"SSDs are not the magic bullet that some would have you believe"

uOpt wrote in post #15805872
Speed is one thing.

Reliability is another one. IMHO much of consumer SATA SSD is still outright crap. And when they fail they fail brutally with the controller down. Regular HDs don't only often allow you to retrieve much of the data after a fail, they also tend to have occasional read errors or other things that you can pick up in S.M.A.R.T. before going down.

Raid over SSDs is an entirely matter still.

Given the speed at which SSDs developed I am very pessimistic about lurking firmware bugs.

There have been cheap SSDs that failed frequently (OCZ, similar crap). There are several excellent companies now, such as Intel and Samsung that make very reliable SSDs with high quality testing processes and solid firmware and controllers. Both have had firmware glitches but.....

The rest of your post sounds like someone who didnt get to experience the IBM Deathstar GXP hard drives that failed in copious numbers in 2000/2001, or the recent Seagate 7200.11 firmware glitch that lock the drive up and cause it to report as 0GB.

Spinning disks have had firmware bugs and poor designs as well. Just like SSDs.

I don't know why but my SSD Transcend 256GB that i installed on my Macbook Pro early 2008 model doesn't perform well, it boots fine but if i keep the laptop without use and go under sleep then it wakes up slow and sometimes hanging for long time before it can be responsive again, so i hope i didn't choose a wrong one. Have no issue with my another MBP 2009 with Samsung SSD, in fact i bought that Transcend because it was cheap and i didn't want to buy one expensive and very fast SSD for my old non SATA3 compatible, it doesn't have SATA2 also, working with only SATA1 [1.5GB/s].

But honestly speaking, this SSD booting so fast over the old manufacturer HDD on m MBP, i am now replaced all my computers with SSD drives.

thedge wrote in post #15808826
The rest of your post sounds like someone who didnt get to experience the IBM Deathstar GXP hard drives that failed in copious numbers in 2000/2001, or the recent Seagate 7200.11 firmware glitch that lock the drive up and cause it to report as 0GB.

Spinning disks have had firmware bugs and poor designs as well. Just like SSDs.

I actually had several of the Deathstars back in the day. I had one of the 7200.11 drives fail with the startup glitch and I left that Seagate was actively deceiving their customers about the specific nature of the bug, and that had left me in the cold (they made it sound like the problem only appears after a lot of stops and starts, but in reality the probability of a startup fail was equal right from the beginning and "what they meant" was that statistically you will only experience the failure if you stop a lot. That's an important difference for those who only do 24/7 on UPSes).

Of course I never use WD drives so I'm not that hard hit by firmware bugs.

Hen3Ry wrote in post #15807679
You're halfway there. Not only are most SSDs crap, so are most disk drives. The drives available as "desktop drives" are mass produced commodity items, with one simple engineering design criterion: to outlast the operating system installed on it. The aren't particularly high tech, and are certainly not designed for reliability, simply because the desktop market is, as Western Digital puts it (when they explain why you shouldn't use desktop drives and expect long term reliability), "a price-sensitive market." This is simply a description of a market that values reliability, but doesn't want to, or simply wont, pay the price necessary for actual reliable disk drives. If you check the prices/types of the disk drives used in commercial server environments, you'll find they are much more expensive than WD Blacks - as for reliable SSDs, HP makes a commercial 800 GB SSD - it sells fro $6400 bucks. There's a reason that enterprise level components cost more than "desktop" drives, and it isn't necessarily speed.

These days, if a drive outlasts Moore's Law, the odds are excellent that you,as a user, won't use it in your next machine because it's too old, or too small, but not because it's too slow.

Here are the two primary rules of disk drive usage:

1. All disk drives fail.

2. The larger capacity the drive, the more data you will lose.

Full Stop.

It really isn't quite that simple. As I said in my post you were replying to, the failure rate of real HDs isn't that different.

However, recovery in the case of failure has a lot of difference:

• SSDs usually go all dead (controller failure) while HDs usually allow you to recover quite a bit of data, often almost all data.
  
• Failed HDs that do experience electronics failures can often be read by exchanging the circuit board. Try that with a SSD.
  
