Buying Hard Drives from the same "Batch" - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 37 Old 07-17-2012, 07:31 PM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
One of the theories I have read multiple times is that you should avoid buying more than 1 hard drive at the same time. The reason is that if there is a "bad batch" you are going to get multiple hard drives from that batch which both have more likelihood of failing compared to drives that aren't from that "bad batch". In short you should spread you hard drive purchases over multiple batches is possible.

I say hogwash.

What I think is that you could just as easily get multiple drives from a "good batch" or "standard batch". I see the argument but I just don't get agree. And I think its something that should be reconsidered and put to rest.

What say you?

Edit: I am, however, a proponent of waiting to buy hard drives until you need them as that keeps your drives younger and cheaper per GB. But this has nothing to do with the "batch" theory. If I needed 10 hard drives I wouldn't have an issue with getting all 10 at once.
assassin is offline  
Sponsored Links
Advertisement
 
post #2 of 37 Old 07-17-2012, 08:45 PM
Senior Member
 
dksc318's Avatar
 
Join Date: Apr 2012
Location: San Jose,CA
Posts: 453
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
The HDD industry shipped 200M drives last year. They can't make that many drives if they have many bad batches. When they introduce next generation technology, it can take as many as ten years to bring to production, such as pattern media and HAMR.
dksc318 is offline  
post #3 of 37 Old 07-17-2012, 09:02 PM
Senior Member
 
mcturkey's Avatar
 
Join Date: Aug 2011
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 43
Quote:
Originally Posted by assassin View Post

One of the theories I have read multiple times is that you should avoid buying more than 1 hard drive at the same time. The reason is that if there is a "bad batch" you are going to get multiple hard drives from that batch which both have more likelihood of failing compared to drives that aren't from that "bad batch". In short you should spread you hard drive purchases over multiple batches is possible.
I say hogwash.
What I think is that you could just as easily get multiple drives from a "good batch" or "standard batch". I see the argument but I just don't get agree. And I think its something that should be reconsidered and put to rest.
What say you?
Edit: I am, however, a proponent of waiting to buy hard drives until you need them as that keeps your drives younger and cheaper per GB. But this has nothing to do with the "batch" theory. If I needed 10 hard drives I wouldn't have an issue with getting all 10 at once.

I don't think the whole "bad batch" thing is derived from manufacturing issues so much as the likelihood that a set of consecutive serial numbers had the same treatment during shipping. I'd be willing to bet that if hard drives were immediately installed into servers by extremely careful technicians with the cleanest power possible running to the servers as soon as they came off the assembly line, they would probably have a less than 0.1% failure rate--probably even lower. The problem comes from shipping. Who knows how the boxes were loaded into the freight container, or how that container handled the ride across the Pacific? Then there's the unloading of the containers, the unpacking of them and individual distribution to the sellers. After that, we've got the unboxing of each drive into whatever custom storage system is used by the respective sellers, followed by the transfer into the box that the drive will be packed into for shipping to you. Now UPS/FedEx gets hold of it, and who knows how they'll treat it.

If there is poor handling of the drives at any of those stages, it can be damaged. Hence why if you buy 10 drives from one "batch", and one fails, there is a higher chance of more drives from that batch failing than from another random batch.
mcturkey is offline  
post #4 of 37 Old 07-17-2012, 10:32 PM
Senior Member
 
dksc318's Avatar
 
Join Date: Apr 2012
Location: San Jose,CA
Posts: 453
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by mcturkey View Post

... The problem comes from shipping. Who knows how the boxes were loaded into the freight container, or how that container handled the ride across the Pacific?....

I sat by a guy from APL (American President Lines) on a flight. They can have 3-5000 containers on a boat. Losing one or two on a voyage is not unusual. They lost a container full of those yellow rubber ducky from China once. Some made it across the north pole to NY, lol.
http://en.wikipedia.org/wiki/Friendly_Floatees

Seriously, when I was a design engineer, we sent a box of new product across the continent and back in a truck to see if the packaging method survived the trip. You can see those black end caps on HDDs. They look simple but do the job. Many HDD designs have "parking" function to unload the heads when they are powered off.
dksc318 is offline  
post #5 of 37 Old 07-18-2012, 03:24 AM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by mcturkey View Post

I don't think the whole "bad batch" thing is derived from manufacturing issues so much as the likelihood that a set of consecutive serial numbers had the same treatment during shipping. I'd be willing to bet that if hard drives were immediately installed into servers by extremely careful technicians with the cleanest power possible running to the servers as soon as they came off the assembly line, they would probably have a less than 0.1% failure rate--probably even lower. The problem comes from shipping. Who knows how the boxes were loaded into the freight container, or how that container handled the ride across the Pacific? Then there's the unloading of the containers, the unpacking of them and individual distribution to the sellers. After that, we've got the unboxing of each drive into whatever custom storage system is used by the respective sellers, followed by the transfer into the box that the drive will be packed into for shipping to you. Now UPS/FedEx gets hold of it, and who knows how they'll treat it.
If there is poor handling of the drives at any of those stages, it can be damaged. Hence why if you buy 10 drives from one "batch", and one fails, there is a higher chance of more drives from that batch failing than from another random batch.

This is the same argument though. You could just as easily get drives that were handled properly.
assassin is offline  
post #6 of 37 Old 07-18-2012, 08:28 AM
AVS Special Member
 
Somewhatlost's Avatar
 
Join Date: Jul 2006
Location: some small blue-green planet thingy...
Posts: 1,792
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 17
I think you are oversimplifying it a bit...
I do not believe it has anything to do with "bad" or "good" batches as such...

it is more along the lines that in any given batch, the drives would have been manufactured/handled/etc in exactly the same way... so their lifespan should be pretty much the same, regardless of if it is 1 day (unlikely) or it is 100 years (also unlikely)...

to the average user, this doesn't mean squat... their drive(s) will in all probability out last their warranty... or they wont, and they will get them RMA'd... either case, "batch" doesn't matter, and all is good with the world...

but for someone in a data center, with a huge raid array, a single failed drive is no problem, but multiple drives failing at the same time, before they manage to replace & rebuild the failed drive, creates huge issues for them... ie, in this case "playing the odds" and doing everything you can to "stack the deck" in your favor would seem to be a good idea...

NOTE: As one wise professional something once stated, I am ignorant & childish, with a mindset comparable to 9/11 troofers and wackjob conspiracy theorists. so don't take anything I say as advice...
Somewhatlost is offline  
post #7 of 37 Old 07-18-2012, 08:34 AM
Senior Member
 
dksc318's Avatar
 
Join Date: Apr 2012
Location: San Jose,CA
Posts: 453
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by Somewhatlost View Post

.......
but for someone in a data center, with a huge raid array, a single failed drive is no problem, but multiple drives failing at the same time, before they manage to replace & rebuild the failed drive, creates huge issues for them... ie, in this case "playing the odds" and doing everything you can to "stack the deck" in your favor would seem to be a good idea...
Google, Facebook place 3 copies of any data, 2 at the same site but different servers, 1 copy at a different site.
dksc318 is offline  
post #8 of 37 Old 07-18-2012, 08:59 AM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by Somewhatlost View Post

I think you are oversimplifying it a bit...
I do not believe it has anything to do with "bad" or "good" batches as such...

I am not sure what you mean with "oversimplifiying". There are a ton of people out there who fret over this very issue...

http://www.google.com/webhp?sourceid=chrome-instant&ie=UTF-8#hl=en&safe=off&output=search&sclient=psy-ab&q=hard%20drives%20from%20same%20batch&oq=&gs_l=&pbx=1&fp=5999ea2cf20f1d2b&bav=on.2,or.r_gc.r_pw.r_qf.,cf.osb&biw=1680&bih=905

Like I said before isn't there just as good a chance (better actually since the failure rates of hard drives are relatively low) that you would actually be better off to grab drives from a single batch?

I don't see how grabbing 10 drives from 10 batches necessarily protects you at all. And unless you have 1000 drives I am not sure it makes any sense.
assassin is offline  
post #9 of 37 Old 07-18-2012, 09:09 AM
Advanced Member
 
aliaskary77's Avatar
 
Join Date: Oct 2007
Posts: 606
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 51
You can argue both sides. its not an issue personally that's worth any thought.. I buy what I need when I need it. the rest is up to the eGods.

