How to create Windows 8 / Windows Server 2012 bootable USB media for deployment on UEFI based systems (by Paul Marquardt)
This article describes a process to create a bootable USB drive which can be used to install the operating system on a UEFI bootable system. Have you ever noticed how long it takes to load the operating system when using a DVD? Transferring your files over to a USB drive makes loading the operating system faster and gives you the ability to load the operating system on devices like tablets that don’t support optical media.
Important: The FAT32 file system only allows file sizes up to 4GB. If you are using a multilingual image, or customized image that has an install.wim file over 4GB in size, you will require the use of 2 partitions, the first being FAT32, and the second being NTFS which will allow you to place your image on the NTFS partition. UEFI requires FAT32 as a boot partition type, while BIOS systems can use FAT, FAT32, exFAT, and NTFS boot partition types. To create more than one partition on a removable device you will need to use a 2012 operating system since Windows 8 will not allow you to create multiple partitions on a removable disk.
You will need the following items to create a UEFI bootable USB drive:
The following steps guide you on how to make the USB drive bootable and get the contents onto the drive.
5. Select your USB drive from the list returned (Disk 2 in this example) by typing Select Disk 2 and pressing ENTER.
6. Remove all data from the disk by typing: clean (Note: This will destroy all of the data on the USB drive) and pressing ENTER.
7. Convert the disk to GPT which is a requirement of UEFI by typing Convert GPT and typing ENTER.
7. Create a primary partition by typing create partition primary and pressing ENTER.
8. Format the newly created partition by typing format fs=fat32 quick and pressing ENTER.
9. List the volumes associated with the system by typing list volume and pressing ENTER. Make note of your newly created volume drive letter. (Volume 6, drive I: in this example.)
10. Exit DiskPart by typing exit and pressing ENTER.
11. At the CMD prompt use BootSect to update the BOOTMGR compatible bootcode onto the GPT partition by typing bootsect /nt60 I: and pressing ENTER.
12. Insert your desired operating system media or mount your ISO.
13. Copy the entire contents of the installation media from the installation media to your USB drive.
You’re now ready to use the USB drive as installation media on your system. Just boot to the USB drive and setup will begin, or just insert the USB drive and allow setup to auto run, or open setup from a file explorer window.
Hi Peter, it tied this with Windows 2012 data center server. I wanted to create a bootable usb for a r720 xd server. I wanted one large 4tb raid 5 with 1ghs but the usb stick would only boot in bios mode and not uefi.
p.s. when copied the 4.8g install.wim to the usb stick it failed.
This doesn't contain one of the needed lines. From source disk @ command prompt (eg your iso is mounted to d:\)... d:\boot\bootsect /nt60 (destination drive letter) to assign BootMgr to the drive. Plus, there are wonky things w/ w8 and 2012. Here's the thing to use: blogs.technet.com/.../the-simplest-was-to-create-a-bootable-windows-server-2012-or-windows-8-usb-key.aspx
JohnMPC, in order to support the file size of the install.wim file a 2nd partition formatted NTFS must be created due to the file size limitation of FAT32. UEFI requires a FAT32 partition for the boot partition, or we would just create a single NTFS partition.
JohnMPC the disk must be GPT. The documentation has been updated to reflect this requirement.
I did not understand the 12 command: do i have to placee my OS windows server 2012 on the USB stick ??; thanks in advance
Paul suggests that a second partition has to be made to handle large .wim files. Is that done in diskpart after the boot sector has been created? If so, is it an EFI or extended partition?