Fixed a Windows-specific issue causing a ghost border around minimized (iconified) process window. Adjusted normalization algorithm to avoid histogram clipping in shadow regions. MacOS-specific updates to enable 32-bit floating point neural network computations (requires MacOS 12 or later). Internal processing is now 64-bit floating point for maximum precision. New tiling algorithm to better handle the edges of images. More intelligent validation/selection of AI definition files.Ĭompatibility updates for PixInsight 1.8.9-2. Ubuntu 18.04 or later, or equivalent (glibc 2.27 or later required).Modern CPU capable of running neural networks (AVX, SSE instruction set extensions).Apple silicon neural engines and GPUs (e.g., M1, M2) are supported natively.GPU/Neural Engine recommended for fast performance, but not required.Photoshop CS4 or later (64-bit only), or.On Windows and Linux machines, GPU acceleration takes a bit more effort. Performance is particularly fast on recent Apple silicon based Macs (M1, M2, etc.) due to their extremely performant built-in GPU and neural engine hardware. On Macs, GPU acceleration of StarXTerminator is automatic, leveraging the “CoreML” software library provided by Apple. Neural network computations can benefit greatly from GPU acceleration. Request a trial before purchasing to make sure StarXTerminator will work on your machine. Older CPUs, and even some modern stripped-down CPUs in mini-PCs and lightweight laptops lack AVX instructions and are not supported. StarXTerminator requires a computer with a modern CPU having “AVX” instructions required to run neural networks. The photoshop version also works with Affinity Photo. StarXTerminator is available as a plug-in process module for PixInsight and Photoshop.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |