Please clarify some things about SSDs for me

[leman]

We read the same article and below is a direct quote of the return rates. 68% higher return rate for SSDs overall. So now you want to ignore some of the SSDs included in the article you cited yourself so the outcome suits you? Hmm.

Yes, I want to ignore the OCZ SSDs because they are an outlier and had 5 times the return quotes of any other SSD. If you want, we can also ignore Hitachi HDDs, which - according to the article are the least reliable of all surveyed HDDs. It won't change a thing.

Just looking at the total picture is akin to saying that cars are unreliable because every second Opel breaks down. Its not the problem of cars, its the problem with Opel

 

[Weaselboy]

Just looking at the total picture is akin to saying that cars are unreliable because every second Opel breaks down. Its not the problem of cars, its the problem with Opel

Using your analogy, if you want to see overall reliability rates for cars, yes... one would include Opel. What you are suggesting is we compare reliability data only for SSDs that are reliable.

 

[leman]

Using your analogy, if you want to see overall reliability rates for cars, yes... one would include Opel. What you are suggesting is we compare reliability data only for SSDs that are reliable.

Ok, if you have 4 companies producing reliable SSDs - which also score fairly close to each other - and one company producing SSDs which scores much worse, what would your conclusion be? That the entire technology is problematic or that the one company is probably doing something wrong? If its the first one, I honestly see no point in continuing this discussion.

 

[Orlandoech]

I also own a OCZ SSD and I had absolutely no problems with it. Still, we are discussing statistics - and those say that over 5% of OCZ SSDs were returned. Of course, it might be that the particular store the data comes from received faulty OCZ products, but I find it rather unlikely.

While I never had to return an OCZ SSD, I don't doubt that statistic lol. OCZ had some major issues with some SSDs. All the SSDs I owned were;

2 x OCZ Vertex Plus (120GB or 128GB?)
2 x OCZ Vertex (60GB or 64GB?)
2 x OCZ Vertex 2 (60GB or 64GB?)
1 x OCZ Vertex 3 Max IOPS (128GB?)
2 x OCZ Vertex 4 (256GB)

And 0 failed and 0 had issues.

 

[Weaselboy]

Ok, if you have 4 companies producing reliable SSDs - which also score fairly close to each other - and one company producing SSDs which scores much worse, what would your conclusion be? That the entire technology is problematic or that the one company is probably doing something wrong? If its the first one, I honestly see no point in continuing this discussion.

Look, if there was a sample size here of 30 companies, and company 30 was 10 or 20 times worse than the company in 29th place, I might buy that. But that is not what we have here. We have a sample size of five companies and you want to toss 20% of the data because of a less than 4% spread. OCZ is now just 2.07% worse than Corsair was in their last survey. Look back through the same survey over the last three years. Each time OCZ is worse than the others by around 2%. So no, tossing the OCZ data makes no sense.

We can argue statistics all day, but you made a snarky post telling us to review your "more serious source", then posted an article that says the opposite of what you claimed, and now want to ignore the data that does not line up with your claim.

I frankly have no idea if this French retail data is worth anything at all, but you are the one who cited it.

Also, I never said the "entire technology" was problematic.

- Intel 0.45% (against 1.73%)
- Samsung 0.48% (N/A)
- Corsair 1.05% (against 2.93%)
- Crucial 1.11% (against 0.82%)
- OCZ 5.02% (against 7.03%)

----------

All the SSDs I owned were;

2 x OCZ Vertex Plus (120GB or 128GB?)
2 x OCZ Vertex (60GB or 64GB?)
2 x OCZ Vertex 2 (60GB or 64GB?)
1 x OCZ Vertex 3 Max IOPS (128GB?)
2 x OCZ Vertex 4 (256GB)

You better never take your laptops to France!

 

[cube]

Not sure what you mean by that, that is a pretty broad brush you are using. For years WD has offered a very good industrial SSD that consumers can also buy. Dunno much about Seagate, perhaps there just was not a business case, or they just not have found the right SSD company to acquire.

Consumers want cheap and high capacity, which have plenty of technological challenges. Hard drives have fallen on their faces recently too, look at the 4K fiasco, and all the trouble folks have been having with green drives when they try to use them in a performance environment (a NAS for example). In that sense neither development is stable as you can pick out an issue with any consumer oriented technology if you care to take the time to look. There are a few solid SSDs and a few solid rotational drive examples too.

The industrial or OEM market is different. Industry must have reliability and or performance as they have no need for a drive defect to ruin their product or service. SSDs have been solid in that application for several years now. So have rotational Hard Drives.

I just don't think there is as much difference in stability as some may think.

What I mean is that in the last months there has been at least 1 consumer release from a serious major player where some serious bug was present.

I didn't say development of consumer vs enterprise was less stable, but that SSD development in general is less stable than that of hard drives, so that those hard drive players would avoid shipping product in large numbers to customers many of which do not have serious backup.

 

[leman]

Look, if there was a sample size here of 30 companies, and company 30 was 10 or 20 times worse than the company in 29th place, I might buy that. But that is not what we have here. We have a sample size of five companies and you want to toss 20% of the data because of a less than 4% spread. OCZ is now just 2.07% worse than Corsair was in their last survey. Look back through the same survey over the last three years. Each time OCZ is worse than the others by around 2%. So no, tossing the OCZ data makes no sense.

