How Many Pieces Against A Queen Win?

This is for my fellow endgame enthusiasts and tablebase experts, what piece combinations generally have forced wins against a queen? I know 2 rooks and a bishop do, but what about Rook + Knight + Bishop or Rook + 2 Bishops, or 4 minor pieces? Any of these piece combinations plus a pawn? What do tablebases say on this?

This is a serious thread! Please go troll in the off topic section instead my guy!

Experience 😂 is what I say

Is this a draw or can black force a win somehow? Obvious if it's black to move he can check the king with the dark squared bishop onto a light square where the queen can be discovered attack by the light squared bishop while checking the king next move, but what if it's white to move? Are the perpetual checks endless?

Tablebase draw when white moves

https://syzygy-tables.info/?fen=8/8/3rk3/3bb3/8/8/8/3QK3_w_-_-_0_1

Tablebase win when black moves.

https://syzygy-tables.info/?fen=8/8/3rk3/3bb3/8/8/8/3QK3_b_-_-_0_1

I would definitely not drawl that, no

Easy win for black if black to move, Bg3+ then wherever the white king moves other bishop can move with check winning the queen. I guess with white to move itself a perpetual check.

I'd say that KRBB is generally insufficient to beat KQ as 50% ends in a draw. Also there are quite a lot of wins by the queen side (20%) which is hard to understand. My explanation is that - starting from random positions - the pieces on the KRBB side are usually uncoordinated (en prise or lost in some moves) while that matters much less on the KQ side. In a realistic game the side having many pieces will know the importance of coordination and not enter this endgame with a chaotic distribution. Which would practically annihilate the 20% KQ win chances. Not sure what it does to the draw percentage.

King and 2 bishops with 2 knights is also enough to win.

That's true. Alot of even equal tablebase positions will have like 60-70% win rate for the player who moves first due to uncoordinated, random piece positions. In QRQR for example, the first player to check usually ends up winning. The same way many KNBK positions are drawn if the original position has the pieces forked. Well if KQKRBB is a draw half the time, then KQKRBN probably definitely is.

Even just 4 knights or just 4 bishops might be enough to win, although I wonder how the color of the bishops effect it. Like 2 dark squared bishops and 2 light squared bishops is better than 3 of one color and 1 of the other color.

I'd say the square colors of the bishops are absolutely essential. It is easy for the queen side to defend on the square color of the single bishop. After all it can sac its queen for it and land in a dead draw!

Definitely. I was thinking more like 2 knights + 2 bishops of the same color, still probably more drawable than 2 knights + 2 opposite colored bishops due to the queen being able to hide out on the opposite color squares of the same colored bishops. I notice tablebases do make the color distinction very important, which makes sense as even 9 of the same colored bishops = insufficient mating material.