Intel SSD Sandforce 330/520 return rates ?

[hhhd1]

Is there any statistics for Intel SSD Sandforce 330 and 520 return rates or failure rate ?

 

[ShintaiDK]

I think they are in the 4-5% range.

After release they changed Intels return rates from 0.1% to 1.73%.

Epic stupidity not to continue with their own controllers.

 

[bryanW1995]

I think that you are wrong. First of all, I know that the earlier return rate wasn't .1%. According to Anandtech, it was .59% as of 3/2/11. That's still awesome, but only 490% higher than your estimate. Additionally, in that report even OCZ had a return rate of under 3%, so I find your claim of "4-5% range" on intel returns as extremely suspect.

 

[ShintaiDK]

I think that you are wrong. First of all, I know that the earlier return rate wasn't .1%. According to Anandtech, it was .59% as of 3/2/11. That's still awesome, but only 490% higher than your estimate. Additionally, in that report even OCZ had a return rate of under 3%, so I find your claim of "4-5% range" on intel returns as extremely suspect.

http://www.hardware.fr/articles/862-7/ssd.html

Try a newer one.

- Crucial 0,82% (contre 0,8%)
- Intel 1,73% (contre 0,1%)
- Corsair 2,93% (contre 2,9%)
- OCZ 7,03% (contre 4,2%)

 

[bryanW1995]

Regardless, your earlier claim of ".1%" for intel was still completely wrong. And you're saying that within 2 months of launch, cherryville somehow magically increased intel's return rate from .59% to 1.73%? I'm going to go out on a limb and say that the higher return rate is due to the 320 series issues and has very very little if anything to do with cherryville. Nice try, though.

I find 2 things interesting about those lists:

1. Anand verified with several manufacturers that the data from March 2011 was accurate, so it seems likely that the more recent link you provided is accurate.
2. OCZ and intel had MAJOR issues over that 13 month period, and the RMA data backs that up. Corsair looks slightly worse but at least it's close, while crucial did very well during that period.

 

[ShintaiDK]

How can it be wrong when the site reports it.

So you say the number is correct, yet you refuse it

Previous one, dated: 25/10/2011

http://www.hardware.fr/articles/843-7/taux-retour-composants-5.html

- Intel 0,1% (contre 0,3%)
- Crucial 0,8% (contre 1,9%)
- Corsair 2,9% (contre 2,7%)
- OCZ 4,2% (contre 3,5%)

 

[bryanW1995]

I apologize on the .1%, I was unaware of that article.

Having said that, you didn't address my comments about the intel Sandforce drive reliability. Can we at least agree that 2 months wasn't long enough to have any meaningful impact upon intel's return rates, and that the likely culprit for the april 2012 report was the 320 series?

 

[ShintaiDK]

I apologize on the .1%, I was unaware of that article.

Having said that, you didn't address my comments about the intel Sandforce drive reliability. Can we at least agree that 2 months wasn't long enough to have any meaningful impact upon intel's return rates, and that the likely culprit for the april 2012 report was the 320 series?

Even Intel employees internally hates the SF based drives. That says something. And no, you cant get a link for that.

The 8MB bug you refer to had very little impact.

 

[jwilliams4200]

Both of you are wrong. If you actually read the hardware.fr return rates article (or the English translation at behardware.com), you will see that the most recent data they have does not cover any SSDs sold after 2011 Oct 1. Therefore it has NOTHING to say about the Intel Sandforce SSDs (and won't, until about May 2013).

Also, the rise in Intel's return rate in their most recent data is probably due to the 8MB bug in the Intel 320. They even mention that in their article.

The returns rates given here concern the products sold between April 1st and October 1st 2011 for returns made before April 2012, namely after between 6 months and a year of use.

 

[n0x1ous]

520's have been absolutely flawless for me in multiple builds.

 

[Coup27]

Even Intel employees internally hates the SF based drives. That says something. And no, you cant get a link for that.

How have you received this information?

 

[bryanW1995]

Both of you are wrong. If you actually read the hardware.fr return rates article (or the English translation at behardware.com), you will see that the most recent data they have does not cover any SSDs sold after 2011 Oct 1. Therefore it has NOTHING to say about the Intel Sandforce SSDs (and won't, until about May 2013).

Also, the rise in Intel's return rate in their most recent data is probably due to the 8MB bug in the Intel 320. They even mention that in their article.

I mentioned that it was likely due to the 320 series' higher returns.

Even Intel employees internally hates the SF based drives. That says something. And no, you cant get a link for that.

The 8MB bug you refer to had very little impact.

How have you received this information?

I'm willing to concede that SOME, possibly even most, intel employees internally hate the SF based drives. They are likely far more complicated than the old intel controller. They are also likely much harder to validate due to the increased complexity. That's probably why they took a year longer than OCZ and others to release the sf-2281 drives. However, Intel employees' dislike of sandforce doesn't necessarily mean that they aren't just as reliable as their older ssd's, it just means that they had to work a lot harder and probably caught a bunch of heat from upstairs about the new ssd's.

 

[lehtv]

How have you received this information?

Just about every thread I see him in, ShintaiDK is posting some apparently inside information, and since he never posts any source, either he's the inside source himself or he's making it up... Which do you think is more likely

 

[Yellowbeard]

How can it be wrong when the site reports it.

If it's on the internet, it must be true.

 

[Emulex]

Any return <30-45 days goes back to reseller.

DOA or buggy = return < 30-45 day.

Or just go meander over to warehousedeals and see who has the most SSD , that's where you returns go when the product works but it is not satisfactory.

Notice the large volume of OCZ? werd.

(oddly other than 2 OCZ DOA's, i've had 3 X25-M failures).

 

[jwilliams4200]

Or just go meander over to warehousedeals and see who has the most SSD , that's where you returns go when the product works but it is not satisfactory.

Notice the large volume of OCZ? werd.

Bad methodology. You need to divide by the number of each make of SSD sold before you compare like that.

 

[bryanW1995]

In the article from AT that I linked, Anand said that he talked to several manufacturers who verified that the data from 3/2011 was accurate. I don't know about the data from other dates, but it seems likely that if their data from 3/2011 was accurate then their more recent data was likely correct, as well.

 

[Emulex]

OCZ makes far less SSD than Intel.

 

[jwilliams4200]

OCZ makes far less SSD than Intel.

How many SSDs did Intel and OCZ sell during a recent 3- or 6-month period?

 

[Emulex]

Let me get numbers on intel oem

 

[bryanW1995]

I don't have sales data on either of them, but at a guess I'd say that OCZ sells a lot more ssd's than intel. Samsung has the apple OEM contract, and most dell/hp/etc computers are bought without ssd (or brands other than intel at least).