IPC
IPC is how much work a CPU completes per clock cycle, which is why a lower-clocked newer chip often outperforms a higher-clocked older one.
Also known as instructions per cycle, instructions per clock
Not measurable from a browser
A browser cannot read temperatures, power draw, fan speed, VRAM capacity, clock frequency or native FLOPS. For a direct reading use a native monitor such as HWiNFO64, GPU-Z or lm-sensors. SystemCheck contributes the other half: whether the condition is costing you throughput, measured as decay and consistency.
Performance is roughly IPC multiplied by frequency multiplied by core count. Frequency has been nearly flat across recent generations, so almost all generational gain now comes from IPC: wider decode, better branch prediction, larger reorder buffers, more cache. A modern 4.5 GHz core can comfortably beat a decade-old 4.5 GHz core by 60% or more on the same code.
IPC is workload-dependent, not a single constant. A chip with a huge L3 cache posts excellent IPC on workloads that fit in it and ordinary IPC on streaming workloads that do not. Vendor IPC claims are averaged over a benchmark suite and can miss badly on your specific code.
You cannot read IPC directly from a browser. That needs hardware performance counters. What SystemCheck measures is the product: actual throughput on a real workload. If your single-core result is far below other machines running the same nominal clock, IPC differences, cache configuration, or memory latency are the usual explanations.
Related terms