WD20EARS or WD20EADS? What about seagate's 2tb drive?

Silencing hard drives, optical drives and other storage devices

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Pierre
Posts: 156
Joined: Sun Jan 17, 2010 3:55 am
Location: Greece

WD20EARS or WD20EADS? What about seagate's 2tb drive?

Post by Pierre » Tue Feb 02, 2010 4:03 am

So I'm out for a 2TB drive...

I was set on the WD20EADS because of the very low - almost non-existent - seek noise and low vibration of the drive, as reported in the reviews in this site...
But then, the WD20EARS comes out which promises more storage area - better performance? - more cache and a future standard operating feature...

But I've read that the WD20EARS is not as quiet as the 32mb cache version...
However, both seem to suffer from the parking/unparking of the hdd head during operation, which makes me doubt about whether choosing anyone of them would be the best thing to do...

So, I'm also looking at the Seagate LP 2TB drive, which is a little bit faster and noisier than the WD20EADS - what about the WD20EARS though concerning noise and performance? - but lacks the problematic Green Power feature, as does with the EARS' new formatting feature...

So I ask myself, which one to buy?
- The quiet WDEADS?
- The "modern" WDEARS which may be noisier?
- Is the EARS noisier than the Seagate Drive?
- Is the Seagate the best balanced solution after all? Not as quiet as the EADS, not sporting the advance format feature of the EARS but better performing and without the hdd head parking issue?

There is the Hitachi 2TB, of course, but I guess this is noisier than all of the above, with performance I'm not that interested in...

So, what do you think?

Pierre
Posts: 156
Joined: Sun Jan 17, 2010 3:55 am
Location: Greece

Post by Pierre » Fri Feb 05, 2010 1:33 am

So I bought the Seagate LP 2TB...the head park/unpark issue, with no general solution given still, scared me off from buying the Western Digital WD20EARS drive, which I really wanted to, to tell you the truth...gaining 100gb of storage with a change of formatting is no little thing...

Also I bought the Akasa AK-HD03-BK hdd cooler to mount the drive on a 5,25 bay: http://www.akasa.com.tw/update.php?tpl= ... AK-HD-03BK

[quote]This passive black aluminium HDD cooling kit effectively reduces temperature, vibration and hard drive noise. Aluminium heatsinks with thermal conductive tape attach to both side of the HDD to improve heat dissipation and enable mounting of the 3.5â€

canadian
Posts: 14
Joined: Sat Sep 04, 2004 10:14 am

Re: WD20EARS or WD20EADS? What about seagate's 2tb drive?

Post by canadian » Fri Feb 05, 2010 3:37 am

Pierre wrote:So I'm out for a 2TB drive...

But I've read that the WD20EARS is not as quiet as the 32mb cache version...
However, both seem to suffer from the parking/unparking of the hdd head during operation, which makes me doubt about whether choosing anyone of them would be the best thing to do...

So I ask myself, which one to buy?
- The quiet WDEADS?
- The "modern" WDEARS which may be noisier?
- Is the EARS noisier than the Seagate Drive?
- Is the Seagate the best balanced solution after all? Not as quiet as the EADS, not sporting the advance format feature of the EARS but better performing and without the hdd head parking issue?

There is the Hitachi 2TB, of course, but I guess this is noisier than all of the above, with performance I'm not that interested in...

So, what do you think?
Hello, Pierre. I am wanting a 1-2 TB drive, too, and was considering those drives as well. The WD Green series is also scaring me off because of the same parking/unparking (+LCC) issues. I wonder if WD realizes they are losing sales because of that issue? Yet, they keep coming out with the same feature. Anyway, I am not sure if I can suggest other websites but it's for info and maybe a guide? The site? Tom's Hardware has a chart which includes benchmarks and other info (temps/power consumption) on these and other drives. I found it useful and interesting.

My concern with Seagate is with their QC and my concern with WD is with their Green drives and the power saving features (parking/unparking/LCC). The WD Green drives supposedly has some utility (that WD doesn't support anymore, though!) that disables or 'changes' how the parking/unparking works and lowers the LCC? But, it sounds really overly complex and not a universal solution, imho. Too many questions, I think.

