Phison Posts Latest Update on SSD Controller Stability

[-]

imaginary_num6er@reddit (OP)

In response, Phison dedicated over 4,500 cumulative testing hours to the drives reported as potentially impacted and conducted more than 2,200 test cycles. We were unable to reproduce the reported issue, and no partners or customers have reported that the issue affected their drives at this time.

[-]

Strazdas1@reddit

so that single social media post was just that, a single social media post and not an actual issue.

[-]

Nickoplier@reddit

Probably is an actual issue.. JayzTwoCents video today shows a huge issue regarding this.

[-]

Strazdas1@reddit

Ive blocked Jayz for his clickbait. What did he find? He would be the only reciever that actually found anything as far as i know.

[-]

Would be that trying to run heavy intensive games, an SSD with a different version controller than others are reporting with issues is encountering the same issue, not just transferring large files, playing a high quality game could cause the SSD to go unresponsive because Windows changed something. Uninstalling the update also doesn't undo this update because apparently there's feature and security update for this problem kb and you can't uninstall the feature that this issue is. And that certain motherboards do not reboot or power cycle the SSD on restart, so it doesn't turn off and back on to unfreeze the SSD, so the BIOS tries to boot up with the SSD still completely unresponsive and just drops to the bios menu with no detected SSD.

[-]

Strazdas1@reddit

So he didnt replicate the problem, he just found something different, that works differently, while doing a different test. Sound to me like the problem still hasnt been replicated, then.

[-]

Alternative_Fan_6286@reddit

what SSD do you use ? are you on windows 11 ?

[-]

Impossible_Jump_754@reddit

I'm shocked!

[-]

TenshiBR@reddit

idk... their have an invested interest to say "we audited ourselfs and found nothing wrong"

still fishy

[-]

III-V@reddit

They said their partners haven't had problems either

[-]

Strazdas1@reddit

in fact noone has been able to replicate the issue. That poster seems to stand alone in the failure.

[-]

TenshiBR@reddit

As a buyer of their products, I hope so.

[-]

hieronymous-cowherd@reddit

From that page's comments section, a Phison Rep also adds:

Phison ran every drive listed in the initial report on X in the exact same workload and cadence. This goes all the way down to the Japanese version of the game! We were unable to reproduce the issue, and to date, we are not aware of a professional test lab that has been able to do so (several partners and other companies have investigated this claim).

By "initial report on X" he means the post on Twitter by @Necoru_cat who reported their tests on 21 drives.

By "game" he is referencing the game Honkai: Star Rail, which was a 50 GB binary @Necoru_cat used for bulk i/o.

Elsewhere, Bleeping Computer reported on the 21st that Microsoft said:

We are actively working with our storage device partners to try to reproduce the issue. At the time of this publication, neither internal testing nor telemetry have identified an increase in disk failure or file corruption

[-]

Albatross_Dazzling@reddit

Personalmente me pasó,ya el administrador de discos no lo reconoce,ni la bios. Podria ser una coincidencia y ser un problema de hardware pero es mucha coincidencia. Tengo un SAMSUNG MZAL4512HBLU-00BL2 no tiene dram

[-]

cjj19970505@reddit

Though the issue is still being investigated, but from the way the issue was reproduced (The usage that triggers the issue is not that rare IMO) according to the original japanese poster (who did a good job reviewing this issue), it should be affecting much more systems during public preview phase, and should be very quickly picked up by some enterprises and then MS withdrawed the updates. For now, I am working for a an F500 company, my PCs (managed by IT) still have this update installed. MS might not care about average consumer, but it will certainly act if the issue hits F500 company.

I wonder if there are any other individual who've done an complete invertigation with multiple SSDs using a different PC setup (as the orignal japanese poster use only his own setup to test multiple SSDs if I remember correctly.)

[-]

Angelworks42@reddit

We programmatically rolled out patches to around 6800 Dell clients this month using Configmgr and there's easily 20+ makes and models of hdd (mostly nvme but still some sata ssd's) in our fleet and haven't noticed any problems. Not a fortune 500 - just a university.

