WoR v1.5 - faster deployments using FFU images

Windows on Raspberry imager version 1.5.0 has been released!

This release focuses on reducing the overall installation time by taking advantage of Microsoft’s FFU image file format.

Downloads: https://www.worproject.ml/downloads
Changelog: https://www.worproject.ml/changelog.htm
Bug tracker: https://www.worproject.ml/bugtracker

Some benchmarks

Windows image build: 17134.191

DISM FFU Provider (FFU) Windows Imaging (WIM) wimlib (WIM)
HGST 7200 RPM SATA 3 HDD (15 GB VHDX) 5 mins 31s 16 mins 7s 17 mins 3s
Sandisk 16 GB UHS-1 USB 2.0 microSD 16 mins 57s 1h 20 mins 48s 1h 33 mins 52s

Image size comparison

Build ISO image WIM image FFU image (7z compressed) FFU image
17134.191 (en-US) 4,63 GB 4,03 GB 4,65 GB 6,09 GB

The official FFU images provided by worproject come with the following binaries:

  • drivers package: WoA64_Drivers_RPi3_20Feb19_USBBoot.zip
  • UEFI: 2019Feb18-GCC5

The tool automatically updates them unless you disable this option on the image selection page.

It is still necessary to configure the UEFI, so follow these steps:

  1. Type exit at the EFI shell and press enter
  2. Go to Device Manager -> Raspberry Pi Configuration -> Chipset Configuration
  3. Set CPU Clock to Max
  4. Go back (by pressing the ESC key), press the Y key to save the configuration
  5. Go to Boot Maintenance Manager -> Boot Options -> Change Boot Order and press enter
  6. Highlight the SD card controller or the USB device (if you’re booting from USB) and use the + key (Shift and +) to move the selection to the top, then press enter again
  7. Go back again and select Reset

Once installed, Windows will boot straight to OOBE. You’ll no longer see the “Getting ready” screen that takes an eternity to complete. So the installation takes even less time.

Minimum required storage space is 16 GB for the currently provided images. WoR does not warn you if you select an 8 GB SD card for example (because I might create compressed images in the future). It will fail later in the process.

Some translations are quite outdated, and a few of them didn’t receive any updates since release 0.2.x. The status can be found here: https://github.com/worproject/WoR-Translations
All contributions are welcome!

1 Like