As for the Hitachi 2GB, I believe it has an extra platter and for me, the heat/temps were higher than the other drives so I'm not considering that drive. I want a drive for storage, mainly, so I want LOW heat/power/noise. Performance is secondary although it would be nice if the read access performance was decent.

I hope this helped.

Vicotnik
*Lifetime Patron*
Posts: 1831
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Re: WD20EARS or WD20EADS? What about seagate's 2tb drive?

Post by Vicotnik » Fri Feb 05, 2010 6:21 am

canadian wrote:I wonder if WD realizes they are losing sales because of that issue?
They are gaining sales as well. I have bought a ton of these drives, mainly because of the head park feature. I love it.

andymcca
Posts: 404
Joined: Mon Jan 11, 2010 8:19 am
Location: Boston, MA, USA

Post by andymcca » Fri Feb 05, 2010 6:37 am

I wonder if you have considered the 1.5TB drives? $/GB they are less expensive (in the USA, anyway). I suppose in Watts/GB the 2 TB drives are superior, though.

I love my WD15EADS :D

Edit: Many rave about the Samsung F2 series, also, but I have yet to try one.

Pierre
Posts: 156
Joined: Sun Jan 17, 2010 3:55 am
Location: Greece

Post by Pierre » Fri Feb 05, 2010 6:44 am

@Canadian

I'm not that concerned with Seagate's QC (quality control,right?)...that's because I've owned many -almost exclusively- seagate drives the last five years and I've never had a bad incident with them...

...of course I make it my business of researching any hardware I want to buy, so when e.g. the 1.5TB Seagate drives were available in Greece I was already aware of the firmware issues they had, I read their support forums daily and I delayed buying the drive until a final fix was out and tested...

So, for me there were no issues....even those who did were able to ship their drives to seagate and have their data saved and copied on the new drive shipped back...

Anyway, I had some doubts about the reliability of this model, because of many reported failures on newegg website...but if those incidents seem to be limited to a given country/region and not verified elsewhere, I suppose it is a minor and limited to specific batch(es) problem...

Anyway, for me Seagate offered better performance overall (but not 7200 performance what I wouldn't make any use of), good thermal and noise behavior and no questionable operating feature...WD's hdds were my first and firm choice...but when I looked around and found no definite fix -and no universally reproduce-able results- I preferred to go on the safest side...
I really wanted the thing to be fixed though, because I was set on the EARS model (I've downloaded the offered utilities etc, burned the necessary cd to format the drives)

I haven't installed the drive, yet...needed to get some work out of the way first...

canadian
Posts: 14
Joined: Sat Sep 04, 2004 10:14 am

Post by canadian » Fri Feb 05, 2010 8:58 am

@Pierre:
Yes, it's difficult to assess the drives based on user opinion sometimes. I haven't had a bad experience with Seagate so I'm inclined to think most are okay but because many users were illustrating potential issues, I thought why risk it. But, the newest ones seemed to be okay but I just like the reported reviews and benchmarks of Samsung drives although they are harder for me to get. The Seagate LP series is probably my 2nd choice after Samsung.

I believe the Samsung F2 (and maybe F3?) series operates a couple of degrees less than the Seagate LP but I guess that's not much difference. The Seagate LP is apparently quieter, though, although I cannot confirm.

@Vicotnik:
I don't know about 'loving' the head parking/LCC issue but I have read of enough people who aren't too enthused about it. If other drives don't use it and can keep lower power consumption and low heat, then I would be concerned about that feature (as in, it's not needed?). I would rather err on the side of caution and just avoid possible issues. Also, as for the newer EARS models, I guess I am inclined to follow a similar guideline and that is to avoid joining the guinea pigs who are taking a chance on the new technology. It sounds like an interesting concept but sounds like more issues can result if you don't align the drive properly.

Pierre
Posts: 156
Joined: Sun Jan 17, 2010 3:55 am
Location: Greece

Post by Pierre » Fri Feb 05, 2010 9:27 am

If the new samsung eco drive at 2TB had been available when I made the purchase I would have liked to test it...I've heard many good words about Samsung drives and I'm now willing to "experiment" with other brands...before, I would have never touched a WD...

