Does it make any difference with emulator and running on device chrome?

Hi There,
I have been using webpage test for a while now.

  1. I have set up a private instance which has only browser support and not the real device like Dulles VA
    So if i use the emulator config on chrome and run the test, would it still be the same in terms of perf data?

  2. What is the config for 3g fast? If i need to customize the input, since i do not find it in my private instance.

Appreciate if anyone could help clarifying it?

1 Like

No, definitely not “the same” as in “equal” (===). A close approximation? Maybe.

Resource Constrained Devices (ie: Phones, Tablets, etc) have more resource constraints than just network bandwidth and latency that is emulated with traffic shaping. They also have constrained CPUs, power sources, memory, GPUs, etc compared to desktop counter parts. Sometimes they also have older browsers (ie: old iPhones that won’t upgrade past a certain iOS version).

So ultimately it depends on your needs for testing. If you just want to ballpark the difference between your fastest and slowest visitor experience, emulation will probably be just fine. However, if you want to test if a real iPhone 6 can load your site, parse the JavaScript successfully, render the DOM quickly, etc then you probably want a real iPhone 6 to do the testing.

So what is the purpose of testing your site for mobile 3g devices? That purpose statement will dictate if you need to invest in some mobile devices for testing.

2 Likes

The latest connectivity settings are here: webpagetest/connectivity.ini.sample at master · WPO-Foundation/webpagetest · GitHub

If you are running on recent code, you may be able to delete the copied version of connectivity.ini and the code will automatically fall-back to using the “sample” configs that are part of the code.

2 Likes