Sequential Write Performance

The sequential write test isn't limited to a small span of the disk, as that usually doesn't make a difference for this performance metric. As always, our averages are of the lower queue depths, but scaling to higher queue depths is also investigated. Bulk file copies and recording uncompressed video are the kind of uses that depend on sequential write performance.

Iometer - 128KB Sequential Write

Even at low queue depths, most SSDs can spread a sequential write workload across multiple channels to achieve very high throughput. All of the PCIe drives perform well above the SATA limit but nowhere near the capacity of the PCIe links. The 512GB 950 Pro seems to again be encountering thermal throttling, and more severely than the SM951s we tested.

Iometer - 128KB Sequential Write (Power)

The average power consumption for writes is still on the high side by SATA standards, but is quite reasonable in light of the performance achieved.

Samsung 950 Pro 256GB

Performance across the sequential write test is mostly constant, but power isn't. The declines in power consumption in spite of increased queue depth are again most likely an artifact of whatever background processing is going on in the drive. On SATA drives, it's usually obvious from looking at the power meter when a drive has completed background processing and is truly idle and ready to run the next test, but with the lack of working power management for PCIe drives on our testbed I'm not so sure here. Overall it's probably a good sign that the drives were able to maintain high performance in spite of whatever else was going on, but given more cooling the 512GB can probably do much better.

Sequential Read Performance

Just like the random read test, sequential reads are tested across the span of a full drive and a representative sample of queue depths. Most operations involving large files (typically images and videos) fall under this test's purveiw, but streaming or playing even the highest resolution videos won't be a challenge for any of these drives. Copying files to another SSD or loading a new level in a video game would be more likely to show noticeable difference from better performance here.

Iometer - 128KB Sequential Read

The sequential read performance is probably the best showcase of what the PCIe 3.0 x4 interface can do. The 256GB 950 Pro attains over half of the link speed, but the 512GB is again bogged down by something—relatively speaking, since it's still more than twice the speed of SATA and faster than even the Intel SSD 750. It's possible that the 950 Pro isn't faithfully implementing the secure form of the NVMe format command and some lingering fragmentation is preventing the 512GB drive from performing as specified. Read operations require less power to be supplied to the flash chip than for writes, but if the thermal throttling is all in the controller it could be showing up here as well.

Iometer - 128KB Sequential Read (Power)

Relative to each other, the 950 Pro's power consumption is in line with the performance they're demonstrating, and proportionally much better than the SATA drives.

Samsung 950 Pro 256GB

At queue depth 1, both of the 950 drives are performing similarly. When moving to higher queue depths the 256GB immediately hits 104% of its rated speed, but the 512GB doesn't improve any. (It's worth noting the differing scale for Transfer Rate in the graphs above.) The puzzling decline in power consumption at higher queue depths shows up again on the 256GB and definitely warrants deeper investigation.

Random Performance Mixed Read/Write Performance
Comments Locked

142 Comments

View All Comments

  • Per Hansson - Wednesday, October 28, 2015 - link

    If you really need it FXi there is always the announced Samsung SM953 drive.
    It's 110mm long due to the inclusion of the tantalum capacitors, otherwise it's very similar to the SM951...
    http://www.samsung.com/semiconductor/global/file/i...
  • R3MF - Sunday, October 25, 2015 - link

    i was under the impression that win7 (install disk) does not support nvme, so i'm curious as to how you went about getting Win7 on a 950?
  • Kristian Vättö - Sunday, October 25, 2015 - link

    There is a hotfix NVMe driver available for Windows 7: https://support.microsoft.com/en-us/kb/2990941
  • blos - Monday, October 26, 2015 - link

    I slipstreamed that hotfix (and almost 200 other fixes) using NTLite to a Win7SP1 image, wrote it back to usb, booted and it would still not allow me to install to a 951 NVMe drive... no drives found.

    Has anybody got this to work?

    Windows 10 installs just fine but I have an unused W7 Pro license and I would really like to use it to active a W10 install.

    I hear the next version of W10 will activate directly from W7 licenses... but hopefully that'll arrive in time before the W10 will really want the activation. Or perhaps re-arming to extend a bit?
  • Per Hansson - Wednesday, October 28, 2015 - link

    Yes it works but you need to integrate it in boot.wim as well.
    I made some details here about it, also nothing an errors that MS is still to fix in that KB article.
    Even though I reported it a long time ago:
    http://www.overclock.net/t/1543242/found-samsung-s...
  • blos - Wednesday, October 28, 2015 - link

    Ah that explains a lot, didn't know that the boot.wim was a different environment :)

    Quick search indicates that boot.wim integration is already available on the NTLite, so I guess it's playtime for this weekend :)
  • catavalon21 - Wednesday, October 28, 2015 - link

    The test system doesn't have a discrete video card. Would a high-powered video card impact the performance of the M.2 PCIe setup?
  • AnnonymousCoward - Wednesday, October 28, 2015 - link

    No.
  • Caramonn - Friday, October 30, 2015 - link

    I've read the review and others and I guess I don't see a reason to get one of these drives yet. Am I missing something? It seems that the real world performance doesn't justify the nearly twice the cost as other Samsung SATA drives. I was really hoping that the rated speeds would translate into actual real world performance, but that doesn't appear to be the case.
  • Tuishimi - Thursday, November 5, 2015 - link

    $350... I could live with that. Sounds like a decent piece of hardware.

Log in

Don't have an account? Sign up now