altitude909
Best Advice Ever: RTFM
I need a new one (SATA, no more the .5 T) and it has been a bit so I was hoping on some recommendations/input
dgatwood
is out. Leave a message.
They all suck. Between myself and my friends, we've lost drives by every single drive manufacturer within the last two years. They are orders of magnitude less reliable than drives built five years ago.
Whatever you buy, buy a second one that's at least twice as big and back up the entire content of the small drive to the large drive religiously every week. Once the larger drive is full of backups, each time you back up, delete the oldest existing backup or two to make room.
Average drive lifespan for me is about 9-10 months of regular use, spread across a wide range of machines, both desktop and laptop. That's terrifying to say the least.
Whatever you buy, buy a second one that's at least twice as big and back up the entire content of the small drive to the large drive religiously every week. Once the larger drive is full of backups, each time you back up, delete the oldest existing backup or two to make room.
Average drive lifespan for me is about 9-10 months of regular use, spread across a wide range of machines, both desktop and laptop. That's terrifying to say the least.
mattr
Resident moody teenager
I recommend Samsung Spinpoint F1 drives. Added another TB (in the form of 2x 500gb F1s) to my computer a few months ago and they've been perfect. Very speedy!
altitude909
Best Advice Ever: RTFM
I hear you there, I just did a "freeze and pray" on a raptor and that was a pretty serious wake up call (thankfully successful, 100% recovery). Albeit, it was going on it's 6th year of heavy use but I would have been up shit creek if I would have lost that data and it died without any real warning. I just keep seeing large DOA/quick fail reports and was hoping there were some trends when it comes to reliability for some of these things
Bristol Posse
Okey Dokey
Samsung have pretty high failure rates comparitively speaking
Seagate Baracuda and WD Caviars are pretty solid performers in the 7200rpm range (Seagates can also be had with a 32MB cache although they run a little hotter than WD
on the faster speed drives Velociraptors are more rleiable, less noisy and run cooler than their lder brother the Raptor. In a good Antec case no noisier than a 7200 RPM drive
Right now I'm experimenting with Intel Solid state Drives. Have to Hard raid em to get a decent capacity but speed is astonishing and since no moving parts, no noise.
Seagate Baracuda and WD Caviars are pretty solid performers in the 7200rpm range (Seagates can also be had with a 32MB cache although they run a little hotter than WD
on the faster speed drives Velociraptors are more rleiable, less noisy and run cooler than their lder brother the Raptor. In a good Antec case no noisier than a 7200 RPM drive
Right now I'm experimenting with Intel Solid state Drives. Have to Hard raid em to get a decent capacity but speed is astonishing and since no moving parts, no noise.
travisisgood
New member
Yeah I was reading about the WD caviar - both green and black - and they were both pretty reliable harddrives. The green wasn't thatt fast when it cames to read/write but it wll save some power (hence its name). and the black was meant moreso for performance.
this may not be the same site that i originally saw, but the results are similar.
http://techreport.com/articles.x/15769/1
this may not be the same site that i originally saw, but the results are similar.
http://techreport.com/articles.x/15769/1
Bristol Posse
Okey Dokey
Caviar Blacks'll giver better boot times, faster loading and possibly higher potential track counts
If you really want to save energy turn off the machine when you're not using it
D
dementedchord
Psychotic State alumni ch
you'ld think this was the case ... though in reality it uses very little current at rest...
i've got three seagates... with the 32M onboard buffer... five year warrantee is hard to beat...
Bristol Posse
Okey Dokey
Agreed
I was just making a silly point that the energy savings of a cavier green over a regular drive compared against perforance drop. you'd be better off using the Caviar black and turning the machine off once in a while if your that worried about a few watts
buckkillr8
David Gilmour's clone
+1
I've never had any problems with either of these drives. I think the high failure rate has to do with the user using them more then the actual drive.
dang donkey
New member
B
boingoman
New member
Nope. Even the retards on this board can't break a hard drive simply by using it. For the most part, drives die, or they don't. And I mean individual drives, not particular brands. Except for one, apparently.
Check out Google's hard drive study for some interesting stats. They use millions of drives, and did a study of 100,000 drives that failed. That kind of number actually means something, unlike reading on a message board that "hey, I have one of those, it's great" or "those suck, I had one die." That kind of crap is totally meaningless.
Bristol Posse
Okey Dokey
Not quite true. Impropper control of: dust management on drives with exposed platters, Vibration damping, Proper cooling and airflow can shorten drive life as can over handling droping the drive or bumping the tower hard and not using an adequate or badly inefficient PSU. Never turning off your PC will also shorten the useful life of your drive (as you burn up hours of drive spooling with no use)
Personally I've built close to 45 rigs in the last 2 years with WD and Seagate and never had a failure.
