Changes in v1.5, first released on 7th of November, 2018:
- Updated BOINC upstream wrapper to v7.13.26016 with various changes and fixes, including: wrapper is now a console application, improved verification of D.net Client to run (security) and process priority control from task settings.
- Updated D.net Client for 64-bit Windows CPU to build 521 (from build 520) for new AVX2 core, IPv6 support and detection for Intel Skylake and Intel Xeon E5/E7 v3.
- Updated D.net Client for 32-bit Windows CPU to build 521 (from build 519b) for new AVX2 core, IPv6 support and detection/core selection changes for Intel Skylake, Intel Xeon E5/E7 v3, AMD Opteron 6xxx, AMD APU, AMD FX, Intel Sandy Bridge, Via Nano and various other processors.
- Updated D.net Client for Windows OpenCL (both AMD and nVidia) to build 521 (from build 520) for IPv6 support, support for any number of OCL platforms, better handling of small blocks in fast GPUs (might not be relevant for us since our WU size increases for speedier cards) and improved GPU initialization and error diagnostics.
- Updated D.net Client for Windows ATI/AMD Stream to build 521 (from build 519) for IPv6 support, better handling of small blocks in fast GPUs (might not be relevant for us since our WU size increases for speedier cards) and improved GPU initialization and error diagnostics.
Changes in v1.4, released on 6th of September, 2014:
- Updated BOINC upstream wrapper to v7.5.26011 with various changes and fixes, including: fixed CPU usage reporting, improved GPU detection, added Intel GPU and OpenCL CPU/GPU detections, Android support, improved dead BOINC client detection (with backward compatibility) and diagnostics framework fixes/improvements.
- Added OpenCL (both AMD and nVidia) and Android support (both PIE and non-PIE versions).
- Initialization failure detection from stderr fixed. It might not have been enabled in some cases.
- Minimum supported Windows is now XP with IE 5.01 due to upstream changes.
- Requires BOINC Client v7.2.28 or later for GPU support due to API requirements.
- Updated D.net Client for Win32 Stream to build 519 (from build 518) for identification of HD7950/HD7970 GPUs and support for GPU device selection.
- Updated D.net Client for 64-bit Windows and Linux to build 520 (from build 518) for two new cores (GO-2c and GO-2d) and detection/core selection changes for AMD Opteron 6xxx, AMD APU, AMD FX, Intel Sandy Bridge, Via Nano and various other processors.
- Updated D.net Client for 32-bit Windows to build 519b (from build 518) for two new cores (GO-2c and GO-2d), Intel Sandy Bridge detection/core selection and identification of new i3/i5 chips.
- Uses only BOINC Client selected GPU instead of all GPUs. This also means our apps should now listen to GPU exclusions set in BOINC Client.
Changes in v1.3, released on 16th of January, 2012:
- Checkpoint files are no longer misplaced after abnormal restart (so they are actually effective).
- Hang detection disabled when the task is suspended (so no more false terminations due to this).
- Wrapper watches and reacts with a temporary exit for an hour to initialization failures found in the stderr output log of the client.
- A random crash on Linux with a SIGSEGV (signal 11) should now be fixed.
- GPU clients now use maximum CPU priority (priority=9) so they should more effectively keep GPUs busy with work (process still has idle priority class).
- Added support for separate ATI/nVidia/CPU core selection prefs in one host. Fallback to old shared core preference if new one is not set.
Changes in v1.2, released on 18th of May, 2011:
- Distributed.net ID and core preferences were not used on Linux.
- Default D.net ID was not used if a custom was not specified.
- Changed default D.net ID to @moowrap.net, where is your Moo! Wrapper account name with certain invalid characters removed.
- Use our domain if there's no domain (missing @) in a custom D.net ID.
- Added a 15min minimum to checkpoint interval.
Changes in v1.1, released on 8th of May, 2011:
- Overactive hang detection code fixed and it now takes into account relative speed of the used card. This should fix many failures and even slowness seen by users especially on CUDA platform.
- Limited retries (no checkpoint or premature exit) to ten (10) so that there's no chance for an endless loop.
- Added support for setting the core used by the client. Allows advanced users to set better core based on their knowledge or client benchmarks.
- There's now a default Distributed.net ID used if one is not specified in preferences. For now it's simply "firstname.lastname@example.org" where is your User ID of your account on this project. Custom ID is still used, if specified in project preferences.
- And some other minor improvements and fixes around the code.
Version v1.0, released on 1st of May, 2011: