How the mighty have fallen. I still remember an ancient 130 MB (yes, MB!) ST3144A drive that worked without problems for at least a decade (might even work now if I plugged it in!).

But now... I won't be buying another Seagate drive any time soon, if ever. Let me explain why (and the reason needs to become more mainstream):

When I got my current 12900K system about two years ago so I could install Windows 11 here and start making Winstep software fully compatible with it, given the good reviews at the time I chose a 2TB Seagate Firecuda 530 NVMe drive as my main system drive (yes, that is a 530R model below but it too suffers from the same exact problem).

At first everything was fine. The drive is Gen4 and was pretty fast.

The first signs something was not quite as it should be came a year or so later when I attempted to copy a 20GB Outlook.pst file to another SSD. The transfer speeds would fluctuate widely, sometimes going *way below* what an SSD should provide. Strangely enough, making a copy of that 20GB file and then moving the copy around between the 530 and other SSDs did not have this problem, and it transferred at full speed.

This was already a clue to what was deeply wrong with the drive, but I did not know it yet. I chucked it to some kind of file fragmentation (in an SSD?!) and did not think much about it, especially because I had already decided to replace the 530 as the main system drive with my 905P Intel Optane that at that point was still in my old system.

And so the 530 became just another drive to store games (theoretically the fastest one).

Some months later I started noticing that loading saves for Horizon Forbidden West (a game I was playing at the time) was taking longer and longer, enough to become annoying. But what eventually made me start searching the net for answers was noticing that my automatic full backups (which are made weekly to external drives) were suddenly taking HOURS longer to complete than they should have.

Using the Performance tab built into the Windows Task Manager I managed to pin point the problem to the 530 drive. Whenever the backup software started copying data from it, through-output speeds would sometimes go down to as little as 10-20 MB/s for extended periods of time (not kidding you, in a drive that advertises 7000 MB/s read speeds).

This is when I went to Reddit looking for answers. And boy did I get answers!!!

So, to cut a long story short: the problem is the Phison E18 controller.

ALL drives with this controller will eventually experience extreme read degradation to the point were they eventually fail. This has nothing to do with TRIM but with data on the NAND cells becoming "stale". The longer those NAND cells go without being refreshed (as in actually being written to), the longer the controller takes to retrieve the data stored in them (probably because it has to resort to a ton of error correction, etc...), slowing the drive down to a crawl.

If the drive is your boot drive, you will start experiencing extreme boot delays, general slowness, BSODs, and eventually the drive will fail to boot entirely.

The only solution is a complete reformat, but this, just like RMAing the drive and getting a new one, is only a TEMPORARY solution. As the data in the SSD becomes stale, the same thing will happen again.

So, if this is a problem affecting the Phison E18 controller and there are other non-Seagate drives using the same controller, then why am I singling out Seagate?

Because Phison released updated firmware that fixes this issue back in July 2024. Kingston released new firmware based on this update for their E18 based drives shortly thereafter. Seagate? Well, Seagate doesn't give a fsk (pardon my French).

They know what the problem is FOR NEARLY A YEAR, there are rumors that beta firmware based on the latest Phison update already exists, but Seagate simply won't release the thing. When contacted they might even RMA you a new drive, knowing full well that this is only delaying the inevitable. Eventually this *will* lead to data loss and Seagate simply doesn't seem to care.

Tech Youtubers unfortunately don't look very interested in this issue or on putting pressure on Seagate to come up with a solution. At this point Seagate is recklessly playing with customer data - and since this is their stance to what is now and has been for months a known problem, then I don't think I should trust them or their drives again.

More information on the issue:

Anyone else who had a Seagate Firecuda 530 2TB NVMe failed/died?

Seagate NEEDS to release Firecuda 530 firmware update NOW

Another month of Firecuda 530 read degradation waiting for Seagate to release the fix from Phison

Plenty of other links in those Reddit posts, and you can just Google with

seagate 530 phison e18 firmware site:www.reddit.com

as the search term for more.