Simply put, A8 only supports single core, A9 supports dual core. The current smartphone CPU architecture is the ARM architecture of the precision reduction instruction set. The mainstream single cell phone in the market is generally A8, the dual core is basically all A9 architecture.
So what is the difference between A8 and A9, apart from the frequency?
In simple words, the biggest hardware difference is embodied in three key technical indicators, the dual-core processor of the CORTEX-A9 architecture has obvious advantages.
1. Integer computing power
CORTEX-A9 's single core CPU processing capacity can reach 2.50dmips*, and Cortex-a8 Peak can only reach 2.00DMIPS. Dmips is the unit used to measure computer chips. The larger the Dmips value, the stronger the CPU's data processing power. That is to say, the CORTEX-A9 dual-core processor is 2.5 times times more capable of data processing than a mathematically calculated point of view. cortex-a8 (mips:million instructions executed per Second, millions instructions for each second to calculate the processing power of the system in the same second, that is, how many millions instructions are executed per second.) SO,CORTEX-A9 single core CPU computing power is superior to the CORTEX-A8 processor.
2. Floating-point arithmetic ability
In the computer, there are two ways to represent real numbers, one is fixed-point, is the decimal point position is fixed, one is floating point, that is, the decimal point position is not fixed, the calculation method is also more troublesome, usually more than the cost of integer operations much more. Therefore, the floating-point operation capability can also directly reflect the performance of a processor is excellent. The same frequency dual-core cortex-a9 in CPU floating-point operation is higher than A8 performance up to 50%. So cortex-a9 does not simply stack the processor core from the CORTEX-A8, but it has a significant boost in the ability to perform a single core operation.
3. L2 Level Two cache
The L2 level two cache is a temporary memory located between the CPU and memory, which has a smaller capacity than memory but is faster to exchange. The data in the cache is a small part of the memory, but this is a small part of the CPU is about to be accessed in a short time, when the CPU calls a large number of data, you can avoid the memory directly from the cache call, basically do not have to wait to speed up reading speed.
In a word, the level two cache meaning is to save the CPU direct read memory time. So the L2 level two cache has a great effect on CPU performance, so increasing the size of L2 two cache can directly affect the CPU efficiency on the basis of constant CPU working frequency.
Different schema CPU configuration L2 values, A8 can configure the maximum value is 1M, and A9 can be configured the maximum value is 8M.
Of course the number of processor configuration level two cache will ultimately depend on the chip manufacturer's product positioning, price-performance and other factors, but the CORTEX-A9 processor L2 the limit is greater, which means that the same is the same as the top-level configuration scheme, The performance of CORTEX-A9 will be far beyond cortex-a8.