5409 | 5416 | AD5 | ADC | BIOS | Boot | Booting | Bootloader | C540 | C5402 | C5409 | C5416 | CCS | Codec | DMA | Dmad | DSK | DSKPlus | Dsplib | EVM | FFT | FIR | Flash | GPIO | HPI | Initialization | Interrupt | JTAG | LOG_printf | MCBSP | RFFT | RTDX | Sampling | STLM | UART | VC540

Hi,

I'm facing the following problem: (i use the C5416 which is not float)

I have to multiply two signed 16 bits words (let's say a and b).
So, 2^(-15)< a < 2^(15), 2^(-15)< b < 2^(15) result will be
2^(-30)< ab < 2^(30).
Thus, the result (axb) will be a double. The thing is that i would like a signed  16 bits
because i'm doing a rifft of the result and because i'm working with 16 bits words. 
So, I thought to normalize the result by dividing by 2^15. But it gives   me a bad result after
ifft.

Is there a simple way to do it? 
Am i doing something wrong?

thx 

getthematrix.      

------------------------------------

Check Out Industry's First Single-Chip, Multi-Format, Real-Time HD Video Transcoding Solution
for Commercial & Consumer End Equipment: www.ti.com/dm6467

(You need to be a member of c54x -- send a blank email to c54x-subscribe@yahoogroups.com )