Current time: 01-23-2021, 03:01 AM Hello There, Guest! (LoginRegister)

Post Reply 
 
Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Understanding my results: slow load times
01-07-2016, 01:02 PM (This post was last modified: 01-07-2016 01:04 PM by Anton Chigurh.)
Post: #14
RE: Understanding my results: slow load times
First of all IMO we muddy the waters quite a bit by testing from various sites, locations, browsers, etc. Because you are introducing multiple variables, such as quality of internet connections, quality of machines, etc. I ONLY test with IE 11 from the Dulles VA location because I know that one to be the most consistent and reliable internet connection and machine I've seen on WPT. (Pretty sure this is Patrick's personal set-up.) I don't like introducing unnecessary variables to the testing, but that's just my preference.

More and more I am seeing FBT problems sometimes even on my own sites, and with nothing changed on my end. I am not sure how reliable the FBT testing in WPT is at this point. Sometimes when you click on the grade, it doesn't show any target FBT, sometimes it does. Plus the host servers themselves are variable, their connection, their traffic and how well they are running can affect FBT. It's a highly variable item in the testing and again, you're dealing in milliseconds there, not seconds.
Quote:The reading I've seen about Clouldflare has been very bipolar (either it's amazing or it's garbage).
I used it for two years and personally never said it's garbage, I simply do not recommend it either. It is proven and even admitted by CF itself that CF interferes with FBT, and really it doesn't do anything for you other than ostensibly, serve your static content to a visitor from a spot geographically closer to that visitor. But when CF is down, you are down and it happens a lot more often than your typical host server down time is.

Good work so far cutting the bloat - From the tests I've run, it shows you have cut your bloat almost in half and the full load times have come down to single digits now. I am seeing anywhere from 6-9 seconds on full load, vs. the 12-19 seconds we were seeing before. This is a much much bigger difference then fixing 500 milliseconds of FBT. (That's on average how much you are missing the target by.) Cutting unnecessary bloat is always good advice and it always helps performance greatly.

Here's one additional test I do when curious about FBT. I pick ONE call out of the waterfall, and run a test on just that link. Always weird to me how FBT gets smaller, the smaller the page load bytes are. I have been told over and over again that there is no relation at all between the kb size of your site, and FBT but never got a explanation of why FBT is always stellar when you're only calling up one file. Take a look - I test ONLY your great big background image:

http://www.webpagetest.org/result/160107_T3_5M4/

And wow, somehow we PASS the FBT test with a A grade when we are loading only a small amount of KB.

Also, just for educational purposes and curiosity, turn off all the plugins then test this site. See what your baseline looks like without all the bells and whistles, just the installation itself. Then of course, turn everything back on.
Find all posts by this user
Quote this message in a reply
Post Reply 


Messages In This Thread
RE: Understanding my results: slow load times - Anton Chigurh - 01-07-2016 01:02 PM

Forum Jump:


User(s) browsing this thread: 1 Guest(s)