Originally Posted by kevs
PS Diagnostic, don't think works, tried it with tech few months, and just now, shut down, start with D key, and then says choose network, ,something about recovery, which does not come up, So don't think Diagnostic works...

It is asking about the network because it wants to logon to the network to download the test software.

Originally Posted by artie505
f I've got this straight, in previous discussions of DriveDx and its ilk you've mentioned that they monitor only those aspects of a drive as are specified by the manufacturer, so, is it possible that kevs's particular problem would not have been detected in a different brand of drive? And in a different vein, is it even possible for DriveDx to detect a problem that isn't specific to the drive itself, as opposed to the other possibilities you've mentioned?

DriveDX etc. can only monitor the S.M.A.R.T. attributes reported by the drive. S.M.A.R.T. attributes are defined in the AT (as is SATA or PATA) interface standards and are the culmination of multiple different manufacturer's systems for monitoring drive health. You can find a complete listing of all the defined S.M.A.R.T. attributes in this Wikipedia article. No one reports all the attributes, and some attributes have never been reported, but there is a set of the attributes that are reported by virtually all desktop and laptop drives. The specific attributes reported by a drive and, perhaps more critically, the pass fail tolerance for that attribute are determined by the engineers that designed the drive.

A few of the attributes recorded by a drive that reflect on its longevity may reflect factors outside the drive itself. For Example, overheating is often the result of external factors such as inadequate ventilation or environmental temperature. CRC, similarly, indicates a problem, but like overheating the source of that problem may be in the drive itself or external. In the specific case of an internal boot drive, it is an indication of a problem that needs to be corrected before it deteriorates further, and the drive becomes unusable. Keys recent experience with both the internal and external drives (CRC errors and overheating respectively) are both examples of S.M.A.R.T. attributes detected and reported by the drive indicating a problem that may or may not be external to the drive itself but could result in the drive becoming unusable. In military systems, I have seen AI based problem analysis systems capable of determining the likelihood of the specific cause of each of those errors, but the computer resources necessary to support that level of problem detection is significant and the software and database to support it even more significant. Maybe someday that will be cost-effective on personal computers, but we are not anywhere near that -- yet.

SUMMARY

While those S.M.A.R.T. or NVMe attributes generally accepted as indicators of drive health and longevity are reported by virtually all drives, it is within the realm of possibility for one drive to report a condition to be analyzed by DriveDX that isn't reported by another drive (even a drive from the same manufacturer, but a different design team). What is more likely to effect the results is not what attributes are reported, but pass/fail limits of that value imposed by the designers and is generally conceded to be the greatest weakness of the S.M.A.R.T. concept. So, what may be a failing attribute score on one drive, could be passing on another, although DriveDX appears to compensate for this to an extent, by permitting the user to set their own reporting levels. Whether there is a possibility of a condition arising from factors external to the drive mechanism itself depends on the particular attribute.

If we knew what it was we were doing, it wouldn't be called research, would it?

— Albert Einstein