OCZ falsely advertising 25nm Vertex 2 drives?

[semo]

It seems that the 25nm edition Vertex 2 drives are advertised the same as the 34nm ones that everyone knows and uses. According to this, the 25nm drives are both SLOWER and SMALLER (in storage capacity) than the original Vertex 2s while their advertised speed and storage capacity are the same as the 34nm drives.

Can someone here verify this? Is this well known?

Update: Anand's responses March, 04, 2011

February, 17, 2011: http://www.anandtech.com/show/4159/ocz-vertex-3-pro-preview-the-first-sf2500-ssd?all=true

I've seen the discussion and based on what I've seen it sounds like very poor decision making on OCZ's behalf. Unfortunately my 25nm drive didn't arrive before I left for MWC. I hope to have it by the time I get back next week and I'll run through the gamut of tests, updating as necessary. I also plan on speaking with OCZ about this. Let me get back to the office and I'll begin working on it

As far as old Vertex 2 numbers go, I didn't actually use a Vertex 2 here (I don't believe any older numbers snuck in here). The Corsair Force F120 is the SF-1200 representative of choice in this test.

Take care,
Anand

February, 24, 2011: http://www.anandtech.com/show/4186/ocz-vertex-3-preview-the-first-client-focused-sf2200?all=true

I've been working with OCZ behind the scenes on this. I've been tied up with the reviews you've seen this week (as well as some stuff coming next week) and haven't been able to snag a few 25nm drives for benchmarking. Needless to say I will make sure that the situation is rectified. I've already been speaking with OCZ's CEO on it for the past week

Take care,
Anand

March, 02, 2011: http://www.anandtech.com/show/4202/the-intel-ssd-510-review?all=true

Correct me if I'm wrong, but I believe everything regarding the 25nm issue has been resolved? OCZ has initiated an exchange program and is covering all associated costs for users that were affected.

I'm still curious to look into the behavior of 25nm NAND however by the time I got back from MWC it looks like the bulk of what needed to happen regarding the 25nm issue already happened.

Is there a remaining aspect of the issue that hasn't been addressed that you'd like me to pursue?

Take care,
Anand

March, 31, 2011. http://www.anandtech.com/show/4253/the-crucial-m4-micron-c400-ssd-review

And for those of you asking about my thoughts on the recent OCZ related stuff that has been making the rounds, expect to see all of that addressed in our review of the final Vertex 3

April, 07, 2011. http://www.anandtech.com/show/4256/the-ocz-vertex-3-review-120gb
In-depth look at the problem on first 4 pages.

 

[Tsavo]

Does your mouse have a scroll button?

 

[Rifter]

yes they are ripping people off, not only that they admitted they were doing it and offered to "exchange" your 25nm drive for a 34nm drive but for MORE MONEY. So they basically said yeah we screwed you but pay us some more and we will fix it.

OCZ will never see another cent from me, ever.

 

[Emulex]

yes let's tar and feather 'em

dey took er jobs!!!

 

[eva2000]

yes they are ripping people off, not only that they admitted they were doing it and offered to "exchange" your 25nm drive for a 34nm drive but for MORE MONEY. So they basically said yeah we screwed you but pay us some more and we will fix it.

OCZ will never see another cent from me, ever.

i made same misunderstanding you did, the paid upgrade is from 25nm 64Gbit to 25nm 32Gbit for capacity reasons not performance. You still get the slower performance after paying the upgrade http://www.ocztechnologyforum.com/f...OCZ-SSD-drives&p=603155&viewfull=1#post603155

 

[semo]

The odd thing is that Anand Lal Shimpi isn't mentioning anything about this. He usually reports on new SSD tech or even news as soon as the NDAs allow but on this occasion he is unusually silent.

I'm surprised nothing was even mentioned when the 25nm Vertex 2s came out and for some reason there was very little discussion about it on the forums as well. I think it's way too soon after the OCZ Core (jmicron controller) fiasco for them to play these games.

