Corsair MP700 Elite SSD review: The retail, budget PCIe 5.0 SSD of choice is officially here

Corsair is quick off the line with a practical next-gen drive

Corsair MP700 Elite SSD
(Image credit: © Tom's Hardware)

Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.

Comparison Products

The Corsair MP700 Elite is one of the first retail drives based on the E31T controller, although we have the Addlink G55 waiting in the wings. (Spoiler alert: It's effectively the same performance, just without the Corsair brand name.) The MP700 Elite isn't intended to hang with drives built on the faster Phison E26 controller, or at least not the higher-end variants that came after the initial launch of 10 GB/s PCIe 5.0 SSDs. The Crucial T705 shows what that latter controller can do, with no comparable drive available with SMI’s SM2508 just yet but with the Adata Legend 970 Pro available using InnoGrit’s competing IG5666 controller.

Drives that might be in more danger include the most popular, high-end PCIe 4.0 SSDs. These consist of the Samsung 990 Pro, the WD Black SN850X, and the SK hynix Platinum P41. It may be more fair to compare the MP700 Elite with the four-channel Crucial T500, but remember that it and the previous three drives listed all have DRAM. There’s some question about whether or not DRAM is needed today, but arguably it's still useful for workhorse drives.

Directly comparable drives would be DRAM-less and these include the Samsung 990 EVO Plus and the Silicon Power US75, the latter representing a host of popular drives including the Teamgroup MP44, the Lexar NM790, and the Addlink A93. At least one of these drives, the MP44, has been found with the E27T and BiCS6 flash instead of the launch MAP1602 and 232-Layer YMTC TLC flash, for which Corsair also has an answer with its Corsair MP600 Elite. It’s also possible to use QLC flash, as is the case with the Crucial P310, but such drives should be less expensive with a heavier focus on capacity. 

Latest Videos FromTom's Hardware
Shane Downing
Freelance Reviewer

Shane Downing is a Freelance Reviewer for Tom’s Hardware US, covering consumer storage hardware.

With contributions from
  • HideOut
    all th is talk about power effeciency, and no graphs or other data to suppport it?
    Reply
  • phxrider
    What am I missing? From what I see, it's barely faster than the PCIe 4 drives, and more expensive. Actually, it's even more expensive than the higher end, better performing Teamgroup 540.
    Reply
  • Mama Changa
    phxrider said:
    What am I missing? From what I see, it's barely faster than the PCIe 4 drives, and more expensive. Actually, it's even more expensive than the higher end, better performing Teamgroup 540.
    Well, the author did say prices were way too high.

    I'd get the T500 ahead of this.
    Reply
  • Maxxify
    HideOut said:
    all th is talk about power effeciency, and no graphs or other data to suppport it?
    682.4 MB/s per watt. This is over 50% more efficient than the 2TB Gold P31, if that puts things into perspective.
    Reply
  • Stomx
    phxrider said:
    What am I missing? From what I see, it's barely faster than the PCIe 4 drives, and more expensive. Actually, it's even more expensive than the higher end, better performing Teamgroup 540.
    When older gen 4.0 drives deliver 7+ GB/s this one with 8.5 does not look particularly attractive
    Reply
  • bit_user
    Maxxify said:
    682.4 MB/s per watt. This is over 50% more efficient than the 2TB Gold P31, if that puts things into perspective.
    I'd like to know how well it does when limited to PCIe 4.0 or 3.0 speeds. There are plenty of systems out there that only support up to PCIe 3.0 and the P31 Gold is a PCIe 3.0 drive.
    Reply
  • bit_user
    Stomx said:
    When older gen 4.0 drives deliver 7+ GB/s this one with 8.5 does not look particularly attractive
    What I find more alarming is the QD1 read performance vs. block size:

    The graph doesn't say, but I'm assuming this is random reads. It falls way behind after 4k and doesn't really catch up until 1M.

    I'd also like to see how random read performance scales vs. queue depth at 4k, 64k, and 1M sizes. If we had that, we wouldn't need the bar charts showing the same data points.

    I think less emphasis could be placed on write benchmarking. Obviously, there needs to be some, but why do I care about peak sequential write at QD8? Who would do that? Why? And do you not see that graph is almost exactly the same as the peak sequential write at QD1?

    @JarredWaltonGPU
    Reply
  • Maxxify
    bit_user said:
    I'd like to know how well it does when limited to PCIe 4.0 or 3.0 speeds. There are plenty of systems out there that only support up to PCIe 3.0 and the P31 Gold is a PCIe 3.0 drive.
    Should be even more efficient. Would be cool to see it tested at lower PCIe speeds.
    Reply
  • bit_user
    Maxxify said:
    Should be even more efficient. Would be cool to see it tested at lower PCIe speeds.
    Yeah, I wouldn't be surprised if it were more efficient, but I really don't know how much of its efficiency advantage comes from higher speeds vs. lower power. That's why I'm wondering.

    FWIW, I find bandwidth per Watt a slightly weird metric, when buying a drive for personal use. I care about performance and power. However, depending on what I'm doing, I'll tend to prioritize one above the other. If it's for a gaming desktop, I'd care about performance first and I'd probably only use power as a tie-breaker. For something like a laptop, I'd prioritize power but probably not to extent that I'd accept a major performance tradeoff.

    You could have two drives with virtually identical perf/W, but very different performance and power levels. I think the metric is basically useless, outside of datacenter applications. Even then, what you'd probably rather see is the number of Joules required to complete a defined workload.
    Reply
  • snapdragon-x
    Where is the Addlink G55 review? So, it's essentially the same product? It's 25% less expensive than the Corsair one where I live.
    Reply