OCZ falsely advertising 25nm Vertex 2 drives?

[semo]

consumer advocacy is one thing... but flaming, ranting, and raving only serve to paint the true picture where some peoples minds tend to gravitate. Many bad facts here.

Some good ones of course.. but far too many bad ones to add credibility to the majority of posts here.

Some basic facts for all the flamers to consider:

1. no one was ever charged for RMA exchange as OCZ quickly changed the original policy before the first RMA was ever sent since we all know OCZ has the tendency to change things, right? lol

2. 25nm drives using 32Gbit nand has eliminated "most" of the lost performance issue due to filling all available channels. Newest firmware is also a tad faster to bridge the gap between original models.

3. OCZ SSD's are consistently aimed at enthusiast types,.. therefore are usually topping the charts in most reviews across the board through more aggressive firmware development.

4. now maybe this should actually be number 1.. but all those who've been duped into slower/smaller drives don't have a clue simply because it's VERY tough to tell the difference without using incompressible data or benchmarks which use numbers to "tell you" they are slower. Sandforce has/will continue to be "all about the averages" from day one. The controller is the key to these drives excellent latency and overall performance. IOW, if the OS consisted primarily of incompressible data.. you might notice in everyday use. There are many who have booted to both and besides the matter of principal, they can't tell the difference AT ALL.

5. and this should be the most obvious here.. ALL review sites have a fine line to walk for the opportunity to recieve beta-releases and MUST use much tact and diplomacy to maintain this relationship with any vendor.

There you go. More fuel for the boiler. :sneaky:

you forgot to mention the (substantial) capacity discrepancies. My stance from the beginning has been that this has nothing to do with technology and is all about consumer rights. OCZ changed the innards of their mainstream flagship product and didn’t formally inform anyone or change their labelling accordingly.

Bait-and-switch tactics have been illegal for a long time and under no circumstance are acceptable.

 

[chong67]

I got a new Agility 2 last week from OCZ.

It is a 32 nm DIE and Windows 7 says it got 55.7 GB of space on a 60GB SSD.

OCZ will not tell me if I got a 25 nm or 32 nm. They said whats important is that I get 55.7 and not 51GB on a 60GB?

So I dont know what to believe?

 

[frostedflakes]

That's the correct size for a 3Xnm drive. You can use the OCZ Toolbox to confirm the die size.

 

[taltamir]

Windows 7 says it got 55.7 GB of space on a 60GB SSD.

This is because Giga is a well defined scientific term (since 1960) means "billion" (10^9). Windows approximates GB to mean 2^30 (so called "binary billion") rather then 10^9 (billion, 1 followed by 9 zeros). Any drive you buy, SSD or HDD will be listed in GB = 10^9 bytes. while windows uses the 2^30 figure (1024^3).
This has been a point of contention for quite a while, in 2007 units were defined to calrify this confusion. GiB stands for GibiByte; Gibi is a binary giga, aka 2^30, but windows does not (yet?) use this clarified notation.

60GB / 1.024^3 = 55.88GiB
the discrepancy between 55.88 and your 55.7 is 180MiB (mibibytes), 100MiB is assigned to a hidden partition used by windows 7. And the other 80MiB are probably accounted for by rounding, some unallocated space at the end (for alignment) and probably some rounding.

This proves you got a V3 veretx2. so you are all good.

That's the correct size for a 3Xnm drive. You can use the OCZ Toolbox to confirm the die size.

The OCZ Toolbox is absolutely unnecessary, just check to see if the size is right. If it was the V2 vertex he would have been a few GiB short.

 

[frostedflakes]

He just seems to be a skeptical guy, figured he might appreciate another source of confirmation.

 

[chong67]

Thank you all. I post this question on OCZ forum and those mod would not tell me the true story.

I did use OCZ toolbox and confirm that it is a 32 die. And I format to 55.7GB out of 60GB.

One guy even open up his SSD to confirm. I am not about to do that.

Then the problem is that you can get a 25 nm and still dont get 55.7GB but less? Are those the one that can exhanged?

And 25nm will short change everyone?

 

[taltamir]

I did use OCZ toolbox and confirm that it is a 32 die. And I format to 55.7GB out of 60GB.

Did you not read my explanation on the difference between GB and GiB?
Windows is saying 55.7GB, this is incorrect. It is actually 55.7GiB not GB partition.
GiB is not the same as GB.

