OCZ Vertex 4 Review

OCZ is really on a roll lately, putting out quality and reliable product like crazy, and improving upon it with regular firmware updates. 

I wasn't in on the initial batch of V4 reviews. Sometimes I am, and sometimes I'm not, so I don't have in house numbers on pre-1.5 firmware. If you read some of the initial reviews, this particular SSD evidently had some issues with living up to performance expectations, but that's all settled now, and I can look at the current form.

The Vertex 4 is more than a numbered evolution of the Vertex 3. Of course the form factor and most properties are the same, but the controller this time is the Indilinx, showing fruits of OCZ acquiring Indilinx back in 2011. News sources elsewhere report that there is some confusion over who made the controller itself, but the firmware is in fact OCZ. I'm not going to get into all that, I'm going to focus on performance, and some of my suggested uses for such a product. 

Unlike the previous review, I was only sent one V4. This means I had to go back and do some fresh benches of various configurations of Vertex drives. This one is a 256GB drive however, and due to the increase in space, I have a few suggestions of what you might do with a drive like this, rather than simply screw it into a desktop to be a boot drive. This also gave me a chance to phase out some older benchmarks, and phase in AS-SSD, and Windows 8. 

STATS of review

For this round, I'm only testing on one system. 

Test System

Vertex 4 Test System

  Vertex Test System "Ivy"
Motherboard Asus P877-V Pro
CPU Intel i7 "Ivy Bridge" 3770K
GPU Intel HD 4000
Memory G.Skill Ripjaws Z F3-12800CL9Q-16GBZL 16GB
Cooler Corsair H100
SSD Control Drive OCZ Vertex 3 120GB ( Single and RAID )
SSD Test Drive OCZ Vertex 4 256GB
PSU OCZ ZT750
OS Windows 8 Preview x64
 
You might notice that I've switched to Windows 8 this time. I'm constantly evolving and improving my site and my processes, and with Windows 8, I noticed something VERY interesting. I published a quick commentary on that here. For my purposes, Windows 8 is a superior testing platform to Windows 7 primarily for three reasons: 
  1. It's the future, get over it. 
  2. It's every bit as stable and high performance as Windows 7.
  3. Finally, Windows 8, once installed, will boot ( as far as I can tell ) ANY system configuration. This cuts down benchmarking and testing time by many factors, and will be the standard moving forward. 

Platform and Release Stats

The Vertex 4 Drive was acquired direct from OCZ for review. The next thing we're getting in the lab will be a return to PCI-E so expect to see me mix it up next time!

Overall Impression

The thing that struck me most about the move to V4 is . . . oh wait, let's hold off on that for now. That's the exciting bit. For now, let's talk about Indilinx. 

Overview

In the interest of cutting down repetitive material, and streamlining this process, I'm going to cut to the parts where the V4 differs from the V3. You're welcome to read that review for more on this platform and form factor. 

Test Platform and Methodology

Main testing was really only done on the new Z77 based system, so that’s what we’ll focus on:

Test software includes ATTO Benchmark, Crystal Diskmark, and AS-SSD. Last time I also included PCmark 7 but I'm phasing out use of it for pure storage benches. I had originally started using it due to plans to have more storage benches, but three is perfectly sufficient in addition to the boot tests, which is where things really get interesting. 

Test Results

ATTO Benchmark

  Vertex 3 Single Drive Empty Vertex 4 Empty Percentage of V3 Performance
Peak Read 557MBs 527MBs -5.6%
Peak Write 505MBs 510MBs +.9%

Based on ATTO benchmarks you'd think that there is little or no reason to choose a V4 over a V3. I may phase ATTO out after this round due to these discrepancies.

Crystal Diskmark

  Vertex 3 Single Drive Empty Vertex 4 Empty Percentage of V3 Performance
Seq Read 486MBs 485MBs +.2%
Seq Write 201MBs 495MBs +146%
512K Read 442MBs 299MBs -32%
512K Write 165MBs 474MBs +187%
4K Read 30MBs 27MBs -10%
4K Write 59MBs 69MBs +17%
4K QD32 Read 179MBs 351MBs +96%
4K QD32 Write 137MBs 374MBs +172%

Overall a pretty massive increase in performance, and also proof of why you shouldn't believe benchmarks as your only source of information, but we'll get into a real world usage in a few moments. 

Crystal Diskmark IOPS

  Vertex 3 Single Drive Empty Vertex 4 Empty Percentage of V3 Performance
4K Read IOPS 7367 6646 -9.7%
4K Write IOPS 14615 17071 +16.8%
4K QD32 Read IOPS 7367 85713 +1063%
4K QD32 Write IOPS 33504 84806 +153%

In case you're wondering if that can be right, it can. I've checked it on two different platforms. Overall, the V4 just crushes it's own predecessor on IOPS. When we get to the boot test, you'll really see a dramatic example of this in action. 

AS-SSD

Up to now, we've seen some conflicting information, from tests that aren't giving us a straight scoop. That's how it is with benches sometimes. We'll look at ONE last one before going to the boot tests. 

  Vertex 3 Single Drive Empty Vertex 4 Empty Percentage of V3 Performance
Seq Read 512MBs 505MBs -1.36%
Seq Write 188MBs 481MBs +155%
4K Read 18MBs 24MBs +33%
4K Write 54MBs 96MBs +77%
4K-64K Thrd Read 180MBs 333MBs +85%
4K-64K Thrd Write 110MBs 307MBs +179%
Accs Time Read .0180 ms 0.149 ms -17.2% ( less is better )
Accs Time Wtite 0.281 ms 0.031 ms -808% ( less is better )
Score Read 249 409 +64%
Score Write 183 451 +146%
Score - Final 550 1075 +95%

A final score jump of nearly 100%? Believe it. The access times on this thing are bananas. But enough about synthetic benches, let's do a boot test!

Windows 8 Boot

Running multiple boot tests, we got some really intriguing results: 

Boot test. Stopwatch started after second bios boot screen, and stopped at the end of the "Metro" ( Now "Windows 8?" ) animation. 

  • Vertex 3 Single- 14.6 Seconds
  • Vertex 3 RAID - 9.9 Seconds
  • Vertex 4 256GB - 9.5 Seconds

That can't be right, can it? It can, it is, I wish I had two V4s to show you, but there you have it. That indilinx really screams through that boot. 

I have video of the these boot tests, so if there is any demand, I'll upload that. 

Test Result Conclusion:

A nearly double performance jump in just one generation, for a reasonable price to boot? There's really nothing negative to say here. 

Other Uses

Have you ever carried around a full size hard drive in order to have data, movies, shows, games, work files, and photos with you? USB thumb drives are both too small, and too slow, and SSDs too expensive? 

This generation really changes that. I slipped the V4 into a SATA 6GBs/USB 3.0 2.5" external enclosure. The whole thing fits in my pocket, and I have 256GB of nearly instantaneous access to a nearly full Steam folder, some other content, and a Windows 8 Preview boot. I highly suggest you try the same thing, and see how easy your life becomes. 

Issues?

Once again, not a thing. 

Conclusion

It bears repeating: A nearly double performance jump in just one generation, for a reasonable price to boot? There's really nothing negative to say here. 

Some perspective for next time

I'm really looking forward to what OCZ/Indilinx gives us next time. A V4 MAX IOPS would be awesome, and inevitable.

Get updates on new projects and images

Subscribe to my newsletter for regular updates on new projects and new image sets.