Search found 3963 matches

by sje
Sun Aug 21, 2016 3:47 am
Forum: Computer Chess Club: General Topics
Topic: Just a moment of your time, please.
Replies: 22
Views: 8580

My appreciation

Let me thank all of you for your kind thoughts.
by sje
Tue Aug 16, 2016 11:42 pm
Forum: Computer Chess Club: General Topics
Topic: Just a moment of your time, please.
Replies: 22
Views: 8580

Just a moment of your time, please.

Just a moment of your time, please. It was back in 1967 when I was a ten year old and my father James introduced me to chess. We were on vacation in Yellowstone National Park and he had given me a dollar to spend at a souvenir store where I bought a small peg chess set made from green and white plas...
by sje
Tue Aug 16, 2016 11:17 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

The lesser evil

As had been mentioned long ago, it is relatively utproductive to increase the signature bit length well past the point where the probability of a false match is far less than the probability of error caused by cosmic ray events. And a cosmic ray event is what has almost certainly happened in one of ...
by sje
Sun Aug 14, 2016 8:11 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

Re: yet another attempt on Perft(14)

Ankan: Would it be possible to use your revised 128 bit signature program to check Symbolic's output for the first 400 work units (000-399)? The results I have for these are at https://dl.dropboxusercontent.com/u/31633927/Perft14Var/PerftSums Ok, will do. Thank you. More file links: On Google Drive...
by sje
Sun Aug 14, 2016 6:40 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

Cosmic rays

Well, be careful with making any changes which might somehow introduce a bug. Whenever I made changes in Symbolic , I then ran many regression tests to build confidence that I hadn't messed up. Independent verification is the only way to detect cosmic ray events. If the verification is done using th...
by sje
Sun Aug 14, 2016 5:19 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

The bottom bits

Again: Peter: 110101101011010011110000110011000101100111100000100110000110010101 Ankan: 110101101011010011110000110011000101100111100000100011100110010101 1. Note that the bottom eleven bits are the same. This strongly suggests that the difference of the totals is NOT due to false positive matches (...
by sje
Sun Aug 14, 2016 4:15 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Perft(14) Weekly Status Reports for 2016
Replies: 75
Views: 27178

Perft(14) Weekly Status 2016-08-14

Perft(14) Weekly Status 2016-08-14 Symbolic has produced more than 54,000,000 perft(7) results so far, about 56.02% of the 96,400,068 needed. Day count: 739 Estimated remaining day count: 580 Estimated total day count: 1,319 Average throughput: 73,072 results/day Effective frequency: 227.72 GHz Work...
by sje
Sun Aug 14, 2016 4:14 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

Re: yet another attempt on Perft(14)

Ankan:

Would it be possible to use your revised 128 bit signature program to check Symbolic's output for the first 400 work units (000-399)?

The results I have for these are at https://dl.dropboxusercontent.com/u/316 ... /PerftSums
by sje
Sun Aug 14, 2016 3:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

Re: Peter Österlund's Perft(14) result

From Peter Österlund 2013.04.02: perft(14) = 61,885,021,521,585,529,237 Which is 10,240 more than Ankan's result. Which is THREE bits different (XOR) out of 66 bits. Which could be a triplet of single bit random flips (cosmic rays). Which shows that independent verification is so important. Peter: ...
by sje
Sun Aug 14, 2016 3:43 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: yet another attempt on Perft(14)
Replies: 33
Views: 14310

Peter Österlund's Perft(14) result

From Peter Österlund 2013.04.02: perft(14) = 61,885,021,521,585,529,237