News PNY Quietly Reduces XLR8 CS3030 SSD's Endurance by Almost 80%

[Admin]

Report from German publication ComputerBase suggests that PNY might have revamped its XLR8 CS3030 SSD in secret.

PNY Quietly Reduces XLR8 CS3030 SSD's Endurance by Almost 80% : Read more

 

[King_V]

SSD makers . .

STOP DOING THIS CRAP

Change the freaking model number!!

 

[centauri0]

I just purchased two of the 2TB ones two weeks ago for mining Chia. I wonder how you can tell the difference.

 

[kal326]

Those are looking like QLC level endurance ratings. I can’t see how they would stick with TLC chips, but drop endurance down to that level. Or swap to QLC chips, but not change the models.

 

[jakjawagon]

I just bought one of these for my new laptop about two months ago. I also bought an Adata XPG SX8200 Pro for my dad's computer just before I found out they changed the parts in that. As far as I can tell I got the good version of the CS3030 (apparently write performance is significantly lower on the new ones), but I have no idea what version of the SX8200 I got.

 

[watzupken]

That's why my advise is, just buy SSD from companies that produce their own NAND, and SSD controller if possible. So I will choose the likes for Crucial, Samsung, and to some extent Western Digital. All these brands like PNY, Adata, Kingston, which are infamous of switching out components don't have much control over the supply of components and will at some point change the specs/ parts. This to me is the third strike, so choose your SSD provider wisely if you get a chance to select it.

 

[Mr5oh]

That's why my advise is, just buy SSD from companies that produce their own NAND, and SSD controller if possible. So I will choose the likes for Crucial, Samsung, and to some extent Western Digital. All these brands like PNY, Adata, Kingston, which are infamous of switching out components don't have much control over the supply of components and will at some point change the specs/ parts. This to me is the third strike, so choose your SSD provider wisely if you get a chance to select it.

Worst part is those brands with a bad reputation for pulling this kind of stuff aren't that much cheaper. My data, time, and peace of mind is worth a couple dollars more.

 

[derekullo]

Before chia less than 1% of people would reach the endurance rating on consumer drives.

It appears that PNY lied about the endurance number in order to boost sales figuring the majority would not come close to using it.

Sounds like a lawsuit ...

 

[GoboII]

Those are looking like QLC level endurance ratings. I can’t see how they would stick with TLC chips, but drop endurance down to that level. Or swap to QLC chips, but not change the models.

This is looking pretty likely... On the spreadsheet for the original spec, it listed "3D TLC Flash Memory." On the new one, it just lists "3D Flash Memory." I'd also imagine that the switch to QLC is likely what gave PNY the storage density to release a 4TB model.

 

[waltc3]

It's really hard to understand articles like this. 360/660TBW is enough "endurance" to last decades--certainly more than 20 years for the average user. These aren't server products, etc. 9999 people out of every 10,000 will replace these with newer products many, many years before the theoretical "wear-out time" of these drives...😉 I'm still using a > 3-year-old 960 EVO 250GB NVMe drive with a 75TBW "endurance" estimate (when I bought it), and at present usage I've got at least 10-12 years life left in the drive. The PNY 250GB drive has an endurance estimate of > 2X that of the Samsung 960Evo 350GB NVMe drive--170TBW!

Endurance estimates are almost always a waste of time--look at the warranties--that's the important metric. Say you've got a TBW of 4096, like the 4TB model, but you have a 5 year drive warranty. If you put on 8 TBW per year (which you'd have to work at doing)--well, gosh, the 4,096TBW endurance estimate means jack if you've put on 44TBW by year 6 and it fails--you can kiss that 4096TBW number goodbye--doesn't mean a thing! OOOOOooops--got to buy a new drive! TBW estimates are marketing hooks for n00bs, basically, imo. Notice how the warranties are of much, much shorter duration than the TBW estimates.

 

[Exploding PSU]

Should an everyday user like me be worried though? I've heard that SSD endurance really doesn't matter for regular users, but 170 TBW for the 500GB version sounds rather low. I mean who knowss what my future usage would be

 

[GoboII]

Endurance estimates are almost always a waste of time--look at the warranties--that's the important metric.

I definitely agree with this; if 100TBW weren't enough for a majority of consumers, the Crucial P1 wouldn't have done well commercially. The point here isn't that the new revision CS3030 won't do well for people, it's that a lot of consumers will probably end up with a different product than they expected.

 

[mrv_co]

Should an everyday user like me be worried though? I've heard that SSD endurance really doesn't matter for regular users, but 170 TBW for the 500GB version sounds rather low. I mean who knowss what my future usage would be

No, definitely not... unless you decide to jump on the THG Chia Faarming band-wagon.

 

[Exploding PSU]

No, definitely not... unless you decide to jump on the THG Chia Faarming band-wagon.

Alright then, good to know. My SSD still has 99% health after 3 years worth of usage it says.
Speaking of Chia, I have a friend who dipped his feet into Chia farming with 1300 AUD worth of equipment. I asked him yesterday on how much money he's making after farming for about 1 month, I mean it's the great new thing right? He's probably making bank with that amount of investment. He gleefully said 0. I thought he was joking but it's true, he made literally no money from his investment. I don't know how it works or whatever, he said there was "difficulty problem" and the "luck factor". But a big fat zero after a month doesn't sound compelling to me.

I mean I can put the same amount of money to the corn farm down the road and probably make 200 AUD at the least in actual farming. He did say he's playing the long game though, I mean heck what do I know, I only read the news not actually farming / mining.

 

[Antonio Java]

I don't suppose anyone has figured out how to tell them apart yet? Perhaps certain serial numbers? The change is recent enough that there might still be stocks of the old TLC version around; I'd like to be certain about what I'm getting.

 

[innova3di]

I just bought one of these for my new laptop about two months ago. I also bought an Adata XPG SX8200 Pro for my dad's computer just before I found out they changed the parts in that. As far as I can tell I got the good version of the CS3030 (apparently write performance is significantly lower on the new ones), but I have no idea what version of the SX8200 I got.

Which version of CS3030 you have? Does it have the huge XLR8 logo with an arrow pointing to the right? Does it also have the "X" like CS3030X included in its part number?

 

[jakjawagon]

Which version of CS3030 you have? Does it have the huge XLR8 logo with an arrow pointing to the right? Does it also have the "X" like CS3030X included in its part number?

I'd have to open up my laptop to check the first one. There's no 'X' in the model number listed in device manager.