Model: ADATA - XPG SX8200 Pro Series 2TB PCIe Gen 3 x4 M.2 2280 Internal Solid State Drive with Flash 3D Nand Technology
Deal History
Deal History includes data from multiple reputable stores, such as Best Buy, Target, and Walmart. The lowest price among stores for a given day is selected as the "Sale Price".
Sale Price does not include sale prices at Amazon unless a deal was posted by a community member.
They reduced the endurance rating which is like reducing warranty period. There is no evidence that they changed anything in components or that it had any impact on the actual longevity. The reason is supposed to be that they were getting a lot of warranty claims from chia miners within the early high endurance numbers. They are honoring the warranty for people who bought with the earlier endurance numbers.
Not even close to same as the Adata.
Sounds then like they either inflated their original endurance ratings or changed the % of overprovisioning for endurance. Chia is irrelevant. 3000 TB of written for chia is not worse than 3000 TB of anything else.
I bought one from my local Best Buy and it has the much slower SM2262G controller. I have emailed Adata CS to see if they can RMA this one and ship me one with the SM2262EN controller.
Sounds then like they either inflated their original endurance ratings or changed the % of overprovisioning for endurance. Chia is irrelevant. 3000 TB of written for chia is not worse than 3000 TB of anything else.
You can speculate all you want with no proof and I think that is fine as long as one states it as their conjecture or suspicion rather than as some known fact like the other poster did. That was libelous.
But you are wrong about understanding the difference with mining or what endurance means in terms of failures in the timeline. Like MTBF it is probability based on a usage model not some brickwall people hit with exactly X writes.
Not all SSDs die when they hit endurance specs. It is a distribution curve based on a lot of parameters. Chip variance, operating conditions where it is used, power supply variations/fluctuations/etc. So, just like hard drives, there will be some premature failures and some will last for a lot more writes than endurance specs. The expected failure are very small in the beginning and rise up as you get closer to the endurance limits. In normal usage, the failure rate doesn't climb up much well past the warranty period. Most people would probably replace them before they actually failed for endurance reasons.
To understand how the usage affects failure on the timeline, say probabilistically 5% of the SSDs fail prematurely by the time they hit 50% of endurance and 10% more by the time they hit 70% of endurance. In typical use, say 90% of the SSDs don't hit 50% pf TBW within warranty period. So, expected failures for warranty claim will be 5% of the 10% or .5% of the units. But let us say just 20% of the SSDs are used for storage based mining 24x7 and reach 70% of endurance within warranty period. Probabilistically, 15% of 20% or 3% of all will claim warranty replacements. The warranty claims have gone up 6x! That can be a substantial hit to original calculation of margins on the units. These are just example numbers.
If you cut the endurance spec by 50% and don't change anything in the units, then a substantial number of them amongst the miners will be ineligible for warranty but most normal users will not be affected at all. And the drives won't fail sooner. The same percentage of early failures amongst typical users with up to 50% usage in old numbers will still get warranty service even under reduced TBW in the numbers above because they will still be under 100% endurance in the new numbers. Again, example numbers to illustrate the math.
You get the idea if you read through and understand the above. That move has just eliminated the outlier users.
It is very much like mobile providers throttling the data speed after a certain threshold that typical users won't hit just to manage the cost of the outlier users.
Same thing would happen in mobile networks if some new mining scheme on mobiles suddenly increased the data usage of phones so that a lot more people started using a lot of data. The mobile carriers would need to reduce the throttling threshold, not because they degraded their networks but the cost of outliers went up.
18 Comments
Your comment cannot be blank.
Sign up for a Slickdeals account to remove this ad.
They reduced the endurance rating which is like reducing warranty period. There is no evidence that they changed anything in components or that it had any impact on the actual longevity. The reason is supposed to be that they were getting a lot of warranty claims from chia miners within the early high endurance numbers. They are honoring the warranty for people who bought with the earlier endurance numbers.
Not even close to same as the Adata.
But you are wrong about understanding the difference with mining or what endurance means in terms of failures in the timeline. Like MTBF it is probability based on a usage model not some brickwall people hit with exactly X writes.
Not all SSDs die when they hit endurance specs. It is a distribution curve based on a lot of parameters. Chip variance, operating conditions where it is used, power supply variations/fluctuations/etc. So, just like hard drives, there will be some premature failures and some will last for a lot more writes than endurance specs. The expected failure are very small in the beginning and rise up as you get closer to the endurance limits. In normal usage, the failure rate doesn't climb up much well past the warranty period. Most people would probably replace them before they actually failed for endurance reasons.
To understand how the usage affects failure on the timeline, say probabilistically 5% of the SSDs fail prematurely by the time they hit 50% of endurance and 10% more by the time they hit 70% of endurance. In typical use, say 90% of the SSDs don't hit 50% pf TBW within warranty period. So, expected failures for warranty claim will be 5% of the 10% or .5% of the units. But let us say just 20% of the SSDs are used for storage based mining 24x7 and reach 70% of endurance within warranty period. Probabilistically, 15% of 20% or 3% of all will claim warranty replacements. The warranty claims have gone up 6x! That can be a substantial hit to original calculation of margins on the units. These are just example numbers.
If you cut the endurance spec by 50% and don't change anything in the units, then a substantial number of them amongst the miners will be ineligible for warranty but most normal users will not be affected at all. And the drives won't fail sooner. The same percentage of early failures amongst typical users with up to 50% usage in old numbers will still get warranty service even under reduced TBW in the numbers above because they will still be under 100% endurance in the new numbers. Again, example numbers to illustrate the math.
You get the idea if you read through and understand the above. That move has just eliminated the outlier users.
It is very much like mobile providers throttling the data speed after a certain threshold that typical users won't hit just to manage the cost of the outlier users.
Same thing would happen in mobile networks if some new mining scheme on mobiles suddenly increased the data usage of phones so that a lot more people started using a lot of data. The mobile carriers would need to reduce the throttling threshold, not because they degraded their networks but the cost of outliers went up.