Tripredacus

The path is C:\users\%userprofile%\downloads They really did screw up the default Explorer view if you ask me, it always shows everything duplicated by default. On my Explorer, it shows Downloads on the left under Quick Access and This PC.

Not sure about it being worthless, I'm sure that the speed increase is technically there. In many circumstances it come down to the user not being able to tell the difference in launching a program with or without Superfetch on an SSD... or even on any sort of disk. Does Superfetch really make Notepad open faster? How can you tell? What about long-to-load programs like Adobe Illustrator or Photoshop?

No I think the myth of it reflecting BIOS boot order stems from "I saw it once, it must always be this way" phenomena. And heck, it might actually work that way for some specific board and/or BIOS version, but it is not designed to work that way and certainly can't be used as a general fact for all computers everywhere. It was proven through observation that Windows Setup will sometimes write the boot entries to other disks. Microsoft never has written up a why or a how to resolve (not even a simple, disconnect your other disks during install KB) and their representatives on forums fall into the trap of "that is how it works sometimes" when it obviously seems like a bug... one that has been present since Vista. The Disk Selection screen does not have anything to do with where the boot files are written, that is the problem. Disk Selection simply is the UI portion for Setup to manually format disks/create partitions and to create the answer file it uses and set the image target. The bits that deal with running BCDBoot (which I don't think it does directly, rather uses some API calls) after installing the image, prior to the reboot into phase two, is done with some other magic. And even this critical phase of installing Windows is not configurable via an XML file create with WSIM. Alas, this issue has existed for such a long time, and has such a small impact on support, even if it were determined why it does this (by testing a butt ton of configurations that no one has time for) I doubt they would even fix it.

It seems like normal usage to me. Do you have SuperFetch enabled?

WindowsPE uses the same method that Windows OS itself does as far as I'm aware. See here for an explanation from MS: https://support.microsoft.com/en-us/help/937251/disk-drive-numbers-may-not-correspond-to-the-sata-channel-numbers-when

What is the actual memory usage with the hardware mapped memory removed from the total?

Predabot changed to DoomN00b

This is incorrect. If the motherboard fails, you can migrate the array to any other board (same model or different) as long as it has the same storage controller on the board. On the destination board, you would enable the RAID with no disks attached, shut it down. Then connect the disks from the other board to the same ports as the original board, boot the system and enter the RAID manager. Presuming that the disks themselves are sound, at the worst you'd have to do a rebuild. So it may not be a simple solution, but besides being able to read a single disk as jaclaz mentions, it is not a total loss to lose the board. But then again, this brings up the other point, RAID is not a backup solution. You lose the array, its a pain but you would have the backup, right? I will note that I have only done array migrations with a RAID1. I have never found RAID0 to be worth the risk for OS volumes and hasn't been a viable speed option since the advent of SSDs.

Then perhaps it is backwards, and that some RAID software will refer to a single disk outside of an array, while still being managed on the controller, as a JBOD disk, even if singular. Terminology aside, there are still some RAID software on desktop boards that do not allow for a single disk to exist outside of an array and documentation for those boards rarely mention such a thing. Usually only find out if it is possible by attempting it, unfortunately.

Well perhaps, but it is, in effect, what a disk on a RAID controller is when out of the array. With a desktop board, you do not get the luxury of having SATA ports not be on the controller when you change the setting in the BIOS. On modern systems, only M2 sockets get this luxury.

Using generalities, a non-specific desktop motherboard with RAID capability will typically have only one storage controller option for all of the disk connections. As opposed to a workstation or server class board which can have two controllers, assigned to specific or selectable ports. In the general use case of a RAID capable desktop board, it would infer that changing the SATA mode to RAID will change all of the ports to that controller. So the SSD in the first post would need to be added as a JBOD, while a RAID0 or RAID1 is used for the two 3 TB disks.

