IS this a reliable SSD?
- Thread starter blip
- Start date
I don't know what controller's in that Cruical drive, but the Samsung is great from all of what I've seen around the web. Pretty cool to have a drive where all of the major electronics is sourced from the same manufacturer. Controller, RAM, NAND...
I'll second the endorsement of Samsung 830 series, I have bought two 256GB drives and one 128 GB drive. My workstation at work has one, as do my home PCs.
Had an export (which runs multiple times per day) of a subversion repository with 90,000+ files and over 20,000 versions go from over 10 minutes to 4 seconds. The only mechanical HDDs I will ever buy is for my NAS holding all my media data.
Cheers
UniversalTruth
Veteran
Is it possible to have at some point in time SSD with lifetime warranties? What's the weak portion in them which causes these relatively high failure rates?
Anyone, do you have data about failures of parts with lifetime warranties, such as memory modules, just as an example.
Of course, lifetime warranty is something really interesting, since nothing can work forever and ever, but maybe the idea is that the product will get obsolete by the time of its first failure.
Anyone, do you have data about failures of parts with lifetime warranties, such as memory modules, just as an example.
Of course, lifetime warranty is something really interesting, since nothing can work forever and ever, but maybe the idea is that the product will get obsolete by the time of its first failure.
You will never get lifetime warranties on devices that are incapable of lasting a lifetime. There are 2 types of hard drives and ssds: those that have failed and those that have not failed YET. Given enough time and usage, they will both fail.
As for failure rates, hard drives have a bath-tub curve on their mortality rate. They generally fail early on in their life-cycle or at the end of their time. For SSD's the flash cells have limited write/erase cycle counts. There's no way around that currently.
As for failure rates, hard drives have a bath-tub curve on their mortality rate. They generally fail early on in their life-cycle or at the end of their time. For SSD's the flash cells have limited write/erase cycle counts. There's no way around that currently.
Never say never... Famous last words, and all that.
I bought 5 pairs of sports socks in Montreal, Canada back in 1994. They had a lifetime warranty. (!)
Best damn socks I ever owned by the way. I had them in my sock rotation for at least ten years and even though they eventually wore out one by one, I made sure to put the last surviving pair away as a keepsake. They earned their retirement, god dammit!
I can only assume that the lifetime warranty was motivated by most people never actually bothering to collect on it. After all, if you make really good socks that last for a long while, you're unlikely to keep the receipt, packaging or anything else related to that purchase for that long. And besides, you probably figure you'd gotten your money's worth out of those socks anyway, and just go buy another pair instead. It's not as if these things cost all that much anyway.
True, but that's actually the case with everything man-made. Entrophy if nothing else will get to it eventually. Shit, even the very elementary particles of matter are destined for destruction if the current theories of physics turn out to be correct.
Blazkowicz
Legend
thanks, I'm gonna call my friend because most probably, that one failure happened. the drive was in a small USB enclosure with no cooling, sold as is.
if we can switch PCBs with a working drive, we can retrieve countless hours of work. but the worst part of it would be buying the same WD green drive!
it could be relegated to being a backup drive turned on once a month. so not everything is lost. in a bigger enclosure with whiny fan or better, just hanging out of a computer case.
This can be done, but it must be the same pcb and there are a bunch of versions. I resurrected one Green by putting a fan on it long enough to pull a couple files off...then it died the rest of the way.
Greens in no fan enclosures have died the most for me. That's why I always do NAS boxes with fans.
Blazkowicz
Legend
there's at least one website where you can order drives from the exact same model precisely for that kind of recovery, I can also place a request on a webforum where I may have a good chance of finding it locally.
I've never cared about heat for drives, but for a fun project - stuffing 6 or 7 IDE HDD in a box for storing music - I will make sure to transplant the hardware in an old "big tower" case.
I've never cared about heat for drives, but for a fun project - stuffing 6 or 7 IDE HDD in a box for storing music - I will make sure to transplant the hardware in an old "big tower" case.
One of the crazy design differences between those "just as good" Greens and the Blacks is that the Green PCB has the components facing the drive housing (presumably using the housing as a heat sink) while the blacks have the components facing outward so air circulates over them.
Heat sinking against the housing is really common these days it seems. Used all the time with 2.5 drives for example, including the velociraptor series.
