OCZ Vertex 4 Review (128GB), Firmware 1.4/1.5 Tested
by Kristian Vättö on August 4, 2012 10:00 AM ESTRandom Read/Write Speed
The four corners of SSD performance are as follows: random read, random write, sequential read and sequential write speed. Random accesses are generally small in size, while sequential accesses tend to be larger and thus we have the four Iometer tests we use in all of our reviews. Our first test writes 4KB in a completely random pattern over an 8GB space of the drive to simulate the sort of random access that you'd see on an OS drive (even this is more stressful than a normal desktop user would see).
We perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time. We use both standard pseudo randomly generated data for each write as well as fully random data to show you both the maximum and minimum performance offered by SandForce based drives in these tests. The average performance of SF drives will likely be somewhere in between the two values for each drive you see in the graphs. For an understanding of why this matters, read our original SandForce article.
.
Neither the 1.4 or 1.5 firmware promised any improvements regarding random read/write performance.
As for the 128GB model, which is new in our tests, its random read/write performance is actually on-par with 256GB and 512GB capacities. This is a pleasant surprise the difference between the 128GB and 256GB capacities in many families can be quite substantial (e.g. Samsung's SSD 830).
Sequential Read/Write Speed
To measure sequential performance we ran a one minute long 128KB sequential test over the entire span of the drive at a queue depth of 1. The results reported are in average MB/s over the entire test length.
It's relieving to look at the graph above. It's now a fact that the 1.4 firmware fixes the sequential read speed at low queue depths issue. Not only does the firmware fix the issue, the sequential read performance is also very good.
OCZ was promising increased sequential write performance for all models with the 1.5 firmware but at least in Iometer, at low queue depths, the difference is negligible. We'll take a look at AS-SSD and ATTO at higher queue depths next.
60 Comments
View All Comments
hybrid2d4x4 - Tuesday, August 7, 2012 - link
I agree with this. I care more about power consumption than incremental performance since I only have an SSD in my laptop for now, and the next one I buy will replace that 128GB F120 and my HTPC and eventually desktop will get them as hand-me-downssswing - Saturday, August 4, 2012 - link
Now it looks like we need a review comparing VTX4-25SAT3-512G & VTX4-25SAT3-512G.M since I haven't been able to find updated specs from OCZ. There's a big price difference but is the performance difference relatively equivalent?Movieman420 - Saturday, August 4, 2012 - link
Actually Ocz has been fore-warned not to repeat the same stunt they pulled when they switched from 34nm 32Gb dies to 25nm 64Gb dies without making it clear that there was a significant performance difference...esp with the 50 & 60gb drives that went from using 8 nand chips (1 per SF channel) to using only 4 chips and 4 channels. Anand took Ryan Peterson to task over the stunt and Ryan promised to differentiate models who's speed change by 5% or more I think.The '.M' drives with Micron nand are slower than the regular V4 with IMFT nand
Here's the thread at Ocz.
http://www.ocztechnologyforum.com/forum/showthread...
menting - Sunday, August 5, 2012 - link
huh? Micron NAND IS IMFT NAND.what was that support guy smoking?
semo - Monday, August 6, 2012 - link
Just to add that not only did the performance suffer but the capacity did as well. They also used Spectek NAND which wasn't what the reviews reported originally.In the end, OCZ managed to bury the issue without even issuing a mass recall of the affected drives. OCZ is a dodgy company and I'm glad that there is enough competition out there for me to avoid buying any of their products.
primeval - Saturday, August 4, 2012 - link
The performance is great and all, but without reliability I just don't see that many people jumping on board. No one wants to replace their SSD frequently.chaudx - Saturday, August 4, 2012 - link
How much of a drop in performance do other drives take when over 50% full?Benchmarking empty drives seems like a waste of time, as most people are going to fill them at least 50%, if not more.
Linuxhippy - Saturday, August 4, 2012 - link
What I really miss is an analysis, how much used capacity affects real-world performance.This way, it looks just a lot like an advertising campaing bought by OCZ ...
The review reminds me a lot about the SandForce benchmarks, where you had to subtract -25% once the drive had been written its capacity for the first time.
AFUMCBill - Saturday, August 4, 2012 - link
Anyone needing an SSD in a video recorder such as the Black Magic Design Shuttle 2 now knows to stay away from this drive. A real time application like that is going to be utterly intolerant of the drive deciding it needs to take a large timeout half way through. I am sure there are other applications in a similar situation to that.Beenthere - Saturday, August 4, 2012 - link
The way these companies rush out half-baked software, firmware and hardware is a crime. They should all be severely punished financially for using consumers as unpaid beta testers.If they continue to do this the executives of these companies should do prison time. This would reduce the exploitation of consumers and bring some sanity and integrity to business. Tolerating the illegally gotten industry monopolies and product price fixing sent a clear message to these criminals that anything goes and as such they are cashing in and using the media to further their unscrupulous behavior.