Recommended Posts

IE6 SP2 opening the neowin front page:

http://img202.exs.cx/img202/4834/ieneowin6dl.png

Not that I'm saying memory leaks are acceptable, but they're a lot more common than you seem to want to believe. Also, most of the firefox ones seem to be in the javascript engine, for what it's worth.

vcv is a MS fanboy. take a look at another vcv inspired thread

https://www.neowin.net/forum/index.php?showtopic=277420&hl=

585515165[/snapback]

I will restate: I do not use IE, I hate it, and I think it's a piece of crap. Please stop trying to derail this thread, I don't appreciate it.

Not that I'm saying memory leaks are acceptable, but they're a lot more common than you seem to want to believe. Also, most of the firefox ones seem to be in the javascript engine, for what it's worth.
Yeah, memory leaks happen, but in smaller numbers. It is common for programs to miss a few things here and there, but not to this calibur.
Extensions:

I have about 20 installed.  I will run again on a clean install.  However, extensions are not responsible for the memory management, FF is.  So this is still a problem regardless of whether or not its the extensions.

585515129[/snapback]

ha, yes please do a clean install. before doing any kind of test like that especially when your going to cluelessly post about your findings with a WARNING! you should do a clean install without any extensions or themes installed...its just common sense. That would be like benchmarking your system after using if for 2 years and after having spyware and virii infected os.

Get the daily nightly from here once this phrase turns into a link (

The Official Win32 2005-02-22 [Trunk] build is out.) which should be in about 2 hours. Do a clean install also...fresh profile!

ha, yes please do a clean install. before doing any kind of test like that especially when your going to cluelessly post about your findings with a WARNING! you should do a clean install without any extensions or themes installed...its just common sense. That would be like benchmarking your system after using if for 2 years and after having spyware and virii infected os.

Wait.. so because I had a lot of extensions install, it's ok for firefox to have that many memory leaks? I don't get that logic at all. Again, extensions do not manage their own memory!

Dude you should know by now you can't say anything about firefox thats not kissing its ass. Now all the lame ass want to be cool kids use that browser and think its the best thing since sliced cheese and makes them cool, so they're not interested in hearing about memory leaks. To them if its not amazingly good news it has to be a lie!

This is just typical Firefox thread.

Someone report Firefox problem, and that someone is now troll and a MS fanboy.

If you want to live in a delusion that Firefox is just perfect, go ahead, but there are people who want to improve it, and this kind of report will contribute for the future of Firefox.

StaticFish, mircleman, joshpo, you're the reason why Firefox is getting a bad reputation as "browser for fanboys".

Have fun with your memory leak, while supernova and vcv is trying to fix the problem.

I only ran firefox for 60-90 seconds and closed it.? Once the process was gone from my list, I looked at the memory leaks.? Firefox can not free memory once it is closed.

585515159[/snapback]

Not that I'm completely sure about this but... Windows may not free memory instantly after an application is closed (usally *.dll files though). Do other applications free memory at once?

ha, yes please do a clean install. before doing any kind of test like that especially when your going to cluelessly post about your findings with a WARNING! you should do a clean install without any extensions or themes installed...its just common sense. That would be like benchmarking your system after using if for 2 years and after having spyware and virii infected os.

Wait.. so because I had a lot of extensions install, it's ok for firefox to have that many memory leaks? I don't get that logic at all. Again, extensions do not manage their own memory!

585515182[/snapback]

No doubt. How many people using FX run it with NO extensions? Especially since it needs an extension for practically everything? That's regardless of the fact of if the extensions are what is not freeing the memory or not?

FF 1.0 fresh, no extensions:

27 leaks(s) found in 1.765 seconds

So it seems to be a problem with mismanaging extensions and/or javascript.

This sucks, I hate using FF w/o extensions :(

585515191[/snapback]

it looks like some of your extensions could be causing the leaks. Why not add one at a time to find out which ones?

