This is a temporary, read-only recovery of the chessprogrammingwiki while a longer-term plan is worked out. Editing is not possible right now, but will be again soon.
Chess Programming Wiki All pages Other namespaces

Node

Home * Search * Node

The Dragon's Head and Tail 1The Dragon's Head and Tail 1


  1. The Dragon's Head and Tail refer ascending and descending Lunar nodes, Image from Guido Bonatti (1550). Foroliviensis Mathematici De Astronomia Tractatvs X. : Vniuersum quod iudiciariam rationem Natiuitatum, Aeris, Tempestatum, attinet, comprehendentes. Adiectus est Cl. Ptolemaei liber Fructus, cum commentarijs vtilissimis Georgij Trapezunt. pp. 119, from Johannes a Lasco Bibliothek (JALB)↩︎

Node, (vertex)
the fundamental unit in graph theory and, applied to computer science, basic unit to structure and link devices of a network as well as data, such as items of a linked list or tree structure. This page is about the node of the search tree in the game of chess, which represents the alternating white and black to move positions and keeps its state dependent on the tree traversal. Inside a depth-first search, nodes are usually counted at top of the recursive search routine, i.e. for the purpose to determine nodes per second. The move is the directed edge which connects an ordered pair of nodes or positions.

Nodes are classified by their topological properties inside the search tree, and in context of alpha-beta and its enhancements, by their type as expected in accordance to the minimal tree before searching this node, or as determined after searching the node.

Contents
  1. Node Types
  2. Topology
  3. Depth
  4. See also
  5. Forum Posts
    1. 1995 ...
    2. 2000 ...
    3. 2010 ...
    4. 2020 ...
  6. External Links
  7. References

Node Types

Node Types of the minimal tree - enumerated as classified by Knuth and Moore:

  1. PV-Node Knuth's Type 1
  2. Cut-Node Knuth's Type 2, also called fail-high node
  3. All-Node Knuth's Type 3, also called fail-low node

PV-nodes are often treated differently by the search - see Principal Variation Search or Internal Iterative Deepening for examples.

Topology

Depth

Name

Heinz

Hyatt

Ambiguity

Horizon Node

0

-

*

Frontier Node

1

0

*

Pre Frontier Node

2

1

*

Quiescent Node

<= 0

See also

Forum Posts

1995 ...

2000 ...

2010 ...

2020 ...

References

Up one Level


  1. That leftmost nodes are always PV-nodes, does not imply each PV-node is leftmost - since we have to deal with real search trees rather than minimal ones.↩︎

  2. Re: simple node definitions question post by Robert Hyatt, CCC, September 13, 2004↩︎

  3. Ernst A. Heinz (1998). Extended Futility Pruning. ICCA Journal, Vol. 21, No. 2, pp. 75-83↩︎

What links here

Contributors: GerdIsenberg.