OCZ Shipping 25nm SSDs, customers not happy over lack of change?

You are right. I dug a little deeper, and all of the OCZ responses tend to be along the lines of "Quit your b**ching. Deal with it." All they had to do was change their model numbers even SLIGHTLY, and it wouldn't be that big of an issue.

I'm doing a Sandy Bridge desktop system build for someone in a few weeks, once Sandy Bridge mobos are available again. Looks like I'll be getting a Corsair F120, instead of an OCZ Vertex 2.

 

I fixed this for you. I am frankly astonished at the way OCZ have chosen to deal with this situation.

 

Oh dear... I got not one but two of these OCZ 60GB SSD drives lying about in my room. Both are shrinkwrapped and unopened but I bought back them in early December 2010 (mainly to beat the UK tax rise in 2011!). Now after reading the articles i'm really confused to whether I have the 34nm or the inferior 25nm version.

I really don't want to use them now since I was saving them for a new Sandy Bridge notebook (which we all now know is delayed again) but I guess I have no choice but to use it and wear it a little to find out. My confidence in OCZ had taken a big knock in light of this news.

 

I thought it is only this month the 25nm starts appearing ?

 

Apparently OCZ officially switched to 25nm last week, hence all the fuss now

 

I just bought an OCZ (see below) from newegg on 2/11. Am I one of the unforunate ones?

OCZ Agility 2 OCZSSD2-2AGTE120G 2.5" 120GB SATA II MLC Internal Solid State Drive

 

Possibly. Have you opened up the drive yet? If the drive is listed as 115 GB (unformatted in BIOS), or 107 GB (formatted NTFS), then you have one of the 25 nm drives. If you haven't opened it up yet, then there's no way to tell for certain, which is a big part of the entire problem.

 

To me, this is the worst part of it.

Companies release new parts all the time, and sometimes those parts suck. That is understandable. But OCZ could have at least changed the model number, so that people could have easily just said "Oh, new Model B sucks... I guess I'll just go back to old Model A". Easy cleanup, no mess.

People would still be complaining (because the Internet is full of ungrateful people that complain no matter what), but at least OCZ would have a defensible position. The way they're doing it... they just look like jerks.

 

If they sell it as 120GB and it's only 115 in the BIOS, that's outright fraud. There's no math trick you can claim that would make that a 120GB drive. Material misrepresentation of the product provided.

 

That's the way every hard drive is and has been since the early days of platter storage. A company sells you a "500 GB" hard drive, but the actual capacity is 465 GB. It's the whole binary versus decimal thing.

 

My BIOS said 500GB for my scorpio black, 80GB for my X25M, 160GB for my HM160C.

We are not talking about Microsoft's number.

 

I'm well aware of GiB vs GB. But if they sell you a 120GB drive and it's not at LEAST 120GB (base 10) of accessible space, it's fraud.

 

The funny thing is computers work in digital information....and hardrives store information in digital format. aka 1024 but yet they "advertise" them in a decimal format....never made sense to me :/

 

No, that is not the case here.

You're talking about decimal-to-binary conversion. A drive advertised as 120GB will show as 111.76GB in Windows, because drive manufacturers define a GB as 1,000,000,000 bytes (10^9), whereas your operating system defines a GB as 1,073,741,824 (2^30).

OCZ Vertex 2 34nm:
Advertised Capacity: 120GB / 111.76GB in Windows
Actual Capacity: 120GB / 111.76GB in Windows

OCZ Vertex 2 25nm:
Advertised Capacity: 120GB / 111.76GB in Windows
Actual Capacity: 115GB / 107GB in Windows

That last part is the problem... people are buying with the expectation of getting a 120GB / 111.76GB drive, but actually getting a 115GB / 107GB drive. It's not related to decimal-binary storage calculations.

The reason that OCZ is doing this is because they are over-provisioning an extra 5GB of memory, to make up for the fact that 25nm NAND memory has a lower write lifecycle than 34nm NAND memory. This wouldn't be a problem if they had just clearly communicated these differences between the 34nm / 25nm drives, creating a different model # for the new 25nm drives, or have a better response to their customers than "Quit your b**ching."

I'm pretty disappointed with OCZ. I bought 3x of their OCZ Vertex/Agility 1/2 drives and was very happy with them, but won't buy from them anymore after this.

 

Vertex 1 drives are even closer to rated capacity.

120GB advertised drives are 119GB in windows.

 

My 30 GB Vertex 1 was actually a 32 GB SSD? Anyone else seen this before? I actually have over 30 GB use in Windows?

