Samsung SM951 512GB M.2 PCIe SSD Review

A place to give your thoughts on our reviews!
Post Reply
User avatar
Apoptosis
Site Admin
Site Admin
Posts: 33941
Joined: Sun Oct 05, 2003 8:45 pm
Location: St. Louis, Missouri
Contact:

Samsung SM951 512GB M.2 PCIe SSD Review

Post by Apoptosis »

The Samsung SM951 M.2 PCIe SSD has been the talk of the enthusiast community for months and it has been without a doubt one killer looking drive. We've been e-mailing and bugging our contacts at Samsung for a sample of this drive since 2014 and we finally was shipped a brand new drive directly from Samsung to try out! Read on to see how it performs!

Image


Article Title: Samsung SM951 512GB M.2 PCIe SSD Review
Article URL: http://www.legitreviews.com/samsung-sm9 ... iew_161689
Pricing At Time of Print: $499.95 Shipped
Find us on Facebook to discover the faces behind the names!
Follow Me on Twitter!
MacGyverSG1
Legit Little One
Legit Little One
Posts: 2
Joined: Sun May 31, 2015 7:49 pm

Re: Samsung SM951 512GB M.2 PCIe SSD Review

Post by MacGyverSG1 »

Great review. I like that it was compared to the Kingston and Plextor drives, but why are the IOMeter tests done using a queue depth of 32? Most people using one of these drives would never go higher than a queue depth of 4 while actually a queue depth of 1-2 is most common. I know the SSD manufacturers use a queue depth of 32 to demonstrate their products performance because it provides flashier numbers for marketing, but it means nothing to most people. A graph showing the performance at different queue depths (0-32) would be great.
User avatar
Apoptosis
Site Admin
Site Admin
Posts: 33941
Joined: Sun Oct 05, 2003 8:45 pm
Location: St. Louis, Missouri
Contact:

Re: Samsung SM951 512GB M.2 PCIe SSD Review

Post by Apoptosis »

MacGyverSG1 wrote:Great review. I like that it was compared to the Kingston and Plextor drives, but why are the IOMeter tests done using a queue depth of 32? Most people using one of these drives would never go higher than a queue depth of 4 while actually a queue depth of 1-2 is most common. I know the SSD manufacturers use a queue depth of 32 to demonstrate their products performance because it provides flashier numbers for marketing, but it means nothing to most people. A graph showing the performance at different queue depths (0-32) would be great.
Thanks for the feedback. I was trying to set my tests up at that level as I knew NVMe takes off at higher queue depths and to try to match the manufacturers rated specifications. Basically to make sure the flashy numbers are correct.

In the future I might re-think the testing methodology a bit. After tens of thousands of people have read the review you are so far the only one to request lower QD testing on the forums or by e-mail feedback. I agree with the importance of it, but wish I had more time in the day!
Find us on Facebook to discover the faces behind the names!
Follow Me on Twitter!
Post Reply