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.

TDA4VM: Does TI optimize use of MMA when given non-optimized input?

Part Number: TDA4VM

In the related ticket it is stated that:

Yes your understanding is correct, MMAv1 ( as part of TDA4VM device) processes 64 output channels together and performs one multiplication for 64 spatial positions per channel in single cycle.

So it means 64x64 = 4K MAC ( or 8K OPs) every cycle and at 1 GHz operating frequency gives 8 TOPS.

Would like to verify,
What happens if require only 32 output channels? Would two channels be processed at the same time?