You are just splitting hairs. I don't want to throw out 20% of the data. I want to throw out the company which does not follow the same pattern as the rest of the data. Looking at these numbers, OCZ return rates were not representative to the SSD state of the art tech in 2012, for whatever reason that might be. It is completely pointless to take an average here.

I frankly have no idea if this French retail data is worth anything at all, but you are the one who cited it.

This is the only one of your points that I can agree with. Still, this was the only source I could find in a quick search. If you have something better, please do share.

Also, I never said the "entire technology" was problematic.

But this is the whole point of this thread! It was argued that SSDs - in their state of the art - are ultimately less reliable than HDDs in their state of the art; and not 'ready' for use. The data I quoted shows that in 2012, according to the return rates of the particular French store, the SSDs were returned much less frequently than HDDs, save for products of one single company.

 

[XoFu]

The OCZ numbers for SSD (5%) are in fact statistically significant outliers (p<0.05, Statistical Grubbs' test). Hence OCZ does not represent the universe of SSD return in this sample, and has to be removed from the group.

It seems in fact that OCZ has something to improve.

I have an OCZ in my MBP with not issues

 

[Weaselboy]

The OCZ numbers for SSD (5%) are in fact statistically significant outliers (p<0.05, Statistical Grubbs' test). Hence OCZ does not represent the universe of SSD return in this sample, and has to be removed from the group.

It seems in fact that OCZ has something to improve.

I have an OCZ in my MBP with not issues

Except this.

...and the test (Grubbs) should not be used for sample sizes of six or less since it frequently tags most of the points as outliers.

 

[Weaselboy]

The data I quoted shows that in (the first half of) 2012, according to the return rates of the particular French store, the SSDs were returned much (somewhat?) less frequently than HDDs, save for products of one single company.

Now see, if you had just said that from the outset, we would not be having this discussion... but you didn't.

Your post with the original link made it sound like everyone here was some sort of rube for considering anecdotal evidence based on forum posts then you offered up as a "serious source" the French survey. You just agreed with my comment that we don't know if the French survey is worth anything at all.

Just to be clear, I have no idea if SSDs are more or less reliable than HDDs, and I have not seen any verifiable data one way or the other. It would appear you have not seen any such data either.

 

[XoFu]

Weaselboy: n>6 says who? if all you have is 5??
Talking about bending arguments.... oh whatever!

OP: I've had SSDs in my macs for over 3 years without fail. Crucial (C300 and M4 and OCZ) so in my experience with a Time Machine and an SSD you should be home free, and enjoy the experience of the SSD

 

[Weaselboy]

Weaselboy: n>6 says who? if all you have is 5??

Page one here and page two here.

 

[XoFu]

Page one here and page two here.

I guess you should look at the conclusions of your second link to be 100% sure you are wrong!

As far as I'm concerned it ends here!

 

[Orlandoech]

You better never take your laptops to France!

Why is that lol?

 

[jesaja]

To the discussion above:
If I decided to buy an SSD or HDD, why would I care for the failure rates of the most unreliable manufacturer? I would buy from one of the top three companies, if not the most reliable one. and I guess everyone who comes here to educate himself before buying would do the same.

So the failure rates of these manufacturers are the ones to compare, because those are what the eventual choices will probably be like.

 

[saturnotaku]

I have an OCZ in my MBP with not issues

As did I. Only reason I replaced it was that the massive 960 GB Crucial M500 came out and I was lucky to get in on the first shipment available.

 

[XoFu]

As did I. Only reason I replaced it was that the massive 960 GB Crucial M500 came out and I was lucky to get in on the first shipment available.

Tell us more about the M500!

 

[Weaselboy]

I guess you should look at the conclusions of your second link to be 100% sure you are wrong!

As far as I'm concerned it ends here!

Wow... and you accused me of bending arguments. Hilarious.

----------

Why is that lol?

it was a joke. It seems like OCZ SSDs and France are a bad mix!

 

[saturnotaku]

Tell us more about the M500!

Not quite as fast as the OCZ Vertex 4 I replaced but not enough of a difference to notice in everyday computing. The M500 is a 7mm drive but works perfectly in a MacBook Pro without the need for the included 9.5mm spacer. If you're buying an SSD for the first time or want something with massive capacity, I'd thoroughly recommend the M500. If you already have a current-generation (SATA 6 Gb/s) SSD and do not need 960 GB of capacity in a single drive, the M500 would be, at best, a lateral move and in some cases a downgrade.

 

[XoFu]

Not quite as fast as the OCZ Vertex 4 I replaced but not enough of a difference to notice in everyday computing. The M500 is a 7mm drive but works perfectly in a MacBook Pro without the need for the included 9.5mm spacer. If you're buying an SSD for the first time or want something with massive capacity, I'd thoroughly recommend the M500. If you already have a current-generation (SATA 6 Gb/s) SSD and do not need 960 GB of capacity in a single drive, the M500 would be, at best, a lateral move and in some cases a downgrade.

Thank you for the info. I was looking at the M500 the other day and saw no feedback in the Crucial site, so I was wondering. I intend to buy one of 480G for my 2010 white MB which I keep as my backup machine, but will wait for prices go down a bit.