Crucial, Western Digital, and Samsung are cheating us swapping drive components!

[c627627]

Crucial and Western Digital were caught swapping TLC NAND for inferior QLC NAND without updating product SKUs (!) or informing reviewers of this change...
Samsung isn't swapping TLC for QLC. Rather Samsung is swapping the drive controller + TLC for a different, inferior drive controller and different TLC.

Proof:
https://www.extremetech.com/computi...atest-ssd-manufacturer-cheating-its-customers


If any of these drives were reviewed by overclockers, I suggest capital letter bold edits to review titles...


Two different versions of the Samsung 970 Plus exist, for example.
Both drives are labeled with the same sticker declaring them to be 970EVO Plus.

One drive is labeled MZVLB1T0HBLR (older, good) and the other is MZVL21T0HBLU (newer, inferior).
Chips are different. One is Phoenix, the other Elpis.
Phoenix drive is older and faster and the Elpis drive is newer and slower.

Bait and Switch laws in the United States prohibit this. Owners can sue and put to test anyone who thinks this is legal.

 

[Janus67]

Yep, it is absolute garbage that companies are getting away with this and have been. I understand that there are some global supply constraints of various fashions for essentially everything, but that doesn't mean it is OK to swap parts around on existing SKUs for more profit/availability without marking as such.

 

[Woomack]

Do they have to inform reviewers? Reviewers are a tool in their hands that helps in sales, not otherwise. Most websites are publishing fake reviews anyway so who cares.
Also, most products have info on the product website that the specification may change without further notice. It's on many product websites for various products, not only SSD. It's nothing new.
Keep in mind that most SSD have in the specifications only the maximum sequential bandwidth and no matter what NAND is used, they offer that bandwidth.

As far as it's not right from the end-user point of view, then most brands do that for years. If you don't like that your product isn't what you expected then you can return it. If the store where you buy hardware didn't update the specification, description or PN then you have the right to return the product.

I also don't get what Crucial products. They have 3 M.2 SSD and 2 SATA SSD and in the linked article they say Crucial but no word what and how. P2 and P5 SSD are without changes and P5 Plus was just released.

Sorry but I see it as creating a new drama as the hardware market is pretty boring recently. I've seen the same topic on some other websites a while ago.

 

[EarthDog]

If it was actually for profits (and not a byproduct of), I would be pissed. As it stands, it feels more like a supply chain and sustainability issue for parts more so than an conscious effort for profiteering (but that's just my guess - perhaps the supply chain issues are a convenient veil, who knows). I hope I'm right, but wouldn't bet my life on it!

Regardless if the label speeds are reached which admittedly 95% of users only look at that anyway and would be OK with, other drives are showing significant decreases in performance in certain situations. I don't know EXACTLY the performance difference in each drive, but that article shows some differences (oddly, one drive was faster until the 'natural' slowdown of the drive where it was then slower.... but not until after 120GB of data was transferred).

That said, I really don't think it changes much for many (good or bad, that), honestly. I don't think it's right, but, I also won't be suing anyone to test and see either. I'm sure a bunch of litigious Americans will bring a class-action lawsuit against them in time if it's remotely worthy. We'll see.

If any of these drives were reviewed by overclockers, I suggest capital letter bold edits to review titles...

We'll add an editor's note at the top for any drives we've covered. FTR, we haven't reviewed any of these.

 

[tRidiot]

I don't think it's right, but, I also won't be suing anyone to test and see either. I'm sure a bunch of litigious Americans will bring a class-action lawsuit against them in time if it's remotely worthy. We'll see.

There'll be a lawyer who will file a multi-million dollar suit, get a large settlement and make a few million bucks. End users will get a check for $0.39 or a coupon for $5 off a year of the manufacturer's respective "data protection service" or some such nonsense....

 

[EarthDog]

Ha! That's a great point. Let's see if it happens and if so, how it pans out.

 

[tRidiot]

I'll anxiously await my check for pennies.

