Review Index:
Feedback

Plextor M6e 256GB SSD Review - speedy PCIe on the cheap

Subject: Storage
Manufacturer: Plextor
Tagged: ssd, plextor, pcie, 256GB

Introduction, Specifications and Packaging

Introduction:

In recent years, Plextor has branched beyond their renowned lines of optical storage devices, and into the realm of SSDs. They have done fairly well so far, treading carefully on their selection of controllers and form factors. Their most recent offerings include the M6S and M6M (reviewed here), and are based on Marvell controllers coupled with Toshiba flash. Given that the most recent Marvell controllers are also available in a PCIe variant, Plextor also chose to offer their M6 series in PCIe half height and M.2 form factor. These last two offerings are not simply SATA SSDs bridged over to PCIe, they are natively PCIe 2.0 x2 (1 GB/s), which gives a nice boost over the current SATA limit of 6Gb/sec (600 MB/sec). Today we are going to kill two birds with one stone by evaluating the half-height PCIe version:

View Full Size

As you can see, this is nothing more than the M.2 version on a Plextor branded interposer board. All results of this review should be identical to the bare M.2 unit plugged into a PCIe 2.0 x2 capable M.2 port on either a motherboard or mobile device. Note that those devices need to support the 2280 form factor, which is 80mm in length.

View Full Size

Here's the M.2 version installed on an ASUS X99-Deluxe, as tested by Morry.

Read on for the full review!

Specifications (from this page):

View Full Size

Packaging:

View Full Size

No frills packaging here. PCIe bracket mounting screw is included, but no half height bracket to speak of.


November 19, 2014 | 08:40 PM - Posted by Anonymous (not verified)

Allyn,

Would it be possible to start running tests which would state how long it would take an OS to load, a game to load, photoshop to startup, etc.?

I ask because every time I read one of your reviews, I'm always saying to myself, "this looks good on paper, but how noticeable would this be compared to what i already have?"

November 19, 2014 | 09:28 PM - Posted by Anonymous (not verified)

It seems with ssds that it can be difficult to include a benchmark that truly charts or demonstrates a clear separation amongst models. Then you have to try to cover as many bases.

I have the same opinion. It doesn't hurt to see more benchmarks. However, they have to have some meaning to them or it just becomes some random test to throw into a review.

November 20, 2014 | 03:25 AM - Posted by meganerd

The problem is that there are many other factors that do not involve the storage subsystem that contribute to these "tests". "Did Photoshop take longer to load because the network was down" is just one example.

The relevant tests that would impact the tasks that you mentioned are mainly the sequential reads. Chances are you are not going to be able to perceive the fractions of seconds difference between most of these devices when starting Photoshop. Measuring this another challenge as well. The differences between many of these devices exist inside of the margin of error of human response time, so you cannot simply use a stopwatch.

What does matter is understanding the workloads that we run. This helps us to focus on the measurements that matter in a given scenario.

If I am doing database work I might care about the random writes, IOPS, and queue depth differences for example.

November 23, 2014 | 10:30 PM - Posted by Allyn Malventano

The commenters have it right here. There are other factors that limit OS boots and application launches. Boots tend to be swayed a lot (seconds), varying by different drivers / option ROM prompt screens of PCIe SSDs, etc. It's just not an apples to apples comparison there. When it comes to application loading, that varies not only by SSD choice, but by number of cores, RAM, etc.

That doesn't mean you can't roughly translate our tests to real world 'feel' of various apps. Probably the best of our tests for pure sequentials are HDTune and our file creation and copy tests. Those will tend to mimic the performance of anything performing bulk reads, like moving large files around, scrubbing large video files, some game loads, etc.

The heavy random IO tests (Iometer) will show how large amounts of parallel IO (higher QD) impacts performance. The lower queue depths (1-2) will show random performance of single-threaded apps. Games with a single thread handling level loads (very likely), that tend to access the game data randomly (varies from game to game), will tend to run at QD=1. This is where SandForce sees a disadvantage, due to high per-IO latency. Higher queue depths show how the SSD will perform during large amounts of parallel IO, such as heavily threaded applications, or multiple simultaneous application launches (windows boot, etc). Basically any time you'd hear a hard drive thrashing badly, that's when the IO's are really stacked up.

