the pawns are also wieghted differently e pawn is worth more than h pawn
Yusupov's book

The big difference is he values a rook at 4.5 instead of 5. I had never counted that way in the past, but it makes a lot of sense.
I use relative value( dynamic value) with relative evaluation value in middle game and endgame, similar to stockfish evaluation table.
In general, Rook relative value is higher ( probably 5)
1. Endgame with multiple open files
2.Endgame with multiple pawn islands
3. Multiple bad pawn structures( double pawns, isolated pawns)
Rook value is relatively decreased ( 4.5) in
1. Middle game
2. Endgame with blocked files
3. Endgame with single pawn island
And also, dont exchange your rook in endgame if you have bad pawn structure or pawn deficit.
Good guidelines. Thanks.
There's a discussion about the "Kaufman" values here (Knight = Bishop = 3.25 pawns, Bishop pair 0.5 pawn bonus etc), I thought these were sort of accepted now:
https://www.chess.com/article/view/the-evaluation-of-material-imbalances-by-im-larry-kaufman