Concerning the advanced formatting feature, have you read of any bad experiences? The only -questionable- experiences I read about was the aligning format taking too much time to complete and mediocre performance (I think on the same case)
...but I've read in WD website that one is better advised to perform the alignment before formatting using any other tool and before using the drive, making the alignment with a bootcd -provided by acronis and released by WD- in order to save time...aligning from within windows would take considerable more time...it also cautioned to disable any power saving features while aligning, cause the process could get stuck...
Now I don't remember being absolutely confident that those who had reported problems with the drive had taken these precautions and followed the advised, time-efficient procedure...
Hard drives makers are also bound to totally replace current media format technology with the advanced format technology by January 2011...from then on, any single drive one buys will have the advanced format feature...

canadian
Posts: 14
Joined: Sat Sep 04, 2004 10:14 am

Post by canadian » Fri Feb 05, 2010 9:56 am

Can Parted/GParted accomplish the same thing as the WD tools or Acronis?

Vicotnik
*Lifetime Patron*
Posts: 1831
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Post by Vicotnik » Fri Feb 05, 2010 10:41 am

canadian wrote:@Vicotnik:
I don't know about 'loving' the head parking/LCC issue but I have read of enough people who aren't too enthused about it. If other drives don't use it and can keep lower power consumption and low heat, then I would be concerned about that feature (as in, it's not needed?). I would rather err on the side of caution and just avoid possible issues. Also, as for the newer EARS models, I guess I am inclined to follow a similar guideline and that is to avoid joining the guinea pigs who are taking a chance on the new technology. It sounds like an interesting concept but sounds like more issues can result if you don't align the drive properly.
I agree about the EARS-drives. The thing I like about head parking is that it makes lower idle power consumption possible. If other drives has lower power consumption while idle I would be more inclined to choose one of these instead. At the time I got most of my WD GPs they were king of the low power while idle.

I also have a bias against Seagate which is why I wouldn't consider one of their drives. I realize there is little logic to this, but in my mind Seagate has a lot to prove until I'm willing to use one of their modern drives again. Loved the Cuda IV and V, have had nothing but trouble with all Seagate HDDs after that.

I realize some has similar experiences with WD. No manufacturer is perfect. :)

canadian
Posts: 14
Joined: Sat Sep 04, 2004 10:14 am

Post by canadian » Fri Feb 05, 2010 12:20 pm

True. I don't recall exactly but Tom's Hardware has tested the power consumption and heat output from those drives, WD EADS and EARS (I think EARS but EADS for sure), Seagate LP and Samsung EcoGreen F2 (F3 is included but only the 500GB version.).

IIRC, the power consumption was really close and lowest ones out of those was WD Green (lowest), Samsung (almost as low) and then Seagate LP. The Seagate LP was also a bit higher with heat generated but I think some part of that is due to the slightly higher RPM (5900 v.s. 5400). It's not a lot but maybe causes both to be a bit higher?

The Samsung Ecogreen so close to the WD Green regarding low power and low heat that I lean towards the Samsung when considering a drive. There are no reports of requiring constant head parks and high Load Cycle Counts.

I don't know if all future drives will follow in WD's footsteps regarding the EARS series and the higher sector size (> 512 bytes). I don't want to be messing around with extra partition requirements and worry about if something is off, my data will be gone.

anakintjc
Posts: 35
Joined: Wed May 16, 2007 4:11 am

Post by anakintjc » Sat Feb 06, 2010 4:37 am

According to WD official specs the EARS edition should be more quiet!

So i don't think it's any louder, probably users 'want' to hear some difference?

The EARS tend to have less problems with the head parking however i read because of this it draws more power in idle...

I just want most bang for the buck and quiet as possible, i think WD15EA(D/R)S are both fine of doing that!

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Sun Mar 07, 2010 5:40 am

I have been considering WD drives too but like almost everyone else the early death due to repeated head parking puts me off.

The drive will be used for bulk storage in a low power NAS/download box (not as the main drive) so I will just use the built-in power saving with a 1 hour time-out. Far fewer stop/start cycles, drive stays powered up while I am using it and is off for the other 17-18 hours a day I don't use it.

Realistically in that situation the amount of power saved by head parking is going to equate to less than £1/year so why take the risk?

speedboxx
Posts: 72
Joined: Wed Dec 30, 2009 9:13 am
Location: Canada

Post by speedboxx » Sun Mar 07, 2010 7:09 am

The head parking issue is overhyped. I used to be concerned about it too, but the WD Green drives have been out for a long time now and there isnt a SINGLE report of a drive failing due to head load/unloads. For data storage I would actually choose the WD Green due to low power consumption, and the low error rate of 1 bit in every 10^15 bits read which is actually enterprise grade. Practically every other consumer grade drive is only 1 bit in 10^14.

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Sun Mar 07, 2010 7:25 am

There are loads of reports of the drives failing. It was worst with Linux systems. The drive reaches the SMART programmed limit for load/unload cycles and reports a SMART parameter failure after just a few months.

It might be fine for desktop use. Might even be fine for NAS or a Linux server. Why chance it for so little gain though?

speedboxx
Posts: 72
Joined: Wed Dec 30, 2009 9:13 am
Location: Canada

Post by speedboxx » Sun Mar 07, 2010 7:32 am

Link to the claims?

Only the older EACS drives count every single head load/unload in SMART. The newer EADS and EARS do not count those cycles in SMART. A SMART reported failure due to head parks doesnt mean that the drive has actually failed. Some people have 1 million+ head loads/unloads and the drive is working just fine. Bad sectors/unrecoverable errors would be a different story.

It's not just WD Green drives, but practically every 2.5" laptop drive out in the market today will park their heads in a similar fashion. I havent found any single report/claim that a drive has failed due to head parking, regardless of whether it is a notebook or a Green drive.

Besides, during the head park process, the heads dont actually touch the platters. I would still buy a WD drive over a Seagate since alot of the Seagate desktop drives still rely on landing zones where the heads make contact with the platters.

whiic
Posts: 575
Joined: Wed Sep 06, 2006 11:48 pm
Location: Finland

Post by whiic » Sun Mar 07, 2010 2:09 pm

canadian wrote:As for the Hitachi 2GB, I believe it has an extra platter
Not only an extra platter but also 7200rpm. It's a Velociraptor killer (which also offers 2000GB at the price of 300GB for the same price)... but not a silencer's best friend.
Pierre wrote:I'm not that concerned with Seagate's QC (quality control,right?)...that's because I've owned many -almost exclusively- seagate drives the last five years and I've never had a bad incident with them...
5 years is a long time. Many were in denial with the IBM 75GXP fiasco since they couldn't believe the most reputable HDD manufacturer could make anything bad, let alone something absolutely horrible.

Sure, modern Seagates aren't quite what "Deathstars" were at their worst period, and their QC issues may already be over. I still wouldn't quite trust them.
canadian wrote:I don't know if all future drives will follow in WD's footsteps regarding the EARS series and the higher sector size (> 512 bytes).
Trust me, they will. Making 512 byte sectors has already become too wasteful as the market expect endlessly improving data integrity, bigger capacities, better linear transfer rates, etc. Larger sectors help to provide all these.
MoJo wrote:I have been considering WD drives too but like almost everyone else the early death due to repeated head parking puts me off.
Early death? WTF?
MoJo wrote:There are loads of reports of the drives failing. It was worst with Linux systems. The drive reaches the SMART programmed limit for load/unload cycles and reports a SMART parameter failure after just a few months.
And the "failing" attribute wasn't the load/unload cycle count attribute itself? (That attribute shouldn't fail, though, as it has threshold of zero, which means the attribute is "for information only" and not any actual health indicator of any kind.)
speedboxx wrote:[what he/she just said]
QFT. (Quoted For Truth)

If MoJo actually has any proof for his/her claim, he/she should present them, otherwise it's just unverified rumours, FUD and scare tactics.

There's a separate thread for Greenpower load/unload cycle count issue.
viewtopic.php?t=51401&sid=674cf4adc5450 ... c1659daf10
"Is there a problem with head parks on WD Green HDDs?"

It starts with a lot of unwarranted hysteria - possibly partially because originally the thread had a way more provocative and scaremongering title - but as you read the topic, you should notice that no-one was actually able to supply any evidence for originally claimed Greenpower mass-death that was used to scare people from buying WD.

