SSD cluster%#$@!

[Aergis]

Recently my old SSD started acting out. I was getting blue screens and after rebooting the bios can't find the drive... So I decide it's time to upgrade to the latest gen SSDs. Mine was a Corsair 1st gen 120gb from a few years ago. So I go for Corsair Force3 120 and it's worse. Bluescreens every 10 mins. Doing more research I find out these have actually been recalled by corsair and I shouldn't have even been able to buy it...

So returned that one and got a OCZ Vertex3. Worked phenomenally for 4-5 days, super fast 500/450! Then same old story. Random bluescreens after 30-45 sec lockups, though only 4 or 5 times per day. RMA for replacement, same thing on the second one. Tried the page file tricks, firmware upgrade, LPM registry tweak, anything I could find online. No help, and it seems to happen the most while flying around in WoW.

Finally convinced newegg to take the thing back for money instead of replacement and I am now trying a Crucial M4, slower but reportedly more stable... Should be here in the mail today.

Did I just get super unlucky or is the SSD market SHIT right now? I've seen soo many problems with the Sandforce controller and no solid leads to fixing it... And across multiple manufacturers?

[Arkesh]

Wow, that's some pretty bad luck. I've been using an OCZ Vertex 2 for about 8 months now and not had any issues (touch wood). The only issue I had been aware of recently was with the Intel SSDs, where they'd drop to 6MB capacity after a power loss.

I'm sure you'll have done this, but tried using a different power and/or SATA cable? Probably unlikely that it's one of those, but maybe a dodgey port or something.

[Talaii]

Wow, that's some pretty bad luck. I've been using an OCZ Vertex 2 for about 8 months now and not had any issues (touch wood). The only issue I had been aware of recently was with the Intel SSDs, where they'd drop to 6MB capacity after a power loss.

I'm sure you'll have done this, but tried using a different power and/or SATA cable? Probably unlikely that it's one of those, but maybe a dodgey port or something.

It's quite possible. I had a drive in my server's RAID array fail over and over again - same pattern: It would fail in the fileserver, I'd test it on my desktop and it'd be fine, try and rebuild the array, drive would fail, etc. Turned out it was the SATA cable, some of the connections at the drive end were loose. Drive heated up, cable lost connection, RAID controller panicked. Switched the cable, all good now.

[RedAces]

The PCGH (a german computer magazine - PC Games Hardware) tested a bunch of SSDs and had weird errors and all... they had the same problem - broken SATA cable. Check if it's working with another cable or maybe test the questionable cable (oh god that sounds wrong - the old cable?).

[Aergis]

I'll try that, I have been using the same cable come to think of it. Though it is a new cable as of 3 months ago with the MB.
I would expect similar behaviour though if it was the cable. The 2 year old SSD was just starting to act up, and was getting it to disappear in BIOS. The Corsair Force3 was a nightmare, could barely get 10 mins without a freeze. The OCZ Vertex3s have been much much better, but still 4-5 times per day it will crash. Twice today when opening my bank bags.... Guessing it's because it has to pull some data which causes some sort of hiccup. Normal job/work stuff like 3ds max photoshop flash it works fine. Though I have come back to bluescreens after leaving the machine for 30 mins or more, but I have sleep/hibernation off for rendering...

So yeah, my first impression if it were a cable would be repetitive. But 3 drives act distinctly different, but they all suck. And the UPS guy didn't come, so I guess I have to wait till monday. I'll try swapping the cable tomorrow and see if it helps.

[Aergis]

Oh, also side note. I noticed that with the vertex3 in a Asus P8P67 Pro was getting 350/300 readwrite. When I swapped boards to a Asus P8P67 WS Revolution it instantly jumped to 500/450... Both were in the Marvell SATA3 ports.

[theckhd]

My OCZ Vertex 3 has been working perfectly on my ASUS P8P67 since I got it in April. I believe I have mine on the native (gray) ports rather than the Marvell controller.

Given that you've had failures with multiple different SSDs from multiple different vendors, you either have the worst luck in the world or it's a problem being caused by something else in your system. Definitely try swapping the SATA cable, possibly swap it off of the Marvell controller as well. Another thing to check would be heat, running it outside of the case would rule that out.

Note that the different failure modes of each drive doesn't rule out the cable. If it's something as simple as thermal expansion causing intermittent loose connections, then each drive could exhibit different symptoms because each drive might have slightly different heating and expansion patterns, even within the SATA connector.

[Sabindeus]

I've been running a few SSDs with Sandforce for a few months now with no issues... I'd probably look at other potential culprits like people said

[Arees]

I've been using a 120GB Intel 320 SSD for several months now with no issues at all. It only supports SATA II and not III though, so its not the fastest out there, but I was extremely impressed by the decreased load times...

[gibborim]

Given that you've had failures with multiple different SSDs from multiple different vendors, you either have the worst luck in the world or it's a problem being caused by something else in your system.

This. Have you tried these drives in different computers? It seems far more likely that the rest of your box is the culprit rather than the one part you keep replacing.

[Aergis]

Quick update for anyone who was interested!
Got my replacement in the mail but havent installed it yet as I am curious to know if you guys are right that it's another culprit.

So I replaced my sata cable, removed my second 2tb files drive, swapped out fresh ram.
It is still happening at about the same frequency.

The MB is brand new as of 2 weeks ago, and not the same one I was using with the first SSD failure ( the 2 year old one ).
The only things remaining as constants are the case, 1200w PSU, a 580GTX, and a stock clocked i2600k


Of those which could be causing issues?

[Arkesh]

My money would be on the power cable, is there another one you can use? The processor maybe could be the issue, but it's less likely and the graphics card won't affect it, to my knowledge.

[Hrobertgar]

Here is a good article on the situation with SSDs:

http://www.anandtech.com/show/4604/the-sandforce-roundup-corsair-patriot-ocz-owc-memoright-ssds-compared

[Jheherrin]

I went with OCZ's PCI Express based SSD (Revodrive) rather than a SATA one. No issues with cables or power supplies

Had no problems with it in the last 6 months since installation. The only issue I had was during installation as my windows 7 did not have drivers for it.

The only downside is that it needs an x4 PCI-x slot on the motherboard, which can restrict your options for multi card setups on the lower end motherboards.