We can roughly calculate the expected improvement based on “non-streaming” factors (8h extra/28h*25h=7.14h of expected improvements that include things like scr

We can roughly calculate the expected improvement based on “non-streaming” factors (8h extra/28h*25h=7.14h of expected improvements that include things like screen refresh rate, decoding and better battery) so therefore 25h new streaming video, minus 12h previous streaming battery, minus 7h expected improvement leaves about 5 hours of streaming performance improvements (WiFi/Network/bug fixes/???). Seems like a lot.