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Mon Mar 08, 2010 6:40 am

speedboxx wrote:Link to the claims?
I'll go one better. Here is the Western Digital support article on the subject:

http://wdc.custhelp.com/cgi-bin/wdc.cfg ... 1232125802

They acknowledge the problem and provide a utility to disable the offending feature.
It's not just WD Green drives, but practically every 2.5" laptop drive out in the market today will park their heads in a similar fashion. I havent found any single report/claim that a drive has failed due to head parking, regardless of whether it is a notebook or a Green drive.
You are correct.

https://wiki.ubuntu.com/DanielHahler/Bug59695

https://bugs.launchpad.net/ubuntu/+sour ... +bug/59695

To quote: "that the current behavior poses a significant risk to the longevity of hard drives used in a wide range of laptop models".
And the "failing" attribute wasn't the load/unload cycle count attribute itself?
Well the manufacturers publish the maximum number of load/unload cycles the drive is designed to survive over it's lifetime. As you get close to that limit it is wise to replace the drive.

Unlike bad sectors a head failing to load is usually fatal and renders the data unrecoverable.

It is probably fine. I'm 95% sure. Why risk it though?

Vicotnik
*Lifetime Patron*
Posts: 1831
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Post by Vicotnik » Mon Mar 08, 2010 8:18 am

MoJo wrote:There are loads of reports of the drives failing.
You seem to define 'failing' as drive with a very large number of loads/unloads reported. To me a drive is not failing unless it's actually something wrong with it. A rouge SMART value is not ideal, but since we here no reports of non-functional drives I feel the problem is not that great.
MoJo wrote:http://wdc.custhelp.com/cgi-bin/wdc.cfg ... 1232125802

They acknowledge the problem and provide a utility to disable the offending feature.
Well, not quite. They provide a firmware update for those who want to change the behavior of the drive to wait longer before positioning the heads in their park position. That's one solution. The other solution is to use software that "does not wake up the drives unnecessarily every 10 to 30 seconds or so, thereby gaining substantial power savings and eliminating superfluous activity." That's the real solution, the firmware update is just a workaround.
MoJo wrote:Unlike bad sectors a head failing to load is usually fatal and renders the data unrecoverable.

It is probably fine. I'm 95% sure. Why risk it though?
Why not? I'd say that since the WD GPs are so popular and we still have yet to hear of any real problem with these drives they are actually less of a risk than many other drives.

Metaluna
Posts: 133
Joined: Wed Mar 12, 2003 2:47 pm
Location: USA

Post by Metaluna » Mon Mar 08, 2010 9:53 am

Vicotnik wrote:Why not? I'd say that since the WD GPs are so popular and we still have yet to hear of any real problem with these drives they are actually less of a risk than many other drives.
I agree. Think of it this way: it would cost WD nothing to tweak their firmware to reduce or eliminate the head park behavior, other than a small idle power hit that most people would never notice or care about. On the other hand, if GP's were dying in droves due to this, that would cost them serious money, both in warranty costs and reputation. The fact that they have released several generations of new drives with about the same head parking behavior despite knowing about the "issue" leads me to believe that they have not observed any significant reliability problems.

whiic
Posts: 575
Joined: Wed Sep 06, 2006 11:48 pm
Location: Finland

Post by whiic » Mon Mar 08, 2010 4:15 pm

MoJo wrote:I'll go one better. Here is the Western Digital support article on the subject:

http://wdc.custhelp.com/cgi-bin/wdc.cfg ... 1232125802

They acknowledge the problem and provide a utility to disable the offending feature.
That link does not support the idea that high load/unload count leads to decreased reliability, just that if HDD is waken from "idle 3" (unloaded idle) state every 20 the power savings from unloading are practically lost anyway (even if you don't disable unload feature).
Equally worthless links for proving increased failure rate associated with high load/unload count.
MoJo wrote:To quote: "that the current behavior poses a significant risk to the longevity of hard drives used in a wide range of laptop models".
I say: no it does not.

Then I can quote: "no it does not."

