Many thanks for posting this article...I especially appreciate (and agree with you) when you say: "Apparently it is downright impossible for Microsoft to release a Windows update, and not get a bunch of end-users running into some kind of trouble."

Intel needs to reboot Optane with AI accelerators plus DRAM, TLC instead of QLC and decompression units. SSDs will get so important in the future because of the next gen consoles, and here Intel could reach a big market. A true next generation SSD.

I updated Windows recently. I have a dedicated Optane 900p boot drive, and a dedicated Optane 900p acceleration drive for my data HDD. Pinning is not supported on data drives, so I didn't encounter this error. Optane acceleration otherwise works fine after updating Windows.

dampflokfreund:

Intel needs to reboot Optane with AI accelerators plus DRAM, TLC instead of QLC and decompression units. SSDs will get so important in the future because of the next gen consoles, and here Intel could reach a big market. A true next generation SSD.

I thought Optane are Xpoint based, not in the MLC-->QLC NAND paradigm. Although the endurance seem to be similar to the Samsung top drive, and performance can be worse in places vs NAND. https://3dnews.ru/assets/external/illustrations/2016/09/01/938764/endurance-final.png

When Microsoft QA team is almost non existent for cost cutting and all their software are tested by normal people like us in their Windows Insider program quality issues like these are bound to happen. Without professional software testing teams how can a company release quality software products ? People should stop participating in Windows Insider program for free. Without people participation, Microsoft will be under more pressure to hire professional software testing people and will improve their software quality.

This is no accident, us end users are now the crash test dummys and M$ uses us for all kinds of untested experiments, you can only turn this crap off in enterprise versions.

Had an Optane H10 in a ROG laptop purchased back in mid-2019. Noticed many different goofy system hangs of one type or another from day one. Having used high quality SSD's and not these Intel 'Rapid' concept products prior, the glitches were easy to spot in everyday use. Sometime after the New Year kicked in, Intel started to notify users on their website that the previous Optane software that works in conjunction with Rapid Storage would be going 'end of life' this summer. In the mean time, the newly rolled-out Intel Optane UWP from the Microsoft app Store remains in beta and is utter garbage. Suffice it to say, I yanked the Optane P O S and popped in a 960 Pro. All is good... and luckily, I did this bit of surgery prior to May 2020 release. Consumer Optane is total garbage IMHO.

Looks like I'm fine with my Intel Optane 900p 😀 4k random read @ QD=1 works pretty well 🙂

Um.... theres nothing even to this news. Windows service updates remove dll's and service binaries from system32 and have done since launch, just reinstall the software and carry on.

Another user and I experienced audio issues after updating to v2004. I went back to v1909 and everything's running smoothly again. More info: https://www.reddit.com/r/Windows10/comments/gxdz2f/psa_may_2020_update_v2004_audio_popsclicks/

There is not a day when I'm not seeing an article about how windows updates ruined a pc through some driver or user acting dumb. Meanwhile I am reading this on my pc updated with Windows 2004 chugging along smoothly since last week. Reinstall the optane driver

toyo:

I thought Optane are Xpoint based, not in the MLC-->QLC NAND paradigm. Although the endurance seem to be similar to the Samsung top drive, and performance can be worse in places vs NAND. https://3dnews.ru/assets/external/illustrations/2016/09/01/938764/endurance-final.png

Yes, but only the cache. The SSD portion is QLC.

As much as I like to hate on Microsoft when they release buggy stuff, their hardware and software partners have a responsibility to ensure that their own products work. Intel had ages to test their products with this release. If neither Intel or Microsoft caught this, I can't imagine that it's wide spread. Either that, or both companies are incompetent.

Thunk_It:

Many thanks for posting this article...I especially appreciate (and agree with you) when you say: "Apparently it is downright impossible for Microsoft to release a Windows update, and not get a bunch of end-users running into some kind of trouble."

There are not that commonly used in desktops but intel did bundle them with CPUs. For those users it may be unpleasant, but they can usually deal with issues. Then there are many laptops with optane, and there, users may be lost.

