The OCZ Vertex 3 Review (120GB)
by Anand Lal Shimpi on April 6, 2011 6:32 PM ESTThe NAND Matrix
It's not common for SSD manufacturers to give you a full list of all of the different NAND configurations they ship. Regardless how much we appreciate transparency, it's rarely offered in this industry. Manufacturers love to package all information into nice marketable nuggets and the truth doesn't always have the right PR tone to it. Despite what I just said, below is a table of every NAND device OCZ ships in its Vertex 2 and Vertex 3 products:
| OCZ Vertex 2 & Vertex 3 NAND Usage | ||||
| Process Node | Capacities | |||
| Intel L63B | 34nm | Up to 240GB | ||
| Micron L63B | 34nm | Up to 480GB | ||
| Spectek L63B | 34nm | 240GB to 360GB | ||
| Hynix | 32nm | Up to 120GB | ||
| Micron L73A | 25nm | Up to 120GB | ||
| Micron L74A | 25nm | 160GB to 480GB | ||
| Intel L74A | 25nm | 160GB to 480GB | ||
The data came from OCZ and I didn't have to sneak around to get it, it was given to me by Alex Mei, Executive Vice President of OCZ.
You've seen the end result, now let me explain how we got here.
OCZ accidentally sent me a 120GB Vertex 2 built with 32nm Hynix NAND. I say it was an accident because the drive was supposed to be one of the new 25nm Vertex 2s, but there was a screwup in ordering and I ended up with this one. Here's a shot of its internals:
You'll see that there are a ton of NAND devices on the board. Thirty two to be exact. That's four per channel. Do the math and you'll see we've got 32 x 4GB 32nm MLC NAND die on the PCB. This drive has the same number of NAND die per package as the new 25nm 120GB Vertex 2 so in theory performance should be the same. It isn't however:
| Vertex 2 NAND Performance Comparison | ||||
| AT Storage Bench Heavy 2011 | AT Storage Bench Light 2011 | |||
| 34nm IMFT | 120.1 MB/s | 155.9 MB/s | ||
| 25nm IMFT | 110.9 MB/s | 145.8 MB/s | ||
| 32nm Hynix | 92.1 MB/s | 125.6 MB/s | ||
Performance is measurably worse. You'll notice that I also threw in some 34nm IMFT numbers to show just how far performance has fallen since the old launch NAND.
Why not just keep using 34nm IMFT NAND? Ultimately that product won't be available. It's like asking for 90nm CPUs today, the whole point to Moore's Law is to transition to smaller manufacturing processes as quickly as possible.
Why is the Hynix 32nm NAND so much slower? That part is a little less clear to me. For starters we're only dealing with one die per package, we've established can have a negative performance impact. On top of that, SandForce's firmware may only be optimized for a couple of NAND devices. OCZ admitted that around 90% of all Vertex 2 shipments use Intel or Micron NAND and as a result SandForce's firmware optimization focus is likely targeted at those NAND types first and foremost. There are differences in NAND interfaces as well as signaling speeds which could contribute to performance differences unless a controller takes these things into account.
25nm Micron NAND
The 25nm NAND is slower than the 34nm offerings for a number of reasons. For starters page size increased from 4KB to 8KB with the transition to 25nm. Intel used this transition as a way to extract more performance out of the SSD 320, however that may have actually impeded SF-1200 performance as the firmware architecture wasn't designed around 8KB page sizes. I suspect SandForce just focused on compatibility here and not performance.
Secondly, 25nm NAND is physically slower than 34nm NAND:
| NAND Performance Comparison | ||||
| Intel 34nm NAND | Intel 25nm NAND | |||
| Read | 50 µs | 50 µs | ||
| Program | 900 µs | 1200 µs | ||
| Block Erase | 2 µs | 3 µs | ||
Program and erase latency are both higher, although admittedly you're working with much larger page sizes (it's unclear whether Intel's 1200 µs figure is for a full page program or a partial program).
The bad news is that eventually all of the 34nm IMFT drives will dry up. The worse news is that the 25nm IMFT drives, even with the same number of NAND devices on board, are lower in performance. And the worst news is that the drives that use 32nm Hynix NAND are the slowest of them all.
I have to mention here that this issue isn't exclusive to OCZ. All other SF drive manufacturers are faced with the same potential problem as they too must shop around for NAND and can't guarantee that they will always ship the same NAND in every single drive.
