Don't seem to have this problem.

youtube.com works fine for me on Vista x64 SP1.

i have lags for some seconds when video playing
Reproduce:
-press "play" on video
-try to scroll page up and down

look like thread issues, deadlocks may be
i use virtualbox windows xp with 1024mb guest memory on my gentoo linux

I have lags on adobe flash page too. Steps to produce this bug

1) Go to http://flex.org/showcase/
2) After the app loads go to some description of an app and make sure there is a
vertical scrollbar there.
3) Scroll using mouse and the whole browser comes to a standstill.
4) Move out of the browser and try to activate the browser window and it freezes again.

A small correction there.. Scroll using mouse wheel.

Some more info
1) OS: Windows xp.
2) RAM : 1 GB.

This problem is really noticeable on http://espn.go.com/ - scroll up and down, and
compare to FF3.

I have the same problem, of bad performance with flash
I tested on youtube, I put a video playing and put my msn window above it and started
moving it above the browser and the video started slowing down and when I click back
to return to the browser it gets frozen for some seconds (like 15 or 20), I tried a
lot of time and this always happen, but this didn't happened on Firefox 3.1 and Opera 9.5
some more info:
1- Windows XP
2- Ram: 1GB
3- Geforce 9600
4- Athlon xp 3200
between, good job on the browser, I loved it

Same for me. Flash slows the whole system down incredibly in Chrome. Especially 
videos. (Native Windows XP)

The biggest problem is while the flash pages are opened (-> the plug-in is loading).

i had this problem too. BSOD'd my XP machine.

got the following Error signature:  
BCCode : 10000050     BCP1 : E162A01C     BCP2 : 00000000     BCP3 : BF837482
BCP4 : 00000001     OSVer : 5_1_2600     SP : 3_0     Product : 768_1  

my processor goes to 100% when i view flash movies, or when the accesed page contains
some flash objects. win xp sp2, 1 GB ram, p4 2,8 Ghz

same as Comment 3  

Reproduce:
-press "play" on video
-try to scroll page up and down
-browser freezes & consumes a lot of cpu. 

win xp (256MB RAM) on virtualbox on linux 

On Youtube, happens when focusing the video (the chrome Browser process starts to take 100% CPU) or when 
closing a tab with Flash content (easy to test using Youtube). Process returns back to normal after ~4-5 
seconds.

This happens on just about any flash video's I've watched. The lag is terrible, it 
mostly effects you if you try to switch to another tab or if you try to scroll. Well 
even if you try to click buttons. Also it lags the video while the audio is playing 
sometimes too. Tried on a few machines, all get the same effect. Windows XP sp2, 2GB, 
amd 64 3200+ and Windows XP sp3 p4 1.7, 512mb.

(No comment was entered for this change.)

Some flash doesn't render at all.  2 examples:  www.playfirst.com, renders fine in IE
and FF.  Joecartoon.com, renders fine in IE and FF.  

[CONFIRMED]

I spend a decent amount of time watching flash videos and chrome is horrible.
Im using xp pro.

Flash crashes 
Page freezes 
Scroll bars unusable as well as other tabs 
Video stutters 
Video out of sync ( sound is 2-5 seconds ahead of clip) 
Video plays even when that page is closed ( 4-10 seconds after page is 
closed) 
Loading favorites or pages with muliple video thumbnails = mostly 
broken thumbnail links 
Very slow 

comment #2: youtube.com works fine for me on Vista x64 SP1.

I'm using XP Pro SP2, and flash is horrible, with several crashes. Maybe it doesn't 
happen in Vista? Can someone confirm? maybe because communication between processes 
is much better in Vista than in XP?

Happens on youtube, everything gets stuck :|

When viewing a page with a flash or a Youtube video I often experience lockups. The 
browser essentially stalls for short periods of time (5-15 secs). The entire browser 
slows down, not just the tab in question (making chrome's task manager inaccessible for the duration of the slowdown).

it got a lot better in this new version (1798), but it still don't have the same 
performance as firefox, but still the increase on performance were amazing, we're 
almost there, good job!

(No comment was entered for this change.)

+1 on short term "lockups"... following A/B test with other browsers this only 
affects Chrome.

XP SP2, U7600 @1.2GHz, 2GB

The slowdown is worst while the video is loading/buffering. As soon as it is 100% 
loaded, the cpuload drops and makes the window scroll fine.

Easy to reproduce with http://espn.go.com/
IE7, FF3, Safari4 work fine
Chrome149.29 is really bad here

More examples have been reported here:
http://groups.google.com/group/google-chrome-help-troubleshooting/browse_thread/thread/2c1642dc143dab84/7b1269d679587bd5?hl=en

Users have tried updating Java, uninstalling and reinstalling the Flash plugin, and
disabling phishing/malwave protecting in Google Chrome, but they are still seeing the
same problems with all Flash videos hanging and/or crashing the browser.

(No comment was entered for this change.)

Confirm this bug on Win XP SP3 Althlon XP2800+ RAM 2GB
Doesn't happen with all browsers but Chrome

CPU usage was 100% when open a webpage with flash and it slows down the browser after
I kill the the flash process ... thing came back to normal once more.

This workaround was reported on GetSatisfaction and in the Google Chrome Help group:
"I noticed that while streaming there are multiple Google chrome
processes listed in taskmanager. The busiest one of the processes (I
suspect it contains the video player) has a thread priority of below
normal. Setting this priority to normal or higher mitigates the
problem. But I am sure it creates other risks, though I have had no
problems with this workaround it so far. I think google makes this
priority lower deliberately so that the other google chrome processes
can continue running in the event of a plugin freeze up. but as you
can see this leaves the video at very low priority and succeptible to
hanging even more when multitasking while surfing."

5 users are reporting that this work around solved their Flash problems.
http://groups.google.com/group/google-chrome-help-troubleshooting/browse_thread/thread/2c1642dc143dab84/4bbf11e18cc42979

I can reproduce this, but not consistently.  I see it most often on www.nytimes.com.  
Sometimes everything is smooth and fine.  But 2-5 times per day, the browser will 
come to a complete stop for 20+ seconds.  All tabs are affected, and I can't bring up 
the task manager.  I can then get a couple of seconds of use before it freezes for 
another half minute.

In those couple of seconds, I can open the Chrome task manager, and see that one 
nytimes.com tab is using 65% CPU.  The Flash plugin is using 0% CPU and ~15MB RAM.  
If I kill the flash plugin, all tabs immediately return to normal performance.  
Everything is very smooth until I open another page that uses Flash.

Possibly related, but closing tabs that use flash is also somewhat slow.  Frequently 
(but again, not always), when I close a tab viewing a page that uses Flash, the 
browser (including all tabs) freezes for 5-10 seconds, then is fine again.

1.5gHz Pentium M, 1.5 GB RAM, Windows XP Pro SP3.  Absolutely no problems on Firefox 
2, Firefox 3, IE7, IE8, Opera or Safari.

This happened with me and it turns out that flash is serving a storm of timer 
messages. I thought our throttling was causing this so I turned off throttling (by 
binary patching at runtime) of WM_USER+1 and that didn't help. WM_USER+1 started 
storming along with WM_TIMERS as well. Turning the throttling back on caused 
WM_USER+1 to appear lot less frequently but WM_TIMER messages continued.

The timer storm subsided when a tab full of YouTube videos was closed. When a page 
has videos, WM_TIMERS or WM_USER+1 keep firing even after the videos have stopped 
playing.

Need to learn more about how flash works in order to fix this correctly.

I've been able to duplicate this on any single core machine I have tried it on. 

Specifically the "browser" process from the chrome task manager will scale to 99% CPU 
. Note that it is not the flash player process that suddenly starts to use 100% but 
what I imagine is the main controlling process. 

By setting the spawned process containing the flash movie player, priority from below 
normal to normal, it prevents the the CPU starvation by the browser process and 
corrects the problem. 

