Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
SSD's are STILL the wild, wild west as far as I'm concerned. Feel free to use them, and DO use them, just keep a backup on a standard HD. Standard HD's are proven reliable. If they weren't nobody would have used them for the last 3 decades. Period.

They're not perfect, but what is? Who ever heard of a car running 30 years without the need of some type of repair. You're safe with an HD, but an SSD is still a crapshoot. My opinions, of course.
 
SSD's are STILL the wild, wild west as far as I'm concerned. Feel free to use them, and DO use them, just keep a backup on a standard HD. Standard HD's are proven reliable. If they weren't nobody would have used them for the last 3 decades. Period.

They're not perfect, but what is? Who ever heard of a car running 30 years without the need of some type of repair. You're safe with an HD, but an SSD is still a crapshoot. My opinions, of course.

I hear what you are saying but current SSDs are WAY better than the original HDDs when that technology started (you do remember the house-brick-sized 10MB first IBM PC drive?). No-one is "safe" with an HDD without a backup, same as with an SSD....
 
With really old ssd drives no one installs any more. They are thrashing them to death no wonder.

I still use spinning disks - for storage and still will. Booting off one - history.

"Thrashing them to death" is what's to be expected. The sudden data loss after power outages or shutdown is to be expected. With a hard drive, the worst that seems to happen is that the index files need repair, but total data loss?

We need more investigations into these types of problems to see if the manufacturers are really making these things real world reliable.

Speed has it's merits, but not if all the data on a drive gets lost at a given instance.
 
Actually the worst thing would be the read/write head damaging the platters and FUBARing the drive. With no power the drive wouldn't park the head properly.
 
"Thrashing them to death" is what's to be expected. The sudden data loss after power outages or shutdown is to be expected. With a hard drive, the worst that seems to happen is that the index files need repair, but total data loss?

We need more investigations into these types of problems to see if the manufacturers are really making these things real world reliable.

Speed has it's merits, but not if all the data on a drive gets lost at a given instance.

But if you are going to install a drive that is going to take a thrashing you would install a drive capable of taking it. Like the Intel types or 840 Pro, these are older consumer grade models they are testing. Plus if you make sure a system has sufficient or even max ram for that model the drives have far less work to do paging to the disk all the time.

And power outages in notebooks are rare unless the battery is broke. For desktop/iMac having UPS backup power particularly if you have poor mains quality is a given no matter whether you have a spinning disk or solid state..
 
The way the Mac kernel has acted has been to cache as much application as possible. This is why once an app is loaded, it launches almost instantly if it's been terminated. Unless constant read/write drive access is needed, the only real time you see the performance gains of SSDs are during initial loads and boot. Other than that, you don't notice the difference, except in price of course.

The stuff about SSDs being rugged isn't really all that good of an argument because the rest of the computer isn't terribly rugged either. If you want ruggedized equipment, yes, you'd pick an SSD over an HD, but who in their right mind goes around tossing their laptop around like it's a baseball?
 
No, I'm talking about the reported instances where someone turns on their system and the SSD is totally wiped of data. I've never seen this on a hard drive, unless it's actually completely failed.
 
A corrupted table on either kind of drive will give a similar symptom. Erasing either drive in any meaningful way takes far more than a failure, I think if an SSD appears to have no data it will be actually a corrupted table, same as for an HDD.

Just because HDDs have physical failure modes because of their technology (moving bits hit other moving bits etc), that does't stop SSD having similar issues electronically, if it doesn't know where the data is, it may as well not exist.
 
No, I'm talking about the reported instances where someone turns on their system and the SSD is totally wiped of data. I've never seen this on a hard drive, unless it's actually completely failed.

I've seen things like that happen in the recent era with HDD's with large drives (>2.2 Tb). In fact I lost over 2Tb off one of my own WD Green 3Tb in a Win 7 Media Center box only two weeks ago. Left me only the 3 most recent bluray rips on the drive out of over 200! I only just restored the data back onto a 4TB on Thursday and its going back to WD for RMA this week.

Took days in total to reformat, verify, copy and then a final chkdsk /r to verify the data took over 12 hours on the 4Tb - a long, long time..

So old spinning disks, even when they are used for what I call their primary usage in this SATA/PCIe SSD boot OS era - data storage can still play up and lose data. Thank goodness I had a backup!
 
Last edited:
I'm not trying to say HDs are perfect. What I'm saying is that when an SSD goes it tends almost to be binary, like an on/off switch.

Most HD failures have had tell-tale signs and you can pick them up testing periodically with something like Scannerz before it's too late. With an SSD it seems to be either something that just fails completely, or in the case of that one strange block going bad, it just goes bad all on it's own. With an HD most bad blocks (at least I think this is true) are from head crashes.

SSDs seem to suffer sudden, silent deaths. HDs, IMHO, tend in most cases to "cry out" if you will, in one way or another, letting the user know there's something wrong.

...but they aren't perfect though. What is?
 
I'm not trying to say HDs are perfect. What I'm saying is that when an SSD goes it tends almost to be binary, like an on/off switch.

Most HD failures have had tell-tale signs and you can pick them up testing periodically with something like Scannerz before it's too late. With an SSD it seems to be either something that just fails completely, or in the case of that one strange block going bad, it just goes bad all on it's own. With an HD most bad blocks (at least I think this is true) are from head crashes.

SSDs seem to suffer sudden, silent deaths. HDs, IMHO, tend in most cases to "cry out" if you will, in one way or another, letting the user know there's something wrong.