I do have some things in the works to better show performance related to specific applications, but it won't be an end-all answer, and will only supplement the other tests.

November 20, 2014 | 08:40 AM - Posted by Prodeous II (not verified)

I have to agree with you. Even though the tests might have some issues as identified by other commenters.

To have a rough idea in real world performance like loading OS, Photoshop, a game, or other. Seeing real performance is much easier for the general population to understand which is better. And it is easier for us to use these "examples" to guide our family/friends with regards to purchase

November 20, 2014 | 09:38 AM - Posted by Anonymous (not verified)

I agree with this too.

IOPS, sequential/random throughput is nice for synthetic benchmarks, but real world examples of performance is more of what consumers will understand.

For example, application and OS load times would help a consumer pick the appropriate SSD for the money.

If one SSD can load Windows in 15 seconds that costs $200, but another can do it in 16 seconds that costs $150 (assuming same storage capacity), then it makes sense to save $50.

Naturally, other features such as capacitors and integrated encryption should also be a consumer's decision factor, but tangible, objective real world benchmarks are needed.

November 20, 2014 | 02:23 PM - Posted by Casecutter (not verified)

I like the other above see all the read/write/mb's, and say... will a couple of hundred dollars provide me a noticeable "kick seat of the pants" improvement to justify the cost?

While here's my aversion to adopting one, do you take an existing install and transfer the OS and other things that makes great use of an SSD, while maintain the other folder and files on the HDD all pointing to the right place? Or is this something you'd be better to start with fresh install on the SSD and restore files to the HHD

I do a good amount of reading and hardware savvy, but the nitty-gritty of the inner workings of the OS and such has alway been something I'd rather not mess with.

Does anyone know of a guide or article that outlines the step's to say install an SSD and cover what you would move/transfer and configure within the OS so files point to the proper drive when saving, updating etc. I've always thought it be great if you could install and SSD, then toss in a disc that prompts you to do this and that, for OS file to transfer automatically, indicate the drive hierarchy you then change to point to the files, etc. Perhaps I don't know what I'm talking about... (ya-thunk), and making a mountain out of nothing, but I believe that like most it's not a super simple thing to get set-up.

November 21, 2014 | 04:31 AM - Posted by Anonymous (not verified)

Have you forgotten about the Samsung XP941 - the obvious direct competitor?

bob Frost

November 23, 2014 | 10:32 PM - Posted by Allyn Malventano

Samsung has not sampled this unit for review.

April 5, 2015 | 10:58 AM - Posted by Robert Osorio (not verified)

Just wanted to let everyone know that I installed this card on my 5 year old Asus P7P55D LE which doesn't have SATA3 ports. My previous boot drive was aSamsung 840, but on this mobo it's been capped at around 250 MB/s according to CrystalDiskMark. With the Plextor M6e, I'm getting 622 MB/s! Nice upgrade for my old workstation.

HOWEVER, let me warn you all of a potential pitfall I ran into. My mobo has two x16 slots (I have no x2 slots and the rest are x1 - the Plextor needs x2 or you get half the throughput). However if there's a video card plugged into the PRIMARY x16 slot (as there was) then the secondary slot on my mobo defaults to x1 mode. So when I ran my initial benchmarks, I got a very disappointing 350 MB/s. I had to read through the terrible English of my mobo manual to figure out what was wrong. I swapped the cards (put the Plextor in the primary slot, and the video in the secondary). This allowed the Plextor to run in x2, and the video card is running in x4 on the secondary, which is just fine as this is not my gaming system.

So bottom line is that this is a nice SSD upgrade for an older system that doesn't have SATA3 ports - assuming you have a REAL x2 PCI-E slot (or better) available. Check your mobo documentation.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.