Recommend logic to to modify starting priority for the Shockwave flash plugin when 
running on a single core cpu. 

This does not occur on firefox or IE. 

Bug confirmed here too.

(No comment was entered for this change.)

I believe the bug referenced by aethergoth regarding the nytimes is a duplicate of 
 bug 792 .

XP SP3, lags w/ various video players (youtube, vimeo) when scrolling w/ mouse wheel.
Celeron 3 GHz, 1 GB RAM.

I confirm that in my case, setting the process from priority 6 (Below Normal) to 8 
(Normal) fixes the problem.

Foolproof repro case: exiting fullscreen mode while watching a YouTube video (by 
pressing Esc) will make all of Chrome hang for a few seconds (very noticeable) on a 
single core machine. Setting the Flash plugin process to normal priority prevents 
this behavior. What's even more interesting is that increasing priority while Chrome 
is hung immediately unblocks it and stops the CPU usage, so it seems like there's 
some kind of race condition going on with Flash continuously generating timer 
messages while it's unable to resume running and the browser process preventing it 
from running by handling those messages and taking up 100% CPU.

I've seen this happen on all single-core machines I've tested on and it's really 
noticeable and annoying. It's so annoying that if I didn't like Chrome so much 
otherwise, I'd stop using it it immediately.

I'd be willing to settle with a quick and dirty fix, which is just to make all 
plugins run at normal priority. Come to think of it this doesn't even seem an 
unreasonable request: pretty much all plugins are user facing, so there is no reason 
why they should be treated as second-class citizens. At least raise the plugins used 
by the active tab to normal priority, like you do with the current tab's process.

www.gamespot.com/xbox360/puzzle/rockband2/video/6194201/?hd=1
XP SP3, AMD Athlon 64 3000+ 1.81GHz 1GB, Gigabyte (ATI) RADEON X300 128MB 1680x1050 
(WideScreen) 16bit
IE6 - Great (though CPU 99%)
Chrome - video plays at a very low frame rate, everything goes slowly, browser gets 
stuck sometimes. CPU at a minimum of 50% but mostly more.
And the priority does not matter, I increased it to any level of priority, 
performance still sucks.

By the way, the Flash plugin remains in the task manager for at least a minute after 
I closed the tab with the flash in it.


Thank you.


PhistucK

Deprecate Area-Plugins label in favor of Area-WebKit and a separate Plugins 
label (reducing number of Area- labels).

See also  issue #387 .  All Flash sites can occasionally lock browser UI for me; Google 
Finance seems particularly egregious.

Hi, we've implemented fixes for these issues.  Can you please try this latest build 
and see if it resolves the issue?  You can download the zip file here: 
http://build.chromium.org/buildbot/snapshots/chromium-rel-xp/2744/chrome-win32.zip

No need to uninstall your current version of Chrome.  Just unzip this and run the 
chrome.exe in the extracted folder.

Please report back to let us know whether it works or not.

Thanks

The new build you provided substantially improves the situation (I have to try hard 
to make it hang as opposed to trying hard to make it not hang when exiting full 
screen mode on YouTube) but it doesn't completely fix the hangs. Repeatedly going to 
fullscreen and back ultimately leads to a few hangs. Setting flash plugin process 
priority to normal, however, still manages to make it go away completely.

If there is a (really) good reason against making the flash process run at normal 
priority, then I'd be willing to agree that this bug is fixed. If the reason is just 
that you're trying to play nice, then I'd say it's not worth it.

Oh BTW, thanks for looking into this. :o)

Confirmed at my side, setting the Flash Plugin process priority from BelowNormal to 
Normal solves the momentary lockup problem in Chrome displaying Flash contents.

So far no side effects under normal use.

-WinXP SP2, Pentium 4 2.8GHz, 1GB RAM

Thanks for looking into this Jam. 

On Chrome 0.2.149.30: 
Flash performance is awful, lots of system lock-ups (sometimes well over a couple 
minutes). Setting flash process priority to Low solved the issue temporarily, and 
would cause flash to crash after extended periods of use. 