The Problem With Ratings
You'll notice that although the three NAND types I've tested perform differently in our Heavy 2011 workload, a quick run through Iometer reveals that they perform identically:
| Vertex 2 NAND Performance Comparison | ||||
| AT Storage Bench Heavy 2011 | Iometer 128KB Sequential Write | |||
| 34nm IMFT | 120.1 MB/s | 214.8 MB/s | ||
| 25nm IMFT | 110.9 MB/s | 221.8 MB/s | ||
| 32nm Hynix | 92.1 MB/s | 221.3 MB/s | ||
SandForce's architecture works by reducing the amount of data that actually has to be written to the NAND. When writing highly compressible data, not all NAND devices are active and we're not bound by the performance of the NAND itself since most of it is actually idle. SandForce is able to hide even significant performance differences between NAND implementations. This is likely why SandForce is more focused on NAND compatibility than performance across devices from all vendors.
Let's see what happens if we write incompressible data to these three drives however:
| Vertex 2 NAND Performance Comparison | ||||
| Iometer 128KB Sequential Write (Incompressible Data) | Iometer 128KB Sequential Write | |||
| 34nm IMFT | 136.6 MB/s | 214.8 MB/s | ||
| 25nm IMFT | 118.5 MB/s | 221.8 MB/s | ||
| 32nm Hynix | 95.8 MB/s | 221.3 MB/s | ||
It's only when you force SandForce's controller to write as much data in parallel as possible that you see the performance differences between NAND vendors. As a result, the label on the back of your Vertex 2 box isn't lying - whether you have 34nm IMFT, 25nm IMFT or 32nm Hynix the drive will actually hit the same peak performance numbers. The problem is that the metrics depicted on the spec sheets aren't adequate to be considered fully honest.
A quick survey of all SF-1200 based drives shows the same problem. Everyone rates according to maximum performance specifications and no one provides any hint of what you're actually getting inside the drive.
| SF-1200 Drive Rating Comparison | ||||
| 120GB Drive | Rated Sequential Read Speed | Rated Sequential Write Speed | ||
| Corsair Force F120 | 285 MB/s | 275 MB/s | ||
| G.Skill Phoenix Pro | 285 MB/s | 275 MB/s | ||
| OCZ Vertex 2 | Up to 280 MB/s | Up to 270 MB/s | ||
I should stop right here and mention that specs are rarely all that honest on the back of any box. Whether we're talking about battery life or SSD performance, if specs told the complete truth then I'd probably be out of a job. If one manufacturer is totally honest, its competitors will just capitalize on the aforementioned honesty by advertising better looking specs. And thus all companies are forced to bend the truth because if they don't, someone else will.
Facebook
Twitter
RSS
153 Comments
View All Comments
erple2 - Friday, April 8, 2011 - link
I believe that the issue is scale. It would not be possible financially for OCZ to issue a massive recall to change the packaging on all existing drives in the marketplace. Particularly given that while the drives have different performance characteristics (I'd like to see what the real world differences are, not just some contrived benchmark), it's not like one drive fails while another works.So it sounds to me like they're doing more or less what's right, particularly given the financial difficulty of a widespread recall.
Dorin Nicolaescu-Musteață - Thursday, April 7, 2011 - link
IOmeter results for the three NAND types are the same for both compressible and uncompressible data in ”The NAND Matrix”. Yet, the text suggests the opposite.gentlearc - Thursday, April 7, 2011 - link
The Vertex 3 is slowerIt doesn't last as long
Performance can vary
Why would you write an entire article justifying a manufacturers decisions without speaking about how this benefits the consumer?
The real issue is price and you make no mention of it. If I'm going to buy a car that doesn't go as fast, has a lower safety rating, and the engine can be any of 4 different brands, the thing better be cheaper than what's currently on the market. If the 25nm process allows SSDs to break a price barrier, then that should be the focal point of the article. What is your focal point?
"Why not just keep using 34nm IMFT NAND? Ultimately that product won't be available. It's like asking for 90nm CPUs today, the whole point to Moore's Law is to transition to smaller manufacturing processes as quickly as possible."
Pardon? This is not a transistor count issue, it's further down the road. I am surprised you would quote Moore's Law as a reason why we should expect worse from the new generation of SSDs. The inability for a company to address the complications of a die shrink are not the fault of Moore's Law, it's the fault of the company. As you mentioned in your final words, the 250GB will probably be able to take better advantage of the die shrink. Please don't justify manufacturers trying to continue using a one-size-fits-all approach without showing how we, the consumer (your readership), are benefited.
erple2 - Friday, April 8, 2011 - link
I think that you've missed the point entirely. The reason why you can't get 34nm IMFT NAND going forwards, is that Intel is ramping that production down in favor of the smaller manufacturing process. They may already have stopped manufacturing those products in bulk. Therefore, the existing 34nm NAND is "dying off". They won't be available in the future.The point about Moore's Law - I think Anand may be stretching the meaning of Moore's Law, but ultimately the reason why we get faster, smaller chips is because of cost. It's unclear to me what the justification behind Moore's law is, but ultimately, that's not important to the actual Law itself. It is simply a reflection of the reality of the industry.