 

That is weird. I checked on my OCZ Vertex 1 and Agility 1 drives, and do see 119GB formatted capacity in Windows.

That means that the unformatted capacity is a full 128GB. Looks like they aren't provisioning that 8GB for garbage collection and wear leveling.

 

Speaking of which, another article from storagereview about this whole thing: OCZ Vertex 2 25nm Review (OCZSSD2-2VTXE60G) | StorageReview.com

I'd be very surprised if there wasn't a class-action lawsuit started within the next few months over this. Assuming OCZ doesn't do the smart thing and fix it with a recall or refund or something, but history doesn't give very good indications of that happening.

 

My 30GB Vertex 1 drive reports 29.8GB. 60GB reports 59.6.

 

That drive is 25nm nand. OCZSSD2-2AGT E120G
All drives marked "E" are 25nm; not to be confused with "EX" which denotes "Extended" versions of the series.

Furthermore, you can download and install the OCZ toolbox here. Unfortunately, if you are running Intel's Rapid Storage Technology 10.x, the toolbox will not work. If it does work, you can tell whether or not you are "unfortunate" by noting the firmware revision.

FW1.24, FW1.25, FW1.26, FW1.27 (mostly) all 34NM

I actually (read: supposedly - my order, invoice, part, and packaging denote it is 34nm) got a 34nm drive and it performs worse than some 22nm. Image here. Of course, this is not the drive's fault - according to the ocz forum staff - it's because the drive is in a notebook.

EDIT: @Tsunade_Hime
Looks like you have the exact same drive, with the exact same problem, as me.
Go dig up the OCZ box. There is a sticker on the back with a barcode and an ocz hologram. Which of the following does it note:
OCZSSD2-2VTX80G
OCZSSD2-2VTXE80G

 

Read that storagereview article. They got 2 of the 60GB drives with the model OCZSSD2-2VTXE60G. One was 32nm, the other was 25nm.

Either way, OCZ is still selling the 25nm drives with much lower capacity than advertised.

 

And much lower speed than advertised.
take a look at ocz's vertex2 page

http://www.ocztechnology.com/ocz-vertex-2-sata-ii-2-5-ssd.html

I don't see any mention of different speeds, do you?
EDITMORE: They are now showing lower usable space

 

you guy's need to start taking screen shots of there website and retailers so you will be able to have the evidence for a class action suit or small claims court. They will start changing all that info soon i bet to cover their tacks. I know form high school in my business class that if a company advertises and price anywhere they have to give you that price no matter what. I assume that goes along with falsely advertising the products specs as well. So get as many sources of false advertisements as possible and you can make a case. Also post on the BBB's site and claim about this so it is on the BBB's record too so you can reference this too for a case. (or so who ever does the class action suit can reference all ur complaints.)

EDIT: i recommend PITA being incharge ^^.....yes i just throw you under the bus ^-^

 

I'm pretty sure screenshots don't count as evidence in a court of law.

 

use a camera take a picture of your screen with their false information on whatever site and make sure yo u have the time and data. Use your phone and show that ur phone is using sprint/at&t network time for each photo.

 

You are correct, on my box, it shows OCZSSD2-2AGTE120G and the toolbox that you linked showed firmware 1.23 which is not on the list for 34nm

I am conflicted, I want to return it, but at the same time, you also said that there are some 34nm that perform worse than the 22nm. However, I do not want to pay the price of a 34nm to get a lower performing SSD (assuming I understood what was said in the forums). So what does everyone think I should do? I just want to get some opinions before I do anything rash

 

Personally, I'd get rid of it and get another brand, like Intel. Support going forward is likely to be defensive and sketchy. Are you sure you want to ride that out?

 

I don't have any OCZ hardware. At all. Their history has driven me away from them well before this incident. I'm just doing armchair analysis of the situation... I have no legal claim to anything in this case, so someone with one of the drives would have to contact a LAWYER (ie, not some random dude on a forum spouting what could be, but is probably not, bull) and discuss options.

That's up to the judge to decide, on whether there's enough other evidence to warrant that they're accurate or not and so on. Most anything can be evidence in a court of law if it was legally gathered. If OCZ is found trying to change things retroactively or hide evidence of some sort, that's DEFINITELY illegal.

Evidence rules are far from set in stone. The best advice is to keep record of everything possible. You can always toss something later when you don't need it, you cannot get it back if it's gone or never existed because you didn't create it.

 

I agree. I am having issues, and it's not even a 25nm drive, and no one will admit there is any issue with my SSD. I mean, come on, look at my bench posted above. How can anyone call those r/w normal?
Can you return it for a full refund? If you bought it from newegg, the answer is no. You should look into your options.

