Hey folks,
I'm a huge fan of this effort, so thank you first and foremost for making this data legible.
Over on the staged site (https://framework-tracker.pages.dev/), the methodology sections (e.g.: https://framework-tracker.pages.dev/run-time/#:~:text=39.21ms-,Methodology,-Each%20framework%20renders) do not yet include enough information. For each test run, I would expect to see:
- Device brand and specs (CPU, memory, disk, GPU (if discrete) and GPU rendering status)
- OS version
- Screen resolution(s)
- A note about CPU core tuning (has anything been done to change default frequency scaling? If not, a line saying that)
- Browser and server software versions
- Other isolation procedures used to ensure that the workload is uninterrupted (e.g., via a link to a larger methodology document).
Given that we're also using Lighthouse tests via Puppeteer, it would also be helpful to make some of the caveats about the network and CPU throttling it can apply somewhat plain.
Hey folks,
I'm a huge fan of this effort, so thank you first and foremost for making this data legible.
Over on the staged site (https://framework-tracker.pages.dev/), the methodology sections (e.g.: https://framework-tracker.pages.dev/run-time/#:~:text=39.21ms-,Methodology,-Each%20framework%20renders) do not yet include enough information. For each test run, I would expect to see:
Given that we're also using Lighthouse tests via Puppeteer, it would also be helpful to make some of the caveats about the network and CPU throttling it can apply somewhat plain.