Difference between revisions of "Guido Schimmels"

From Chessprogramming wiki
Jump to: navigation, search
(Created page with "'''Home * People * Guido Schimmels''' '''Guido Schimmels''',<br/> a German software developer contributing to FSF GNU and simil...")
 
 
Line 28: Line 28:
 
=Forum Posts=
 
=Forum Posts=
 
* [https://www.stmintz.com/ccc/index.php?id=19790 Re: Hash Table Size Versus Performance] by [[Guido Schimmels]], [[CCC]], June 02, 1998 » [[Transposition Table]]
 
* [https://www.stmintz.com/ccc/index.php?id=19790 Re: Hash Table Size Versus Performance] by [[Guido Schimmels]], [[CCC]], June 02, 1998 » [[Transposition Table]]
* [https://www.stmintz.com/ccc/index.php?id=20057 Question to Bob: Crafty , Alpha and FindBit()] by [[Guido Schimmels]], [[CCC]], June 05, 1998 » [[Crafty]], [[BitScan]]
+
* [https://www.stmintz.com/ccc/index.php?id=20057 Question to Bob: Crafty , Alpha and FindBit()] by [[Guido Schimmels]], [[CCC]], June 05, 1998 » [[Crafty]], [[DEC Alpha]], [[BitScan]]
 
* [https://www.stmintz.com/ccc/index.php?id=20868 Re: Zero-width Window Null Move Search] by [[Guido Schimmels]], [[CCC]], June 18, 1998 » [[Principal Variation Search]], [[NegaScout]]
 
* [https://www.stmintz.com/ccc/index.php?id=20868 Re: Zero-width Window Null Move Search] by [[Guido Schimmels]], [[CCC]], June 18, 1998 » [[Principal Variation Search]], [[NegaScout]]
 
* [https://www.stmintz.com/ccc/index.php?id=23672 Failing low at the root] by [[Guido Schimmels]], [[CCC]], August 03, 1998 » [[Root]]
 
* [https://www.stmintz.com/ccc/index.php?id=23672 Failing low at the root] by [[Guido Schimmels]], [[CCC]], August 03, 1998 » [[Root]]

Latest revision as of 10:22, 3 November 2019

Home * People * Guido Schimmels

Guido Schimmels,
a German software developer contributing to FSF GNU and similar projects, such as GNU TeXmacs [1], and the ROX Desktop [2]. He further is contributor of the Ultimate++ framework [3]. In the late 90s and early 2000s, Guido Schimmels was engaged in computer chess programming, and made a couple of posts in CCC.

PVS vs. NegaScout

Guido Schimmels in a CCC post on the difference of PVS vs. NegaScout [4]:

The difference is how they handle re-searches: PVS passes alpha/beta while NegaScout passes the value returned by the null window search instead of alpha. But then you can get a fail-low on the research due to search anonomalies. If that happens NegaScout returns the value from the first search. That means you will have a crippled PV. Then there is a refinement Reinefeld suggests which is to ommit the re-search at the last two plies (depth > 1) - but that won't work in a real program because of search extensions. NegaScout is slightly an ivory tower variant of PVS (IMHO).  

PVS

value = PVS(-(alpha+1),-alpha)
if(value > alpha && value < beta) {
  value = PVS(-beta,-alpha);
}

NegaScout

value = NegaScout(-(alpha+1),-alpha)
if(value > alpha && value < beta && depth > 1) {
  value2 = NegaScout(-beta,-value)
  value = max(value,value2);
}

Forum Posts

References

Up one level