I have never lost a hard drive in 25 years until last month, where i lost 2. One seagate I dropped very lightly onto a carpet, and a WD black that has lost data twice and is being RMA'd. It's all a crapshoot.

note: the post above is my opinion. as such, when reading any recommendations from me, please do you research and seek out other recommendations and make up your own mind on your next course of action. i mean, most reasonable adults should know that, but it seems this should be stated anyways.
aliaskary77 is online now  
post #10 of 37 Old 07-18-2012, 09:57 AM
Member
 
wsume99's Avatar
 
Join Date: Oct 2011
Posts: 82
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
I think this only applies to a situation where you are using the drives in a protected array where you are trying to avoid having multiple drive failures within a very short time frame thereby reducing the likelihood of losing data due to multiple drive failures at the same time . So your goal should either be to have drives that never fail (which is essentially impossible) or to have as much variation as possible in drive life. I don't think getting all your drives from a single batch will accomplish the goal of having as much variation in drive life as possible. The reduction in risk of data loss may not be that great but I think there is at least a valid argument here. Whether or not the hassle of staggering your purchases justifies the small benefit is debatable IMO.
wsume99 is offline  
post #11 of 37 Old 07-18-2012, 09:58 AM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,451
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 151 Post(s)
Liked: 151
Quote:
Originally Posted by assassin View Post

Like I said before isn't there just as good a chance (better actually since the failure rates of hard drives are relatively low) that you would actually be better off to grab drives from a single batch?

But that's the point, if you buy all your drives from the same batch, then you'll get a set of drives with similar failure characteristics. Theoretically you're better off if the drives your array have different failure characteristics. If you get ten drives from the same batch, and one fails, then the odds are higher that another drive will fail at the same time (or while the rebuild is going on) than if you'd bought ten drives from ten different batches.
Quote:
I don't see how grabbing 10 drives from 10 batches necessarily protects you at all.

It protects you because drives from different batches will have been through different manufacturing/transportation history and are more likely to have different failure characteristics, such that you are less likely to have two drives fail at the same time thus killing your array and all your data. Take for example if there's some manufacturing defect that goes undetected, if you bought all your drives from the same batch with the same defect odds are high more than one of those drives will fail at the same time and kill the array and everything on it. Where as if you bought the drives from different batches, your odds are much better that they don't all have the same defect and that not more than one will fail from it.
Quote:
And unless you have 1000 drives I am not sure it makes any sense.

All that said, I agree, I don't think it's really worth a lot of consternation. IMO it's more trouble to try and buy drives from different batches for the small decrease in likelihood that multiple drives will fail in the array before you can replace/rebuild.

See what an anamorphoscopic lens can do, see movies the way they were meant to be seen
stanger89 is online now  
post #12 of 37 Old 07-18-2012, 09:58 AM
AVS Special Member
 
Somewhatlost's Avatar
 
Join Date: Jul 2006
Location: some small blue-green planet thingy...
Posts: 1,792
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 17
Quote:
Originally Posted by assassin View Post

I am not sure what you mean with "oversimplifiying".
by oversimplifying, I just mean breaking it down into there being "bad" batches (and presumably "good" batches...)
Quote:
Originally Posted by assassin View Post

Like I said before isn't there just as good a chance (better actually since the failure rates of hard drives are relatively low) that you would actually be better off to grab drives from a single batch?
I don't see how grabbing 10 drives from 10 batches necessarily protects you at all. And unless you have 1000 drives I am not sure it makes any sense.
define "better off"?
do you mean that the drives wont fail?
or that drives, bought from the same batch, and then put into use in the same RAID array (real RAID where all your data goes "poof!!!" if parity +1 drives fail at the same time, not JBOD+Parity), wont fail at similar times?

note, for the record, for my 2 tiny RAID's (2 drive RAID & 4 drive raid 5) setups, I just buy all my drives all at once as the quantity of drives in question is so small that the risk is so small it is irrelevant... I wouldn't actually start to care until I was dealing with 10 drives in raid 5 or 20 drives in raid 6 in a single system... and even then, as a home user, I far prefer the benefits of a jbod+parity (unRAID in my case, but Flex or Snap would work just fine too) setup for large scale storage like that, so the whole batch thing still doesn't even apply to me... but that doesn't mean that it isn't a valid concern for those who do need to deal with large RAID array's... tsunami's have no effect on me either, but that doesn't mean someone living near an ocean should ignore them...

anyway, I do believe there is some truth to the theory, and if you are in charge of some huge RAID arrays you should probably follow it... but I would bet it doesn't really apply to anyone here... (unless they do that sort of thing for work, and they bring their work home with them).

