Well compared to simply using the "raw" latency increase under load number and describe this as a cost or price (so everybody intuitively understands lower is better), the frequency approach drags in a not strictly necessary division. That raw period/duration nicely avoids all the issues that appear once the period/divisor gets (close to) zero... Also, I like to think about delay in terms of havin a budget (for specific use cases) and I think it useful to be able to decompose that aggregate budget into terms for individual delay introducing steps.
But people clearly did not flock to to raw RTT numbers, so my intuition apparently is not useful guidance... > On Jul 6, 2021, at 16:46, Michael Richardson <m...@sandelman.ca> wrote: > > Jonathan Morton <chromati...@gmail.com> wrote: >> My idea was simply to express delays and latencies as a frequency, in >> Hz, so that "bigger numbers are better", rather than always in >> milliseconds, where "smaller numbers are better". The advantage of Hz >> is that you can directly compare it to framerates of video or >> gameplay. > > Marketing people will get this better. > They already know that higher-Mhz is better :-) > > > _______________________________________________ > Bloat mailing list > Bloat@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/bloat _______________________________________________ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat
