I’m running some tests on a local Nginx web server I created. The web page I’m testing on had 10 images. I compared the results from using plain HTTPS and SPDY, and there doesn’t seem to be any difference at all of web page load time. Can someone advise me on why this is happening?
For my tests I was using the London Chrome WPT with a dial up DSL connection.
Do you have an example test we could look at?
I believe I saw you post to the SPDY discussion list as well and it looked like Will Chan explained what you were likely seeing but if you want us to look here as well, links to test results will likely help.
My bet is that the 10 images are bandwidth constrained and the latency savings from SPDY just aren’t kicking in.
Hi Patrick. Yes that was me. I was able to get the correct performance by switching my ISP.
Thanks for inquiring.
Another question I have. I noted that the servers on webpage test have a 2 minute timeout. Is there anyway to increase this timeout from the users side?
The public instance has the 2 minute timeout to prevent the test agents from being starved/monopolized. My guess is that if you are testing at speeds where that is a problem (dial-up?) then you’re likely bandwidth constrained and SPDY isn’t going to show you any benefits.
If you want to set up a private instance then you can specify longer timeouts: https://sites.google.com/a/webpagetest.org/docs/private-instances