More: http://walyou.com/worldwide-places-hit-by-the-blue-screen-of-death/

I'm presuming based on your other topic that you have figured this out already, but here is a relevant reply. When installing Windows on a system with multiple disks/arrays, I would recommend that you only have the OS disk/volume present at the time of installation. The reason for this is twofold: 1. Windows Setup will sometimes write boot data to another disk/volume than the one the OS is installed onto. The issue with this is that if the other volume is removed or has a problem, the OS won't boot. 2. Windows PE will assign disk numbers based on first-come first-served enumeration. Because all hardware is different, even between different systems with identical hardware, the system may make available disks in an uneven order. As such, there are situations where a card reader or RAID volume on a controller gets enumerated before the SATA ports, and this can cause an issue in some installation scenarios because Disk 0 ends up being something other than what you would think. I have never seen an ODD get Disk 0 designation, and have yet to see a NAND/NVMe in M2 slot be anything other than Disk 0... Because you cannot control enumeration of disks to Windows, the best practice is to remove the possibility of the incorrect device being detected as Disk 0. When using motherboard RAID in the scenario outlined above, the controller needs to be capable of creating a array with 2 or more member disks and 1 disk set to JBOD. Unfortunately, specs for RAID on motherboards rarely have a matrix of possible combinations documented, and there are definately instances of desktop motherboards being incapable of this configuration. The only way to know if it is capable is to attempt to do it, or find someone who has tried or has the board. For an Intel chipset board with Rapid Storage controller, you set the SATA mode to RAID. Then you would reboot and enter the RAID software and create 1 array using the two disks, and set the other disk as a JBOD member disk that is separate. If it allows this configuration, then you can either disable the array, or disconnect the disks physically, to install the OS. After the OS is installed, re-connect/enable the array disks and boot into the RAID setup to verify the array is healthy. Then boot into Windows, go into Disk Management and you'll be able to format the virtual disk that is detected.

If you have access to hardware (v)RAID, there is no reason to use the software equivalent. You do not mention the RAID level, presuming RAID1 it is possible to boot a degraded disk in another machine, provided that machine has the same RAID controller and you connect it to the same SATA port. However, this is not a scenario that is considered in a pro/con argument regarding RAID. It is because such an action is only considered if there is a motherboard failure. Even so, I would not recommend using original degraded disks on another computer, instead use a copy of said disk. Hot swap in RAID is not for making member disks portable, but to allow for powered replacements. If it is a RAID0, you cannot use a degraded member-disk in any other system. When RAID is enabled in the BIOS, you use the board software to create your array. The board software then presents this as a virtual disk volume, which Windows sees as only 1 device.

You can do it using DISM /set-ScratchSpace, but be aware that your ramdisk size is dependent on the amount of system memory available (and addressable if using 32bit). Example: Dism /image:C:\test\offline /set-ScratchSpace:128

WELL! I just got an email now for a PM. But I had gotten PMs over the past week that I didn't know I had until I got here. So it works sometimes.

Bump. Not receiving email notification for PMs.

If you are having this error, you'll need to run MGADiag.exe and click copy, then paste the report into this thread. Then we can tell you what the problem is. http://go.microsoft.com/fwlink/?LinkId=203456

Interesting video showing the "proper" procedure of doing a hard disk reshell. Apparently a bit old, however may be worthwhile to watch for anyone interested. Note that this is a situation where they change out the shell of a hard drive because the spindle bearing is bad. They are not doing any other things like swapping PCBs. It is not clearly seen but the original PCB from the first drive is what is mounted on the new shell. So this is not a case of just moving the platters from one drive to another. They show some tools at the end also, but none of that is really explained.

Found this, if anyone wants to check it out: Ref: https://twitter.com/mattifestation/status/1063141786779115520

Yes, we do pride ourselves in at least trying to figure out actual solutions, rather than the typical "just reinstall Windows" or "update your antivirus" type posts you see a lot of other places.

RAID is not a backup solution, never equate the two.

I have accumulated all sorts of weird adapters over the years as well. I rarely try to figure them out, just put them in a bin of mystery connectors.

2nd line from the bottom in his post (third if you count the smiley)

A picture would be nice, even if you can find a stock photo of what you have. From your existing post "kind of port that a flash drive could go into" makes me think that one end is a USB Type A receptacle. However, the other side, the RJ45, you don't specify if it is a port or a connector. I'm sure there are various uses for such a device. A quick looksee on Amazon (ASIN: B005DDI51U) for example, says this particular adapter is to connect a USB ADSL modem to a router. If the thing you have is not the same as the thing from Amazon, then I used this phrase in google search: RJ-45 USB Type A adapter