Incremental update of PST values slower than not incremental?

[JVMerlino]

Finally got around to converting the PST portion of my eval to be incrementally updated with make/unmake, rather than doing it all from scratch whenever Evaluate() is called. And my engine is now about 10% slower....

When you do a depth=x search, do the old and new version produce the same number of nodes searched?

They do. Same nodes, same score, same PV.

[JoAnnP38]

C# multi dimensional tables are slow. Use some function to generate an index to a one-dimensional array, something like return 64 * (6 * x + y) + z; this will help both with incremental and non-incremental tables, and perhaps it will shift the proportion between one and the other.

Agreed, that why I use my own class UnsafeArray2D<T> which uses fixed memory and pointer to access the array and I have to do the index arithmetic myself. I just used the multi-dim indexer because that's what C# allows you to overload. Unfortunately, while this benchmarks out fantastic, in actual use a lot of time is spent invalidating the cache when very large lookup tables which starts to dwarf .NET's array indexing and bounds check overhead. Amdahl's law bites me in the rearend again! I hate that guy!

[syzygy]

[comment removed because I mistook the person to whom I responded for the TS]