surface.

    Course, that would also mean that care would need to be taking so that the tests in and of itself are repeatable.

    I only mention it because I would be VERY interested to see how this compared to the Intel 750 series.

    Thanks.

Log in

Don't have an account? Sign up now