FIR decimation filter error

I'm keep getting this error when using FIR decimation block

This error message means that the calculated word length needed for a fixed-point multiplication operation in your model (in this case, 50 bits) is larger than the maximum word length allowed by the hardware platform you've specified in your "Hardware Implementation" settings within your simulation environment

all i did was connect a Sine wave block from DPS toolbox, and a data type converter to fi(1,25,23), any idea?

[wen]

nvm gotta the data wrong