When I tried machine with AVX512_VNNI, The int8 model is still faster than fp32 model FP32: SGEMM INT8: no message 2. I used a machine that is not support VNNI to run CNN last time. I changed the machine, the following is the information of CPU Architecture: x86_6...
新发布的Geekbench6将VNNI/int 8 AI图像处理引入到测试中,并且实际上归类为整数测试,总分受到整数项目影响大于浮点项目,多线程方面降低并行度,强化高吞吐带宽影响,高带宽或者带宽效率更高的系统将在多核测试中占优,而某品牌DDR5带宽实际堪比DDR4,将在多核测试中落后,苹果的超高带宽系统将因此在多个子项中大幅提高成绩...