By using this site, you agree to our Privacy Policy and our Terms of Use. Close
CGI-Quality said:
fatslob-:O said:
Not quite 15 TFlops and a bummer that it's using slower clocked HBM2 modules but it's pretty much beast ...

I worked out the math according to the specs and here's what I got ... 

5120(# of CUDA cores)*2(FMA op)*1.455(boost clock) = 14899.2 GFlops which is just shy of 14.9 TFlops ...