OCZ is offering an RMA upgrade program, where consumers who purchased/got tricked into purchasing 25nm drives can pay an inflated price-per-GB to upgrade to a 34nm drive.

"only"

source: http://www.ocztechnologyforum.com/f...OCZ-SSD-drives&p=602194&viewfull=1#post602194

 

I agree with you, I knew what had to be done, I just needed that push forward, oh well. I guess I will have to stick with my 500 gb 5400 rpm hard drive till then.

 

That was exactly what I did when I spent a few hours with those support staff when I briefly owned a faulty Vertex 2.

I don't mind lemon but a company with such business attitude, no thank you.

 

I bought the drive on February 11, 2011 and received the item on February 15, 2011. When I checked into my newegg account, there were two options, refund and replacement. The refund option was not allowed because it was under manufacturer warranty, but the replacement option was available and I do not want the replacement.

I spoke to a representative of Newegg through live chat, explained the situation, gave my S/N, item #, invoice #, UPC # and he would have it checked out. I gave them a link to the OCZ forums of the situation that was posted above which could help my situation. I am hoping for a positive response answer from them. I will keep everyone informed.

 

This is not actually OCZ's fault. The extra memory loss is due to RAISE, which is an integral part of the Sandforce Duraclass system. In other words, the loss is specifically due to the Sandforce controller, and is nothing that OCZ has any control over (unless they switch over to another controller, but that's a completely different ball of wax).

Edit - specifically, it's due to RAISE on denser chips (64 Gbit versus 32 GBit). Of course, the less dense chips are more expensive, so technically, yes, it's OCZ's fault for using cheaper chips instead of raising their prices. They still have no control over the technical aspect of RAISE on denser chips, which is a Sandforce function.

With 1.23 firmware, you definitely have a 34 nm drive. Any firmware previous to those listed will be 34 nm.

As for what to do with it, personally, it really depends on your usage patterns. Unless you deal with large files (on a small SSD!) all the time, the sequential numbers are largely useless to you, and what you're interested in are the 4K read/write numbers. Those are still largely unaffected, and you'll still have a very snappy and responsive drive. Now, if you wish to return the drive to "punish" OCZ for their business practices, that's a perfectly understandable reason, but in terms of actual performance, you still have one of the better 60 GB drives currently out (any other Sandforce drive will pretty much perform the same, so the only really "better" drives would be a C300 or Samsung 470).

It's not an upgrade to a 34 nm drive; it's a "sidegrade" to a different 25 nm drive with lower density NAND so that RAISE doesn't take up as much space (16x 32 Gbit chips instead of 8x 64 Gbit chips). I don't know that we have performance figures on how the 2 (well 3, actually, since we're all comparing to 34 nm in the end) compare performance-wise, though. Oh, and that link has now been updated so there's no charge for the changeover, although you'll still have to pay shipping, I'm sure.

 

He has a 25nm drive. The list shows 32nm shipping firmwares for anything ordered after november 2010, and is irrelevant considering the "E" on his product - which clearly denotes it as 25nm

 

No, he doesn't. The "E" does not necessarily denote a 25 nm drive, as pointed out by Storagereview here. They actually cracked open their drives and looked inside. Both are "E" drives, and one of them is 34 nm NAND, while the other is 25 nm NAND. Any firmware prior to 1.24/1.25 _cannot_ be 25 nm NAND... it doesn't support it. Firmware support for 25 nm NAND wasn't added until then. In fact, people are getting so panicky about this issue that they're returning _34 nm_ drives to OCZ to be replaced with 25 nm drives...

 

According to another poster, the OCZSSD2-2AGT E120G is definitely a 22 nm because the E denotes the 25 nm. However, reading your response and looking into the OCZ website forums again. There was one particular statement that made me rethink everything.

FW1.24, FW1.25, FW1.26, FW1.27 (mostly) all 34NM so if your came
with that FW originally you need to leave this thread.

If my firmware is 1.23, logically, it would make sense that it would be 34 nm because every firmware after 1.23 is mostly 34nm. For now, I am just going to wait for a response from Newegg whether it is 22 or 34 and if I am still able to return it because I do not like how they tried to pull a fast one.

 

I can pretty much guarantee to you that it's a 34 nm drive. As I said, returning the drive simply because you don't like their business practices is a perfectly valid reason; after all, part of the point of a capitalist economy is the ability to "vote with your wallet", so to speak. I'm not trying to influence you in that respect. I personally probably would not buy OCZ either for the time being (although I'm not going to dump the 240 GB Vertex 2 I bought from them last December either!). If, however, they make sufficient amends over this matter, and come out again with a high quality product at a very good price point, I might consider them again... a few years down the line.

