No more tedious warez hunting!
This website hosts big collections of Commodore 64 games, demos, music and magazines for download. Every archive contains files that are directly usable on a C64, no need for futher file conversion or extraction. (C64 Emulator usage is also possible.)
Recommended to use with IDE64 cardridge, SD2IEC or other similar mass storage or PC-Link solution. You can extract these archvies on a bigger computer, and then copy to your hard disk with fusecfs (on Linux), or host it via PC-Link or copy to SD card for your SD2IEC drive.
Last update: 2nd of April, 2021: TDD mags, demos, party, HVSC
open15,9,15,"XE+":close15 or if you have a DOS Wedge: @XE+.
: The GPEN-BFR-2048 model is demanding. Due to its size and high resolution, it requires significant VRAM (Video RAM) . Running it on a system with a limited GPU might be slow or impossible. A modern, powerful GPU is recommended for a smooth experience.
Trained specifically on high-fidelity 2048×2048 resolution images to preserve maximum detail. gpen-bfr-2048.pth
Blind face restoration (BFR) is the process of recovering a high-quality (HQ) face from a low-quality (LQ) input without knowing exactly what type of degradation corrupted the image. The degradation could be a combination of noise, blur, compression artifacts, or downscaling. : The GPEN-BFR-2048 model is demanding
| Attribute | Value | |-----------|-------| | | PyTorch checkpoint ( torch.save ) | | Size on disk | ≈ 2.1 GB (fp32) – ~1.1 GB when saved with torch.save(..., _use_new_zipfile_serialization=False, pickle_protocol=4) and torch.save(..., dtype=torch.float16) | | Top‑level keys | 'encoder', 'mapper', 'generator', 'args' | | encoder | state_dict of a ResNet‑50 (BN layers stripped) | | mapper | 2‑layer MLP (512 → 512) plus LayerNorm | | generator | StyleGAN2 weights (including the new 2048‑pixel synthesis blocks) | | args | Namespace containing training hyper‑parameters, input resolution, output resolution, and a version string ( GPEN-BFR-v2.0-2048 ). | | Compatibility | Requires PyTorch ≥ 1.8 and CUDA ≥ 11.0 (or CPU‑only fallback). The checkpoint can be loaded on any device with the same architecture (ResNet‑50 + StyleGAN2). | A modern, powerful GPU is recommended for a
gpen-bfr-2048.pth is a pre-trained weight file for the GAN Prior Embedded Network (GPEN) , specifically designed for high-resolution Blind Face Restoration (BFR)
The logic is brilliant:
The Browse links point to the collection's original location thus they don't reflect contents of the archives here!
The collections have been created using the ai64 - batch file extractor (v1.4, files in 2021 are with v1.5). The conversion is an automatic process, but errors are still possible. Feel free to report errors and I'll try to invesitage them. The process is not designed to be error-free, it's designed to do most of the work.
Please contact me if you know a good download source of Commodore 64 programs that should be available here for download.Collections are © copyright by their original maintainer as mentioned above.
Original works are © copyright by their original authors as seen in the files.
© 2010-2021 -
Lion/Kempelen/ex-Chromance/ex-Chaos