|
Hi all; Im still stuck when using the rfft function eventhough Ive followed every advice ive found in the mailing list archive concerning the rfft. Ive compared the result with Matlab and i still find that the result of rfft is wrong. I really need it to work in order to continue my work. I extremely need yr help . below are the steps Ive taken: 1) ive included dsplib.h and Tms320.h 2) ive included 54xdsp.lib 3) ive set my input data in the cmd file as follows .fftdata: {} > IDATA PAGE 1,align(32) 4) could anybody please tell me what i did wrong. my code is as follows: #include <std.h> #include <stdio.h> #include <math.h> #include "dsplib.h" #include <tms320.h> #pragma DATA_SECTION(AA,".fftdata"); short AA[16]={0}; float A[16]={1.0000, 0.7071, 0.0000, -0.7071, -1.0000,-0.7071, 0.0000, 0.7071, 1.0000, 0.7071, 0.0000, -0.7071, -1.0000,-0.7071, 0.0000 , 0.7071}; main() { int i=0; fltoq15(A,AA,16); unpacki_16(AA); cbrev(AA,AA,(ushort)16/2); rfft(AA,16,1); q15tofl(AA,A,16); for(i=0;i<16;i++){ printf(" %f\n",A[i]); } return 0; } I thank you in advance for yr cooperation. Sazali |
rfft problem
Started by ●October 23, 2002
Reply by ●October 24, 20022002-10-24
|
On Wed, 23 Oct 2002 08:31:34 +0200, you wrote: >Hi Brian, > >this is not an exactly answer to your question but it may help you. >Take a look at TI' Application Report spra366 - "Generating Efficient Code >with TMS320". There you'll find some c-coding guidelines. >Organizing your code in a propper way let the compiler produce efficient >code. Thanks for the pointer, I've read that before but its been awhile. What I'm really trying to find out is how much success (or trouble) people have had in taking a 'standard' piece of 'C' code and dropping it into their CCS project. I've had a variable amount of success, some of the problems being mine and others being things that I think the CCS compiler should at least warn about. Thanks, Brian ----------------- Brian C. Lane (W7BCL) Programmer www.shinemicro.com RF, DSP & Microcontroller Design |
