| IMPORTANT POINTS |
|---|
| An NVIDIA 3090 GPU with 24GB VRAM is highly recommended and will get the best performance and highest resolution outputs. |
| An NVIDIA 2080 GPU with 8GB VRAM would be the bare minimum hardware spec and will only give you small resolution images. |
| If you have an older GPU or one with less than 8GB VRAM, do not bother. You will only be disappointed and complain "it doesn't work". |
| You MUST have an NVIDIA GPU for the Machine Learning related modes to work in Visions of Chaos. If you do not have an NVIDIA GPU stop now and don't bother with the rest of this page. |
| The required programs, Python libraries and machine learning models require around 300 GB of disk space to install. If you do not have the patience for long large downloads then machine learning is not for you. Due to the rapid pace of advancements in these systems there will be frequent updates that break previous libraries needing new Python enviropnments to be created. Again, if the thought of having to wait for some downloads and updates after a new release bothers you, then do not go any further. |
| If you do not have reliable and fast Internet then I do not recommend you continue. Machine learning requires many huge files that need to be downloaded and large updates happen frequently as new systems emerge. Having bad Internet is only going to frustrate you. |
| Version numbers are important. If these instructions specify an exact version number make sure you download and install that version. Some versions listed may not be exact, but if you see an exact version, download and install that version. |
| These steps only seem to work on Windows 10 or Windows 11 64-bit versions. I have not yet been able to get them working on earlier versions of Windows. |
| Update Your GPU Drivers |
|---|
| Select Help->Update Graphics Card Drivers from Visions of Chaos. That will show you which model GPU you have and open the NVIDIA download page. |
| Once the download page opens, select your model GPU and click Search. The recommended driver will appear. Download it, install it, reboot. |
| CUDA |
|---|
| Download CUDA v11.2.2 March 2021 from here and install it. |
| NOTE: by default the Nvidia installer wants to install extra drivers etc, you only need the compiler and libraries options checked, ie |
 |
| cuDNN |
|---|
| 1. Download cuDNN v8.1.0 (January 26th, 2021) for CUDA 11.0,11.1 and 11.2 from here. |
| 2. Extract the cuDNN.rar file to a temp folder. |
| 3. Copy the 7 DLLs files from the RAR file into the C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2\bin\ folder. |
| Add CUDA to Path |
|---|
| 1. Hold the Windows key and press Pause to bring up the System Control Panel. |
| 2. Click Advanced system settings. |
| 3. Click the Environment Variables button. |
 |
| 4. Highlight Path in the lower "System variables" section and click Edit. |
 |
| 5. Click New and add C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2\bin as follows. |
 |
| 6. Click OK, OK, OK to close all dialogs. |
| Verify nvcc command line |
|---|
| You should now be able to go into a command prompt window, type nvcc and get the following response. Make sure you open a fresh command prompt window. If you try and use an existing command prompt that was open prior to adding the path entry the new path is not updated in the already open command prompt window. |
 |
| If you cannot get nvcc.exe working (nvcc fatal means it is working) on a command line, do not go any further before fixing it. It is needed for the later Python library installs. |
| Visual Studio Community Edition |
|---|
| Download Visual Studio Community Edition from here. |
| Start the installer and check "Desktop development with C++" on the first tab |
 |
| Click the "Individual Components" tab, scroll down and check the most recent "C++/CLI support" checkbox |
 |
| Click Install. |
| Once the install is finished, you do not have to start Visual Studio or create an account to sign in. Just close it. |
| Add Visual Studio to Path |
|---|
| 1. Hold the Windows key and press Pause to bring up the System Control Panel. |
| 2. Click Advanced system settings. |
| 3. Click the Environment Variables button. |
|
| 4. Highlight Path in the lower "System variables" section and click Edit. |
|
| 5. Click New then click Browse and browse to your VS Hostx64\x64 directory. |
 |
| Mine is C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\Hostx64\x64 , but your directory may have a different version number or drive letter, so don't just copy and paste my path. Browse to your Hostx64\x64 folder to be sure your PATH is correctly pointing to the right location. |