Now that obviously proves a LOT because I quoted it. Alternatively I could quote
Penn Jillette wrote:BULLSHIT!
MoJo wrote:Well the manufacturers publish the maximum number of load/unload cycles the drive is designed to survive over it's lifetime.
Read them again and you should notice that they promis at least N load/unload cycles. At least means minimum number of load/unload cycles the drive should be able to take. Practice has shown that HDD manufacturers have been very conservative in these published minimum sustained cycle count specs. Greenpowers exceeding 10 times rated cycles, laptop HDDs reaching 10 million cycles, etc.
MoJo wrote:It is probably fine. I'm 95% sure. Why risk it though?
Seagate firmware is probably fine (by now). I'm 90% sure. Why risk it though?
Vicotnik wrote:The other solution is to use software that "does not wake up the drives unnecessarily every 10 to 30 seconds or so, thereby gaining substantial power savings and eliminating superfluous activity." That's the real solution, the firmware update is just a workaround.
It's funny how Linux users have so much difficulty admitting that their OS isn't perfect and that their OS does something totally useless every 20 seconds that negatively affects power saving, increases unload count and makes extra clicks that annoys the silencers. Obviously Ubuntu is not to be blamed.
Metaluna wrote:Think of it this way: it would cost WD nothing to tweak their firmware to reduce or eliminate the head park behavior - - - The fact that they have released several generations of new drives with about the same head parking behavior despite knowing about the "issue" leads me to believe that they have not observed any significant reliability problems.
Indeed. WD has not "fixed" the firmware so it's reasonable to think there is nothing worthy of fixing. To not fix something that is a known issue with a known cause (since it's intentional behaviour) for several drive generations can only be concluded that the behaviour doesn't cause increase in warranty returns (which is something WD would like to keep as low as possible).

Seagate bricking firmware on the other hand was assessed with firmware updates, and as the cause for bricking wasn't completely clear for manufacturer, the fix didn't quite solve all bricking issues. Fixed or not fixed, they took it seriously and attempted to fix it. WD has not done any changes to the unload timer, only offering optional tool for those who like to reconfigure their drives.

Vicotnik
*Lifetime Patron*
Posts: 1831
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Post by Vicotnik » Mon Mar 08, 2010 10:12 pm

whiic wrote:It's funny how Linux users have so much difficulty admitting that their OS isn't perfect and that their OS does something totally useless every 20 seconds that negatively affects power saving, increases unload count and makes extra clicks that annoys the silencers. Obviously Ubuntu is not to be blamed.
Does Linux do this to every HDD in the system or just the system drive? I will set up a file server soon running Linux with a few WD GPs. It doesn't matter if the one with the system on gets accessed often; there will be torrent downloads and stuff going on so that drive will probably never have time to park anyway. But if the other drives were affected it could be a bother.

What's Linux doing to the drive btw? Is there a service or something that can be stopped?

Getting a bit off topic here, sorry for that.

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Tue Mar 09, 2010 8:02 am

Vicotnik wrote:You seem to define 'failing' as drive with a very large number of loads/unloads reported. To me a drive is not failing unless it's actually something wrong with it. A rouge SMART value is not ideal, but since we here no reports of non-functional drives I feel the problem is not that great.
Don't misunderstand me. The drive is not failing when the load/unload count gets very high, however it is approaching the design limit.

It's like the cam belt in a car. You replace it after so many thousands of miles, even though it isn't broken. You do it because a broken cam belt would at best be annoying and at worst trash the engine (mainly in diesels). Cam belt manufacturers have got pretty good at knowing how long the belt is likely to last and at what point the chance of it failing becomes significant enough to warrant replacing it.

When a HDD approaches the design limits, be it load/unload counts, hours powered up, sector re-allocations etc. then it is a good idea to replace it if the data is valuable. Unfortunately some WD drives in particular configurations reach this point rather quickly.

Make of that what you will.

Vicotnik
*Lifetime Patron*
Posts: 1831
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Post by Vicotnik » Tue Mar 09, 2010 8:59 am

MoJo wrote:Don't misunderstand me. The drive is not failing when the load/unload count gets very high, however it is approaching the design limit.

It's like the cam belt in a car. You replace it after so many thousands of miles, even though it isn't broken. You do it because a broken cam belt would at best be annoying and at worst trash the engine (mainly in diesels). Cam belt manufacturers have got pretty good at knowing how long the belt is likely to last and at what point the chance of it failing becomes significant enough to warrant replacing it.

When a HDD approaches the design limits, be it load/unload counts, hours powered up, sector re-allocations etc. then it is a good idea to replace it if the data is valuable. Unfortunately some WD drives in particular configurations reach this point rather quickly.