...but they aren't perfect though. What is?

Can't disagree with any of that, when SSD do go it's very sudden. Though I advise absolutely everyone to have their systems backed up to an external or a time capsule which are almost always good old spinning disks!

As an example today my replacement Mac Pro arrived this morning - a 4,1 single socket which after AHT got flashed to a 5,1 and fitted a hex core 3.33ghz w3680 Xeon and 24gb of ddr 3 1333 ram. I can notice the difference with the superior single core performance with the newer and faster Xeon but I am currently doing the initial setup using a normal HDD. My older 8 core 3,1 with ddr2 800 ram, PCIe SATA 3 card and twin 840 evo drives blows this 6 core away in terms of how quick it feels in every single other aspect by miles.

Once you go SSD for booting an OS you never go back - or not for long :D
 
Last edited:
With an HD most bad blocks (at least I think this is true) are from head crashes.

Head crashes spray the magnetic coating all over the inside of the drive and write off the drive.

With an HDD reading the data back from the platter is an analogue activity so it is subject to retries, increasing read amps etc - there are several ways to try and get the data off when it doesn't come immediately.

SSDs are all electronic, the data is there, or it isn't so yes they will be more prone to sudden death.
 
how to check ssd for bad blocks? are there other ways except scannerz? I get random lockups for 3-4 seconds on my macbook pro retina 13" with yosemite. After rebooting my system it gets solved. Could it be the ssd?
 
how to check ssd for bad blocks? are there other ways except scannerz? I get random lockups for 3-4 seconds on my macbook pro retina 13" with yosemite. After rebooting my system it gets solved. Could it be the ssd?

Hate to tell you, but that's probably Yosemite just being Yosemite!!!!!

Scannerz can check for bad blocks on the SSD and usually the manufacturer provides some software that will do likewise. If they auto correct properly it shouldn't be a problem. If the SSD can't recognize a bad block, then it's a serious problem because it would imply a firmware bug.

Like I said, though, it's probably Yosemite.
 
...I should have also added that before blaming your hardware, if you've moved to Yosemite and find yourself having problems, you should probably go to the App Store and read the user comments, or visit the Yosemite section of this site.

There are a lot of problems with Yosemite.
 
thanks for the answers. Im relieved that its not a hardware issue. Lets wait and hope for the updates to roll in and fix the issues.
 
We don't really know if it isn't a hardware issue. However, any time a new OS is installed and the system suddenly develops some types of problems, the first thing to blame is the OS not the hardware.

I always clone a new OS onto a spare volume to test it out before installing it. The first release of Mountain Lion was a bug filled nightmare and Yosemite looks like it's as bad or worse.
 
We don't really know if it isn't a hardware issue. However, any time a new OS is installed and the system suddenly develops some types of problems, the first thing to blame is the OS not the hardware.

I always clone a new OS onto a spare volume to test it out before installing it. The first release of Mountain Lion was a bug filled nightmare and Yosemite looks like it's as bad or worse.

None of them are perfect - even snow leopard had teething problems. Though nothing can possibly compare to OS X 10.0 which is my worst zero release of all time, topping even any Windows I can think of since 3.0. It truly was an absolute dog!

I am still holding off Yosemite on both my boxes for most likely the same reasons you are :D


Though SSD's are the best thing since we stopped booting off floppy discs. I used to do that with Macintosh and Windows :D
 
Last edited:
We don't really know if it isn't a hardware issue. However, any time a new OS is installed and the system suddenly develops some types of problems, the first thing to blame is the OS not the hardware.

True in your own system but when some several million users upgrade then some of them will suffer coincidental hardware issues, some will have had latent issues they may not have been aware of etc etc...
 
True in your own system but when some several million users upgrade then some of them will suffer coincidental hardware issues, some will have had latent issues they may not have been aware of etc etc...

That's true, but the probability of someone installing a new OS and the hardware going bad at the exact same time isn't high. That doesn't mean it can't happen, it's just not likely.
 
That's true, but the probability of someone installing a new OS and the hardware going bad at the exact same time isn't high. That doesn't mean it can't happen, it's just not likely.

Actually its a dead certainty that some machines will have issues within say week before or after they choose to upgrade - that is the problem with millions of installs of Yosemite going on, some installs <will> coincide with hardware issues, internet issues, router issues etc etc.

I work for a telco, we see it all the time.

Lets say 10million Macs are 4yrs old this year (out of the 75million in circulation). Assume only 10% (1 million), will have a hardware issue in this year. That means roughly 80,000 will have an issue in each month, say only 25% upgrade to Yosemite in that month - that is 20,000 machines having an issue in the same month they update to Yosemite.....5,000 in the same week, 700 on the same day.

If you add in all the other IT stuff that needs to work on a daily basis (and therefore goes wrong on a continuing basis), then the above figures can be multiplied several times over.
 
I really can't argue with your logic, because it's true, but in Yosemite's case I think the OS itself stands a better chance of being the source of the problem.

10.10.0 had thousands of reviews in the App Store, and since the update to 10.10.1 they got rid of them and started over. As of the time of this posting the top and bottom ratings are as follows:

97 - 5 star ratings (Excellent review)
93 - 1 star ratings (Terrible review)

Granted, a lot of people just don't like the way the OS looks, but others are identifying a myriad of other problems as well. A near 50% "Terrible" rating is pretty bad in my opinion.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.