| 6. Click OK, OK, OK to close all dialogs. |
| Verify Visual Studio command line |
|---|
| You should now be able to go into a command prompt window, type cl and get the following response. Make sure you open a fresh command prompt window. If you try and use an existing command prompt that was open prior to adding the path entry the new path is not updated in the already open command prompt window. |
 |
| If you cannot get cl.exe working on a command line, do not go any further before fixing it. It is needed for the later Python library installs. |
| CMake |
|---|
| Download the latest Windows x64 Installer of CMake from here. |
| Install CMake making sure you check the "Add CMake to the system PATH for the current user" checkbox. |
 |
| All other settings can be left at the defaults. |
| Verify CMake command line |
|---|
| You should now be able to go into a command prompt window, type cmake and get the following response. Make sure you open a fresh command prompt window. If you try and use an existing command prompt that was open prior to adding the path entry the new path is not updated in the already open command prompt window. |
 |
| If you cannot get cmake.exe working on a command line, do not go any further before fixing it. It is needed for the later Python library installs. |
| Git |
|---|
| Download the latest 64 bit installer of Git for Windows from here. |
| Install Git accepting all the numerous default settings. |
| Verify Git command line |
|---|
| You should now be able to go into a command prompt window, type git and get the following response. Make sure you open a fresh command prompt window. If you try and use an existing command prompt that was open prior to adding the path entry the new path is not updated in the already open command prompt window. |
 |
| If you cannot get git.exe working on a command line, do not go any further before fixing it. It is needed for the later Python library installs. |
| Image Magick |
|---|
| Download the latest 64 bit HDR version of Image Magick from here. |
| Start the installer and select the following install tasks. |
 |
| You can leave everything else at the defaults. |
| Python |
|---|
| Download Python v3.9.x from here. Newer 3.10.x versions won't work. Make sure it is the 64 bit version. If the first setup page shows 32-bit, cancel and download the 64-bit version. Any 3.9 version will work. |
| Install it making sure you check the "Add Python to PATH" checkbox. |
 |
| If you also install Python outside the "Program Files" directory (eg to C:\Python\) it saves potential security rights hassles in the pip install stage. |
| Create LIB environment variable for Python\Libs directory |
|---|
| 1. Hold the Windows key and press Pause to bring up the System Control Panel. |
| 2. Click Advanced system settings. |
| 3. Click the Environment Variables button. |
| 4. Click the lower New button in the "System variables" section. |
| 5. Call the variable LIB and point it to the Libs directory under your Python install directory |
 |
| 6. Click OK, OK, OK to close all dialogs. |
| Enable Long Filename Support |
|---|
| 1. Press the Windows key and type regedit. |
| 2. Click Registry Editor to open the Registry Editor. |
| 3. Expand the left hand tree to HKEY_LOCAL_MACHINE\CurrentControlSet\Control\FileSystem |
| 4. On the right hand pane, double click LongPathsEnabled and change the value from 0 to 1 |
 |
| 6. Reboot for the change to take effect. |
| Check and Enable Machine Learning in Visions of Chaos |
|---|
| 1. Start Visions of Chaos (if you already have Visions of Chaos open, close it and restart to ensure all path updates are seen) |
| 2. Click the Options menu and then select Options to open the Options dialog |
| 3. Under the Machine Learning tab, click the "Check Machine Learning" button |
| At this point you should get a message that Python, nvcc, cl and cmake have been detected. If not, go back up and fix the installation for the error. |
| You should now have a bunch of sub menus under Mode->Machine Learning to explore. |
| As you start the various machine learning related modes for the first time you may be prompted to download model files or setup virtual environments. Let these processes run to ensure you have all the required machine learning related files. |
| Because Visions of Chaos keeps its required Python packages in virtual environments, if you already have Python packages/libraries installed they will not be affected by this install process. |