xentr_theme_editor

  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

OCZ Vertex 4 512GB SSD Review

Status
Not open for further replies.
xentr_thread_starter
NON-TRIM Environment Testing

NON-TRIM Environment Testing


In order to replicate a non-TRIM environment we first modified our testbed so that it would not pass on the necessary cleaning commands. Meanwhile, to artificially induce a degrade state we ran eight hours of IOMeter set to 100% random, 100% write, 4k chunks of data at a 64 queue depth across the entire array’s capacity. At the end of this test, the IOMeter file is deleted and the drive was then tested. This will replicate drive performance after extended heavy usage prior to any self maintenance routines kicking in and is indicated by the “Dirty” results below.

In order to allow each drive’s self-maintenance routines to kick in, we then wait 30 minutes (Dirty + 30 results) with the system at idle and rerun the tests.



Real World Results


For a real world application we have opted for our standard Vista load time test.

no_trim_boot.jpg


Keeping an SSD's performance in top shape isn't an easy process. Some drives like Crucial's M4 and Corsair's Performance Pro have some aggressive background garbage collection routines that run during idle periods while SandForce-based products have a set amount of storage space reserved for wear leveling or implement a software solution (like Intel's Toolbox application). OCZ's Vertex 4 on the other hand takes a different path. The use a type of passive ITGC routine that runs in the background that does seem to work, but not all that well according to our results since performance was still lagging after some 30 minutes of idle time.

It is unfortunate that OCZ has still not taken a page from the original Vertex by giving their new drive a software solution – aka “wiper.exe” - to this age old problem. This is a missed opportunity, as Intel does offer a non-trim solution for their drives and the difference is noticeable. Simply put, even the best “Idle Time Garbage Collection” routines – which this drive does not appear to have – are a much slower solution than a software solution at getting a drive out of a degraded state. Luckily, there is one mitigating factor in the Vertex 4’s favor: “quantity has a quality all its own” and that extra large cache combined with a very powerful controller means that even when degraded the results are still very, very good.

It also bears mentioning that providing a software solution similar to Intel’s shouldn't be that difficult for OCZ to implement as they do “own” the controller and create its firmware in-house.
 
Last edited by a moderator:
xentr_thread_starter
SATA 2 Performance

SATA 2 Performance


In a perfect world everyone investing in a new solid state drive would have access to a SATA 6Gb/s controller which could pass on the TRIM command. In reality not everyone has this and for many the decision comes down either giving up TRIM – never a good idea with most controllers – and running it off a secondary controller; or taking a performance hit and running in SATA 2.0 mode.

These tests will consist of some of our real world and synthetic benchmarks run on our standard 1155 test-bed; but the drive will be attached to an SATA 2 port.

For synthetic we have opted for the newcomer to our charts: Anvil Storage Utilities Pro. For real world we have opted for our Adobe test. These two tests should give you a very good idea of the level of performance impact you can expect from running a modern SATA 6 drive in compatibility mode.



s2_anvil_w.jpg


s2_anvil_r.jpg


s2_adobe.jpg


It truly is doubtful that a customer purchasing such a powerful, large and expensive SSD would tie it at the hip to an antiquated interface like SATA 2. However, if you do decide – for whatever reason – to just that, the Vertex 4 will acquit itself admirably well. It will be bordering on sacrilege, but doing this will still give any rig a boost .
 
Last edited by a moderator:
xentr_thread_starter
Conclusion

Conclusion


With the Everest 2 controller, OCZ has shown what it takes to be a winner in the SSD market. They have not only eliminated the inherent weakness of relying solely upon third party manufacturers for their high end drive models, but at the same time have created a new controller which is over-brimming with potential. Simply put, with the release of the Vertex 4, OCZ is very much on their game and looking to innovate in any way possible.

From a synthetic benchmark standpoint, the Vertex 4 is a bit of an enigma. Many expect a next generation drive to be head and shoulders above its predecessors but other than a few instances, this drive barely made it out ahead of the Vertex 3 MaxIOPS and ran shoulder to shoulder with Intel’s excellent Cherryville. Normally, these would be noteworthy achievements but in the grand scheme of things, you’d be forgiven if you expected a bit more.

Now make no mistake about it, the Vertex 4’s firmware is still in its infancy and has a ton of growth ahead of it so expect regular performance updates as more refined versions roll off of OCZ’s development line. We’re also hoping that some of the rough edges will be evened out since we did experience some minor hiccups throughout testing.

While the Vertex 4 certainly may not be a synthetic benchmark junkie, it does everything right on other fronts. Rather than hoping to win you over with massive numbers, it focuses upon long term reliability and implements improvements where they count while letting the competition fight over the imaginary “performance crown”. Emphasis has been placed upon real world performance results and this is exactly where it excels against the competition. From our Photoshop benchmark to Windows boot times, it led the pack again and again while costing substantially less than Intel’s 520 series.

This isn't to say that we don't have some reservations about the Vertex 4 and its current positioning. The name may give the illusion that it will provide the generational leap its forefathers did but we just didn't see that kind of past versus present difference this time around. More importantly, enthusiasts will have to ask themselves whether the Vertex 4 really presents an upgrade path for them. The answer to that isn't as straightforward as some may think since OCZ's newest flagship fares well against some drives released in the early quarters of 2011 but it doesn't provide any large performance improvements over the market's current darlings: the Vertex 3 and Intel 520.

Regardless of the lack of serious forward movement in performance numbers, the Vertex 4 shows us that progress can indeed be made within a market that's been stagnant for all too long. Not only has the Everest 2 controller allowed OCZ to untie themselves from SandForce’s grip but they can now offer something none of their competitors can match: a full ground up architecture developed exclusively for their new drives. This gives OCZ free rein over upcoming firmware revisions and allows them to target their focus by concentrating on delivering further optimizations for the Vertex 4.

One of the most important things to take away from this review is OCZ’s intent. The Vertex 4 wasn’t created with a goal of introducing a revolutionary performance jump. Rather, it is supposed to take the next logical, evolutionary step towards a more competitive and accessible SSD market. In this respect, it wins hands down by concentrating upon real world benefits through controller innovation and a great pricing structure.


Pros:

- Equally great compressible and incompressible file performance
- Excellent write performance
- Very good read performance
- In-house design with all the potential benefits this entails
- Massive capacity
- Good price
- Proves that once again OCZ are true innovators not adverse to risk
- 5 Year warranty


Cons:

- Firmware still requires refinement
- OCZ Toolbox needs a bit of an update
- Still too expensive for most users.
- Less than optimal non-trim environment performance
- Semi-plastic case

dam_good.jpg
di.png

 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top