4537a72275
Pure downsampling is slower, but may or may not be more accurate. Though probably not worth it. It did help me realize a minor error, though. The decimator's volume is twice as loud as it should be. Signed-off-by: Christopher Snowhill <kode54@gmail.com>
19 lines
416 B
C
19 lines
416 B
C
#include <stdint.h>
|
|
#include <stdio.h>
|
|
|
|
int main(void) {
|
|
fprintf(stdout, "static const float dsdtofloat[256][8] = {\n");
|
|
|
|
for(size_t i = 0; i < 256; ++i) {
|
|
fprintf(stdout, "\t{ ");
|
|
for(size_t j = 0; j < 8; ++j) {
|
|
if(j) fprintf(stdout, ", ");
|
|
fprintf(stdout, "%s", ((i << j) & 128) ? "+1.0f" : "-1.0f");
|
|
}
|
|
fprintf(stdout, " }%s", (i < 255) ? ",\n" : "\n");
|
|
}
|
|
|
|
fprintf(stdout, "};\n");
|
|
|
|
return 0;
|
|
}
|