Using evaluation hash score as current bestscore (initial guess)

[xr_a_y]

I think it is clear the adjusted value could technically utilize tighter bounds. How much tighter depends on the information gained from hash. On the other hand, a position found in your hash may be a more difficult one on average, so perhaps less tight bounds would be better after all.
Most likely these are things you could tune, but whether or not that is worth it I cannot tell you, I have not tried it myself.

Why not just try it in Minic?

I will

I'm busy tuning history score for move ordering for now ...

This worked ! I got something out of that ... next release ...

[silentshark]

I think it is clear the adjusted value could technically utilize tighter bounds. How much tighter depends on the information gained from hash. On the other hand, a position found in your hash may be a more difficult one on average, so perhaps less tight bounds would be better after all.
Most likely these are things you could tune, but whether or not that is worth it I cannot tell you, I have not tried it myself.

Why not just try it in Minic?

I will

I'm busy tuning history score for move ordering for now ...

This worked ! I got something out of that ... next release ...

Just come into this late. I must admit, I'm not sure about what this is doing either

[xr_a_y]

Here is the idea :
- first write all margins under the form : a*depth+b
- second
* if static score is from eval : use the usual "a"
* if static score is from hash : only use a proper "b" (and a little "a"), make use of (depth-h.depth)

[D Sceviour]

upper hash < static - use the lower score
upper hash > static - could still be a safe prune

lower hash < static - looks like a safe prune
lower hash > static - worth extending so make best_score = hash_score - to raise cutoff value and re-look at the position a little wider. use the higher score