 

[BTA]

OCZ has always been a shady company, this does not surprise me.

 

[razel]

1st I noticed Anand doesn't do non review SSD articles often. The last was probably the CES previews. 2nd an OCZ top executive and Anand appear to be 'close.' Not business-wise, but I remember seeing a candid shot of an OCZ top dog's office in one of the Vertex reviews. He wasn't even looking, smiling at the camera or even looking to invite the guest. It looked like a shoot-from-the-hip type photo. Kinda funny actually.

If you're interested in another reputable view about this Vertex mess, check out storagereview. They're about to release a benchmark comparison between the two Vertex's. They have already released a few numbers in their forums and they're disappointed OCZ hasn't changed the size or updated the bench #'s.

I think they should at least update their website. It's not hard these days to change text as opposed to graphics on an SSD. I bet it was harder for them to change the firmware. Of course, if I ran the business and it's too late to change the labels. I'd at least put a ghetto correct size sticker right over the GB. As ghetto as that sounds, at least they'll get some respect out of it.

 

[GoSharks]

Class action suit in the making.

 

[Concillian]

If you're interested in another reputable view about this Vertex mess, check out storagereview.

I just read their 25nm overview article.

Interesting that they mention 25nm NAND is only ~10-15% cheaperthan 34nm NAND for the same capacity. Wonder if this is after the market increases OCZ has been noting for 34nm NAND.

NAND is not the only component in an SSD, and pieces like the PCB, cover, controllet, and assembly won't change, which means we may only see 5-8% SSD price drops at a given capacity by moving from 34nm to 25nm? Pretty disappointing.

 

[Voo]

Just read the storagereview article and how come they don't mention anywhere if their IOmeter tests are using random patterns or zeros (which would be the default and therefore completely useless) - seems like the latter though.

Also they don't distinguish between GB and GiB, which just causes more confusion for some people *sigh*. Buying a 60GB drive and getting 55.x GiB is just fine, because that's just what you paid for. Buying a 60GB drive and getting 51.xGiB is fraud, not more not less.

 

[taltamir]

NAND is not the only component in an SSD, and pieces like the PCB, cover, controllet, and assembly won't change, which means we may only see 5-8% SSD price drops at a given capacity by moving from 34nm to 25nm? Pretty disappointing.

nand is the bulk of the cost. a controller costs 5$ according to reputable sites and the rest another 10$ at most. almost all the cost is the exhubrant price of nand, which is due to severe lack of stock not due to being more expensive to make. The market exploded overnight and there are simply not enough nand fabs in the world to meet demand. This is why the move to 25nm is important, it gives you twice the nand per die/waffer...

 

[frostedflakes]

The StorageReview articles are very helpful. Looks like what others speculated is true, the performance drop is due to OCZ only using four of the eight possible channels on the 60GB Vertex with 25nm. The larger drives should continue to use all eight channels and have no performance drop (can anyone confirm this, are there any reports of 120GB 25nm drives with performance issues? So far it seems like only 60GB is affected). Of course anything <180GB will still have less than advertised capacity.

I also thought it was interesting that they say 25nm is only 10-15&#37; cheaper. I wasn't expecting it to be 50% cheaper, but I figured it would have been a bit more than only 10-15%. As production is scaled up, though, prices should come down.

 

[flamenko]

The reason for size changes will be the same with all SandForce drives. 34nm flash uses 4GB die whereas 25nm flash uses 8GB die. For the most part, 34nm is now becoming extinct.

SF RAISE requires the use of a full die in order to function, regardless of whether it is 4GB or 8GB die. This is where you lose the additional 4plus GB to every new SandForce drive that uses RAISE. Its the small price we will pay to watch bigger capacity drives get introduces eventually at lower prices.

Just imagine.... a new flash can carry 8 die for a total of 64GB.

Its no different whether it is OCZ or another manufacturer. OCZ was first out of the gate and stepped on their own words in this case however...

