Bad Drives-Best RAM

Hi,

A twofer if I can for my curiosity, and for others here too I think. The first I'll be looking for an answer after I post this, but I'll appreciate any personal, hands on info I can get. The second would be for those unfamiliar with the better names used by the Discussion members here.

So, One, are there any symptoms or signs that an HD is failing? Quirks, problematic behavior, loss of speed?
Do they generally go quick or can it take awhile? If it takes awhile, what would the best utility to keep track of it be? Would Apple Hardware Test show a problem?
Disk Utility is ok but it seems to lack in depth.

Two, I'd like to know what everyone's preference for RAM is. What manufacturers are considered the best? What do you consider the best?

Thanks,
Steve

MP 8-core 2.93GHz, 32GB RAM, Rad4870, 3xOCZ120, 3xIntel80, 4xVR300, Mac OS X (10.6.2), 30" HD Cinema Display, 20" Cinema Display, iPod touch, G5Quad (JIC)

Posted on Jan 19, 2010 6:17 PM

Reply
36 replies

Jan 20, 2010 6:27 PM in response to Samsara

Hey, I built my own computer once - long time ago. I couldn't tell you one thing about the RAM. It all came as little 8 KBit x 1 DRAM chips on a big card - all 64 KBs of it. The box with the backplane was bigger than my Mac Pro and was made out of copper and steel. Weighed a ton. Two 8" floppy disk drives were housed an another box the same size. Power supplies had iron transformers in them with huge capacitors to filter the DC. They were referred to as S-100 systems and were the second generation of personal computers The OS was called CP/M and ran in 16 KBs including BIOS and drivers. The CPU was a single 8 KHz Z80. Each of my floppy disks held a whopping 360 KBs of data.

Now those were the good old days!

Jan 21, 2010 6:26 AM in response to Kappy

Now those were the good old days!

I bet in a lot of ways they really were.. Even when I came in there was an excitement of being on the threshold of something revolutionary. And, man,... has it become so. And no end in sight! Someone learning computers today can still experience the thrill of being on the edge of a lot to come in the future.

Jan 21, 2010 7:25 AM in response to Kappy

Ignorance must surely be bliss

Ah, but isn't that part of Apple's doing. Hiding the "innards" with a beautiful and simple interface designed just for that purpose in mind? A "point and shoot" computer you only had to go click with.
I would be very surprised if many people knew about RAM or much else that goes on with a computer. Very, very few casual users I run into do. Very few users use their machines for heck of a lot, mostly I would say for the Internet since that seems almost essential now. No, like a lot of things, people are disinclined to look under the hood unless there are problems.
And that I would say that is one very good reason we are here. Because when people have been going along contentedly without a problem, not knowing what could even cause one, and one happens,... that's when a person needs the most help. Particularly because they are clueless and ignorant. It's after that when a persons real interest to learn is revealed.
But ignorance hasn't been bliss for me. Far from it, and maybe that's true for everyone. No one likes to feel stupid or ignorant. And speaking for myself only, I feel both at some time each and every day.

Jan 21, 2010 10:51 AM in response to The hatter

I started with the dumb terminal, an Ads, with a 300 baud telephone handset modem. That was my first "computer." I used it for doing JCL programming and running certain stat programs while doing my dissertation. It was a step up from using IBM Selectric typewriters converted to punch card stations. While in grad school I started with old punch card machines (can't remember the names,) then the university installed time-sharing to a large RCA mainframe and put in teletype terminals from which we could submit jobs written in Fortran or Basic. Even had paper tape readers/writers. These were my earliest introductions to using computers. I first learned to program in a language called ZACII (pronounced like "zack-see." Zero Address something, something - I don't remember. It was a very primitive "advanced" machine (independent) code language. The year was 1970.

My first personal computer was an Osborne I. Twin 5" 90K SSSD floppies and a 5" display all packaged into a sewing machine sized enclosure. Weighed a mere 19 pounds, I think. It was a portable computer. Came with two different Basic languages, a spreadsheet called SuperCalc, WordStar I, and DBase I. I taught myself two different Basic dialogs, Pascal, PL-I and PL-II, and Z80 machine code. I already knew Fortran. All but one of the Basics and the Z80 language were from Microsoft. Microsoft's documentation at the time was almost arcane. I also used to be a first-class DBase programmer. And, none of my degrees are even related to computers. My BA is in English Lit and my doctorate is in Finance/Real Estate.

Jan 21, 2010 10:58 AM in response to Samsara

Oh, so true. I can't even begin to understand the level of technology on the periphery of our knowledge. I can comprehend it only at a very basic level. Not being trained in information technology nor ever having worked in the field I can only understand a small piece of it. I wish I knew more, but I'm retired and that is not on my list of things to do. User uploaded file

Jan 21, 2010 11:04 AM in response to Samsara

Yes, and I think that's at the heart of every business catering to the technology consumer. The less they know the better. But I do find it telling that most people know more about how their toaster works than the do about their TV, radio, stereo, car, or microwave oven. Heck, most people don't even know what a microwave is! The scary part is consumers don't care about their ignorance.

When this nation was founded it was founded on the premise that people would have the opportunity to become educated participants in government. Our earliest educational system was premised on that goal. But politicians have ever since destroyed that idea. Modern public education seems based more on the limited goals of integration and selective ignorance. But then this is another topic altogether.

Jan 21, 2010 2:24 PM in response to Samsara

Google labs released a hard drive study in 2007 based on the thousands of drives that they have used.

