OCZ Vertex 4 128GB SSD Review and 1.4RC FW Comparison – SSD Steroids for Your Vertex 4

If you thought the most exciting improvements with OCZ’s new Vertex 4 firmware were reserved for the 256GB and 512GB models, you’d be wrong. As it happens, the most interesting and substantial performance improvements are to be had with the 128GB Vertex 4.

We happen to have a 128GB Vertex 4 on our bench in order to find out what makes it and the new Vertex 4 1.4 Release Candidate firmware worth a look.

INTRODUCTION

The Vertex 4 possessed many desirable characteristics at launch, but also many puzzling attributes. For example, despite it’s brutal speed, it could barely compete with it’s contemporaries in Vantage testing. New firmware addressing some initial shortcomings was promised, and now OCZ has delivered. Though still in a pre-release stage, OCZ has made the upgrade available for the adventurous enthusiast. Does the new firmware increase performance? Is it worth the upgrade from a stable 1.3FW release to a release candidate? To find out, we’re putting a 128GB Vertex 4 though the gauntlet with both firmwares.

SPECIFICATIONS AND PRICING

The 128GB Vertex4 isn’t the fastest member of the new OCZ flagship family. With sequential reads of up to 535MB/s and sequential writes up to 200MB/s, the “junior 4” falls behind the performance of it’s big brothers. All three are said to have nearly the same random performance though: 85,000 random write IOPS and 90,000 random read IOPS (95K for the 512GB). Each is also given a “max IOPS” figure: 120,000. Digging a little deeper, it appears that this number is generated from 512 byte random accesses at high queue depths.

The 128GB Vertex 4 SSD is currently listing at Amazon for $149.99, or around $1.20 per usable GB. The 256GB and 512GB units can be found in the neighborhood of $300 and $650, respectively. Each Vertex 4 is given a five year warranty to bring it in line with Intel, Plextor, and OWC’s premium offerings.

INTERIOR COMPONENTS

The 4’s Everest 2 platform is composed of a Marvell processor strapped with OCZ’s 1.3 version firmware. Riding shotgun on the PCB is 1024MB of DRAM, or put another way, more memory than came standard on most laptops before 2005. The 128GB doesn’t make use of all that cache, and only the 512GB model makes use of the whole GB. The package is completed with sixteen 64gbit ONFi NAND modules, each 8GB in capacity. After formatting, users are left with a little over 119GB of the 128GB onboard, which amounts to zero percent over-provisioning.

After voiding the warranty and removing the case screws, the PCB is available for pictures. The plastic housing hides one of the most intricate and visually interesting PCBs around, packed to the gills with traces and surface mount components. The Indilinx-badged controller is surrounded by eight flash packages and 512MB DRAM. Given some of the characteristics of the drive, the controller could be a variant of the new Marvell 88SS9184.

The opposite side, just as visually interesting, holds the remaining 512MB Micron DRAM IC and the other eight Intel 29F64G08ACME3 NAND packages. Some drives may have identical Micron flash in the future, but so far most shipping drives appear to use the Intel-branded IMFT flash. Either way, the Vertex 4 will have 3,000 PE (program/erase) cycle flash in it, regardless of whether it says Intel or Micron on the flash.

If you compare the PCB shots from this sample and the original Vertex 4 512GB we reviewed in April, you’ll notice some visual differences. The golden traces are much more visible and the Indilinx branding on the controller is not as white on this 128GB version.

The V4 makes good use of the 3K PE cycle NAND with it’s 128bit ECC (Error Correcting Code). With the more advanced ECC engine, the Marvell processor and Indilinx firmware should be able to get every last bit of endurance out of lower-PE flash.

18
Leave a Reply

avatar
9 Comment threads
9 Thread replies
0 Followers
 
Most reacted comment
Hottest comment thread
7 Comment authors
TeefattLes@TheSSDReviewdaCeltJaredThirsktinkererguy Recent comment authors
  Subscribe  
newest oldest most voted
Notify of
Jack
Guest
Jack

i wanted to see some updates on 256gb drive lol but this is great. the HD tune showing a weird behavior though, would probably wait for the releases of next firmware, which no one knows when..

Jack
Guest
Jack

just a thought, how come the 4k write is so high, 150 mb/s.. thats the highest so far lol

Les@TheSSDReview
Guest

The article explains the optimizations and board used that helped to attain these results. They are indeed not normal and we thought, as a bit of a change, we might throw such in for interest.

tinkererguy
Guest
tinkererguy

With 1.4 firmware now released (no longer labeled release candidate), my informal test results of the 256GB are looking quite nice, so far, looking forward to hearing if the same holds true for the 128GB:
http://tinkertry.com/ocz-vertex-4-firmware-1-4/

Les@TheSSDReview
Guest

We intend on posting the 256GB SSD updates soon enough.

Redcell
Guest
Redcell

“once you’ve upgraded to 1.4RC. Flashing to 1.4RC involves a destructive update (all data is wiped in the process), and the drive must not be the primary system drive.”

im understand why. flashing to 1.4RC involve destructive update,
but im not understand, why the drive, cant act as the primary system drive?
pls explained?
thanks

Les@TheSSDReview
Guest

You cannot upgrade while it is a boot drive or it will destroy your system. You can upgrade and then use it as a boot drive.

Barrett Archer
Guest

I think the author was saying if you’re running this drive right now, and it’s running 1.3, and your operating system is running on the drive running 1.3, you can’t update to 1.4, because flashing it would wipe out everything on the drive running 1.3, including the operating system.

Dave
Guest
Dave

Many people reporting Vertex 4 is having issues being detected on cold boot which is also causing waking from sleep bsod.

Barrett Archer
Guest

Way to be punctual and terse about it! 🙂

jack
Guest
jack

also i still dont get how firmware seems to boost the performance up for everything, its clear in CDM as well as ATTO or ADSSD, but not the vantage.. is it something to do with that issue with HDtune?

Les@TheSSDReview
Guest

This has us a bit stumped as well but, in testing as many SSDs as we have over the years, we have seen variances where SSD compatibility was an issue for on reason or another without a clear understanding of why.

Jack
Guest
Jack

im wondering if its because the latency tripled as shown in the picture AS SSD and that probably cause the poor performance on the overall realworld vantage