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,
I notice that both C7x DSP and C66x DSP have been contained in one TDA4VM SoC, my confusion is:
1. What differences between C7x DSP and C66x DSP? Why do we need both of them?
2. What is the most possible use of C7x DSP in a self-driving vehicle?
3. What is the most possible use of C66x DSP in a self-driving vehicle?
4. Plus, what is the most possible use of 4 non-isolated R5F cores in a self-driving vehicle?
PS: It's the best if you can show an example of reasonable distribution of TDA4VM's resources in a self-driving(like AVP) car.
1. C7x is latest generation DSP from TI and very high level the key difference is (A) It has a Neural Network accelerator (B) It has much wider SIMD data path compared to C6x (512 bit vs 64 bit) (C) More advanced ISA and additional instructions units.
2.C7x DSP with a hardware accelerator is mainly intended to use for Neural Network acceleration
3. C6x DSP is mainly intended to use other computer vision tasks like pre processing and post processing in a neural network based work flow or otherwise as well
4. R5s are used to control multiple hardware accelerator available on TDA4VM like optical flow, Streo disparity Engine, Imagine pipe, and also other peripherals
Please refer the TI's SDK deliverables (www.ti.com/.../PROCESSOR-SDK-J721E) on TDA4VM to understand examples of resource distribution, SDK comes with examples like Auto Valet Parking.
Thanks,
Pramod
Hi Ken Lan, I believe the response addresses your question, and we can close the thread. if anything pending please let us know
Thanks,
Pramod