this was a gauntlet of three Craftys versus an assortment of engines - this time control was base 6.6 sec inc .11 seconds ( adjusted to match Bob's hardware where he tests with base 10 sec inc .17)
No GUI here, this uses c shell scripts and few command line apps written by Bob that manage engine vs engines matches with the lowest possible overhead- just set a few parameters and type submit and it's off and running - Bob has it setup so that that he can automatically feed various servers that were waiting for work...
A 4 minute video demonstrating Bob's custom scripts and apps to run his GUI less crafty testing.
I have modified his scripts to run round robin(RR) matches and to show progress throughout the match. In this video , it was a 4 round RR using two test positions , each side playing with black and white. A very short micro time was chosen test to demo the process. In this case it was 1 sec base/ .1 sec inc. There were 11 engines, playing 4 games each with the other engines. Each engine played 40 games in total. The RR, 220 games in total, was completed in less than 5 minutes on a 12 core Mac Pro. The script has played 30,000 games consecutively without an error or crash.
MikeB wrote:this was a gauntlet of three Craftys versus an assortment of engines - this time control was base 6.6 sec inc .11 seconds ( adjusted to match Bob's hardware where he tests with base 10 sec inc .17)
No GUI here, this uses c shell scripts and few command line apps written by Bob that manage engine vs engines matches with the lowest possible overhead- just set a few parameters and type submit and it's off and running - Bob has it setup so that that he can automatically feed various servers that were waiting for work...
MikeB wrote:this was a gauntlet of three Craftys versus an assortment of engines - this time control was base 6.6 sec inc .11 seconds ( adjusted to match Bob's hardware where he tests with base 10 sec inc .17)
No GUI here, this uses c shell scripts and few command line apps written by Bob that manage engine vs engines matches with the lowest possible overhead- just set a few parameters and type submit and it's off and running - Bob has it setup so that that he can automatically feed various servers that were waiting for work...
Summary: 13 engines play each other 5000 times, with alternating colors, with a one of kind 2500 posiiton book crafted from the top (+3200) CCRL games which were decisive. Estimated completion time 2 weeks.
Objective: To set a high end baseline for future crafty engine testing.
Summary: 13 engines play each other 5000 times, with alternating colors, with a one of kind 2500 posiiton book crafted from the top (+3200) CCRL games which were decisive. Estimated completion time 2 weeks.
Objective: To set a high end baseline for future crafty engine testing.
Summary: 13 engines play each other 5000 times, with alternating colors, with a one of kind 2500 posiiton book crafted from the top (+3200) CCRL games which were decisive. Estimated completion time 2 weeks.
Objective: To set a high end baseline for future crafty engine testing.
Phase 2A has started, a 7800 game run that will take about 6 hours - same link as before for realtime updates. Also added a link that will just show the Phase 2A results.
Phase 2B will start soon, a 15600 game run that will take about 15 hours - same link as before for realtime updates. Also included a link that will just show the Phase 2B results.
Phase 2C will start soon, another 15600 game run that will take about 15 hours - same link as before for realtime updates. Also included a link that will just show the Phase 2B results.
Phase 3 will start soon, another 15600 game run that will take about 15 hours - same link as before for realtime updates. Also included a link that will just show the Phase 3 results.