NOTE: As one wise professional something once stated, I am ignorant & childish, with a mindset comparable to 9/11 troofers and wackjob conspiracy theorists. so don't take anything I say as advice...
Somewhatlost is offline  
post #13 of 37 Old 07-18-2012, 10:24 AM
Senior Member
 
csmith's Avatar
 
Join Date: Apr 2001
Location: Charlotte,NC
Posts: 290
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 11
I've had many drives fail over the years, in an enterprise environment (24x7) RAIDs. I've had 5 SATA Hitachi enterprise drives fail (out of 18) over a 1.5 year period. I had 32 750GB SATA Seagate drives before that and had about 4 of them fail over 4 years. And finally had 20 75GB SCSI Seagate drives with only 1 failure over a 15 year period.

I've always ask for the drive set on the server to be from different batches, but don't always get them. Case in point the Hitachi drives, luckily they didn't all fail at the same time. A bad batch may get through QA, with components just on the edge of tolerances, but fail when deployed for a period of time. I've seen this with ICs, back in the 80s when developing circuit board test programs. The individual component would pass an IC test, but fails when placed on circuit board or later in the field. Changing component vendors (i.e. TI vs Motorola) for the same component can cause issues too.

But overall, it's always good policy to backup no matter what... smile.gif
csmith is offline  
post #14 of 37 Old 07-18-2012, 11:00 AM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by stanger89 View Post

All that said, I agree, I don't think it's really worth a lot of consternation. IMO it's more trouble to try and buy drives from different batches for the small decrease in likelihood that multiple drives will fail in the array before you can replace/rebuild.

I get what you are saying. But couldn't I make the same argument that there are different characteristics that are an improvement in a certain batch? That's what I am saying. And given the likelihood that a drive will fail and that multiple drives will fail at the same time (extremely low) I think this is a non-issue yet see it emphasized all the time. That's what I don't get.

It would make more sense if you would spread your drives out in 6-12 months increments so that the drives have different ages. But that's not feasible but to me is a much stronger argument.

In the end I think we are all agreeing here that this is largely a non-issue for HTPC where you have 2-20 or so drives.
assassin is offline  
post #15 of 37 Old 07-18-2012, 11:22 AM
AVS Special Member
 
Somewhatlost's Avatar
 
Join Date: Jul 2006
Location: some small blue-green planet thingy...
Posts: 1,792
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 17
Quote:
Originally Posted by assassin View Post

I get what you are saying. But couldn't I make the same argument that there are different characteristics that are an improvement in a certain batch?
not an improvement per se, just a difference...

Quote:
Originally Posted by assassin View Post

In the end I think we are all agreeing here that this is largely a non-issue for HTPC where you have 2-20 or so drives.
totally agree...
but the theory is still valid...
its just not relevant

NOTE: As one wise professional something once stated, I am ignorant & childish, with a mindset comparable to 9/11 troofers and wackjob conspiracy theorists. so don't take anything I say as advice...
Somewhatlost is offline  
post #16 of 37 Old 07-18-2012, 11:22 AM
AVS Special Member
 
Dark_Slayer's Avatar
 
Join Date: May 2012
Posts: 2,587
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 203 Post(s)
Liked: 309
Quote:
Originally Posted by assassin View Post

I say hogwash.

I also say hogwash, and echo the sentiment to naysayers of Sandforce and OCZ SSDs.

