Releases: Timmoth/grandchesstree
GrandChessTree-0.1.5
- Added silent mode, when set to true it won't output the realtime metrics to the console (see below config)
- Client now reports thread count, allocated memory, MIPS to the server
- API request / response bodies use binary encoding to minimize bandwidth requirements.
{
"api_url": "https://api.grandchesstree.com/",
"api_key": <api_key>,
"workers": 24,
"worker_id": 0,
"task_type": 0,
"mb_hash": 1024,
"sub_task_cache_size": 2046,
"sub_task_launch_depth": 3,
"silent": false
}
GrandChessTree-0.1.4
Added allocated cache memory readout to the client (note it will probably be less then specified as it takes your input as a max value)
Improved fast task result compression / reduced bandwidth requirement.
{
"api_url": "https://api.grandchesstree.com/",
"api_key": <api_key>,
"workers": 24,
"worker_id": 0,
"task_type": 0,
"mb_hash": 1024,
"sub_task_cache_size": 2046,
"sub_task_launch_depth": 3
}
GrandChessTree-0.1.3
Added commands:
's' + enter to start / stop
'r' + enter to reset client side stats
Reduced bandwidth requirement of client submitting result batches
{
"api_url": "https://api.gran
8000
dchesstree.com/",
"api_key": <api_key>,
"workers": 24,
"worker_id": 0,
"task_type": 0,
"mb_hash": 1024,
"sub_task_cache_size": 2046,
"sub_task_launch_depth": 3
}
GrandChessTree-0.1.2
Added the sub_task_launch_depth (defaults to 3)
{
"api_url": "https://api.grandchesstree.com/",
"api_key": <api_key>,
"workers": 24,
"worker_id": 0,
"task_type": 0,
"mb_hash": 1024,
"sub_task_cache_size": 2046,
"sub_task_launch_depth": 3
}
GrandChessTree-0.1.1
TGCT client v0.1.1 has just been released (including updated docker images)
https://hub.docker.com/r/aptacode/grand-chess-tree-worker/tags
Massive performance improvements to the subtask generation / caching (Thanks for the help @Chester-alt !)
More detailed output (see attached)
Subtask cache configuration option (sets the size of the global subtask cache in MB)
{
"api_url": "https://api.grandchesstree.com/",
"api_key": <api_key>,
"workers": 24,
"worker_id": 0,
"task_type": 1,
"sub_task_cache_size": 2046
}
Contributors
Assets 12
GrandChessTree-0.1.0
Refactored DB schema to PerftTaskV3
GrandChessTree-0.0.8
Added 'mb_hash' parameter which allows you to specify the size of each workers hash table in MB - it's memory hungry so the more the better!
{ "api_url": "https://api.grandchesstree.com/", "api_key": "<api-key>", "workers": 4, "worker_id": 0, "task_type": 1, "mb_hash": 1024 }
GrandChessTree-0.0.7
Added the 'perft nodes' task type.
Added "task_type" property to the config. E.G
{ "api_url": "https://api.grandchesstree.com/", "api_key": "<api-key>", "workers": 4, "worker_id": 0, "task_type": 1 }
set task type to 0 to work on the full perft stats tasks or set it to 1 to work on the perft nodes tasks.
The worker output table doesn't show in the client app by default anymore.
Hit 'd'+enter to toggle the worker nodes report table on / off
GrandChessTree-0.0.6
- renamed releases to include build version.
- Added tgct engine to release.
- Added the ability to calculate unique positions reachable at a given depth
GrandChessTree-0.0.5
Added the ability for the server to have multiple tasks. The client will automatically choose the quickest task.