Make of that what you will.
I understand what you're saying. I'm just stating my opinion regarding these drives. :) I don't use my WD GPs in such a way that they are constantly reloading the heads. And even in situations when that might happen, there are no reports of failing drives due to that. So I feel pretty safe. As safe as I can be anyway; with HDDs you can never be 100% safe so backups are always important.

whiic
Posts: 575
Joined: Wed Sep 06, 2006 11:48 pm
Location: Finland

Post by whiic » Tue Mar 09, 2010 6:30 pm

MoJo wrote:The drive is not failing when the load/unload count gets very high, however it is approaching the design limit.
It certainly is approaching it's end of life, even if it might take 50 years to actually reach it.
Mojo wrote:It's like the cam belt in a car. You replace it after so many thousands of miles, even though it isn't broken. You do it because a broken cam belt would at best be annoying and at worst trash the engine (mainly in diesels). Cam belt manufacturers have got pretty good at knowing how long the belt is likely to last
Either HDD manufacturers don't know durability of their products as well as car part manufacturers do, or they know they are extremely tolerant to load/unload cycles and only give "drive will sustain at least N cycles" instead of "drive will have 50% mortality rate before reaching N cycles". They really might not even know it because tens of millions of cycles are quite many unload cycles to reach during testing. And it's pretty pointless to take testing that far as by that time, in real environment, HDD should already start suffering from all sorts of other old-age problems.

Also, there's no reason to replace a HDD before it dies or becomes obsolete (capacity, performance, noise). You replace cam belt to prevent engine damage (worst case) or at least sudden failure on the road (annoyance). If you keep back-ups of HDD contents, a dead HDD
A) costs exactly the same to replace than to replace it "just to be certain" and
B) is not more annoying to replace than it is to replace a functional HDD to another.

Do also note that if you replace functional drives, you end up replacing them more often and as a result
A) it really does cost more even though a single replacement costs exactly the same and
B) because you replace HDDs more often, the amount of effort you make swapping funtioning HDDs will be multiplied.

This of course assumes you keep back-ups. If you don't... then you are an
IDIOT.

Cam belt comparison fails badly. Especially considering that HDDs have high infant mortality rate. Old age effect kick in somewhere after 3-10 years of use. Replacing HDD before 5 years will probably only make reliability worse as the replacement will have horrible reliability for the first few months of use.
MoJo wrote:When a HDD approaches the design limits, be it load/unload counts, hours powered up, sector re-allocations etc. then it is a good idea to replace it if the data is valuable.
But do you know the limit the drives can take? No. You really don't know, even if you think you know.
MoJo wrote:Unfortunately some WD drives in particular configurations reach this point rather quickly.
Proof?

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Sat Mar 13, 2010 7:08 am

whiic wrote:IDIOT.
I have no further interest in discussing this with you.

sxr71
Posts: 29
Joined: Tue Nov 01, 2005 11:56 am

Post by sxr71 » Wed Mar 17, 2010 3:15 pm

MoJo wrote:
whiic wrote:IDIOT.
I have no further interest in discussing this with you.
I don't think he called you an idiot unless you don't backup your data, in which case I would agree.

fjf
Posts: 192
Joined: Wed Apr 06, 2005 9:44 am
Location: Europe

Post by fjf » Thu Mar 18, 2010 12:11 am

Regardless of the backing up, the tone of this (and other) discussions here is too harsh. Too many people being right all the time.

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Fri Mar 19, 2010 3:23 am

fjf is right, that is why I have stopped responding to whiic. He's just ranting off on some tangent now.

puddnhead
Posts: 57
Joined: Wed Feb 09, 2005 10:39 am

Post by puddnhead » Thu Mar 25, 2010 5:45 am

MoJo wrote:fjf is right, that is why I have stopped responding to whiic. He's just ranting off on some tangent now.
He didn't call you an idiot, like you are making out. He didn't call you anything. He said people who don't do backups are idiots. Which is pretty much true (I would have chosen "fool" instead of idiot ... but ... close enough).

To me it's you who is choosing to create all the drama in this exchange, by snipping out the part of the sentence in your reply that made it clear he wasn't addressing you directly, and then acting all offended about something that you should know by now was never said.

Post Reply