I maintain several system that use Optane in one capacity or another, 3 of them are my personal systems, all of them on 2004. The Optane pinning problem best I can tell is an Intel problem and I can absolutely confirm that it is not limited to 2004. I saw the identical pinning error when migrating from 1809 to 1903/1909. If it is happening again when migrating to 2004 then Intel still has not bothered to address the issue. Here is a July 2019 thread where a user has the identical problem: https://www.dell.com/community/XPS/Intel-Optane-Memory-Pinning-error/td-p/7341264 if you search around you can see many support threads from 2019 with this well known bug. My personal systems that use Optane are a 4X VROC 0 905P X299 workstation, an MSI P75 Creator that uses a combination of a SATA SSD and an 800P Optane drive and a NUC8i7HVK with 2 800P in RAID 0. The Optane pinning bug happened when I migrated to 2004 but I was expecting that since I had no faith whatsoever that Intel would have fixed this old bug. Since the cancellation of the 805P and M15 drives Intel's intentions with support has been pretty clear. As far as bugs with Optane excluding the pinning bug go I have seen literally nothing. All 3 of my systems are fully functional and the 6 or so that I support have had no issue other than the pinning bug. As far as Optane goes, they came to market too soon with it and used it in the wrong way. They should have come to market with a product that had competitive sequential speed (because that actually sells products even if this metric usually does not matter much) while still crushing the competition with its random IO performance. The price and form factor was also a huge issue. The U.2, PCIe and 22110 M.2 form factors are all outside of what typical users build with so they were never considered by most users on from factor alone. The 800P drives (awesome OS drives BTW) were wwwaaayyy too expensive for their capacity. The H10 drives are also terrible in design. Instead of a controller that manages the cache on the actual drive, you need a board that can bifurcate the port and the caching is offloaded to the CPU. The H10 drives are more like a riser with 2 SSDs on it than they are an actual single drive. Had they put the actual caching processing on the H10 drives themselves they would have been stupid expensive so there was no way to win with this design. There was a brief period of time where the price of a 58GB 800P + a 2TB SATA SSD was cheaper than a high end 2TB M.2 NVMe SSD while also being faster. IMO this was literally the only use case Optane has ever had where I felt comfortable recommending it which is hilarious because this setup is not officially supported by Intel and when you ask them about it, they claim that it does not work (they are full of crap). You can read about Intel's official stance in this thread where I brought up how stupid it is that they do not support the best use case for Optane: https://forums.intel.com/s/question/0D50P000049rf12SAA/what-happend-if-use-intel-optane-memoryssd?language=en_US

I also had problems with this on a Dell laptop. Reinstalling optane from add or remove programs fixed this. Found some old thread about this and and I am not sure if this is a Intel or MS problem. Still, could be worse. Could be Apple where features are removed from OS and you have to be scared about every update. Friend does software development for both Windows and OSX. I dont think that he has ever said anything positive about Apple.

nosirrahx:

The Optane pinning problem best I can tell is an Intel problem and I can absolutely confirm that it is not limited to 2004. I saw the identical pinning error when migrating from 1809 to 1903/1909. If it is happening again when migrating to 2004 then Intel still has not bothered to address the issue. Here is a July 2019 thread where a user has the identical problem:

Intel could address the problem by keeping all driver dll's in the DriverStore like GPU vendors have been doing for some time. But its really just a matter of reinstalling the software XD

These all to frequent problems always perplex me. The updates aren't secret - new builds are released to the Windows Insider / beta rings continually. The RTM version is out for weeks before public availability. Not one user with an Optane drive had this issue prior to release? Not one MS developer? Intel didn't even test with their own product along the Windows Insider course? Who is really at fault here if no user, MS, or Intel saw the issue (or even cared enough about a fringe product to test)?

SSD_PRO:

Not one user with an Optane drive had this issue prior to release? Not one MS developer? Intel didn't even test with their own product along the Windows Insider course? Who is really at fault here if no user, MS, or Intel saw the issue (or even cared enough about a fringe product to test)?

They knew, it has been being reported for about a year. Intel and MS point the finger at each other and then do nothing since only a small number of users have Optane products.

There are many issues with 2004. I experienced BSODs on my X570 build doing nothing but sitting on the desktop wallpaper. Yes, I did a clean install, even then it gave me BSODs. Reference (Dutch): https://answers.microsoft.com/nl-nl/windows/forum/windows_10-other_settings/build-2004-mei-update-veroorzaakt-bsods/77b1e8c3-c7f7-4c0d-b93e-019bcad45cf2?tm=1591699295392 Reference 2: https://docs.microsoft.com/en-us/windows/release-information/status-windows-10-2004