Vertex 3 BSOD

[zanzabar]

on topic with the thread, why would any1 use the jmicron sataIII controller for raid or at all.

OCZ are not a retailer. They manufacture a product with their name on it. That includes the PCB, the case, NAND and SF controller.

• They messed up on the case - wrong size does not fit in laptops
• They messed up the PCB boards - post production botch fixes
• They screwed people over with 25nm nand. They said 34nm was no longer available and now they are posting on their forum asking members to push retailers to start selling V2 drives with guaranteed Intel 34nm NAND that they refuse to stock
• They push out drives with RC firmware without any validation

The Core was blamed on MS & JMicron. The V1 was blamed on Indilinx. Now it seems Sandforce are the problem?

Let's see if Intel bring out a SF drive. If they do you can guarantee it will be just as reliable as their own controllers. Why? Because Intel will make their own firmware and it will be properly tested before it goes to market.

http://www.anandtech.com/show/4256/t...review-120gb/5

Before we get to the Vertex 3 we have to talk a bit about how validation works with SandForce and its partners. Keep in mind that SandForce is still a pretty small company, so while it does a lot of testing and validation internally the company leans heavily on its partners to also shoulder the burden of validation. As a result drive/firmware validation is split among both SandForce and its partners. This approach allows SF drives to be validated heavier than if only one of the sides did all of the testing. While SandForce provides the original firmware, it's the partner's decision whether or not to ship drives based on how comfortable they feel with their validation. SandForce's validation suite includes both client and enterprise tests, which lengthens the validation time.

The shipping Vertex 3s are using RC firmware from SandForce, the MP label can't be assigned to anything that hasn't completely gone through SandForce's validation suite.

on the sata SSD drives, ocz dose not make them, they do not design them, they do not make the firmware directly, they do not make controllers that have gone into any of the drives at the time and there is no RTM firmware when they launch the drives.

ocz gives awesome support for their drives and ssds are new, sure there were problem with the jmicron having no cash so it was not suited to mlc, then the indi drives they had used naya nand for a while and im not sure on the sandforce but they dont seam to die more than others (or i dont hear about them.) when ocz goes and puts in an order for the controller they want and what kind of nand its not like they have time to evaluate every batch of nand, they also cannot order bins as the chips are not binned, so its like going to frys and trying to pickup an overclocking kit but u cannot do anything but look at rev numbers or a corsair stick. the only company that makes their own is samsuing, and only crucial (as they are a subsidiary of micron) makes the nand for their drives.

SSDs are new and there is no OS that works right with mlc and an onboard controller or one without dedicated cash. i have had a core v2, an agility and an intel g2. the core v2 was crap as a boot drive but it worked great as a read only drive for core differential partitions for hyperV, then the agility was awesome but ive killed 3 of them, and then intel will shudder like the core v2 if u read and write to it at the same time.


so ocz replaces drives quickly, windows is not friendly and the competition has its own problems.

[Gilgamesh]

OCZ are not a retailer. They manufacture a product with their name on it. That includes the PCB, the case, NAND and SF controller.

+1

...and, true or false, I don't know:
http://blog.macsales.com/9438-not-al...tory-continues

[LexDiamonds]

+1

...and, true or false, I don't know:
http://blog.macsales.com/9438-not-al...tory-continues

That is a pretty scathing piece of journalism there... reminds me of the whole remarked memory blanks scandal from like 10 years ago before ocz initially got their act together as a legit company.

[Tony]

Just some facts:

34nm has been superseded by 25nm nand, in the quantity we required we were unable to source, we decided to push to 25, a mistake was made, we entered into a program of total soak replacement...OCZ soaked up all charges both ways and will continue to do so until all drives that customers are unhappy with are swapped to what they are happy with.
We spend a pretty penny making 34nm drives and next to no etailers want to stock it as all they want to push on you guys is CHEAP...if you want them, we have them, go to your local etailer/retailer and ask them to stock them, be prepared for a shock when you get a no from the etailers as they are unwilling to pay the additional cost for the drives.

FW2.06 is the same FW all SandForce partners are using, we are seriously pushing to get the issues this FW has kicked up resolved, you have zero idea how serious I for one take this matter...SF are being pushed like you would not believe.

Jmicron days are long gone, at that time the ONLY other option was Intel and they screwed customers just as much with no TRIM. We worked to bring new tech to the channel as soon as possible, we partnered with Indilinx and then SandForce, both while not perfect solutions are very very good, we stand behind both and will continue to use both going forward. Indilinx now comes under OCZ total control, we build our own FW now, you have seen Arowana which we have given to our partners, this could be just the start of what is coming.

Regards Intel making their own FW for SF drives..you are having a laugh, seriously mate wake up. SandForce created duracalss, there is NO WAY they will release code, not today, not for anyone, unless you bought SF and their IP (and they are worth a pretty penny )
Stop dreaming and get serious.

There will be a fix for FW2.06 and the various other names our competitors use for their version, we got the Apple issues fixed for Vtx2 (it took a while), the pressure applied for this new issue is 20x more...

Believe me or not im not coming here again defending what we are doing, we sell 150K+ drives a month now, if 1000 people post on my forum they have issues its still less than 1% of just a months sales. Remember you will rarely see positive posts on a support forum.

Last..at least we have a support forum, many competitors do NOT and do not allow you a public voice at all.

Im done now, you want any other info come post/ask on my forum.

[Callsign_Vega]

Just some facts:

34nm has been superseded by 25nm nand, in the quantity we required we were unable to source, we decided to push to 25, a mistake was made, we entered into a program of total soak replacement...OCZ soaked up all charges both ways and will continue to do so until all drives that customers are unhappy with are swapped to what they are happy with.
We spend a pretty penny making 34nm drives and next to no etailers want to stock it as all they want to push on you guys is CHEAP...if you want them, we have them, go to your local etailer/retailer and ask them to stock them, be prepared for a shock when you get a no from the etailers as they are unwilling to pay the additional cost for the drives.

FW2.06 is the same FW all SandForce partners are using, we are seriously pushing to get the issues this FW has kicked up resolved, you have zero idea how serious I for one take this matter...SF are being pushed like you would not believe.

Jmicron days are long gone, at that time the ONLY other option was Intel and they screwed customers just as much with no TRIM. We worked to bring new tech to the channel as soon as possible, we partnered with Indilinx and then SandForce, both while not perfect solutions are very very good, we stand behind both and will continue to use both going forward. Indilinx now comes under OCZ total control, we build our own FW now, you have seen Arowana which we have given to our partners, this could be just the start of what is coming.

Regards Intel making their own FW for SF drives..you are having a laugh, seriously mate wake up. SandForce created duracalss, there is NO WAY they will release code, not today, not for anyone, unless you bought SF and their IP (and they are worth a pretty penny )
Stop dreaming and get serious.

There will be a fix for FW2.06 and the various other names our competitors use for their version, we got the Apple issues fixed for Vtx2 (it took a while), the pressure applied for this new issue is 20x more...

Believe me or not im not coming here again defending what we are doing, we sell 150K+ drives a month now, if 1000 people post on my forum they have issues its still less than 1% of just a months sales. Remember you will rarely see positive posts on a support forum.

Last..at least we have a support forum, many competitors do NOT and do not allow you a public voice at all.

Im done now, you want any other info come post/ask on my forum.

I think there is some confusion as some of OCZ's staff on the support forum were saying that 2.06 was not the issue at all. You clearly state that it is.

So if it is a firmware issue, that means a new firmware could be issued to resolve the problem and the drives aren't necessarily bad? Or did the process of applying 2.06 ruin the drive(s)? If you RMA the drive and get one back with 2.06 on it, is there a chance that people will still have the same issues?

It all comes down to wasted time waiting for RMA's, reinstalling everything etc. I for one will never again apply an SSD firmware update until well after it has been released and people have tested it thoroughly. This is the second time an SSD firmware update has bit me in the ass. They obviously do not go through much testing.

Oh and don't forget that for every user that actually knows why their Windows is BSOD'ing and the V3 is dropping offline and finds your particular support forum, there are at-least ten others out there with similar issues that don't know what to do.