Waranty is a pretty good indication of a companies faith in the durability of their product and WD and Samsung have good warantees.
Samsung are not highly regarded in the PC builders community because of their reputation for drive failure
Last edited:
dgatwood
is out. Leave a message.
Funny you should mention that. I tried unsuccessfully to recover data from a friend's dead Samsung 250GB drive just last week. The heads not only crashed, but did platter damage such that the heads sounded like they were fine when we started recovering data (and even read data), but then when we left it sitting there to try to copy further data, it got bits of magnetic material under the heads, which then tore the rest of the surface to shreds. I didn't discover what had happened until we disassembled the drive. Totally f*cked.
dgatwood
is out. Leave a message.
In my case, it tends to be things like "those suck, I had three die". That's not nearly as meaningless. Three Seagate drives in two weeks, mind you. Three out of the three Seagate drives I had in regular use at the time, all of which were about the same age. And some Seagate models right now have a known firmware problem that is taking out drives in droves (like close to half of the drives failing within the first year), albeit with the data recoverable (and I think Seagate is even footing the bill for recovery in this case).
And I've talked to folks who have dealt with certain other drives by other manufacturers that quite literally failed consistently after a given number of powered-on hours within a range of just a couple of percent. If this were uncommon, it wouldn't be so sad, but it really isn't nearly as uncommon as you might think. Certain models of drives just plain are unusable in terms of failure rates, and it moves from one manufacturer to another rapidly enough that it's pretty hard to track.
Unfortunately, Google's study doesn't reveal brands, but even that would mainly tell you what I've already said: the only good hard drive is the drive encased in a lead-lined vault somewhere in the hills of eastern Kentucky that is powered down, not spinning, and contains the fifth backup copy of your data.
The two really interesting things the Google study showed are:
1. S.M.A.R.T. is practically useless. A large percentage of drives showed no signs of failing until they went catastrophically bad.
2. With modern fluid-bearing-based drives, cold temperatures are worse than hot temperatures. Optimal is around 38C (100 degrees Fahrenheit), but failure rate is nearly flat from 36C-45C (97-113 degrees Fahrenheit). The worst failure rate came from drives running below 20C (68 degrees Fahrenheit).
The big problem with it is that all the data is, at this point, nearly useless as a metric for judging drive reliability of today's drives, as they use different mechanisms. Also, because AFAIK their study was about servers that presumably use server-grade drives, this really can't be generalized to the context of typical desktop or laptop hard drives. Bear in mind that server-grade drives use different firmware and may even have significantly different hardware than the equivalent desktop drives in some cases (particularly where things like parking ramps and frequency of head parking come into play).
Au contraire. In my experience, what shortens the life of disks most severely is the constant power cycling of the drive spindle and parking of the heads associated with spinning the drive down. If you want to truly maximize the drive life, disable all power management and leave the machine running 24x7. I still have drives running after well over a decade that were treated in this way, and that's not at all uncommon. In particular, parking heads puts a lot of stress on the heads and head arms.
Read this:
http://68kmla.net/forums/viewtopic.php?p=78086&sid=d922cfbb249658ba7725c983df64632c#78205
altitude909
Best Advice Ever: RTFM
Cosigned. I have a half dozen 2 gig samsungs that are over a decade old, all are in 24/7 win2k data loggers. My raptor that just died was purchased in 2003
dgatwood
is out. Leave a message.
The old globegate server's drive dates back to either early 97 or early 98, shortly before the factory Quantum drive in that machine ate itself.... The ancient drive is a 9GB Seagate SCSI drive, if you believe that....
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: SEAGATE Model: ST19171N Rev: 0024
Type: Direct-Access ANSI SCSI revision: 02
The remarkable thing is that judging from the used prices I'm seeing, I could pull that drive, and as a working pull, could apparently sell it today for just about what it cost new in the late 90s.
altitude909
Best Advice Ever: RTFM
I'll believe that in a second. Our old machines (running win98 and Labview 3 mind you) are FRICKING PENTIUM 100's and for all I know, they were assembled in Micheal Dell's dorm room. They all have DAQ cards in them that cost 3x more than the computer at the time and have been doing exactly what they were designed to do for a LONG TIME. I think a PSU went bad on one but beyond that, I cant really recall anything else going south. Now since then, I have stacks of crappy dead P3 main boards (the wonderful bunko capacitor days) and piles of dead multigig hard drives that make such a racket when turned on that I am literally expecting them to fly apart. The commoditization of computer hardware is not the best thing in the world
I ended up with a WD RE3, a .5 TB external and Acronis so hopefully I'll be able to sleep a little better at night after the nightmare last weekend. Lesson learned
I ended up with a WD RE3, a .5 TB external and Acronis so hopefully I'll be able to sleep a little better at night after the nightmare last weekend. Lesson learned