So there you have it. I dont know if this has been explained elsewhere except in my own article. hmmm.

 

[RhoXS]

Its easy to be a Monday morning quarterback but, nevertheless, it is hard to understand why OCZ does not now take some potentially effective steps to salvage their reputation. IMO, if they simply make a public statement sincerely apologizing for their mistake, include no lame excuses, and immediately change the labeling (even if it is just a sticky lable on the drives and boxes), it will do much more good than doing nothing and remaining silent. The silence is just breeding animosity and distrust. However, I will be surprised if this happens because I do not think it is in OCZ's corporate culture to think or act this way.

Aside from my own experience with a failed popwer supply, only days past the end of the warranty, and an opinion formed from various threads here and there, I perceive OCZ has an arrogant culture that does not really subscribe to doing anything more for their customers than aboslutely necessary. They lost my business last year because of poor customer service when I called them about a failed power supply, just days past the expiration of the warranty based on the sales receipt. Their actions were fully in accordance with their warranty but they just did not capitalize on an oppurtunity to make a small reasonable concession that would have cemented me as a customer. Adding insult to injury, they also told me the warranty starts on the day on manufacture, not the day of sale.

In other words, where there is smoke, there is fire. As a result, at least for me, OCZ products went from something I did not hesitate to buy last spring to a business I would not consider now. The power supply incident was my break point but their handling of their 25 nm SSDs just solidly reaffirmed my OCZ no buy decision.

 

[watzup_ken]

Just saw their statement.

http://www.techpowerup.com/140553/O...uccessful-Transition-to-2X-nm-NAND-Flash.html

The transition was a "success" and they also succeeded losing their customers. :sneaky:

 

[jwilliams4200]

The reason for size changes will be the same with all SandForce drives. 34nm flash uses 4GB die whereas 25nm flash uses 8GB die.

That is not a fundamental cause, just a botched transition by OCZ and/or bad Sandforce controller design (I cannot tell which).

As you can easily see from the pictures in the storagereview article, there are still 8 flash packages in the 64GiB SSD. With a properly designed controller and circuit board, they could obviously still use 8 channels, one per flash package. But for whatever reason (trying to save money by keeping old double-sided circuit board? controller limitation?), the new 25nm OCZ Vertex 2 SSDs are half the speed of the 3xnm ones.

Also, consider that the 64GB Micron C400 has sequential speeds already quoted on Micron's web site, and they are faster than the 64GB C300. If it were not possible to make 64GB SSDs with 25nm flash with the same number of channels as with 34nm flash, then how has Micron done it? You could claim that Micron is fudging the numbers, but you'd have a hard case to make.

 

[GundamF91]

I predict this will cost OCZ dearly when people think about buying other SF driven SSD. There are plenty of fish in the sea in this SSD market, OCZ really shouldn't have pulled a fast one to get some short term cost benefit.

 

[flamenko]

That is not a fundamental cause, just a botched transition by OCZ and/or bad Sandforce controller design (I cannot tell which).

As you can easily see from the pictures in the storagereview article, there are still 8 flash packages in the 64GiB SSD. With a properly designed controller and circuit board, they could obviously still use 8 channels, one per flash package. But for whatever reason (trying to save money by keeping old double-sided circuit board? controller limitation?), the new 25nm OCZ Vertex 2 SSDs are half the speed of the 3xnm ones.

Also, consider that the 64GB Micron C400 has sequential speeds already quoted on Micron's web site, and they are faster than the 64GB C300. If it were not possible to make 64GB SSDs with 25nm flash with the same number of channels as with 34nm flash, then how has Micron done it? You could claim that Micron is fudging the numbers, but you'd have a hard case to make.

What I stated was the simple truth. The 4 plus GB discrepancy is accounted for through RAISE requiring an entire die. A die used to be 4GB (32nm)and is now 8GB (25nm). This is your size difference and will be on any SF drive using the new 25nm NAND.

