We are running a survey to help us improve the experience for all of our members. If you see the survey appear, please take the time to tell us about your experience if you can.
Dear,
I am developing projcet using cortex-A8, I optimize some algorithm code on cortex-A8 cpu.
The code shall be ported to cortex-A15 in near future, I got some information about BEON and VFP from website,the different NEON / VFP version betweem cortex-A8 and cortex-a15.
cortex-a8: SIMD/VFPv3
cortex-a15:SIMD2/VFPv4
could you tell me the detailed difference of noen version between cortex-A8 and cortex-a15?
Thank you in advance.
BRS,
Meng
some complements: algorithm is optimized by NEON instrution.
Thanks again.
Hi Meng,
ARMv7-A/R Architecture Reference Manual, section A1.4.1 "Instruction set architecture extensions". Both Cortex-A8 and Cortex-A15 implement the non-trapping variants.
Really simply put, SIMDv2 and VFPv4 has a fused multiply-accumulate and IEEE754-2008 half-precision (16-bit) floating point support. That's the big, big differences.
Ta,
Matt