Benchmark by memory, cache, network, concurrency, language, silicon factors

[obriensystems]

Benchmarks bound by for example on chip cache

ObrienlabsDev/blog#91

see work (newer to older)

current generic repo - https://github.com/ObrienlabsDev/performance
https://github.com/ObrienlabsDev/cuda
https://github.com/ObrienlabsDev/collatz
(above extends below)
https://github.com/obrienlabs/benchmark
https://github.com/obrienlabs/collatz/blob/master/collatz-nbi/src/main/java/cloud/obrienlabs/collatz/process/ComputeImpl.java
parallax microcontroller machine language reference
https://github.com/obrienlabs/electronics/blob/master/propCollatz_13_20120317h.spin#L244

Criteria

• Java BigInteger is 10x slower than using native long (but tops out at 63 bits)

[obriensystems]

VS 2022 optimizations

https://learn.microsoft.com/en-us/cpp/build/reference/o-options-optimize-code?view=msvc-170