[alfaunits]

IMO, the only big issue here is that OCZ did not mention the drives were different and people thought they were getting a better thing (size and performance wise) than what they were getting (based on both specs and benchmarks).
Whether it is 25nm or 34nm by itself is not important as long as different specs are noted (i.e. smaller useable size and speed).

OT: Is it possible to disable DuraClass somehow? Say we want to hammer the drive and don't care for warranty?

34nm has been superseded by 25nm nand, in the quantity we required we were unable to source, we decided to push to 25, a mistake was made, we entered into a program of total soak replacement...OCZ soaked up all charges both ways and will continue to do so until all drives that customers are unhappy with are swapped to what they are happy with.

[johnw]

Im done now, you want any other info come post/ask on my forum.

Just so long as you don't ask a question that OCZ does not like. If you ask too difficult of a question, the thread will get closed.

[LexDiamonds]

I just want to throw out that both Tony and Ryder have maintained a consistant presence on XS for YEARS. This is a great thing, but alas, two men cannot move mountains. Overall, I think its safe to say the sandforce partership has hurt OCZs rep thusfar rather than nurtured it. Hopefully this will change in the future as SF matures as a company.

[antiacid]

to answer the original question, drop the R0 120x2 array and get a 240gb drive instead, maybe a M4 if you can. R0 with ssd is just not worth the hassle, ever. Especially for flight sims

[vardirox]

Oh and don't forget that for every user that actually knows why their Windows is BSOD'ing and the V3 is dropping offline and finds your particular support forum, there are at-least ten others out there with similar issues that don't know what to do.

There is an informative review from a buyer on Amazon and he compares the hardware and the just as important software and technical support. After reading this and because I have several AMD systems, I would rather buy a OCZ Vertex than an Intel 510/320.

http://www.amazon.com/review/RHAU08S...#RHAU08SR6YQ7G

This Intel SSD will surely disappoint many users. There are severe limitations, specifically in using the Intel SSD Toolbox. I strongly suggest to go to the Intel website, download the 41 page Toolbox users guide, and read it before buying.

The toolbox will not do the following:

1. It will not allow to optimize (TRIM) if this drive is installed in a non-Intel computer. If your PC is AMD or other than Intel, this funtion will not work.

2. It will not allow to optimize (TRIM) if you're drive has encryption or is going to be encrypted. So for corporations, or end-users using encryption, forget it.

3. It will not allow to optimize (TRIM) if you have a dual boot, or multiboot system. If the tool finds other bootable partitions, it won't allow you to perform this function. So if you have Win7 and WinXP in dual boot, forget using this function.

4. It will not allow to optimize (TRIM) if you have a RAID configuration.

More problems:

This is a terrible disfunction on the part of Intel. Many users will want to clone their laptop hard drive to improve performace and battery life. All laptops these days do not come with the OS installation CD's, so a fresh install is out of the question to install without AHCI or RAID enabled in non-Intel and Intel based systems. The HD images provided by the laptop manufacturer do not allow for such alterations at the time of install. If done after install, expect BSOD's at boot.

For non-laptop situations, there is another problem. If you are already in RAID or AHCI, you'll need to perform a fresh install and install without if possible. If you need F6 to install a non standard SATA driver, you're probably already going to have problems. So cloning your old HD is out of the question in many home pc configurations.

One could attempt to hack into the registry and disable RAID of AHCI in order to try and let Windows use a generic SATA driver, but who wants to do that just get a feature to work in the Intel SSD toolbox?

The Intel SSD Toolbox is touted as an added value, especially for XP users since XP does not itself support TRIM. It allows for scheduled TRIM, which is nice if it likes your configuration and allows you to.

More bad news:

You cannot flash the Intel SSD from the toolbox. There is no such function. Why have a toolbox that cannot flash the drive? OCZ does. The tool also doesn't provide the flash version on the SSD. If you look in Windows device manager, it also won't display the flash version either. So, you have to download an iso from Intel, burn a bootable CD, have it search for the SSD at boot time, to tell you that you don't need to flash it????. A waste of time and a wasted CD.

