This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi, TI experts,
My understanding of DCL_runDF22_C5(immediate) DCL_runDF22_C5(partial) is to save clock cycles in calculation compare to a full calculation of DCL_runDF22_C4. but the clock cycles they add together(29+60=89) is more than C4 version which is only 71.
Can you explain?
Thanks!
Hi Shengnan,
The purpose of partial and immediate is for scenarios where you would want to have a saturation clamp (e.g DCL_runClamp_C1), as the full implementation of DF22 (C1/C4) does not have a clamp built-in. In some circumstances, users could also design their program to invoke the partial function in a non time critical section, and subsequently complete the immediate calculation during the time-critical section. In this perspective, we would save some clockcycles during the time-critical section compared with running the full implemention in time-critical section.
Nonetheless, the external assembly implementation should be plenty fast (C1 for full implementation, C2 and C3 if you need a clamp) for most applications. Just be sure to also include the .asm file itself when using these external assembly functions, located under the source folder of DCL.
Best,
Sen Wang