Despite being a firefox fanboy myself, I do agree with vcv regarding the Memory Leak problem in Firefox or, in a larger view, the Mozilla Platform. Memory definitely should be managed by the Platform, not by individual extensions. Imagine the Gecko Runtime Environment (GRE) became something like Java Runtime Environment (JRE) in Mozilla 2.0 Platform, and the memory leak problem still persists. That is why Java has garbage collector - to free unreferenced memories.

Anyway, the memory leak problem is getting fixed while we speak. It is just not that easy as one may imagine trying to track down memory leak problems.

-bYtE,Feb 22 2005, 08:55]Not that I'm completely sure about this but... Windows may not free memory instantly after an application is closed (usally *.dll files though). Do other applacation free memory at once?

585515196[/snapback]

Yes.

Firefox Nightly (2/21) / no extensions:

14 leak(s) found in 1.562 seconds

Edited by vcv

IE opened - 2 web pages, then closed

21 Leaks in .873 seconds

If you use aggressive settings - then 31, etc..

I would suggest you fully understand what the tool is telling you - before you put on your chicken little outfit, and start screaming about memory leaks, memory leaks ;)

Yes memory leaks can be a pain in the as_ - and I agree they sux.. But before you go around yelling about the sky falling, you might want to verify the test software your using is any good, and that you understand how it works, and what it is telling you, etc..

So you want from 6000 something to 27 -- hmmm something wrong with your testing methods? ;)

IE opened - 2 web pages, then closed

21 Leaks in .873 seconds

If you use aggressive settings - then 31, etc..

I would suggest you fully understand what the tool is telling you - before you put on your chicken little outfit, and start screaming about memory leaks, memory leaks ;)

Yes memory leaks can be a pain in the as_ - and I agree they sux..  But before you go around yelling about the sky falling, you might want to verify the test software your using is any good, and that you understand how it works, and what it is telling you, etc..

So you want from 6000 something to 27 -- hmmm something wrong with your testing methods? ;)

585515224[/snapback]

I fully understand how the program works, I've been using it on my own software for a long time.

It went from 6000 to 27 after getting rid of extensions. That's not a problem with the testing methods, that's a problem with the extensions.

When you view sites with a lot of high res images, Firefox will use a lot of memory. I've seen it as high as 240MB.

When you close tabs it will not automatically free all of that memory. Shortly after you close the last instance of FireFox it will. It is not a "true" memory leak. It is not even a bug, it is done for performance-enhancing purposes.

The same situation happens on Epiphany on Linux. If you clear your cache it will free up the (image cache) memory instantly.

Therefore my advice would be to clear your cache after you've done a lot of wallpaper browsing and are then ready to return to Neowin.

It's not a big deal.

Not that I'm completely sure about this but... Windows may not free memory instantly after an application is closed (usally *.dll files though). Do other applications free memory at once?

585515196[/snapback]

The memory is always freed by windows when the process terminates. It is not zeroed out, or cleared, or anything.. But it is freed and made available for reuse.

So, theoretically you can get away with not freeing your memory, and letting windows do it for you when your program exits, but it goes without saying that this won't work if you expect your program to run for any prolonged amount of time.

Even the smallest leak matters when the program could potentially be running for hours/weeks/months at a time. Let's say a web server leaks 4 bytes on every request, over time this will bring the server to a crawl as it runs out of memory and turns to paging.

I've noticed the same thing with Firefox. If you keep the same session running for a few days, it's not uncommon for it to use absurd amounts of memory.

When you view sites with a lot of high res images, Firefox will use a lot of memory.  I've seen it as high as 240MB.

When you close tabs it will not automatically free all of that memory.  Shortly after you close the last instance of FireFox it will.  It is not a "true" memory leak.  It is not even a bug, it is done for performance-enhancing purposes.

The same situation happens on Epiphany on Linux.  If you clear your cache it will free up the (image cache) memory instantly.

Therefore my advice would be to clear your cache after you've done a lot of wallpaper browsing and are then ready to return to Neowin.