I believe transistor count IS the issue. The more transistors Intel (or whomever) can pack in to a memory module for the same cost to them (thereby increasing capacity), the more likely they are to do that. It is a business, after all. Higher density can be sold to the consumer at a higher price (more GB's = more $'s). Intel (the manufacturer of the memory) doesn't care whether the performance of the chips is lower to some end user. As you say, it's up to the controller manufacturer to figure out how to take into account the "issues" involved in higher density, smaller transistor based memory. If you read the article again, Anand isn't justifying anything - he's simply explaining the reasons behind why RIGHT NOW, 25nm chips are slower on existing SF drives than 34nm chips are.
It's more an issue of the manufacturers trying to reuse "old" technology for the current product line, until the SF controller optimizations catch up to the smaller NAND.
gentlearc - Saturday, April 9, 2011 - link
Once again, why do an article explaining a new product that is inferior to the previous generation with no reason why we should be interested? AMD's Radeon HD 6790 was titled "Coming Up Short At $150" because regardless of the new technology, it offers too little for too much. Where is the same conclusion?Yes, this article was an explanation. Anand does a 14-page explanation, saving a recommendation for the future.
"The performance impact the 120GB sees when working with incompressible data just puts it below what I would consider the next-generation performance threshold."
The questions remains. Why should the120GB Vertex 3 debut $90 more than it's better performing older brother?
mpx999 - Sunday, April 10, 2011 - link
If you have a problem with speed of flash memory then a good choice for you are drives with SLC memory, which doesn't have as much speed limitations. Unfortunately manufacturers severy overprice them, as SLC drives are much more than 2 times more expansive than MLC ones at the same amount GB, despite the fact that the flash is only 2 times more expansive. You can buy reasonably priced (2x MLC version price) SDHC cards with SLC flash, but you can't get reasonably priced (2 x MLC version price) SSD with SLC flash.taltamir - Thursday, April 7, 2011 - link
"After a dose of public retribution OCZ agreed to allow end users to swap 25nm Vertex 2s for 34nm drives"Actually OCZ lets customers swap their 25nm 64Gbit drives for 25nm 32Gbit drives. There are no swaps to the 32nm 32Gbit drives
garloff - Thursday, April 7, 2011 - link
Anand -- thanks for your excellent coverage on SSDs -- it's the best that I know of. And I certainly appreciate your work with the vendors, pushing them for higher standards -- something from which everybody benefits.One suggestion to write power consumption:
I can see drives that write faster consume more power -- that's no surprise, as they write to more chips (or the SF controller has to compress more data ...) and it's fair. They are done sooner, going back to idle.
Why don't you actually publish a Ws/GB number, i.e. write a few Gigs and then measure the energy consumed to do that? That would be very meaningful AFAICT.
(As a second step, could could also do a mix, by having a bench run for 60s, writing a fixed amount of data and then comparing energy consumption -- faster drives will be longer in idle than slower ones ... that would also be meaningful, but that's maybe a second step. Or you measure the energy consumed in your AS bench, assuming that it transfers a fixed amount of data as opposed to running for a fixed amount of time ...)
Nihil688 - Thursday, April 7, 2011 - link
Hello all,I am kinda new to all this and since I am about to get a new 6GB/s Sata3 system I need to ask this
The main two SSDs that I am considering are the Micron's C400 or the OCZ Vertex3 120' version.
I can see that their sequential speeds in both write and read are completely different with V3 winning
but their Random IOPSs (always comparing the 120GB V3 and the 128GB C400) differ with C400 winning in reads and V3 winning with big difference in writes.
I must say I am planning to install my windows 7 OS in this new SSD I am getting and what I would
consider doing is the following:
-Compiling
-Installing 1 game at a time, playing, erasing, redo
-Maybe Adobe work: Photoshop etc
So I have other hard drives to store stuff but the SSD would make my work and gaming quite faster.
The question is, C400 gives 40K of read which is more important for an OS whilst V3 gives better overall stats and is only lacking in random reads. What would be more important for me? Thanks!
PaulD55 - Thursday, April 7, 2011 - link
Connected my 120 Gig Vertex 3 ( purchased from New Egg) , booted and saw that it was not recognized by the BIOS, I then noticed the drive was flashing red and green. Contacted OCZ and was told the drive was faulty and should be returned. New Egg claims they have no idea when these will be back in stock.