 

[The_Jizzler]

I'm with Woomack on this one. How's it any different than RAM manafacture's changing chips on DIMM or a motherboard revision? Happens all the time. Maybe not to the degree that the drive manafacture's have gone to, but if the performance is still in spec....

 

[EarthDog]

I'm with Woomack on this one. How's it any different than RAM manafacture's changing chips on DIMM or a motherboard revision? Happens all the time. Maybe not to the degree that the drive manafacture's have gone to, but if the performance is still in spec....

That's what I'm saying though. Performance HAS changed (that article shows one example).

But this goes to show how many(most?) users perceive performance... box specs (which don't seem to change in this case). You've been a member here since before SSDs were a thing, but still fell back on box specs. Box specs are best-case scenarios and overall performance isn't listed on the box. For example, the 'feel' of an SSD in non-transfer type operations is in the 4k R/W ability. If something like that suffers, it sucks. But if it's only affecting the post cache write performance, that really doesn't bother a huge amount of people frequently. That doesn't make it right.

I can honestly see both sides of this. Either way, it shouldn't be happening.

 

[Janus67]

Especially in cases where they change from TLC to QLC, the endurance of those chips is far worse than the previous version as well.

 

[EarthDog]

Will someone summarize what each company did/changed?

 

[c627627]

Source: Slashdot

The WD SN550 Blue -- which is currently one of the best-reviewed budget SSDs on the market -- has undergone a NAND lobotomy. While the new SSD variant performs on-par with the old drive that WD actually sampled for review, once you exhaust the SLC NAND cache, performance craters from 610MB/s to 390MB/s. The new drive offers just 64 percent of the performance of the old drive.

"The 2TB Crucial SSD I purchased for my own video editing work is one of the bait-and-switched units, and it's always had a massive performance problem -- as soon as it empties the SLC cache, it falls to what I'd charitably call hard drive-level performance. Performance can drop as low as 60MB/s via USB3.2 (and ~150MB/s when directly connected via NVMe) and it stays there until the copy task is done. The video upscaling projects I work on regularly generate between 300-500GB of image data per episode, per encode. Achieving ideal results can require weaving the output of 3-5 models together. That means I generate up to 1.5TB of data to create a single episode. God help you if you need to copy that much information to or from one of these broken SSDs. It's not literally as bad as a spinning disk from circa 2003, but it's nowhere near acceptable performance."

 

[tRidiot]

Interesting... I've noticed when copying files from my 500GB 860 EVO SATAIII it will transfer very fast for the first few seconds, maybe up to a gigabyte of data - at 500+MB/sec, then slows down to like 120-150MB/sec. This is going to my platter drive in the same machine. I assumed it was something with the platter drive not being able to keep up the write speed. I dunno. But the behavior sounds more like this.

 

[EarthDog]

Interesting... I've noticed when copying files from my 500GB 860 EVO SATAIII it will transfer very fast for the first few seconds, maybe up to a gigabyte of data - at 500+MB/sec, then slows down to like 120-150MB/sec. This is going to my platter drive in the same machine. I assumed it was something with the platter drive not being able to keep up the write speed. I dunno. But the behavior sounds more like this.

That's just the cache running out. Normal behavior.

 

[tRidiot]

That's just the cache running out. Normal behavior.

That was what I expected. It's not a major problem... I wish I could afford massive storage with SSDs. I'm already running out on this 12TB, time to look for a second one.

 

[Culbrelai]

All it takes is someone wealthy to sue them or a large legal firm getting wind of it. I'm sure there will be a lawsuit if there isn't already.

The crazy thing is this isn't new behavior. About ten? years ago both Kingston and PNY were caught bait and switching their SATA SSDs.

https://www.extremetech.com/extreme...itching-cheaper-components-after-good-reviews

It seems like the only people who weren't caught doing this are Sabrent, SK Hynix and Intel, the latter of whom are exiting or have already exited the SSD business which is a shame, I have one of their 750s AIB SSDs and its been a champ. It has written 41 TB and still has 93% health.