They sold you a 60GB drive, it contains 60GB of space. 60GB = 55.88GiB (windows 7 uses up the missing 0.1 for a hidden partition so you see 55.7GiB)
You got a good drive; not one of the bad ones

One guy even open up his SSD to confirm. I am not about to do that.

Not necessary

Then the problem is that you can get a 25 nm and still dont get 55.7GB but less? Are those the one that can exhanged?

Yes, if you had a problem drive it would show up as less. IIRC it would be 49GiB

And 25nm will short change everyone?

The issue has absolutely nothing to do with 25nm. It specifically has to do with the vertex2.

He just seems to be a skeptical guy, figured he might appreciate another source of confirmation.

So you tell him to use an OCZ tool? And besides which, he said he had already used that tool, and he just didn't believe it.

 

[poofyhairguy]

I just bought a 3.5inch 120GB Vertex 2 on Newegg.

Then I came to this thread.

Damn.

 

[taltamir]

I just bought a 3.5inch 120GB Vertex 2 on Newegg.

Then I came to this thread.

Damn.

how much space does windows tell you it has?
120GB = 111.76GiB
So if windows tells you its 111.6GB (Incorrectly name GiB as GB. 100MB is taken by overhead, hidden partition, etc) then you got a good one. If you are several GiB lower than that, then you should get a free replacement from OCZ

 

[poofyhairguy]

I will get Windows 7 from my University and check.

 

[frostedflakes]

Did you not read my explanation on the difference between GB and GiB?
Windows is saying 55.7GB, this is incorrect. It is actually 55.7GiB not GB partition.
GiB is not the same as GB.

They sold you a 60GB drive, it contains 60GB of space. 60GB = 55.88GiB (windows 7 uses up the missing 0.1 for a hidden partition so you see 55.7GiB)
You got a good drive; not one of the bad ones


Not necessary


Yes, if you had a problem drive it would show up as less. IIRC it would be 49GiB


The issue has absolutely nothing to do with 25nm. It specifically has to do with the vertex2.



So you tell him to use an OCZ tool? And besides which, he said he had already used that tool, and he just didn't believe it.

He actually didn't, he just said it was a 32nm die. For all I know he got that info from an OCZ rep or whoever he had already talked to about this issue.

 

[groberts101]

frostedflakes, I don't work for OCZ but I've studied enough data from the forum and experimented more than most with their products(Sandforce especially) so here it is in a nutshell.

25nm drives which are RMA will not format to the required .93 multiplication factor in Windows so shorted capacity models will show 51GB instead of the normal 55. The bigger issue stemed from halving of the channels used with the use of 25nm nand itself as the lessor of the 2 evils.

whether or not you have 32/34nm or 25 could be easily seen in AS SSD in the sequential tests(can use CDM3 as well). The 32/34nm 60's usually run at around 95-100MB/s using that incompressible data type(vids are similar) since they eliminate the extra speed these drives can add through compression.

The 25nm 60's will be about 15% slower at around 85'ish MB/s.

If the drive in question has not moved to the newest 1.32 firmware yet, it's worth the hassle as many have seen slight speed improvements. Not to mention all the fixes Sandforce has finally taken care of, many of which are making their way into all other mfgrs Sandforce models/firmware as well.

 

[taltamir]

whether or not you have 32/34nm

its 34nm. People keep on calling it 32nm because CPUs are using 32nm and they get confused.

 

[semo]

I just bought a 3.5inch 120GB Vertex 2 on Newegg.

Then I came to this thread.

Damn.

This is why I think Anand and other review sites should have covered this a long time back. I wonder how many people have bought a drive and don't know that they might not have what they expected.

Don't worry though. The thing with this issue is that once you know about, you will be fine (see taltamir's posts). OCZ won't come out to publicize how hard they screwed up, but if you approach them then they will sort it out for you (you are catching them red handed so there's no other way they can react).

 

[curlysir]

it seems as though you lack enough understanding of facts or SSD tech in general to form any sort of clear or consice argument other than "I was screwed by OCZ". Good to know since we didn't quite "get it" from the first wave of you little personal smear campiagn.

and I wish I was getting paid,.. cause with as much as I've helped others with OCZ related issues.. I should be! lol

ever stop to think that maybe I post about Sandforce controlled drives simply because I've got good experience with them? apparently not. :hmm:

and trust me here.. if the floor stays full of SH*T, then I won't want any part of it due to people like you constantly screwing forums like this one with all the irrelevant flaming. So say something relavant or just go away, not just for my sake of course.. but everyone else's too. 🙂

You did get paid in a way. I sure didn't get a free Vertex 3 to try out. I know your are not a OCZ employee and do know that you are vocal on the OCZ forum, but you do need to be upfront about this fact on the Vertex 3. The problem I and a lot of people have with OCZ is not their product (I have a OCZ Vertex LE that I really like) but the way they tried to change to the 25nm chips without changing model # or anyway of knowing the difference.

