Speedtest.net - higher ping in Firefox

[thewhat]

Just something I noticed... this is a typical result (Firefox 24, Opera 12, Chromium):

It can be anything between 0-10ms, usually it's around 4ms.
Download and upload speeds are the same.

With Firefox, Flash starts its own process (in Task Manager), whereas with Opera and Chromium it doesn't. So that could explain why it behaves differently. But still, it seems like a big-ish difference, doesn't it?

Pingtest.net doesn't show any difference, it only happens with Speedtest.net.

 

[Bubbaleone]

A 5 ms variation in latency isn't even detectable by the human senses. Latency (ping) is just one common measurement of three, that can be used to determine your internet connection quality (from your geographical location), and is not a function of your browser. If you doubt that, run the same ping command from the command prompt and you'll see the same round-trip time. Run the same ping command at different times over a 24 hour period and you'll see a much larger variation, depending on the amount of traffic on the backbone at a given time. Ping is the time it takes for a packet to travel from your computer to a server and back, packet loss is the percentage of packets sent to a server that never arrived, and jitter is the variance in ping measurements where zero means all were the same. 30-40 ms latency is very acceptable but if you're having problems you need to check packet loss and jitter as well.

.

 

[taq8ojh]

5ms is not even margin of error. It's NOTHING. Stop worrying about things that are not problems in the first place.

 

[Dr. Canny]

At this point, since the time difference is so minimal, choosing which browser of those three to use comes down to preference and usability.

 

[thewhat]

Just to clarify why I made this thread, before it gets derailed even further.
It's not about which browser you should use or whether 5ms is noticeable or not, but rather:
Is this actually a thing? Is it reproducible?
Does it perhaps affect non-Flash content? (Or even Flash content outside of Speedtest?)
Why does it happen?
Etc.

 

[Ketchup]

Why don't you give is the average of repeated tests on each browser for the answer? As Bubbaleone mentioned, with one test in each browser, 5 ms is well within margin of error. Not enough to make an affirmative answer one way or the other.

 

[taq8ojh]

Just to clarify why I made this thread, before it gets derailed even further.
It's not about which browser you should use or whether 5ms is noticeable or not, but rather:
Is this actually a thing? Is it reproducible?

I haven't noticed anyone even try to, let alone really derail it. What are you on about? People are trying to explain it to you.

As for whether it's reproducible, uhm, you will have to try for yourself. I am not sure what are you trying to reproduce, specific ping with Speedtest in specific browser? But you're free to keep rerunning the test until you either lose sanity, patience, or reach enlightment Sooner or later you will probably see the same number appear again.

 

[thewhat]

Why don't you give is the average of repeated tests on each browser for the answer? As Bubbaleone mentioned, with one test in each browser, 5 ms is well within margin of error. Not enough to make an affirmative answer one way or the other.

It's clear from my first post that I made more than one test.
I didn't keep a record of them, though. It's just something I noticed while measuring download speeds.
Probably I made around 20 tests in both Firefox and Opera, less in Chromium.

I am not sure what are you trying to reproduce, specific ping with Speedtest in specific browser?

Yes. I thought it was evident. I guess it wasn't, oh well.
But as I said in my second post, I'm interested in this if it affects something other than Speedtest. Because Speedtest alone doesn't interest me much, it just happens to have ping measuring in included.

 

[Bubbaleone]

Just to clarify why I made this thread, before it gets derailed even further.
It's not about which browser you should use or whether 5ms is noticeable or not, but rather:
Is this actually a thing? Is it reproducible?
Does it perhaps affect non-Flash content? (Or even Flash content outside of Speedtest?)
Why does it happen?
Etc.

Since you feel your thread's been derailed, you might consider that neither your first, second, or third post make clear what it is that you're actually asking about. Going back to your first two posts; If you're asking if it's possible to reproduce an average mean latency variation, over several 24-hour periods, the answer is yes. If you're asking if latency affects Flash performance the answer is generally no, not unless you start getting into the 150ms range and even in that range the affect wouldn't be very noticeable. Bandwidth, on the other hand, directly affects how well Flash performs. As to your last questions:

"With Firefox, Flash starts its own process (in Task Manager), whereas with Opera and Chromium it doesn't. So that could explain why it behaves differently. But still, it seems like a big-ish difference, doesn't it?

Pingtest.net doesn't show any difference, it only happens with Speedtest.net."

Opera and Chrome both have built-in Flash, but Firefox uses the Adobe Flash plugin which runs as a seperate process in Task Manager. If you're asking why Opera and Chrome opted for built-in Flash and Firefox didn't, you'll probably get a better answer from the Mozilla forum. I've run all three of the most recent versions of each and haven't seen any discernible difference in Flash performance between them. So you need to better define/explain what you feel is the "big-ish" difference in behavior that you're seeing in Firefox vs Opera and Chrome, and what exactly your use of pingtest.net and speedtest.net has to do with that.

.

 

[thewhat]

If anyone cares, this is what I found:
http://support.mozilla.org/en-US/questions/926258
(4th post)

It appears that the JS code in Speedtest is interpreted differently by (newer versions of) Firefox, which results in higher ping reports.

Mystery solved, I guess.

 

[R0H1T]

If anyone cares, this is what I found:
http://support.mozilla.org/en-US/questions/926258
(4th post)

It appears that the JS code in Speedtest is interpreted differently by (newer versions of) Firefox, which results in higher ping reports.

Mystery solved, I guess.

Umm what, do you even know that FF24 is probably the fastest of the three in terms of JS(check various benchmarks on other tech sites) & the thread you linked to is what a year old

IMO you're looking for something that isn't even there & whatever it is you're trying to prove is lost by your argument about ping tests with a godawful 5ms variation D:

 

[taq8ojh]

IMO you're looking for something that isn't even there

My thoughts exactly.

If you are concerned about stability(?) of your connection, why don't you find any IP address within reasonable low distance, and keep pinging it for 24 hours, checking for packetloss and average (and maximum) ping? That will tell you a lot more than one time ping done through some lousy Java application.