 

[jwilliams4200]

What I stated was the simple truth. The 4 plus GB discrepancy is accounted for through RAISE requiring an entire die. A die used to be 4GB (32nm)and is now 8GB (25nm). This is your size difference and will be on any SF drive using the new 25nm NAND.

Neither simple, nor truth.

On an 8-channel controller like the SF1200, you should be able to use all 8 channels as long as you have either at least 8 packages or 8 die (depending on how many die are in each package and how they are addressed).

On the circuit board of the 25nm 60GB OCZ Vertex 2, you can clearly see that their are 8 flash chip packages. Therefore, with a proper circuit board design and a proper 8-channel controller design, you can use all 8 channels. OCZ just botched it somehow. That does not mean that all Sandforce controlled SSDs will botch it. And it clearly does not mean that all 25nm flash SSDs will be slower than 34nm flash SSDs, since the 64GB C400 (with 25nm flash) has sequential specs that are higher than the 64GB C300 (with 34nm flash).

EDIT:

On re-reading your comment, it looks like you are making the claim that the lower capacity of the 25nm flash drives is a direct result of the number of flash chips on the circuit board. I suspect you are repeating a misleading claim from OCZ's forum. Both the 25nm and 34nm SSDs have the same total number of bytes of flash memory on the circuit board. But the overprovisioning (aka, reserved space) on the 25nm OCZ Vertex 2 SSDs is larger, which is why the available capacity is smaller. The misleading claim seems to be that the only reason the capacity is smaller is because an entire chip of flash memory must be reserved. While that may be true (I don't know those sorts of details about the SF1200 controller), I think the more important reason that they increased the reserved space is because the 3K erase/program cycles of the 25nm flash necessitate more reserved space in order to maintain the SSD expected lifetime.

 

[taltamir]

since the 64GB C400 (with 25nm flash) has sequential specs that are higher than the 64GB C300 (with 34nm flash).

random specs on it are higher too.

 

[jwilliams4200]

random specs on it are higher too.

The only performance specs on the Micron C400 that I see on the Micron website are for sequential read and write.

Where are you seeing specs for non-sequential read and write for the C400?

http://www.micron.com/partscatalog....cts/parametric/solid_state_storage/client_ssd

 

[taltamir]

The only performance specs on the Micron C400 that I see on the Micron website are for sequential read and write.

Where are you seeing specs for non-sequential read and write for the C400?

http://www.micron.com/partscatalog....cts/parametric/solid_state_storage/client_ssd

IIRC I saw it mentioned in a preview on some website. lacking something more concrete though we can disregard it for now. Your argument is rock solid even without it.

 

[thescreensavers]

yes they are ripping people off, not only that they admitted they were doing it and offered to "exchange" your 25nm drive for a 34nm drive but for MORE MONEY. So they basically said yeah we screwed you but pay us some more and we will fix it.

OCZ will never see another cent from me, ever.

Exactly. They also said that the price supposedly went down from the 34nm, to the 25nm. RMA department says 170 bucks for the 25nm drive which is the MSRP for the 34nm drive. Price drop my ass. Same price for less

 

[eva2000]

The odd thing is that Anand Lal Shimpi isn't mentioning anything about this. He usually reports on new SSD tech or even news as soon as the NDAs allow but on this occasion he is unusually silent.

I'm surprised nothing was even mentioned when the 25nm Vertex 2s came out and for some reason there was very little discussion about it on the forums as well. I think it's way too soon after the OCZ Core (jmicron controller) fiasco for them to play these games.

I think everyone was caught off guard with ocz's switch last week to 25nm - everyone including the media, and distributor and retailers. Will be a pain for retailers dealing with customers affected by this too as they wouldn't have know such a difference either.

Question would be, if OCZ knew Micron 25nm wouldn't meet vertex 2 rated specs. If they didn't then would be something to take up with Micron ?