Contribute
Register

General NVMe Drive Problems (Fatal)

For anyone interested:

One interesting point..... ALL the good drives are available in H.K suppliers even I bought one a few days ago
it is STILL using the original build components..
Seems they are only farming the crap out to Western countries....

For part of my university course I spent a year of in depth research on Flash controllers & SSD, as part of a paper on forensic "subversion".
There was all sorts of things going on... hidden Flash chip commands, hidden controller commands
Highly confidential Datasheets about "Binning" and sub grading of chips...
This was about the time Apple bought out one of the best companies researching Flash chips; an israel company...
What I learned put me of Flash technology for a decade...
 
Last edited:
@c-o-pr
my NVME died at upgrading stage to Monterey with my laptop Dell XPS 9570
Toshiba XG5 1TB NVMe, known as kxg50znv1t02 by Kioxia
 
@c-o-pr
my NVME died at upgrading stage to Monterey with my laptop Dell XPS 9570
Toshiba XG5 1TB NVMe, known as kxg50znv1t02 by Kioxia
Hey hey, sry to hear this.

Per observations of @hardcorefs above, and my own experience, it seems that flash storage, especially NVMe, will be an ongoing pitfall. We have to assume Apple has its own measures and designs with only tangential regard for 'compatibility'. This would be a big concern except sun is setting on Apple as a PC.

OTOH, there's nothing much different about this stage of evolution reliability compared to other previous eras: SSD technology now works well and allows big performance gains over disks, which took decades to perfect and suffered routine epic failure. I've always been impressed at how accepting people are of catastrophic PC failure. They usually blame themselves and take a perverse joy in the experience, as if it's amazing. So there's nothing to complain about re SSD reliability in the big picture.

But with Apple's guidance, we are facing a big in the idea expoused by PC industry — and pursued by hackintosh hobbyists everywhere — that a reliable personal computer can be approached as a constellation of interchangeable parts. This idea is patently false. But who cares? The industry happens to not care at all because all the epic failure creates further demand which keeps production and incremental advances in motion. Failure (death) is good for you!
Today I will blather suggest the as the industrial revolution incrementally advanced over several centuries to produce the utter mayhem we call urban life, and a looming ecological crisis of our own fabrication, that as we move on, the personal computer may be causing a looming crisis of the mind. I will jump ahead a mile or so and observe that our inability to confront the nature of purpose of design is our greatest hazard, and therefore the greatest opportunity of sentience.

I have been surprised over my life too see Apple create a popular public engagement with computers as marvelous new creative gizmos, build an new empire based on style, lose it by leaning too far towards the taste fetish of the soft drink industry, fall to the point of near collapse in a roiling market of competition, then bounce back to supremacy at a level of prosperity that exceeds all previous human expectations. But what's weirdest of all is the societal region of Apple's headquarters, and the state of societies in general, are more fraught than ever. Whereas the long history of human calamity has traditionally been about a world of "natural forces", we now enter a new world where calamity is designed and apportioned, thanks to personal computers. It's at once wrecking us and saving us from its wreckage.

The hackintosh user is a peculiar creature in this new world, in that we prefer a coherency of design we can only get through tight central planning, but we want to feel more involved than the central planners prefer, even if it means far less reliable devices.

There's maybe a vague argument that the hackintosher seeks greater usefulness of his device, but to me not worth noting the details. We assemble PC parts-bin systems and boot a proprietary OS that is designed in every way to the intention to curtailing our freedom, because we feel it makes us freer!

Rather than being contradictory, I think there's some essential lesson in this that smashes the value of common sense. The gap between insanity and genius is eternally a strange space.

So @ww7, did your drive die of natural causes?
Or is it one on a list of suspects that confirms a prejudice?
Or is it known to work but surprised you by dying anyway?
Did it catch on fire?

Clearly, everybody who has a computer has computer problems, which, if this site is an indication, are not only incredibly common, but also fun!

People always want to tell me about their computer problems and I always think "Did it catch on fire? No? Then you don't really have a problem. You have a normal computer."

Best of luck on future purchases.
 
Computers fail because they are generally not treated with respect ,particularly with respect to the Chemistry & physics.
I've generally had >60,000~70,000 hours out of disk drives before their first failure. others have had less.

I found that for some reason on the latest security update (i'd been running 12.3.0 happily)
that my "RUST" game now stops every 10 minutes and system performance collapses for 30 seconds(120fps->~5fps), then it starts again.

So I spent the weekend hacking my case to add in a spinning rust. (its a long story)
move the game & its cache to the disk drive and now it does not do it.

So I suspect that Apple is possibly "tuning" something in the SSD algos to gain even more performance after getting kicked in the performance nuts by the new intels.
This is neither Chemistry or physics, but Apple being Apple,
Maybe this weekend, Ill go out and get another 64gb ram ,and copy the game to a ramdisk to see what the fps is and if it still does it.
 