[labs.google.com/papers/disk_failures.pdf]

engadget’s post about the study says “Google studied a hundred thousand SATA and PATA drives with between 80 and 400GB storage and 5400 to 7200rpm, and while unfortunately they didn't call out specific brands or models that had high failure rates, they did find a few interesting patterns in failing hard drives. One of those we thought was most intriguing was that drives often needed replacement for issues that SMART drive status polling didn't or couldn't determine, and 56% of failed drives did not raise any significant SMART flags (and that's interesting, of course, because SMART exists solely to survey hard drive health); other notable patterns showed that failure rates are indeed definitely correlated to drive manufacturer, model, and age; failure rates did not correspond to drive usage except in very young and old drives (i.e. heavy data "grinding" is not a significant factor in failure); and there is less correlation between drive temperature and failure rates than might have been expected, and drives that are cooled excessively actually fail more often than those running a little hot.”

Jan 21, 2010 2:34 PM in response to Chronic fatigue

And SMART reading are only really accurate using the vendor's own utility. There is no generic or universal - though I am getting the feel that the open source SMART Utility comes closest for Mac OS.

Which was intriguing.

Spare blocks being used, or a decrease in the pool of spares, etc., now wasn't that one of the primary indicators?

And if you click on Disk Warrior's Manual Test, it writes a line of data from the drive to the system.log containing the number of spares to begin (and there can be spares used from the factory), along with how many used since.

Jan 21, 2010 6:54 PM in response to Chronic fatigue

Thank you, CF.

Reading their conclusions, despite what they say about the effectiveness of scan errors to predict failure and the odds of it happening, I would still take note of them and begin to take precautions.
We find, for example, that after their first scan error, drives are 39 times more likely to fail within 60 days than drives with no such errors. First errors in reallocations, offline reallocations, and probational counts are also strongly correlated to higher failure probabilities.

And they go on to say,
Despite those strong correlations, we find that failure prediction models based on SMART parameters alone are likely to be severely limited in their prediction accuracy, given that a large fraction of our failed drives have shown no SMART error signals whatsoever.


failure rates are indeed definitely correlated to drive manufacturer, model, and age; failure rates did not correspond to drive usage except in very young and old drives (i.e. heavy data "grinding" is not a significant factor in failure); and there is less correlation between drive temperature and failure rates than might have been expected, and drives that are cooled excessively actually fail more often than those running a little hot.”

Good that that confirms our own observations, and if one wanted the name of a good HD maker there's enough of an experience pool here to draw upon. Interesting that heat and heavy use don't contribute all that much to a drives poor health, that maybe cold could be a worse factor. (My Mac seems ice cold when I go to use it in the morning).


Wish we had a better way solid way to predict drive failure it seems. But it's good that there are some fairly easy to spot symptoms, or warnings, if the drive allows us enough time to back up.
Easy in that our drives may simply disappear from the desktop (really bad news if it's the boot drive and no backup system drive), or Open and Save dialog boxes indicate the computer is having difficulty locating certain HDs you knew were there just a minute ago.
There's more but either of those would be enough for me to clone everything on that disk. And if it's the System Disk, I'd use the clone I made and transfer it back to a different internal drive in the Mac.

I wonder how SSDs will go compared to standard drives. I imagine they might go more quickly once the process starts.
Funny thing about SSD failure rates is that up front we are told they will not last as long as regular disks, and in two ways. They slow down over time and then, if you're lucky, they die altogether when they become just too slow to use anymore. Cool thing is that since they've been out, it's been too soon to see anyone report any significant speed loss yet.

Jan 21, 2010 7:25 PM in response to Chronic fatigue

I think from what I've read about SMART that if the SMART status says your drive is failing, then there's an 80-90 % likelihood the drive will fail. If the SMART status says a drive is OK then there's only a 50-60% probability it won't fail. In effect you can trust SMART to tell you the drive is failing.

Unfortunately, you can't trust all manufacturers to enable all or many of the parameters for SMART on their drives, and as Hatter said SMART is only as good as the data provided to it.

Jan 22, 2010 2:40 AM in response to Samsara

Hi Steve,

I don't know about elsewhere, but here in the UK Kingston RAM is generally regarded as the best alternative to Apple's own brand.

Most authorised Apple resellers seem to sell it, and it is the only brand of RAM sold by Macupgrades (probably the primary UK Mac upgrade specialist).

It is more expensive in the UK than other brands, but still considerably cheaper than Apple's.

I might be wrong, but I believe that Kingston is one of the few manufacturers, which actually builds its products to comply with Apple's own specifications, including thermal sensors.

Before I purchased 16GB of RAM for my 8-Core Mac Pro, I spent ages trawling the web for feedback, and complaints about Kingston memory were virtually non-existent, something which could not be said about most, if not all, other manufacturers. At that time there was a lot of reported dissatisfaction with 4GB modules (OWC and Crucial in particular), and so I decided to avoid them in favour of 2GBs.
Perhaps 4GBs have improved since then?

Anyway, I decided to pay the extra for the Kingston brand, and I purchased 2 KTA-MP1066K4/8G 1066Mhz DDR3 ECC Kits with thermal sensors, and to date I am completely happy with my purchase.

Regards,

Bill

Jan 22, 2010 6:54 AM in response to Biltan-Wales

Hi Bill,

Kingston was the RAM I first chose some years ago. It seemed to dominate advertising in the magazines, was highly talked about, and for me, it was what the Macs in my shop were using.
I can't remember why or when I changed to another brand, but as far as I know Kingston is still on the better RAM list.
and to date I am completely happy with my purchase.

Just the thing people hear and take into consideration. Thanks. 🙂

Steve

This thread has been closed by the system or the community team. You may vote for any posts you find helpful, or search the Community for additional answers.

Bad Drives-Best RAM

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.