From bad to worse:

Performance wise, worse than the competitors. Although sequential read and writes are a little faster, other performace tests are close to or slower than other SSD manufactures. On a Win7 Ultimate system with 6GB ram, intel motherboard, and performing the Intel toolbox optimizations, boot and shutdown were a few seconds slower than an OCZ Vertex2. I assure everyone that the SSD was aligned correctly (4k) and verified.

So, in conclusion:

The toolbox will only work as advertised in an Intel based PC, without multi-boot partitions, without RAID configuartion, and without encryption. Most HD clonings will reveal that the tool won't work well.

Two thumbs up for Amazon, two thumbs down for Intel. This SSD is an old dog. That's the scoop.

[Ao1]

SandForce created duracalss

and they can keep it

This SSD is an old dog. That's the scoop.

shill

[Brahmzy]

R0 with ssd is just not worth the hassle, ever.

How wrong you are.
First, there's no "hassle" - R0 is cake, done correctly.
Depending on usage, gains are huge.

[Tony]

Just so long as you don't ask a question that OCZ does not like. If you ask too difficult of a question, the thread will get closed.

you can ask any question you like as long as it does not contravene forum rules on competitor products and you are not rude to the staff.

In all my years running the forum ( yes I do run it) i have edited no more than 10 threads and all were for competitor product links.

If you want to show comparison benches or talk about someone elses product do it on that competitors forum or on a neutral forum like this one...

simple rule due to the forum being manufacturer owned, corsair have even stricter rules on their forum as do others...

I feel you guys feel you can say what you like where you like...sorry that is not going to happen.

[Ao1]

Tony, can you please explain how DuraClass works? Specifically:

Are drives configured for a 3 year or 5 year life span?

Is the write limitation applied to all SF based SSD's?

How much data are you allowed to write per day before DuraClass kicks in?

Does DuraClass distinguish between non compressible and compressible data? i.e. you can write 2GB per day non compressible and 10GB 100% compressible data?

How are writes subsequently controlled? For example assuming DuraClass is configured to ensure a 3 year life span that equates to ~0.64 on the wear out indicator per week.

According to the experiment I ran in the SSD Write Endurance thread I managed to put 9% wear on the SSD in 7 days before throttling kicked in. If I don't use the SSD for the next 14 weeks will throttling then be turned off?

Does the SSD need to be powered on/ idled once throttled to register the drop in writes to enable throttling to be deactivated or could I unplug it and be good to go in 14 weeks when I reconnected it?

I believe it is possible to check a SMART attribute to see if writes are being throttled. Is this feature available on all drives or only drives from a certain generation? Can you please explain how to check this?

Why does DuraClass throttle reads?

Thank you in advance.

(Can always ask on the OCZ forum if you prefer)

[Callsign_Vega]

Luckily Newegg is doing an advance RMA on my two 120GB MaxIOPS V3's and I will have the replacements here in a couple of days. I just hope they have firmware 2.02 on them or I might end up right back where I started. I'll just Ebay the Corsair Force 3's as it looks like their performance is a bit slower.

[johnw]

you can ask any question you like as long as it does not contravene forum rules on competitor products and you are not rude to the staff.

Not true. There was a thread about the brand and quality of flash used in Vertex 2 SSDs that got closed a while ago, apparently because OCZ did not want to answer detailed questions about the quality of flash used in their SSDs.

[Ao1]

Tony, further to my earlier post I can see in the Vertex 2 FW release notes for revision 1.11 from 1.10:

Modified SMART attribute 230 to indicate “lifetime throttling was active for last write” instead of “lifetime throttling is currently active”

Attribute 230 does not appear in the OCZ toolbox or any of the 3rd party SMART monitoring tools I have tried. How are you supposed to get a reading of that attribute?

[Callsign_Vega]

How wrong you are.
First, there's no "hassle" - R0 is cake, done correctly.
Depending on usage, gains are huge.

Brahmzy are you sticking with 2.02? Just curious if you are to the challenge of 2.06!

[Brahmzy]

Brahmzy are you sticking with 2.02? Just curious if you are to the challenge of 2.06!

Definitely stayin' with 2.02 man. These things have been fast and rock-solid for me - don't wanna change that.

I think we'll see new SF firmware soon enough.