Hello Sir,
I am using ALTFP_LOG megafunction in my application, so when i verify this ip core, i get abnormal results for particular integer values. If i give particular inputs of few numbers to ALTFP_LOG, i was unable to get the accurate values to the given inputs. For example, i'm using counter as input data block to generate some values for the inputs of ALTFP_LOG megafunction, or say that i'm using ALTFP_CONVERT to convert Integer-to-Float from the counter due to input format of ALTFP_LOG megafunction. and for the input values respected output values are given below.. x - ln(x) 1 - 0 2 to 4 - 1 5 to 12 - 2 13 to 33- 3 34 to 92- 4 so on... but if we see the value of ln(92) = 4.5217885770490403096412170747265, but here we see 4 as output.. as we check for the value for some hundred's or thousand's number, the IP ALTFP_LOG megafunction output will not be correct. So i would like to know how accurate this IP core is? what i must do to get expected or accurate result.. Please kindly do the needful..Link Copied
Did you ever look at the float output numbers?
For more complete information about compiler optimizations, see our Optimization Notice.