Ben's Bargains Serving Fresh Deals 24/7

come on baby.. $120 by black friday! (I've been predicting this for a while)

5900 rpm? that sucks.

and seagate. They're great but I'm still reading about failures.

Can we trust Seagate after the 1TB debacle?

If you're like me and run in RAID5 config, the error rates mean it is likely that after 1 disk fails, putting in a replacement disk will fail during the rebuild effort. RAID6 or JFS RAIDz2 address that better than other solutions like RAID10.

If you just need a scratch disk for temporary files, this could be fine.

I'll be waiting 6+ months to see how these really perform.

I'm sticking with WD RE's for raid.

#4: Absolutely every hard disk manufacturer has had issues at one time or another. Are you proposing that we don't buy hard disks any longer?

The warranty period is useless except for failures on brand new drives. I will never return a drive that has some of my data on it. The only exception to this is if I have used some serious encryption on the entire drive.

Judge each drive model on its own merits. Try not to buy brand new drive models - let someone else experience the failures and publish their dissatisfaction so you don't get caught.

Over the years, I have had fewer failures with Seagate than WD. But that's like attempting to predict the future stock price of a company by its previous performance - a rough guide at best.

Finally it's under $150, I'm in for 2.

Seagate: Company Fail

does anyone do any kind of a burn-in on new drives (a few hours/days?) Anyone have any specific programs to recommend for something like this?

Thanks.

#9, there's been no such thing since the mid 80's, just like cars.

#9 I just run all of the tests using the manufacturer's utility. Usually, a "long" test will test every sector's validity. If any bad sectors are found they will show up in the drive's SMART data and the program will (hopefully) also tell you of any bad sectors it may have found.

I generally check the SMART status of my drives weekly if not more frequently. If a drive develops even one bad sector I RMA it, and if it develops many in a short period of time then obviously back your baconnaise up and RMA it ASAP.

In for 4 to upgrade my Drobo!

Get the retail version from Fry.s when they have them for $159.

Funny... Seagate was the king up until about 1 year ago...Then the problem with the drive that needed the firmware update issue and now Seagate is the worse drive manufacturer ever.

#4: #6 is right. Good job searching wikipedia for buzzwords though.

#6: You're paranoid. What could they possibly want that's on your drive? Do you really think the support people when given a drive for a warranty claim laugh maniacally when they see there's data on it? If you're paranoid about that, then you shouldn't have a mailbox on any of the free webmail services. They can read your mail.


#11: As it turns out, SMART isn't.

#9, if you have a 64-bit processor, get Advanced Clustering Breakin and use it to exercise the whole box. It stress-tests memory, CPU, and drives, simultaneously. This means it's also putting your power and cooling subsystems under heavy load, so it's a good all-around stability check.

If you just want to thrash the disks, Microsoft SQLIOSIM is a good start, also try any of the disk benchmarks in continuous mode. Some of those don't actually test for errors, so check SMART afterward and see if there are any pending or remapped sectors.

7200 rpm or death! What's up with this slow carp being considered ok now? Then they just brand it as 'green' to cover up the fact it's just slow.

Agreed #15 about #6's paranoia. What exactly do you keep on your drive - kiddy pron? They're welcome to my saved games - or yeah maybe the odd email about meeting up this friday for beer. But I also run mine in a raid 5 stripe, so they can't rebuild from just one drive - just get striped bits of files. Yeah, you could bit-level dump the drive and yes you'd get some of the files less than the block size (64k usually) on one drive - but really - I don't think anyone would go that far.

*Maybe* they'd do it for a single drive that they know the data is all there, but probably not for a striped drive. You'd have to be really dedicated to sort through all that junk when there's lots of easier picking...

Some of you are just trying to be tards. There was a specific firmware issue with a specific series that is no longer relevant.

7200RPM is entirely unnecessary for most uses of 2TB on a PC, or were you planning on running Yahoo's webserver with this? If you're talking performance, you're talking about multiple drives and several gigs of main memory that's catching all the small I/O but the larger I/O in the realm of gigs of data isn't going to fit in main memory and can't be processed faster than the ~ 100MB/s, so all you're really talking about is shaving a few percent off first-load times after a reboot.

And for that, you're talking about SSD not 7k2 RPM HDD.

Unlike when you take your whole computer to some geek squad dork with no integrity or sense of responsibility (Or really I mean liability tied to that individual), if any hard drive manufacturer did anything with data they found on a drive, it'd be the end of their HDD business. If you think one bad series of firmware is bad, imagine how fast such news would traverse the web.

Retrieving data off a failed hard drive is beyond cost effective to a manufacturer. The technicians doing the work are paid more for the job than a brand new drive costs at retail, let alone tying up extremely expensive equipment and cleanrooms.