So @ww7, did your drive die of natural causes?
Or is it one on a list of suspects that confirms a prejudice?
Or is it known to work but surprised you by dying anyway?
Did it catch on fire?
The first boot of Monterey hung for a long time, I reset and let it boot again in verbose mode, which also caused it to hang. The third time, the NVME stopped being detected and is no longer recognized in the notebook or on the PC.
 
The first boot of Monterey hung for a long time, I reset and let it boot again in verbose mode, which also caused it to hang. The third time, the NVME stopped being detected and is no longer recognized in the notebook or on the PC.
Ah, yes, bad

What model drive was it?

If it's a model that's known to work with mac, have you considered static safety of your work area? This usually is not a big deal, but in some very dry environments it's a real hazard.

In my case, macOS ate my Sabrent drives where I think something about the way they are driven irritates an untended edge-case in the controller design, maybe involving power mgmt. My Sabrents were new PCI gen4 models on a new mainboard series, so many places things could be wrong.

My limited personal experience, plus forum consensus, is that WD drives are best bet for hackintosh right now.

I have found Crucial SATA SSD to be solid over 10 or more devices over years and years so I'd give their NVMe a shot if price warranted but idk.

Good luck
 
Last edited:
What model drive was it?
As I say in previous post, I had native NVMEe shipped with laptop Dell XPS 9570
Died: Toshiba XG5 1TB NVMe, known as kxg50znv1t02 by Kioxia

I bought Samsung 980 Pro, hope for fixes for it
 
I have a 980 Pro which developed hard errors a month after installing, but I only found out because Apple APFS Replicator failed when I was making a bootable copy using CCC.

I then ran CCC normal backup (no system files) and it's copier told some files had read errors. I checked and they were corrupted. I recovered them from another backup. If they had been key system files then the build would have become unstable and very tricky to diagnose, so I was lucky there.

Samsung Magician (Windows) reported about 80 hard errors via SMART so asked Samsung to replace, which they did.

The new drive worked well for several months and I moved to Monterey which was working well. Then Samsung Magician offered another firmware update (there has been a previous) which I installed. From there Monterey boot began stalling at Trim for that drive! Since then I've migrated back to a WD SN750 which has always worked great and is as fast as the 980 Pro (10900 is PCI3).

Web customer (e.g. Amazon, Newegg) reviews rave about WD SN 850 as top performer across the board, and that is where I would go for latest builds, such as z690.
 
Top Google for Toshiba XG5 1TB returns a TweakTown review with a detailed tech description:

//A close-in view of the drive's second-generation Toshiba TC58 NVMe controller manufactured by TSMC. We speculate the controller is an 8-channel design with between three and five CPU cores.//

I don't recall seeing Toshiba mentioned in Drive Hall of Fame nor the Anti-Hackintosh Buyer's Guide.

When I built with a new Rocket 4 that died after a week, I found a blog review from a justifiably irritated Sabrent NVMe user who lost all his data. Nonetheless I might have assumed my case was bad luck, except that when a new replacement Rocket was installed, it died even faster than the first one! That was under Big Sur.

As I had some experience getting to know the first Rocket 4, which impressed me because of the rate at which it could move files (first NVMe) and concerns for thermals, I was watching it's temp with istat menus graph, and temp would soar under certain loads. For example, if I reset Spotlight to re-index, and ram large file reads at the same time, temps would surge up to 90C (open air) then the drive would quit. The Asus Hero M_2 NVMe slot heat spreader appears not designed to carry the heat that drive could generate as it would still surge to 70s. When the Rockets died, they would go into a thermal runaway condition where a Sabrent toolless enclosure would get so hot I couldn't touch it, but the drive was not functioning.

WD SN750, 980 Pro and 970 Plus never go beyond about 55C open air.

Sabrent started offering these huge add-on heat-sinks that look completely insane, for $30 and after much interaction with their tech support I came away feeling like figured they don't understand their they don't understand their own products and they target gamers, who are conditioned to expect their kit to be unreliable, so they know they're on the bleeding edge, it's part of the gamer mystique and allure.

Unfortunately, due to secrecy of industry, we will probably never know why some drives die under Mac and others work. In true high-tech libertarian fashion, the devil takes the hindmost and every man for himself. As someone else wrote here in so many words: "That failure of a standard PC module to operate in a standard slot completely contradicts the point of PC compatibility!"—(like, like, omg, like)

I hear GE is getting ready to offer portable nuclear reactors!
 
The first boot of Monterey hung for a long time, I reset and let it boot again in verbose mode, which also caused it to hang. The third time, the NVME stopped being detected and is no longer recognized in the notebook or on the PC.

I have a Toshiba XG3 THNSN5512GPU7 that dies after hard shutdowns (e.g. powercuts or hold power button) despite being on Firmware 57DA4105 which was supposed to fix this problem: "- Fixed issue encountered in some system where drive is unable to detect after dirty power down".

To revive it I put it in a NVME to USB caddy and power on and off for various durations, e.g. 10mins on for 30 seconds off. I got the idea because old SATA SSDs can be revived in same way. So you could try that with your XG5.
 
Back
Top