• Even in the infamous Seagate 7200.11 case - sure the data can't be read by the customer, but the data isn't lost. You could get the data back by either sending it to Seagate or by issuing the right commands over the disk's serial port yourself. No such recovery was even attempted by the OCZ drive that I used for games (recoverable data, otherwise I wouldn't have used a OCZ). Dead SSD is dead SSD. Firmware bugs in SSDs will usually affect new block allocation and block mapping and hence scramble the data real good.
  
• Let's not talk about WD's firmware.
  
• I have seen harddrives announce their upcoming death in S.M.A.R.T. (which of course few people look at before read errors, but they could if they cared). Never seen it for a failing SSD. It's unable to read from one moment to the next.
  

thedge wrote in post #15808826
There have been cheap SSDs that failed frequently (OCZ, similar crap). There are several excellent companies now, such as Intel and Samsung that make very reliable SSDs with high quality testing processes and solid firmware and controllers. Both have had firmware glitches but.....

The rest of your post sounds like someone who didnt get to experience the IBM Deathstar GXP hard drives that failed in copious numbers in 2000/2001, or the recent Seagate 7200.11 firmware glitch that lock the drive up and cause it to report as 0GB.

Spinning disks have had firmware bugs and poor designs as well. Just like SSDs.

With all due respect, OCZ has seriously cleaned up their act. You should check out their current lineup.

kaptnkain wrote in post #15813333
With all due respect, OCZ has seriously cleaned up their act. You should check out their current lineup.

That train has left the station, sorry.

The thing isn't just that my (only) OCZ SSD broke. The problem is that they wouldn't even accept a RMA without running *two* random programs, one to detect the SSD model, one to flash the firmware. One program only ran booted into DOS and the other one only under windows (bad for Mac and Linux users).

Sorry, but if the thing stops delivering block from one moment to the next it is unlikely to be helpful to flash the firmware. And even if it was, I expect *them* to go through the procedure. The thing breaks inside warranty, I deserve a RMA without requiring me to make a USB boot stick and track down a windows box.

They suck.

BTW, having a stack of broken OCZ branded PSUs in the basement.

uOpt wrote in post #15813918
That train has left the station, sorry.

The thing isn't just that my (only) OCZ SSD broke. The problem is that they wouldn't even accept a RMA without running *two* random programs, one to detect the SSD model, one to flash the firmware. One program only ran booted into DOS and the other one only under windows (bad for Mac and Linux users).

Sorry, but if the thing stops delivering block from one moment to the next it is unlikely to be helpful to flash the firmware. And even if it was, I expect *them* to go through the procedure. The thing breaks inside warranty, I deserve a RMA without requiring me to make a USB boot stick and track down a windows box.

They suck.

BTW, having a stack of broken OCZ branded PSUs in the basement.

I have a 50% failure rate with OCZ. Bought two, one worked for a week, maybe two. Not again.

uOpt wrote in post #15811606
It really isn't quite that simple. As I said in my post you were replying to, the failure rate of real HDs isn't that different.

However, recovery in the case of failure has a lot of difference:

• SSDs usually go all dead (controller failure) while HDs usually allow you to recover quite a bit of data, often almost all data.
  
• Failed HDs that do experience electronics failures can often be read by exchanging the circuit board. Try that with a SSD.
  
• Even in the infamous Seagate 7200.11 case - sure the data can't be read by the customer, but the data isn't lost. You could get the data back by either sending it to Seagate or by issuing the right commands over the disk's serial port yourself. No such recovery was even attempted by the OCZ drive that I used for games (recoverable data, otherwise I wouldn't have used a OCZ). Dead SSD is dead SSD. Firmware bugs in SSDs will usually affect new block allocation and block mapping and hence scramble the data real good.
  
• Let's not talk about WD's firmware.
  
• I have seen harddrives announce their upcoming death in S.M.A.R.T. (which of course few people look at before read errors, but they could if they cared). Never seen it for a failing SSD. It's unable to read from one moment to the next.
  

Yes, but I'm writing above specifically of rotating drives. Recovery is a different issue, especially once the owner sees the prices for taking his hard drive into a clean room to recover the data. I think we're in agreement. This is really why I keep nothing on SSDs except operating Systems, apps, and cache/paging - though occasionally I save an image I'm currently working on in my picture directory, especially if it gets large. All small, always backed up, and easy to restore.

The fact that it cuts down on Windows boot time significantly for me is worth the price of admission.

uOpt wrote in post #15813918
That train has left the station, sorry.