You are fully entitled to your opinions but just be upfront about your relationship with OCZ.

 

[groberts101]

You did get paid in a way. I sure didn't get a free Vertex 3 to try out. I know your are not a OCZ employee and do know that you are vocal on the OCZ forum, but you do need to be upfront about this fact on the Vertex 3. The problem I and a lot of people have with OCZ is not their product (I have a OCZ Vertex LE that I really like) but the way they tried to change to the 25nm chips without changing model # or anyway of knowing the difference.

You are fully entitled to your opinions but just be upfront about your relationship with OCZ.

I make no entrance to any thread around here trying to hide anything. I've mentioned it from the very first post around here and don't feel the need to put a banner up to promote the fact that I was given the opportunity toi test the V3. Simple fact is that I share what I know from experience. Key word there is "share".

And yes,.. I am VERY vocal with my input on that forum. I'm a consumer first and foremost and never, ever, stray from that reality. I payed over $2000 for my OCZ SSD's(not bragging in the least) and you can bet yer (you know what) that I'm concerned from that perspective. Warranty is good and fast which is all I can rely on at this point.

I just want most to get off my back just because I promote the Sandforce controller. It just so happens that I(and several others I've installed these things for) have had excellent results with OCZ products which is specifically why I still recommend them(though I will be quite honest and say I would not recommend the newer 25nm V2's). Of course this was before all the 25nm debacle, but just saying.. "why wouldn't I have good things to say with that many drives running without issue?" I personally feel that Sandforce is the biggest culprit with poor compatibility across the board and OCZ just takes more heat because they sell more of them than all the others.

The fact that they tried to pull a fast one?.. just adds to all the ammo the critics already had. I had a buddy that had 1 Intel drive out of 3 die on him but that doesn't mean I'm going to run around the forums screaming that Intel has a 33% failure rate, right?

It's a perspective thing for sure and I'm simply giving mine. And for the record, IMHO it was actually BECAUSE of my brutal honesty and from the hip attitude, that I was given the chance to beta-test. Would you want someone with major Sandforce testing/study experience and a no-holds barred attitude to test your product to improve upon it?.. or someone who will just be so happy to get a free drive that they'll say anything you want to hear? I bring a bit of a different perspective to the beta-tester forum and I'm proud of the fact that I haven't strayed much at all from my original perspective. 😎

 

[groberts101]

its 34nm. People keep on calling it 32nm because CPUs are using 32nm and they get confused.

I wrote 32/34nm because OCZ has used both on Sandforce controlled drives in the past. Was not a mistake or assumption at all. 🙂

 

[Voo]

I wrote 32/34nm because OCZ has used both on Sandforce controlled drives in the past. Was not a mistake or assumption at all. 🙂

Considering that there's no defined industry standard for labeling process nodes, it doesn't really matter what we call them anyhow. If we want we could start calling it "gummy bear" process 😉

Anyway about OCZ: The major critic point isn't reliability or anything, just the simple fact that they used bait-and-switch tactics to save money at the cost of the consumer.
Just saying "Hey nobody paid anything for the RMA" misses the point completely. First of all they did and do pay with their time (backup data, remove drive, send in, use other drive for that time, get it back, install software,..) and second OCZ only complied in that regard after the outrage in lots of forums and they saw the PR desaster they had at their hands.

And not to forget that ~10% performance loss between the 32nm and 25nm drives.

My personal decision after all that is to still recommend OCZ if they're clearly the best choice from a technical pov, but to sure as hell go with someone else if there's no clear cut decision (uhuh what an unbending position 😉 ).

 

[groberts101]

while I definately agree with most of that(not sure about the "gummy bear thing though, as I was simply responding to Taltimir's correction) you are taking a few things out of context. Many of my responses were intended to clarify some misinformation and was the main point with many of my posts. And most of what you wrote there I've pointed out several times and on various threads so I see all angles.