On Chromium 0.3.155.0:
Flash performance has improved significantly, didn't have a single lock-up yesterday.  
Won't be able to test it extensively as you can't seem to be able to toss over the 
bookmarks from google chrome into chromium, which is a bummer, but that's a whole 
other issue. Also, I was able to reproduce the "Repeatedly going to 
fullscreen and back ultimately leads to a few hangs." as mentioned by alin. However, 
I had to use hotkey Esc to exit full-screen mode in order to reproduce this hang - 
could not reproduce it using only the mouse and clicking the "close full-screen 
button" and then the "enter full-screen button" repeatedly.  

(No comment was entered for this change.)

This should be fixed in recent build 2744:
http://build.chromium.org/buildbot/snapshots/chromium-rel-xp/2744/chrome-win32.zip

and also in the dev channel release 0.3.154.

Please try this build (by downloading and unzipping the file and running "chrome.exe" 
in the extracted folder) or try the dev channel release.

Please let us know if this solved the problem for you.  Thanks for the report!

WOW, good job its working flawlessly now on windows xp
i loved it =)

In the dev channel release 0.3.154 (on win xp sp3, athlon xp2500+, 1gb ram) the performance for flash video and the pdf plugin is much better now - no 
hanging system anymore :-)

But the 100% cpu hang is still occurring for me on a flash game:

Steps to reproduce:
1. use a single core/cpu machine and play "bubble tanks" http://armorgames.com/play/58/bubble-tanks (wasd to move, mouse1 to shoot)
2. every some seconds the game will become unresponsive for a short time (ca. 0,5 to 5 seconds) and the main chrome process will consume near 100% cpu 
(attached screenshot)

To fix - changing the flash process's priority (2. process in screenshot) from 6 to 8 still resolves the issue.

	chrome.png 17.2 KB   View   Download

Just tried "Bubble tanks" http://armorgames.com/play/58/bubble-tanks
And noticed the "BelowNormal" (e.g. priority 6) Chrome.exe process under the main chrome process 
in Task Manager.

Setting the priority to Normal (e.g. 8) fixed laggy stops and slowdowns..

Windows XP SP3, P4 2.8Ghz, 2GiB RAM, Flash 10, Chrome Official Build 2200
(Chrome/0.2.149.30)

gaspard.leon: We just checked in a fix for this earlier today.  The next build will 
have this fix.

http://www.gamespot.com/xbox360/puzzle/rockband2/video/6194201/?hd=1
XP SP3, AMD Athlon 64 3000+ 1.81GHz 1GB, Gigabyte (ATI) RADEON X300 128MB 1680x1050 
(WideScreen) 32bit
IE6 - Great (though CPU 99%)
Chrome - video plays at a very low frame rate (frames are dropped), everything goes 
slowly. CPU at a minimum of 50% but mostly more for the plugin and generally - 100%.
And the priority does not matter, I increased it to any level of priority, 
performance still sucks.


Moreover - the bandwidth of the streaming video is not being counted in the task 
manager.


Thank you.


PhistucK

 Issue 3072  has been merged into this issue.

As pointed in  issue 5428  - the performance is still suboptimal.
May be it is suboptimal only in a single processor architecture and not in a multi 
processor architecture - but not all of us have the later.

Look at this one - http://www.battleon.com/msg-GuardianMax.asp
Scroll up and down - this is really messed up.

That's some awful performance right there phistuck. 
Lagg is extremely noticable when scrolling that site, maybe half a second or so per 
scroll. 

Win XP Pro SP3, Athlon 2400+ 2,01 GHz, Flash 9, Chromium 0.5.155.0 (Build 6152),
1600x1200 desctop resolution

Result in firefox: no noticeable lagg
Result in IE: very slight lagg

I'll comment in  issue 5428  separately since those are actually two different problems 
(even though they look the same).  This bug was for windowless plugins, the other is 
for windowed.