one to wonder about ??

I'm not sure I've put the diagram in correctly but black's last move was to capture a pawn on b5 which I thought was a blunder because it trades a pawn for a rook so I took his rook and eventually won by resignation. 

Computer analysis says nothing about black's move ...39 Rxb5 but credits me with a blunder ?? for taking his rook!

go figure huh?  

thisi sjust the begining of the computer revolt i have seen this befoe we are all doomed omg

Apparently this is what the computer was thinking about ... ( 40. Kg4 Rb2 41. g3 Rb7 42. Rc5 Rg7+ 43. Kf3 Ra7 44. Rc6+ Kf7 45. Kg4 Ra5 46. Rc2 Ke6 47. Rb2 Rc5 48. Rb6+ Ke7 ) ( 40. Rxb5 Kd7 41. e5 f5 42. e6+ Kxe6 43. Re5+ Kd6 44. Rxf5 )

whatever... at my level I'll take the free rook every time!! 

Don't believe the Chess.com computer. One time it told me a move maintaining equality was a blunder, and it instead suggested hanging my queen and giving my opponent mate in 8.

sounds like they need to fix this thing or dump it ... has this been discussed before?

Hehe.. I have also been wondering about the computer analysis. Sometimes it also claims that no blunders were made (where it lists the number of blunders, mistakes, and uncertainties), even though this was clearly the case.. and surely enough when I go through the analysis, it writes that a blunder was made.. 

um. yeah, Rxb5 is correct. The fact that computer sees some theoretical line in which somehow not taking the rook allows you to take it in a better way 13 moves deeper is not worth bothering with.

These things are true:

A) Computers play strong and will (over and over) point out kickass tactics to you! Use them.

B) Computers can miss larger strategic ideas that are VERY helpful/simple to humans. Don't take computers as gospel. People who specialize in beating computers make them look really silly because some of the mistakes are pretty glaring. I'm not sure what happens with the very latest programs.