 

[Woomack]

"The 2TB Crucial SSD I purchased for my own video editing work is one of the bait-and-switched units, and it's always had a massive performance problem -- as soon as it empties the SLC cache, it falls to what I'd charitably call hard drive-level performance. Performance can drop as low as 60MB/s via USB3.2 (and ~150MB/s when directly connected via NVMe) and it stays there until the copy task is done. The video upscaling projects I work on regularly generate between 300-500GB of image data per episode, per encode. Achieving ideal results can require weaving the output of 3-5 models together. That means I generate up to 1.5TB of data to create a single episode. God help you if you need to copy that much information to or from one of these broken SSDs. It's not literally as bad as a spinning disk from circa 2003, but it's nowhere near acceptable performance."

A crucial example still says nothing about the SSD model. It's only comment from a single person about some 2TB SSD. I reviewed all of their SSD (literally all as their product list isn't long) and I had no problems and I have never heard that their SSD didn't match the specs. If it's an external SSD then more things may cause a described problem and believe me, if it's X8 SSD then inside is the P1 M.2 SSD, the same as the standard M.2 one, which hasn't changed since the release.
I have seen a topic about Samsung and WD a while ago but it still affects a single product, not all, while this forum thread (already old considering when the issue appeared) suggests that these brands do that all the time in more series.

It looks like no matter what components are inside, it has to meet the official specs and specs are not including bandwidth in every scenario or IOPS. Some series have described IOPS but usually only sequential read and write. Especially cheaper series usually have only sequential read/write and in most cases, it doesn't matter what is inside, it usually meets the bandwidth in the general specification.
I'm not saying it's good that they change components under the same PN but they have the right to do that, as long as the SSD works as specified. If it's not, then the user has a right to return the product, buy something else, and end the problem.

So what, 2-3 models of different brands had that, and we need a lot of noise on the forums right now when it happened some months ago?

There also a factor that someone mentioned earlier. There is a shortage of some components so most brands are using something else to keep production and acceptable prices.

 

[c627627]

It's understandable that in times of parts shortage they may not have a choice but to use something else.
Is it okay to ask that if they are using something else then they should then call it something else, just like the law says?

This is about products that perform [much] worse when tested, no one is complaining because other parts that perform the same are being used, right?
I would also like to know which Crucial model is alleged to have been illegally swapped. I own the X8.

There is nothing wrong with using different parts... They just can't have the same name if they test to perform worse. So a company selling a product in the United Sates is subject to penalties if they break this law:

 

[EarthDog]

As I see it (as well as Woomack), the problem with a lawsuit is that they are still performing as advertised. Few advertise 'extended write performance'. It's high-level things... sequential R/W and Random R/W IOPS. Typically when looking at the box and website, it's a best-case scenario and with an asterisk, they point you to the benchmark used to come up with that value. It's not often you see details of sustained write performance. That said, Samsung actually does. Below is an image from the Samsung 980 Pro page... in quotes is the asterisk/fine print...

¹ Performance may vary depending on the SSD’s firmware version and the system hardware & system configuration. Performance measurements are based on IOmeter 1.1.0. The write performances were measured with Intelligent TurboWrite technology being activated. The sequential write performances after Intelligent TurboWrite region are: up to 500MB/s(250GB), 1,000MB/s(500GB) and 2,000MB/s(1TB and 2TB).

If the performance drops below that as advertised, they have much more firm ground to stand on. I'm not sure about the other drives and what their box/website says about performance. I don't think it's right, but, that's how they can get away with it. They are literally offering exactly what the box and website says. If not, there may be a problem.

I'm surprised I haven't heard of any legitimate attempt at class action lawsuit honestly.

 

[c627627]

The person complaining about Crucial said "That means I generate up to 1.5TB of data to create a single episode." and talks about a significant decrease in performance. Either that's testable to be true or not...

Are you saying that in Samsung's case, the decrease in productivity, though measurable, falls within advertised specs?
And that people buying the originally released version should consider themselves lucky to have gotten a product that performs better than advertised?