I'm not sure which method really leads to the lowest component temps, air flow often isn't terribly good around a harddrive, speaking in general. Enthusiast chassis may have fans blowing right at the drive cages of course, but that's not always the case. Those chips are encased in epoxy resin, which is a terrible heat conductor, and air is pretty terrible too so if there's not much flow the chip will get rather warm. The aluminium of the casing is at least able to soak up the heat and spread it decently well...
I'm not sure which method really leads to the lowest component temps, air flow often isn't terribly good around a harddrive, speaking in general. Enthusiast chassis may have fans blowing right at the drive cages of course, but that's not always the case. Those chips are encased in epoxy resin, which is a terrible heat conductor, and air is pretty terrible too so if there's not much flow the chip will get rather warm. The aluminium of the casing is at least able to soak up the heat and spread it decently well...
Blazkowicz
Legend
the drive that failed was in a very small enclosure made of back plastic. the enclosure electronics is on a single, very small PCB with connectors soldered on, nothing like the old-style ones with IDE ribbon and molex cable.
here is the model. it's very small and good looking, but air can't even escape.
a slick black hard drive coffin :
http://www.newegg.com/Product/Produ...ark=&IsFeedbackTab=true&Page=2#scrollFullInfo
here is the model. it's very small and good looking, but air can't even escape
.a slick black hard drive coffin :
http://www.newegg.com/Product/Produ...ark=&IsFeedbackTab=true&Page=2#scrollFullInfo
Velociraptors sink to the external heat sink around the drive, not to the hard drive itself. Greens have no external heat sink.
If the type of hard drive heatsinking on your setup makes a huge difference in temps then I'm sorry but you have it setup wrong...
My Green hardly ever rises above 100F..which is barely above body temp...lol.
Time to stop finding excuses to somehow "prove" your non-winnable case....
There's really no warranty solution for user stupidity...maybe it's time to buy hard drive insurance...lol
My Green hardly ever rises above 100F..which is barely above body temp...lol.
Time to stop finding excuses to somehow "prove" your non-winnable case....
There's really no warranty solution for user stupidity...maybe it's time to buy hard drive insurance...lol
Last edited by a moderator:
Thanks for the reminder.
Sandisk is REALLY big in the flash market, I'd be surprised if it's a bad drive. Actually, are there ANY bad SSDs anymore? I don't really think so; people bitch and whine over OCZ, but if what they made really was bad - like you'd commonly see before Intel entered the scene - they'd go bankrupt immediately.
It's probably a drive using generic firmware issued by the manufacturer (or if it's a Marvell drive, Sandisk prolly has its own engineers working on custom firmware since from what I understand, Marvell itself doesn't put much work into its own firmware.)
It's probably a drive using generic firmware issued by the manufacturer (or if it's a Marvell drive, Sandisk prolly has its own engineers working on custom firmware since from what I understand, Marvell itself doesn't put much work into its own firmware.)
I have 3x Vertex 2 60GB drives (34nm variety) around. I think I had one cause a BSOD once but that's been the only problem. Also have an 80GB Intel 320 and a 120GB Crucial M4. Thankfully no problems with any of them! And I can't tell a practical speed difference between any of them.
My WD Green experience has been fine too. I have 3 models. One is the earliest WD10EACS, which is a dog, slowing to <40MB/s writing the outer tracks, and causes stutters if you try to use it for gaming because of the head unloading. I had that drive in a fanless USB enclosure attached to my router for almost 2 years. I didn't tweak the head loading until recently so it has a zillion load/unload events. But it works fine still.
The most recent failures I've seen are several Seagate 1.5TB and a couple of 150GB 3.5" WD Raptors. Though I do have one of that Seagate 1.5TB batch still working fine in a Synology DS111.
My WD Green experience has been fine too. I have 3 models. One is the earliest WD10EACS, which is a dog, slowing to <40MB/s writing the outer tracks, and causes stutters if you try to use it for gaming because of the head unloading. I had that drive in a fanless USB enclosure attached to my router for almost 2 years. I didn't tweak the head loading until recently so it has a zillion load/unload events. But it works fine still.
The most recent failures I've seen are several Seagate 1.5TB and a couple of 150GB 3.5" WD Raptors. Though I do have one of that Seagate 1.5TB batch still working fine in a Synology DS111.