So in a nutshell here:

Do I like OCZ for the products and technical help of I've had such good luck with for more than a year straight? YES

Do I think it's a horrid way to do business if boiled right down based purely on "proven" facts? YES of course.

Would I still continue to buy OCZ products if this threads premise was proven by a third party source as 100% accurate? Well.. that's a tougher one as I often gamble and may still consider the risk given the possibility of the best performance I could buy using the newest Sandforce controlled drive. I'm a hardcore raider/OC type of guy, so I would probably go to the highest bidder(so to speak) who offered the best performance, "free drives" or not. lol. If that would cost me the option to be a beta-tester?.. then so be it.

Performance is my key requirement not business ethics or whether or not my nand will last as long as the next guys since I'll surely upgrade before that anyways.

Would I get up on my high horse and go on a campaign of consumer awareness about the pitfalls of doing business with OCZ? Not unless it touched home at a personal level and left me jilted like so many others. It hasn't.. so here I am on the "other side" apparently(sarcasm intended).

So I do.. "get it".. and my replies should not be misconstrued as if I'm trying to sell something here. Because if that's the case then others are trying to push just as much from the other direction. So basically,.. it's the "pot calling the kettle black" here. I gain nothing but the ability to share what I've learned so far through firsthand experience with this particular mfgr. They ALL have issues and I've not had any to date with OCZ so I'll advise based on that experience. If some don't like it, and apparently this place is like a haven for them, that's just too bad since there will be MANY using the Vertex 3 around here in a short while that will appreciate the advice whether others like this company or not. 🙂

 

[taltamir]

I wrote 32/34nm because OCZ has used both on Sandforce controlled drives in the past. Was not a mistake or assumption at all. 🙂

then it was a mistake... made by OCZ Rather then you.
There is still no 32nm NAND process. That the name is rather arbitrary doesn't matter, the process has a name and OCZ managed to confuse it.

 

[groberts101]

then it was a mistake... made by OCZ Rather then you.
There is still no 32nm NAND process. That the name is rather arbitrary doesn't matter, the process has a name and OCZ managed to confuse it.

If we're going to keep splitting hairs here just to be right then google Toshiba 32nm nand. This is the nand that OCZ has used in past SSD's and has said it many times in more than a few threads over there.

They just didn't halve the channels used by the Sandforce controller or slow anything down to the point that it was an issue. After all, it's not called the Vertex 2 32nm SSD, right?

Anywho, believe what suits you in the end and I'm wrong.. you're right.

 

[taltamir]

If we're going to keep splitting hairs here just to be right then google Toshiba 32nm nand.

Wow, you ARE right. http://www.electronista.com/articles/09/02/24/toshiba.32nm.flash.in.fall/

Toshiba has named their NAND 32nm. IMFT and others call it 34nm. Funny.

 

[semo]

We have an in-depth coverage of the issue now.
http://www.anandtech.com/show/4256/the-ocz-vertex-3-review-120gb

You can see Anand was a bit soft on OCZ:

OCZ ultimately took care of those users who were left with a drive that was slower (and had less capacity) than they thought they were getting.

What about users that didn't think they were getting slower and smaller drives? OCZ won't do anything for users unless the they find out what's happening and complain.

I also don't agree with this:

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.

Of course I don't expect to get proper specs out of a box and model number. However, I do expect that all those boxes contain the same stuff that I've read about on AT and other review sites!

And finally, I think this story should not have been burried in a review article but stand out by itself on the main page with a catchy title. I'm glad Anand pushed OCZ to fix some of the issues. It is what we expect of him and why I was disappointed initially that this issue wasn't getting due attention

 

[groberts101]

Wow, you ARE right. http://www.electronista.com/articles/09/02/24/toshiba.32nm.flash.in.fall/

Toshiba has named their NAND 32nm. IMFT and others call it 34nm. Funny.

Sorry.. I was wrong Tal. It's actually Hynix and is said to cause some more of the discrepency we see some users complain of as well. Makes it even tougher to help others troubleshoot their speed issues and distinguish the reasons why it's present on some "premium hardware/mobo's".

and here's some more fuel for this fire.
http://forums.anandtech.com/showthread.php?t=2156257

 

[poofyhairguy]

Ok, OSX is telling me I have a full 120GB. I guess its a good drive, I am gonna use it.

EDIT: Tests are showing its a 25nm drive based on write speeds. Honestly though this is such a big upgrade from my old Samsung 64GB SSD (and I was able to get it for the current after rebate price with no rebate) so I am not gonna complain.