Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple is killing our harddrives prematurely
- Thread starter dukebound85
- Start date
- Sort by reaction score
That is a known issue I thought.
Apple probably put it in to prevent damage, like before posters have mentioned.
And why should Apple really care, the HDD will last more then 3 years. Then Apple Care will be gone and the HDD manufacturers are glad that you will need a new HDD.
Apple probably put it in to prevent damage, like before posters have mentioned.
And why should Apple really care, the HDD will last more then 3 years. Then Apple Care will be gone and the HDD manufacturers are glad that you will need a new HDD.
In 3 years SSDs will exceed the capacity of HDDs, be cheaper, and probably more reliable, so a lot of people will have moved on by then.
well yea, as it hasnt hit the max yet. its the rate at which these cycles add up that alarms me
i still fiind this unacceptable as it really is wearing the hdds out way too fast, esp when its on my desk 99% of the time...
http://discussions.apple.com/thread.jspa?messageID=6498085
http://forum.synology.com/enu/viewtopic.php?f=118&t=10355
http://www.dslreports.com/forum/r21566852-WD2500BEVE-in-PPC-Mini-High-Load-Cycle-Count
https://forums.macrumors.com/threads/669798/
in other words, this is NOT GOOD
at any rate, since the drive is only rated for 600,000 load cycles,
I have 339 power on hrs and 38,609 load cycles
This means with current usage, I will only have 220 days of power on time before my drive is in failure criterion
this should be alarming to all of you. the consensus is that this is an OS issue, not the drive. WTF APPLE
edit: the method i used to fix this doesnt work upon awaking from sleep as i have to rerun the script
is there an automated program or setting somewhere?
Load_Cycle_Count isn't one of the Pre-fail attributes, so it's take it with a grain of salt...
As for the frequency of head parking... I admin a bunch of MacBooks with a whole slew of different manufacturer's hard disks in 'em. I also admin a bunch of other laptops running Windows and Linux. I haven't noticed any unusual failure levels amongst the Macs in the last couple years -- they seem to have their drives die at about the same rate as the machines running other OSs. (Anecdotal evidence, I know... but still...)
As for the frequency of head parking... I admin a bunch of MacBooks with a whole slew of different manufacturer's hard disks in 'em. I also admin a bunch of other laptops running Windows and Linux. I haven't noticed any unusual failure levels amongst the Macs in the last couple years -- they seem to have their drives die at about the same rate as the machines running other OSs. (Anecdotal evidence, I know... but still...)
I have 2 MBs, 2 PBs, and a MBP in my household and no HDD failures to date. The PBs and the MBP have never been upgraded. One of the PBs (12") gets heavy abuse by a ADD child in college. In other words: the ADD hasn't hurt her HDD and neither has Apple. 
-GDF
-GDF
I have had to replace many hard drives within Macbooks. But never in any Windows PC or Laptop. At first I thought the case was just poorly designed to withstand normal shock and vibration but I guess this guy has found the real cause. I have had to replace the hard drives on several of my Mac laptops as well. They all tend to die in between the years 1-3 of ownership.
Believe it if you want or not but anyone with a Macbook be prepared for a dead hard drive in the future. And yeah they are easily replaced and relatively cheap, once your drive dies, the data is lost which is a huge hassle.
Believe it if you want or not but anyone with a Macbook be prepared for a dead hard drive in the future. And yeah they are easily replaced and relatively cheap, once your drive dies, the data is lost which is a huge hassle.
There are some interesting points made in this thread: http://forums.somethingawful.com/showthread.php?threadid=3113140&userid=0&perpage=40&pagenumber=1 . They contend that the failure rate from load cycling is very low, much lower than sudden motion head crashes. This is my speculative reasoning, but the 600K ramp spec metric may not relate to wearing out the ramp from head parking but rather is a rough correlation to drive failure. In other words, two or three sigmas out on the curve may show drives dying after 600K load cycles, but that doesn't necessarily mean load cycles contribute to drive failure. If anything, the head parking profile seems like a hedge against drive failure (parking the head more reduces the vulnerable timeslices while the drive is in use). The sudden motion sensor is another hedge to shock damage, but more hedges may be better.
Using hdapm seems effective in reducing the annoying head parking noise in some drives for machines with low risk of sudden shock damage. I expect that you are mitigating the risk of load cycle drive failure in return for increasing the risk of motion-related failure. I'm going to leave the setting of my drive alone since my laptop is moved around quite a bit.
I'm not saying the head parking profile on Macs aren't overly aggressive, but the anecdotal evidence for drive failures makes it difficult to draw any clear conclusions. Personally, I've seen many more drive failures on Windows machines (maybe a dozen across 6-8 machines) compared to 0 across 7 Macs. However, that sample size is too low to tell me anything for sure.
Using hdapm seems effective in reducing the annoying head parking noise in some drives for machines with low risk of sudden shock damage. I expect that you are mitigating the risk of load cycle drive failure in return for increasing the risk of motion-related failure. I'm going to leave the setting of my drive alone since my laptop is moved around quite a bit.
I'm not saying the head parking profile on Macs aren't overly aggressive, but the anecdotal evidence for drive failures makes it difficult to draw any clear conclusions. Personally, I've seen many more drive failures on Windows machines (maybe a dozen across 6-8 machines) compared to 0 across 7 Macs. However, that sample size is too low to tell me anything for sure.
What brand do we get for 'good' drives then?
Looking at buying a macbook, and outsourcing a 7200rpm (and larger) external drive.
with certain HDDs the clicking can be maddening!
