Is this a bug in double precision data type?

Try this

n=7; eye(n)*inf+ones(n)

I got

ans =

Inf NaN NaN NaN NaN NaN NaN

NaN Inf NaN NaN NaN NaN NaN

NaN NaN Inf NaN NaN NaN NaN

NaN NaN NaN Inf NaN NaN NaN

NaN NaN NaN NaN Inf NaN NaN

NaN NaN NaN NaN NaN Inf NaN

NaN NaN NaN NaN NaN NaN Inf

What I am expecting is the off-diaginal ones are 1's, rather than Nan's

[Dingyu Xue]

Sorry I made a mistake. The answer is correct. To take the same effect, the infinite term inf should be made to a large finite value, say 1e30. The command should be

>> n=7; A=ones(n)+1e30*eye(n)

Answer 1

[Josh]

It's not a bug. What's happening is you're multiplying Inf * 0, which is undefined (it's equivalent to trying to calculate 0/0). Try something like this, it will just set the diagonal elements to Infinity without multiplying the off-diagonal elements by Infinity as well:

n = 7; diag(Inf * ones(n, 1)) + ones(n)

[Dingyu Xue]

Thanks, I made a mistake.