Had my first bad SSD experience

Juddog

Diamond Member
Dec 11, 2006
7,851
6
81
Long Version: for the past 2-3 weeks or so, I've had random BSODs.

This is my watercooled windows 7 rig in my sig. Before the BSODs over the past few weeks my system was running great. I tried tweaking down the memory speed, adding more voltage to the core, bought a new video card.

Still the BSODs continued to plague me. I had a plan to get rid of my old OCZ Vertex (1st gen, the kind that required a destructive upgrade to firmware) and replace it with a second-gen Intel SSD, which were known for being rock solid stable but not very fast.

I wasn't psyched about the upgrade really but mostly was looking to upgrade to have more disk space (120 GB on OCZ Vertex vs. 160 GB on Intel). To get the Intel SSD ready, I wanted to run a complete nuke on the drive via the Intel utility to make sure it was completely clear, but kept getting an error message that the drive was secure locked.

Having read through some forums, it recommended a custom Linux boot called "gparted", so I finally got a USB stick with the gparted live boot on there, booted up, and finally was able to "secure erase" the Intel drive, then cloned over the OCZ Vertex to the Intel drive. My plan was to then erase the OCZ drive and use that in an old laptop I had, to give the laptop some pep.

So I go to secure erase the OCZ drive and notice it's not even an option. I do the regular full format and then notice the drive has some red marks (errors) on it. The OCZ SMART table on their website only seems to exist for the Vertex2 + models, so the field values don't make a lot of sense. I'm wondering now - should I just throw the drive out, or is there any hope for an old SSD that has mostly depleted NAND cells?

Every since swapping the SSD the computer has been rock solid stable, so it was definitely the drive.

Cliff Notes:
* computer randomly BSOD'ing, found it was the OCZ Vertex (first gen)
* Replaced SSD with newer gen, works like a champ
* What do I do with this old Vertex that seems to be mostly depleted?
 

bbhaag

Diamond Member
Jul 2, 2011
7,517
3,172
146
Kind of stinks you lost the Vertex but those first gen models were always flaky anyway I thought. At least you got your data over to the Intel before you lost the drive completely.:thumbsup:
 

Juddog

Diamond Member
Dec 11, 2006
7,851
6
81
Kind of stinks you lost the Vertex but those first gen models were always flaky anyway I thought. At least you got your data over to the Intel before you lost the drive completely.:thumbsup:

Yeah true, reading through the newegg reviews on the first gen Vertex had a lot of people all complaining about random BSODs cropping up (I started reading reviews on the Vertex afterwards when searching for the SMART codes and proper field values).
 

fzabkar

Member
Jun 14, 2013
173
46
101
Can you post the SMART data? Does the following document help?

Kingston® SF-2000 Based SSD SMART Attributes:
http://hddguardian.googlecode.com/svn/docs/Kingston SMART attributes details.pdf

I have a theory (based on some testing at HDD Guru) that the reported high failure rates of OCZ SSDs may be due to them being designed with a low NAND flash supply voltage (2.8V versus 3.3V). Would you indulge me by measuring some supply voltages on the PCB? I would need to see a photo to show you what to measure. You would only need a cheap digital multimeter (~US$5).

BTW, when I asked for similar feedback at OCZ's forum, my posts were at first closed then subsequently deleted, so I'm wondering whether I struck a nerve. A similar request at Tom's Hardware remains unanswered. :-(
 
Last edited:
Feb 25, 2011
17,000
1,628
126
Bummer. I was getting hard freezes on my laptop w/ SSD, but it turned out to be fixable. (Well, fixable after I spent $100 on a directory rebuild tool, but I didn't have to reformat.)
 

Juddog

Diamond Member
Dec 11, 2006
7,851
6
81
Can you post the SMART data? Does the following document help?

Kingston® SF-2000 Based SSD SMART Attributes:
http://hddguardian.googlecode.com/svn/docs/Kingston SMART attributes details.pdf

I have a theory (based on some testing at HDD Guru) that the reported high failure rates of OCZ SSDs may be due to them being designed with a low NAND flash supply voltage (2.8V versus 3.3V). Would you indulge me by measuring some supply voltages on the PCB? I would need to see a photo to show you what to measure. You would only need a cheap digital multimeter (~US$5).

BTW, when I asked for similar feedback at OCZ's forum, my posts were at first closed then subsequently deleted, so I'm wondering whether I struck a nerve. A similar request at Tom's Hardware remains unanswered. :-(

I don't have a multimeter unfortunately.

I'll try to boot up tomorrow to the Linux distro and see what data comes up from the SMART report.