I know it's not the same kind of testing this guy did but I'm sure the helpdesk call volume would have gone up if people's disks or data were being corrupted or damaged.

[-]

UsernameAvaylable@reddit

according to the original japanese poster (who did a good job reviewing this issue)

Did they? I mean, is there any verification that it was not just a creative writing experience?

[-]

Strazdas1@reddit

if it worked as simply as stated in original posts we would have millions of failures being reported because moving a large file to another drive is a very common use case.

There were many who tried to replicate issue but i havent seen a single success.

[-]

FinBenton@reddit

If tech tubers were instantly able to replicate this, how is it possible that they couldn't?

[-]

ElectricalTip2318@reddit

4500 hours? 187 days? Mine got corrupted I had to reformat the SSD, is not briked but all data got corrupted. So who's hiding the numbers?

[-]

soulless_ape@reddit

Im surprised it got this big just by a post on Twitter. If we had an issue using Phison's controllers, our engineering team would go back and forth with their guys debugging the errors and testing new settings, and even patching the firmware if needed. Issues get solved relatively fast.

[-]

throwaway9gk0k4k569@reddit

This is some classic media narrative framing.

Why is the focus here on Phison?

This is Microsoft Windows problem.

These drives are not destroying data in Linux. This wasn't a problem before Microsoft fucked it up. The hardware isn't failing.

[-]

UsernameAvaylable@reddit

If the drive gets read / write commands and dies from it, its a drive problem, not a problem of the software those IO requests come from.

And do we know they really do destroy data anywhere? Are there 3rd party verifications?

[-]

gondezee@reddit

I work as a hardware engineer at a tier one HW company. I’m hesitant to use Phison on my builds at home.

[-]

bizude@reddit

Why are you hesitant?

[-]

gondezee@reddit

lol at the downvotes.

I’m not gonna break confidentiality and get into the specifics. Generically: we qualify parts to use in our designs. Some work no problem, some need tweaking to the part itself or firmware running on it to get working well across a variety of conditions. Sometimes that process goes poorly and leaves a bad taste in your mouth. It could have been the specific FAE, isolated product teams or section of their portfolio, not sure. I choose to take my personal business elsewhere though.

[-]

Reactor-Licker@reddit

“I’m not going to break confidentiality”

But you brought it up in first place…

[-]

gondezee@reddit

There’s a difference between identifying and detailing specific devices and issues had, the circumstances and application as installed, any interactions we may have had with a vendor… and saying I’ve used a vendor professionally and wouldn’t use them personally.

[-]

Firefox72@reddit

If moving 50gb of data was all it took for this to happen then this would be a massive issue that would blow up overnight.

[-]

Nicholas-Steel@reddit

Yeah, it would've have fairly quickly spiraled in to impacting tons of people like the issue that downed airports and several other places not too long ago.

[-]

kindaMisty@reddit

…right. I guess people who had their drives bricked from this update must have been dreaming. This isn’t very reassuring. Microsoft doesn’t have an answer for this, Phison doesnt either. Guess we just brush this one under the rug?

[-]

Impossible_Jump_754@reddit

You have to realize a lot of people go on the internet and lie.

[-]

S_A_N_D_@reddit

I'm willing to bet that many of them aren't even lying. Rather they're just misattributing the failure to the update.

Just by sheer numbers, there are likely to be a number of people who's drive failed right after the update though the cause was independent of the update and the timing was pure coincidence.

[-]

Strazdas1@reddit

Very likely. Ive seen people misatribute RAM instability issues to anything from software devs being lazy to PSU power instability. People misattribute causes all the time.

[-]

0xdeadbeef64@reddit

You have to realize a lot of people go on the internet and lie.

Sadly, this is true but lying means it's a knowing act of deceiving. I think more than a few just have preconceived opinions resistant to facts and critical thinking worthy of r/confidentlyincorrect/

[-]

Strazdas1@reddit

There are no people who had their drives bricked from this update. It was just FUD.

[-]

Flimsy_Swordfish_415@reddit

hardware breaks all the time, windows receives updates all the time. figure it out