 

I had the Vertex 2 on my sights but in light of this im going with Intel or Corsair instead.

Consumer affairs should get involved in this. Its sort of bait and switch.

 

Well, they'll be releasing a tool next week to tell you what version your drive is - or you can pry open one side (the side with no warranty void if removed sticker) and count chips - but that's probably not the best idea.

 

Seems to be the case.

Let's see if they catch us by using 64Gbit chip(so we can add to the bottom line)

Oops, got caught. Let's see if we can still keep the bottom line by charging 10 bucks more and swap for the 32Gbit chip

Oops, it doesn't look pretty, we may be in for big trouble. Let's drop the 10 bucks charge and stop the 64Gbit chip version alltogether.

 

I can verify that with my OCZ Vertex II's bought in December 2010. It's also marked with an "E" on the model number and packaging so OCZ's model identifier is not a good way to see if you had the 34nm or 25 nm version which just adds more to the confusion.

EDIT - Confirmed with the Toolbox that my OCZ Vertex in December came shipped with FW1.25 so I guess i'm one of the lucky ones. But I probably avoid OCZ next time and stick with the Intel SSD's after this.

 

Oh look, another incident where OCZ shoots themselves in the foot.

 

E = extended, all the drives with extended storage are technically E drives.

50GB -> 60GB-E

 

I just got my Vertex 2 about a week ago, but no packaging, got it through a friend.

Maybe that's why he sold it cheap..

 

So now OCZ is sending out engineering samples of Vertex 3's
...with 32nm NAND
That seems pretty dishonest to me, considering the consumer products will feature 25nm NAND.

Source: OCZ Vertex 3 Pro Preview: The First SF-2500 SSD - AnandTech :: Your Source for Hardware Analysis and News

 

I think that the reason the OCZ Vertex 3 samples are being sent out with 32nm flash memory is because these OCZ Vertex 3 32nm engineering sample drives were manufactured before OCZ made the switch to 25nm. I don't think there was any kind of deception going on there.

People are rightfully upset about this whole OCZ Vertex 2 25nm issue. But some of those people think that there is some kind of grand conspiracy going on, where OCZ is trying to do some sort of bait-and-switch and "trick" people into buying an inferior product.

More likely, I think what is going on is:
(A) OCZ made a process switch from 34nm to 25nm.
(B) That process switch ended up causing unforeseen problems with their product.
(C) OCZ screwed up the customer relations around dealing with that problem.

To me, it is far more likely that OCZ screwed the pooch on customer relations, rather than the alternative theory that they crafted this grand conspiracy to trick people into buying inferior product.

 

kent1146,

more likely is that OCZ simply needs to fund their V3 SSD development and decided that the customers should make a contribution in both $$ and a worse performing (inferior!) product.

There is no way they did not know about this - they simply choose to ignore the consequences of their actions and at the same time, hope they could get away with it.

Not only is OCZ to blame here, but also SandForce too. They should/could? have changed the controller to properly match the nand it was being mated to.

No excuses for both of them - afaik, they are not even trying to 'make this right' - except for the official statement of 'we'll work with each case individually' for a price - Hah!

 

They're starting to respond to customers actually
That exchange program is now free of charge because of the customer uproar.
It's just too bad they seem so intent on shooting themselves in the foot to start things off. Many people have said it; if they actually said this is what's happening and here are the new capacities/speeds from the start, no one would have been surprised/cared.

Anyway, it's nice to see they're actually bending to complaints.

 

As I explained, that doesn't sound like the case. 34nm ->25nm 32Gbit NAND is not a problem for them and that the 25nm chip should be cheaper as well. But they deliberately choosen to use the 64Gbit and reduce the chip count(thus cost). Don't tell me they don't know the consequence as that is SSD ABC.

 

Either way, the product that will eventually ship as Vertex 3 is highly unlikely to be anything like that production sample.

For a start the review is of a Vertex 3 Pro production sample and has the enterprise class SF-2500 controller. SF-2000 as far as I am aware is still a complete disaster. We know the NAND is going to be different too. When you think about it, the only thing thats going to be the same or similar is the PCB and SATA connectors. And if that is not the case, this drive will cost a fortune relative to how much more a Vertex 2 Pro costs over an equivalent Vertex 2. The target market for this drive isn't gonna be anyone here thats for sure.