Throttling Method
To simulate/emulate CPU & Network Performance for different types of Devices, Web Performance Testing Tools use different throttling methods.
There are various different throttling techniques such as Simulated & Applied Throttling. Each has its own pros. and cons. and for this reason, ⚡️ SpeedVitals offers both throttling modes.
You can change the Throttling Method after clicking the ⚙️ More Options button on the test page.
Simulated Throttling
In Simulated Throttling, the test is done using an unthrottled CPU & Network Connection and based on the data obtained, it simulates the provided throttling.
While Simuated Throttling is accurate for most web pages, it suffers from inaccuracy in some edge cases. Most Lighthouse Powered Tool including Google Page Speed Insights uses Simulated Throttling for both Network & CPU Throttling.
| PROS | CONS |
|---|---|
| ✅ Fairly accurate for most cases | ❌ Can have inaccuracies in rare cases |
| ✅ Test is completed much faster when compared to Applied Throttling | ❌ Web Vitals Timings on Snapshots can be inaccurate |
| ❌ Web Vitals Timings on Page Load Video can be inaccurate |
Applied Throttling
Applied Throttling is also known as Request-level throttling or Devtools Throttling. In Applied Throttling, actual throttling is applied during the Page Load and for this reason, it is more accurate than Simulated Throttling.
WebPageTest uses Applied Throttling for CPU Throttling.
| PROS | CONS |
|---|---|
| ✅ More accurate than Simulated Throttling | ❌ Not as accurate as Packet-level Throttling |
| ✅ Web Vitals Timings on Snapshots are more accurate | ❌ Tests can take longer to complete when compared to Simulated Throttling |
| ✅ Web Vitals Timings on Page Load Video are more accurate |
For accurate markings of the Web Vitals Timings on the Page Loading Snapshots and Page Load Video, it is highly recommended to use Applied Throttling.