Maybe it's because I prefer to keep my head in the sand over component reliability, but I've owned several computers and hard drives over the past couple decades and I've never had a drive fail before it's connector tech was too slow to survive the upgrade cycle. I don't seem to keep tech long enough or stress test it enough for any of the failure probability arguments to be statisitcally relevant. Also, the initial 1TB, 1.5TB, and 2TB Seagate 7200rpm, SATA II line was supposedly a "bad batch" with lots of failures. I own and still use a 1.5TB drive from the "bad batch" that has 5000+ hours on it. If I needed 2 of them, I would have bought two on the same order.

I actually like the skepticism about product reliability, because I usually wind up getting a great deal on those products and use them without issues biggrin.gif
Dark_Slayer is offline  
post #17 of 37 Old 07-18-2012, 12:51 PM
AVS Special Member
 
damelon's Avatar
 
Join Date: Sep 2007
Location: Woodbridge, VA
Posts: 1,723
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 113
Quote:
Originally Posted by assassin View Post

I don't see how grabbing 10 drives from 10 batches necessarily protects you at all.

This concept is exactly why I agree with your point of view. A certain % of hard drives are going to be faulty. If you have 10 hard drives to through 10 different means of shipping/delivery I'd think you have just increased the odds of something bad happening to at least one of them, or that you would make it even more likely you would encounter a "bad batch" if you really worry about it. I've never had a hard drive arrive DOA, and in the last 15 years I've only ever had 2 that I needed to return.
damelon is offline  
post #18 of 37 Old 07-18-2012, 01:52 PM
Advanced Member
 
flocko's Avatar
 
Join Date: Dec 2011
Posts: 707
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
I personally think it is luck of the draw . No more , no less

What we consider a " batch" and what a hard drive supplier considers a "batch" would most likely be WAY different . So does anyone know or think they know what those two numbers may be ? Do you think that 20 HDs is a batch to the public and a 1000 to the supplier ?

I think this is an important number in both regards as that may determine the outcome of the debate at hand
flocko is offline  
post #19 of 37 Old 07-18-2012, 02:08 PM
AVS Special Member
 
Somewhatlost's Avatar
 
Join Date: Jul 2006
Location: some small blue-green planet thingy...
Posts: 1,792
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 17
Quote:
Originally Posted by damelon View Post

This concept is exactly why I agree with your point of view. A certain % of hard drives are going to be faulty. If you have 10 hard drives to through 10 different means of shipping/delivery I'd think you have just increased the odds of something bad happening to at least one of them, or that you would make it even more likely you would encounter a "bad batch" if you really worry about it. I've never had a hard drive arrive DOA, and in the last 15 years I've only ever had 2 that I needed to return.
and that is exactly why some people are missing the point, this theory/fear/paranoia/whatever you would like to call it, has absolutely NOTHING to do with "bad batches" of HDD's

all HDD's ever made will fail eventually, that is a fact, hopefully it wont fail until far after its useful life has passed, but all HDD's will fail eventually, this "theory" or whatever you want to call it, is all about trying to control when they will fail, not if, and not having too many in one huge RAID array failing at the same time.

NOTE: As one wise professional something once stated, I am ignorant & childish, with a mindset comparable to 9/11 troofers and wackjob conspiracy theorists. so don't take anything I say as advice...
Somewhatlost is offline  
post #20 of 37 Old 07-18-2012, 02:10 PM
Senior Member
 
mcturkey's Avatar
 
Join Date: Aug 2011
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 43
It comes down to odds. Say there are 10 "batches" of 100 drives. Nine batches will experience, say, a 2-3% failure rate. The tenth batch experiences a much higher failure rate of 15% or worse, due to the poor handling it received along the way. If you were one of the unlucky ones who had batch #10, you're hosed. On the other hand, if you had purchased your hundred drives evenly across those batches, your overall failure rate would be much closer to the 2-3% average than the 15% from unlucky batch 10.

