Such a study is very difficult ot caary out, see: http://www.chessbase.com/newsdetail.asp?newsid=3465
the best player of all time
You're right but with 2008 engines probably more accurate results than those of 2006. I read somewhere that the engines are up to 3000 or so playing strength. I don't know when they are strong enough to give results with some validity. To figure that out, some gold standard would be necessary; that is some way to use an engine with various games with known analysis done apart from a computer and then see how the computer does, but such a thing does not exist--indeed it is the computer that is supposed to do that. Perhaps it is just a matter of power. If the computer can assess every known move to the end of a game then we would know but that is not possible now and may not be possible until a new level of computing exists and that level of computing would ruin the game. So...your and the articles point is well taken.
I found this
http://www.truechess.com/web/champs.html
here
http://chessmind.powerblogs.com/
after picking it up on main board from
http://blog.chess.com/SonofPearl/gert-ligterink-on-ivanchuks-chess-obsession
very interesting, it analyzes who was best player of all time