The thing isn't just that my (only) OCZ SSD broke. The problem is that they wouldn't even accept a RMA without running *two* random programs, one to detect the SSD model, one to flash the firmware. One program only ran booted into DOS and the other one only under windows (bad for Mac and Linux users).

Sorry, but if the thing stops delivering block from one moment to the next it is unlikely to be helpful to flash the firmware. And even if it was, I expect *them* to go through the procedure. The thing breaks inside warranty, I deserve a RMA without requiring me to make a USB boot stick and track down a windows box.

They suck.

BTW, having a stack of broken OCZ branded PSUs in the basement.

Hen3Ry wrote in post #15814480
I have a 50% failure rate with OCZ. Bought two, one worked for a week, maybe two. Not again.

I dunno, their priorities seem to have changed since RP got kicked out.

Hen3Ry wrote in post #15814480
I have a 50% failure rate with OCZ. Bought two, one worked for a week, maybe two. Not again.

I wouldn't have had a problem with it as much as I did if it wasn't for the IMHO insane RMA situation at OCZ where huge amounts of my time are wasted on a dead drive before they even grant me a return.

I needed to make a USB boot stick, bring up a spare computer with it to run one of their programs, then take my gaming computer apart (my only windows box) to connect the SSD there to run the other program. Only to then report to them that their stupid program simply hangs on the dead SSD and does nothing. That got me a RMA number.

This insanity of requiring the customer to spend time to try to fix it themselves with custom software is also something that regular harddrive companies do, however I have never seen it this badly with two programs requiring two different OSes. But the HDs are cheaper so you can just trash them if it isn't worth it.

kaptnkain wrote in post #15814619
I dunno, their priorities seem to have changed since RP got kicked out.

No kaptnkain, that was in 2012.

RS only took the CEO position in the middle of October 2012.

uOpt wrote in post #15811606
It really isn't quite that simple. As I said in my post you were replying to, the failure rate of real HDs isn't that different.

However, recovery in the case of failure has a lot of difference:

• SSDs usually go all dead (controller failure) while HDs usually allow you to recover quite a bit of data, often almost all data.
  
• Failed HDs that do experience electronics failures can often be read by exchanging the circuit board. Try that with a SSD.
  
• Even in the infamous Seagate 7200.11 case - sure the data can't be read by the customer, but the data isn't lost. You could get the data back by either sending it to Seagate or by issuing the right commands over the disk's serial port yourself. No such recovery was even attempted by the OCZ drive that I used for games (recoverable data, otherwise I wouldn't have used a OCZ). Dead SSD is dead SSD. Firmware bugs in SSDs will usually affect new block allocation and block mapping and hence scramble the data real good.
  
• Let's not talk about WD's firmware.
  
• I have seen harddrives announce their upcoming death in S.M.A.R.T. (which of course few people look at before read errors, but they could if they cared). Never seen it for a failing SSD. It's unable to read from one moment to the next.
  

SSDs can be caught before complete failure.... In some cases. But yes, generally complete failure. I have also had SSDs that showed SMART errors prefailure.

Yes, 7200.11 drives didn't have "data loss" and could be rescued with serial commands to the controller (which is what I did) but guess what happens when some drives in a RAID5 pool of five 7200.11 drives got the firmware bug, which is what happened to me? Sure, the "data" was there, but a lot of corruption.

kaptnkain wrote in post #15813333
With all due respect, OCZ has seriously cleaned up their act. You should check out their current lineup.

I have, and there's nothing to justify my money going to them vs Intel or Samsung, who have a much better track record.

thedge wrote in post #15817661
Yes, 7200.11 drives didn't have "data loss" and could be rescued with serial commands to the controller (which is what I did) but guess what happens when some drives in a RAID5 pool of five 7200.11 drives got the firmware bug, which is what happened to me? Sure, the "data" was there, but a lot of corruption.

Corruption? How so? The firmware bug means it won't power on, so you should have a cleanly degraded (and hence not corrupted) array.

But yes in general I wish I could pass the drive fail errors through the layers of raw devices to instantly mark a filesystem for readonly use only, like Linux does by default for plain drives. If it would do that you could deal with multiple drive errors in a RAID much better (aka if a second drive entirely fails during recovery and the first drive might be able to give out it's data one more time then you could recover that way).