I wish I had more solid numbers to back this up, but the experiences of many different folks who have ordered multiple drives from Newegg, Amazon, and other places have pretty closely aligned with this.
mcturkey is offline  
post #21 of 37 Old 07-18-2012, 03:22 PM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by mcturkey View Post

It comes down to odds. Say there are 10 "batches" of 100 drives. Nine batches will experience, say, a 2-3% failure rate. The tenth batch experiences a much higher failure rate of 15% or worse, due to the poor handling it received along the way. If you were one of the unlucky ones who had batch #10, you're hosed. On the other hand, if you had purchased your hundred drives evenly across those batches, your overall failure rate would be much closer to the 2-3% average than the 15% from unlucky batch 10.
I wish I had more solid numbers to back this up, but the experiences of many different folks who have ordered multiple drives from Newegg, Amazon, and other places have pretty closely aligned with this.

Right. But your odds are higher of getting one of the those drives if you sample from 10 different lots. In your example you would be 100% assured of getting one of those drives if you were doing this for 10 drives from 10 different lots.

I think, as I suspected, that this is largely a non-issue for users of this forum.
assassin is offline  
post #22 of 37 Old 07-18-2012, 03:48 PM
AVS Addicted Member
 
Mfusick's Avatar
 
Join Date: Aug 2002
Location: Western MA
Posts: 23,615
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 183 Post(s)
Liked: 929
I am with you Assassin.

HOGWASH

-

"Too much is almost enough. Anything in life worth doing is worth overdoing. Moderation is for cowards."
Mfusick is online now  
post #23 of 37 Old 07-18-2012, 04:04 PM
Senior Member
 
mcturkey's Avatar
 
Join Date: Aug 2011
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 43
Quote:
Originally Posted by assassin View Post

Right. But your odds are higher of getting one of the those drives if you sample from 10 different lots. In your example you would be 100% assured of getting one of those drives if you were doing this for 10 drives from 10 different lots.
I think, as I suspected, that this is largely a non-issue for users of this forum.

For those who are only going to be getting a handful of drives at a time, sure, it's not a big deal. But I think you misunderstand how the math works on this.

Going with my previous numbers, if you buy a single "batch", you have a 10% chance of ending up with 10-15 drive failures, or a 90% chance of 2-3 failures.. If you buy across all batches, then you could reasonably expect 3-5 drive failures. For small numbers of drives (less than 10), this becomes statistically irrelevant, as there are too many other variables that could mess up a drive. However, when you are dealing with significant quantities of drives, it is much safer to buy from multiple sources to reduce the risk of having a large batch fail.
mcturkey is offline  
post #24 of 37 Old 07-18-2012, 04:10 PM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by mcturkey View Post

For those who are only going to be getting a handful of drives at a time, sure, it's not a big deal. But I think you misunderstand how the math works on this.
Going with my previous numbers, if you buy a single "batch", you have a 10% chance of ending up with 10-15 drive failures, or a 90% chance of 2-3 failures.. If you buy across all batches, then you could reasonably expect 3-5 drive failures. For small numbers of drives (less than 10), this becomes statistically irrelevant, as there are too many other variables that could mess up a drive. However, when you are dealing with significant quantities of drives, it is much safer to buy from multiple sources to reduce the risk of having a large batch fail.

No, I understand the math completely.

I just don't think it applies to the average HTPC or PC home users yet --- like so many things --- gets perpetuated for no good reason.
assassin is offline  
post #25 of 37 Old 07-18-2012, 04:19 PM
Senior Member
 
mcturkey's Avatar
 
Join Date: Aug 2011
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 43
Quote:
Originally Posted by assassin View Post

No, I understand the math completely.
I just don't think it applies to the average HTPC or PC home users yet --- like so many things --- gets perpetuated for no good reason.

As I stated, it's not statistically useful or relevant for small quantities for an "average" HTPC or PC home user. But let's not go about saying it isn't true simply because of the relevance to one group. There actually are a number of folks who peruse this sub-forum for whom this consideration is valid and warranted.
mcturkey is offline  
post #26 of 37 Old 07-18-2012, 04:21 PM
Advanced Member
 
aliaskary77's Avatar
 
Join Date: Oct 2007
Posts: 606
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 51
Quote:
Originally Posted by assassin View Post

No, I understand the math completely.
I just don't think it applies to the average HTPC or PC home users yet --- like so many things --- gets perpetuated for no good reason.

hell, there goes the whole ocz/crucial thread out the window! :P

note: the post above is my opinion. as such, when reading any recommendations from me, please do you research and seek out other recommendations and make up your own mind on your next course of action. i mean, most reasonable adults should know that, but it seems this should be stated anyways.
aliaskary77 is online now  
post #27 of 37 Old 07-18-2012, 04:24 PM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by mcturkey View Post

I stated, it's not statistically useful or relevant for small quantities for an "average" HTPC or PC home user. But let's not go about saying it isn't true simply because of the relevance to one group. There actually are a number of folks who peruse this sub-forum for whom this consideration is valid and warranted.

