04-30-2013 07:32 PM - edited 04-30-2013 07:41 PM
I've got a 480GB m500 in my main pc on a marvell 9128 controller (EVGA SR-2 motherboard). This is actually my second m500 having already RMA'd the first with amazon.
First drive, I used "paragon migrate OS to ssd". Seemed to work well, although benchmarks were a bit below what has been posted on various reviews for the m500 devices. Still, performance wasn't bad. After about a week, the system blue screened (windows 8) and would no longer boot. Scandisk ran and repaired some disk corruption, but the system still would blue screen during boot from that point on.
I reconnected my spinning disk and booted and repaired the SSD, still no boot from the ssd though. Ran HD tune pro to read the SMART data which reported errors "(C4) Reallocated Event Count" was in a warning state. I forget the number. That's when I RMA'd the drive.
Replacement came in. For some reason the migration would not work this time; but a fresh install of win8 on the drive worked. It ran fine again for about a week; performance on this drive was slightly above that of the first but still a good bit below published benchmarks. Today I rebooted after an install and got the all too familliar bluescreen and auto repair. Auto repair failed, system restore failed.
I again booted from the spinning disk and ran HD tune to read the smart data. Again, I see "(C4) Reallocated Event Count" is at 16 and is throwing a warning.
Is there a bad run of m500 devices going around? I'm more than a little concerned, as I know I can get a refund from amazon for this one but I ordered three more 240's for my laptops from newegg which is usually a little more of a pain if I have to rma...
There doesn't seem to be ANY information at all on the crucial support site for m500; m4 is the most recent I can find.
Any suggestions other than get a refund and try the "fourth generation" of ssd's whenever those come out?
05-01-2013 12:45 AM - edited 05-01-2013 01:00 AM
Try check the SMART attributes in a program such as CrystalDiskInfo or Hard Disk Sentinel. Been a while since I checked, but HD Tune was not programmed to read SMART attributes of SSDs last time I checked.
Maybe try a different SATA cable and check your RAM?
ID (Dec) ID (Hex) Description
1 1 Raw Read Error Rate
5 5 Re-Allocated Sector Count
9 9 Power-On Hours Count
12 C Device Power-Cycle Count
13 D Soft Error Rate
170 AA Reserved Block Count
171 AB Program Fail Count
172 AC Erase Fail Count
173 AD Average Block-Erase Count
174 AE Unexpected Power Loss Count
181 B5 Vendor-Unique Data
183 B7 SATA Interface Downshift
184 B8 Error Correction Count
187 BB Reported Uncorrectable Errors
188 BC Command Time-Out Count
194 C2 Enclosure Temperature
195 C3 Vendor-Unique Data
196 C4 Reallocation Event Count
197 C5 Current Pending Sector Count
198 C6 SMART Off-line Scan Uncorrectable Sector Count
199 C7 Ultra-DMA CRC Error Count
202 CA Percent Lifetime Used
206 CE Write Error Rate (errors/MB)
Notes: 1. Attributes are preliminary and subject to change.
2. Attributes are applicable to M500 firmware revision MU02
05-01-2013 01:42 AM
My M500 has a value in reallocated event count and appears to function fine. It's never crashed and chkdsk and h2testw are perfectly happy with it. So it in itself isn't an indication of a problem. The drives have an insane amount of over provision so its a fairly insignificant amount of failure, assuming it's not increasing. The value on mine for that attribute and the accompanying attribute AA has been there since day 1 and has never increased. I wonder if it replaces the factory bad block count that seems to be no longer present?
As suggested above, you should use a more modern SMART monitor. For instance, the line you have highlighted in blue is not what HD Tune says at all. It is in fact utterly harmless.
Speed wise, the Marvell 192x controller can't run an SSD at anything like its top speed unfortunately. It's closer to sata 2.5 than 3. Also, it has been known to be unreliable for some people. So I would try avoiding using it and instead using your onboard intel sata 2 controllers. The Intel controller has higher random performance anyway so counter-intuitively is likely to have better real world performance.
07-19-2013 01:50 AM - edited 07-19-2013 01:53 AM
Are you actually having problems with the drive? As disscussed earlier in the thread, the drive has large areas of over provision to compensate for some SMART errors.
I haven't seen an M500 screenshot yet that doesn't have some reallocated sectors right out of the factory (including mine). I'm guessing they did away with the 'factory bad block' attribute and are just leaving them in reallocated. If that figure is not growing and the drive is otherwise functioning fine then I wouldn't worry about it. http://www.softpedia.com/get/System/System-Miscellaneous/H2testw.shtml is good for testing flash memory is OK.
07-22-2013 10:58 PM
Thanks! Yes, on one forum I have already read about this. Thay sad it's because of tight capacity of this drive
After that I entirely filled the disk with some of data and deleted. Looks like OK.
And thanks for the link