• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FRONTPAGE Corsair MP700 2 TB PCIe Gen5 x4 NVMe M.2 SSD Review

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Overclockers.com

Member
Joined
Nov 1, 1998
Featured-scaled.jpg
With PCIe Gen 5 SSDs finally rolling out into the mainstream market, Corsair was kind enough to send out its flagship MP700 for review. This represents the very best Corsair (or anyone) has to offer for M.2 storage. It boasts the new NVMe 2.0 interface, 2 TB of capacity, and a blazing fast 10,000 MB/s sequential read and write speeds. In this review, we see just how much of an improvement this PCIe Gen5 drive is when compared to its Gen4 predecessors. Finally, we will answer the long-standing question: Is it worth the extra cost to upgrade?
Click here to read more!
 
I have always thought it strange that people put so much emphasis on sequential tests and seem to minimize random one. The author of the review thought the random result were an anomaly because they were bad for the MP700. It's possible the results are real and that the controller is not optimized for random read/writes?

I now pay more attention to random results because they better reflect real world use. I just replaced one drive with another. The 2nd drive only had slightly faster sequential read/write results than the 1st but much faster random read/write results. In use the 2nd drive booted must faster than the 1st and was much faster at the desktop. This showed that the much faster random read/write speeds made the difference.
 
It's possible the results are real and that the controller is not optimized for random read/writes?
Anything is possible with new drives, but we actually found an issue in our testing.

Cliff's - CDM was updated for this review, and we figured out one of the benchmark's tests changed after digging deeper into the anomaly. It appears the MP700 ran a different test than all the other drives in the 4K random. The other results appear to be within variance, but we're double-checking everything.

We'll add an editor's note and an updated chart ASAP. :)
 
Yes the third test default used to be RND4k Q32 T16 for CDM v7.0.X and the new default vor v8.0.4 is RND4k Q32T1. So the test was run on single thread. I'll be re-running the benchmarks and updated this review shortly.
 
I re-ran the Crystal Disk Mark benchmark with the original default settings and these are the results. We will be switching the format slightly for the next NVMe review and I will run this drive with those settings too so that we are back to a level playing field. (as level as technology will allow that is)

CMD-SEQ1M-Q8T1-Updated.png
CMD-SEQ1M-Q1T1-Updated.png
CMD-RND4K-Q32T16-Updated.png
.
CMD-RND4K-Q1T1-Updated.png

Sorry for the inconvenience.
 
Back