Who is buying hundreds of hard drives?

Even those people that have 20+ drives likely did not buy them all at once because they likely added them on an as needed basis. And even then I am not sure its relevant.

So I would think the people that this applies to on this forum are maybe literally a handful.

BTW I was referencing people on this site and others that are PC (personal computer) users. So again, I don't think its relevant to those groups.

There could be some IT people from google on here getting ready to purchase their next batch of 2000 hard drives. But I doubt it.
assassin is offline  
post #28 of 37 Old 07-18-2012, 04:25 PM - Thread Starter
AVS Addicted Member
 
assassin's Avatar
 
Join Date: Jul 2004
Posts: 12,961
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 241
Quote:
Originally Posted by aliaskary77 View Post

hell, there goes the whole ocz/crucial thread out the window! :P

You mean the thread that has some actual data behind it? Even if its a poll it is actual data and not just a theory or a "feeling".
assassin is offline  
post #29 of 37 Old 07-18-2012, 05:54 PM
Senior Member
 
mcturkey's Avatar
 
Join Date: Aug 2011
Posts: 343
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 43
Quote:
Originally Posted by assassin View Post

Who is buying hundreds of hard drives?
Even those people that have 20+ drives likely did not buy them all at once because they likely added them on an as needed basis. And even then I am not sure its relevant.
So I would think the people that this applies to on this forum are maybe literally a handful.
BTW I was referencing people on this site and others that are PC (personal computer) users. So again, I don't think its relevant to those groups.
There could be some IT people from google on here getting ready to purchase their next batch of 2000 hard drives. But I doubt it.

I suspect it's more than a handful of people here who would be buying a larger number of drives, but again, the point is that just because it's not applicable or relevant for everyone does not mean you should automatically dismiss it as not being valid. If that were a fair or logical way of approaching things, then damn near everything stated on this forum is irrelevant, because it does not apply to everyone. Even in the Echo extender thread, many folks have different desires and expectations for what it will do. Should we exclude those who are not interested in specific aspects of it simply because they are not a majority? It's fine to have opinions and perspectives, but it becomes a problem when we start dismissing the opinions, perspectives, and experiences of others, just because they may not be the most relevant to a certain segment of the forum population. You have put a tremendous amount of time and energy into your guides, and I respect that. I have no use for them personally, but I recognize that there is a large segment of the AVS crowd that actually benefits greatly from them. That's part of what makes forums like this work so well - a variety of opinions and backgrounds. I've pretty clearly stated that what applies to someone buying a few drives does not apply to someone buying fifteen, or twenty, or a thousand, and I think it's important to recognize the difference there.
mcturkey is offline  
post #30 of 37 Old 07-18-2012, 06:08 PM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,451
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 151 Post(s)
Liked: 151
Quote:
Originally Posted by assassin View Post

Right. But your odds are higher of getting one of the those drives if you sample from 10 different lots. In your example you would be 100% assured of getting one of those drives if you were doing this for 10 drives from 10 different lots.
I think, as I suspected, that this is largely a non-issue for users of this forum.

In this context getting one bad drive is not a problem while getting two bad drives is a disaster.
Quote:
Originally Posted by assassin View Post

I just don't think it applies to the average HTPC or PC home users yet --- like so many things --- gets perpetuated for no good reason.

Doesn't mean it's not a valid consideration, or that it's "perpetuated for no good reason". There are good reasons to diversify your drive purchases if you're buying them for an array of some sort. Whether those good reasons, those considerations are important enough for the average AVS Forum member to worry about is up to the individual.

See what an anamorphoscopic lens can do, see movies the way they were meant to be seen
stanger89 is online now  
Reply Home Theater Computers

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off