MSN homepage causing high CPU

[hews310]

MSN UK homepage seems to be making the processor go crazy on Firefox, constant 28%+ usage (higher on my laptop), US site seems less bad but still jumps up every few seconds

http://uk.msn.com/

http://www.msn.com/

[xan K]

confirmed. it appears to be caused by the scrolling "latest news" text band. focus your mouse pointer on top of it to make stop and you'll see cpu usage back to normal.

[hews310]

confirmed. it appears to be caused by the scrolling "latest news" text band. focus your mouse pointer on top of it to make stop and you'll see cpu usage back to normal.

Yes that's it, the image change seems to cause the CPU on jump on MSN US.

[hews310]

I don't think it's a FF issue, getting the same on other browsers including IE9.

[Cyberbeing]

It doesn't happen with Chrome Nightly (~2-4% CPU w/ Adblock | ~40% w/o Adblock) vs Firefox Nightly (~50% CPU maxing out its single-thread w/ Adblock | 60-80% split between Firefox & PluginContainer w/o Adblock), so a bug should probably be filed to see if the devs can find a way for Firefox to not use 15x that of Chrome when the flash ad on MSN is blocked.

[hews310]

It doesn't happen with Chrome Nightly (~2-4% CPU w/ Adblock | ~40% w/o Adblock) vs Firefox Nightly (~50% CPU maxing out its single-thread w/ Adblock | 60-80% split between Firefox & PluginContainer w/o Adblock), so a bug should probably be filed to see if the devs can find a way for Firefox to not use 15x that of Chrome when the flash ad on MSN is blocked.

I have alerted MS via the MSN feedback tool, as it uses 40-50% in Internet Explorer 9 as well so I think it's something they need to fix, if anyone knows a better way of alerting MS and/or MSN please do so. This is going to be a potential annoyance/drain a lot of laptop batteries for a lot of people as it automatically goes to the MSN UK page when you log out of Hotmail/Livemail.

[kilara1988]

I get about 2-3% CPU in Chromium too,but only about 15% in FF Nightly.Both with adblock on.I don't know how you guys get 50% on FF.

[Stifu]

I get about 2-3% CPU in Chromium too,but only about 15% in FF Nightly.Both with adblock on.I don't know how you guys get 50% on FF.

Something tells me they don't use the same computer as you do.

[kilara1988]

I know that.And yet the chrome usage is the same as theirs.So i guess their CPU is not that different than mine.I have i5-430m @ 2.27(2.53 with turbo boost).

[hews310]

I get about 2-3% CPU in Chromium too,but only about 15% in FF Nightly.Both with adblock on.I don't know how you guys get 50% on FF.

Get 5-6% CPU total usage (adding up processes) on Chrome/Iron so that is double what you get (Athlon 64 II x2 250 3GHz).

[LordStriker]

IE 9: 6-8 %

Fx 9a1: 11-13%

Maybe ain't bug but Fx can be more optimised on something specific? Actually, I don't have a clue what is going on.

[Stifu]

I know that.And yet the chrome usage is the same as theirs.So i guess their CPU is not that different than mine.I have i5-430m @ 2.27(2.53 with turbo boost).

It's not that simple. You wrongly assume that results will be similar (or proportional) from one computer to another. Someone could have a Chrome that's faster than yours (for certain things), but a Firefox that's slower than yours, or vice versa. And it's not just about the CPU, many other criteria are involved (RAM, possibly GPU...). For example, we could imagine that Firefox works particularly well with your video card, hence your low CPU usage (as the GPU would do more work, and faster). Or just that what Firefox does on that site is more optimized for your type of CPU.

Those are just random examples. Just highlighting the fact others are not necessarily doing something "wrong" to get worse results.

[hews310]

IE 9: 6-8 %

Fx 9a1: 11-13%

Maybe ain't bug but Fx can be more optimised on something specific? Actually, I don't have a clue what is going on.

Was that with the MSN UK page, the US page doesn't have the scrolling news ticker which causes the really high CPU, the US page does have a spike(on mine, 26%) when the featured image cycles.