Apple will do anything to:
1. increase battery life in notebooks.
2. decrease heat in notebooks.
all the HDD clicking in Apple's notebooks is more about power management and the two things stated above that it is about protecting the drive in a shock or a fall. the latter is the responsibility of the SMS (sudden motion sensor), and it is not what is responsible for all the clicking.
Apple will do anything to:
1. increase battery life in notebooks.
2. decrease heat in notebooks.
all the HDD clicking in Apple's notebooks is more about power management and the two things stated above that it is about protecting the drive in a shock or a fall. the latter is the responsibility of the SMS (sudden motion sensor), and it is not what is responsible for all the clicking.
hey thanks for your solution doug, got it to work, though i have to redo it if i wake from sleep
glad i caught it somewhat early with 38k
glad i caught it somewhat early with 38k
Some good information in this thread and the threads that are linked off.
I checked my SMART status and have 2074 power on hours and 95127 load cycles, this is in 20 months of usage.
I checked my SMART status and have 2074 power on hours and 95127 load cycles, this is in 20 months of usage.
*SIGH*
When will people realize there is no such thing as "my hardware X is reaching such usage and itll die soon"
No such thing when it comes to computers. Ever.
I've seen drives die a week after they're installed and drives that are 15 years old still running.
Parking the drive in a laptop is the best invention ever. Seriously. Models of computers that do this at work see wwaaaaaayyyy less harddrive failures then the models that dont. And we have thousands and thousands of laptops.
When will people realize there is no such thing as "my hardware X is reaching such usage and itll die soon"
No such thing when it comes to computers. Ever.
I've seen drives die a week after they're installed and drives that are 15 years old still running.
Parking the drive in a laptop is the best invention ever. Seriously. Models of computers that do this at work see wwaaaaaayyyy less harddrive failures then the models that dont. And we have thousands and thousands of laptops.
To cover their asses? You'd have to ask them to get a real answer.
Yes, you want fewer rather than more load cycles, but unlike, say Reallocated_Sector_Ct, Load_Cycle_Count nearing its threshold doesn't mean that disaster is near.
SMART indicators are meant to be just that: indicators. They're not a hard and fast rule, and they're not even always correctly implemented (I've seen some SSDs report Load_Cycle_Count values, for example).
If it worries you, feel free to bump up your head park time (or even disable it all together). You'll burn more power and put your data at risk for a bit longer, but hey -- if it makes you feel better there's not too much harm to be done.
I'd actually recommend WD. Their Scorpio series of notebook drives are cheap, large, fast, and are pretty power efficient.
If you've got the money, get an Intel X-25M or (even better) an X-25E. Then you won't have to worry about head crashes, load cycles, or drive temps. Plus your Mac will become much, much more responsive given the awesome seek times and access speeds. Personally I'm waiting until July/Aug. to get an Intel SSD, but if you've got the cash to burn now I highly recommend it. Until I go to SSDs though, I'm gonna continue to stick with WD drives, as I've had great experiences with them in the past.
Then buy a desktop and stop whining.
It amuses me that people assume they know the hardware better than Apple does. When Apple engineers the product, do you really think they say "OK, how can we make sure the HDD fails as soon as possible?"
No. They say "OK, how do we balance HDD life with protecting the HDD from bumps and jolts that could ruin it?" Parking the head in a mobile HDD makes perfect sense - if you drop or bump your machine while the head is moving, say good bye to your drive.
Then buy a desktop and stop whining.
was this necessary? and i have a desktop
it amuses me that people think apple knows hdds more than the hdd manufactuerers
the hdd makers they are the ones that test their products....
does apple make hdds? didnt think so
no, at least it didnt for me
youre kidding me right?
a hdd is a tad bit more impoartant than a batt as a batt doesnt store anything
It seems to have reduce the number of time the drive parks, which seems to be a good compromise.
it amuses me that people think apple knows hdds more than the hdd manufactuerers
the hdd makers they are the ones that test their products....
does apple make hdds? didnt think so
The hard drive manufacturers do not say "your HDD will fail on the 600 001th load cycle". As has been pointed out, their numbers are only one benchmark of what makes a drive fail.
Even we assume you're right, and your drive will fail as soon as it hits 600 001 load cycles, is that Apple's fault? Is that "premature"? Only if you insist on comparing it to a desktop, which is foolish. A drive in a notebook should not be expected to last as long as a drive in a desktop. Notebook computers make compromises to be able to deliver on size and portability. That means higher wear, for starters, which means that components are likely to fail sooner.
All you've done is whined that your notebook drive has more load cycles than your desktop. If it's that big of a deal, then stick to desktops and stop complaining. Your drive will almost certainly (in theory) last longer. But if you're going to buy a notebook computer, then accept that notebooks come with compromises relative to their desktop counterparts.
Apple knows that by having the drive spin down more frequently it will wear out more quickly. But they also know something which you seem to be incapable of figuring out: that notebooks go through more movement, bumps, falls, and generally more wear and tear, than desktops do. So in order to protect the HDD - and their customers' valuable data! - they compromise on HDD life in favour of HDD protection.
Now, if you can prove that HDDs in Apple notebooks tend to fail more often than HDDs in other notebooks, you might be on to something. But comparing the load cycles in a notebook to a destop is foolish. It completely ignores the fact that notebooks and desktops are completely different machines, with different design needs and limitations. It would be like comparing the fuel economy of a Toyota Prius to a Lamborghini Gallardo. Yeah, the Prius will go further on one tank, but that's not really the point, is it?