How to use Multi PV and why is it useful?

[halestorm]

I keep reading that Houdini 102 doesn't have MultiPV but 103 will. I know it's used for analysis, but how exactly would a user actually use this feature in other engines to study or analyze a game?

[Uly]

It's a way of asking the engine what other moves besides the best one they like, so you know which one you should investigate.

But for interactive analysis where you visit several positions several times, the searchmoves feature is more powerful for this (unless the engine has a learning algorithm).

[Swaminathan]

You can analyse your own games and see if your move matches with those of engines, if not at which rank it matches the engine's in multi-pv.

sometimes my move matches 2nd best move given by the engine, sometimes the 4th - I'll see the 1st best and understand why it's so much better than the one I had made. Also evaluation differences.

Very useful learning tool.

[Lostark]

sometimes my move matches 2nd best move given by the engine, sometimes the 4th - I'll see the 1st best and understand why it's so much better than the one I had made. Also evaluation differences.Very useful learning tool.

In fact that's the way I train regularly and when there is a team of two or more engines then I can find out which engine is preferring my moves (in my case Junior 7 ) and I can draw solace from that...

I also do a bit of psychology training and ask my students about all possible moves in a given position and after a brainstorming session I ask them - " Can someone tell me what would a Super GM play, what would an IM play and what would a 2000 rated player play in this position ?
(of course I would introduce them using the names of well known GMs and IMs and also some local player at 2000 rating range)...
That way the boys get excited and if some one keeps touching the SGM and GM levels regularly we get a reward system to give a personal motivation....
I am sharing this in the hope of getting further inputs from experienced trainers out here

Thx.