It's not a big deal.

585515234[/snapback]

No. You don't seem to understand how this works. This is memory profiled when the program is closed, when it's supposed to free ALL memory. This has nothing to do with total memory usage, but rather amount of memory allocated vs. freed.

When you view sites with a lot of high res images, Firefox will use a lot of memory.  I've seen it as high as 240MB.

When you close tabs it will not automatically free all of that memory.  Shortly after you close the last instance of FireFox it will.  It is not a "true" memory leak.  It is not even a bug, it is done for performance-enhancing purposes.

If by "close the last instance" you mean quit the Firefox process completely, then that is indeed a true memory leak, and a bug. Not freeing your own memory and leaving Windows to clean up your mess after you terminate is hardly justifiable, even for "performance-enhancing purposes."

And even if you mean that it keeps the images in the ram cache (as opposed to disk cache,) after you close the the tab/window, that is still poor design.

I think your still learning... and your original statement should have been. "Is this normal for FF?" not "OMFG LOOK AT TIHS ISSUEZ!!!" You didnt even properly/throughly test your theory/statement before posting. However, Its nice to know now due to this thread to check extentions before I use them. (BTW: Netscape used to do the same thing with the browser hogging memory until you closed all windows, they had a neat little "Maximum Amount of memory" or something like that in settings of the browser. Maybe FF/Moz should impliment that back?)

I think your still learning... and your original statement should have been. "Is this normal for FF?" not "OMFG LOOK AT TIHS ISSUEZ!!!" You didnt even properly/throughly test your theory/statement before posting. However, Its nice to know now due to this thread to check extentions before I use them.
I don't follow you. I found memory leaks in Firefox, and I'm still learning? I made no theory, and my statement still stands.. that firefox has some major memory management issues.

From 1400 to... wait 14000 to 30 +/-... Thats a huge plummit in "memory leaks". I use "" due to how the term is being thrown around. I'm not going to sit here and argue all day, point being, its good you found them. Thanks for pointing them out. Next time make sure you double check your "statements" to make sure they are well thought out. Considering 20ish found is normal for most apps, I wouldnt quite consider this issue MAJOR... however they still should be addressed.

When you view sites with a lot of high res images, Firefox will use a lot of memory.  I've seen it as high as 240MB.

When you close tabs it will not automatically free all of that memory.  Shortly after you close the last instance of FireFox it will.  It is not a "true" memory leak.  It is not even a bug, it is done for performance-enhancing purposes.

The same situation happens on Epiphany on Linux.  If you clear your cache it will free up the (image cache) memory instantly.

Therefore my advice would be to clear your cache after you've done a lot of wallpaper browsing and are then ready to return to Neowin.

It's not a big deal.

585515234[/snapback]

Its a big deal when you only have 256mb ram :(

Dude you should know by now you can't say anything about firefox thats not kissing its ass. Now all the lame ass want to be cool kids use that browser and think its the best thing since sliced cheese and makes them cool, so they're not interested in hearing about memory leaks. To them if its not amazingly good news it has to be a lie!

585515189[/snapback]

Thats pretty much what I said in the first page...

The FX analists will never ever admit somethings wrong :laugh:

Theyll try and blame it on something else, and may I ask what a "nightly build" is gonna fix? certainly not the rest of the 20 million or so peoples Fx browsers .. :whistle:

I bought this memory issue up once before and was told a wide range of things it couldve been, but not once did anyone say I might be right or be on to something here...

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Thanks for the advice guys, I'll give my current GPU a clean and then run the benchmarks to see how similar it is. Hopefully a bit of a clean-out will help it last a bit longer.
    • Pretty sure those will still be in this game. The series' well-explored psychosis themes will return as well.
    • "performance issues? what performance issues?!"
    • Microsoft making much needed change to Windows 11, 10 Patch Tuesday security updates by Sayan Sen Recently, Microsoft delivered its latest Defender patches for Windows 11 ISOs. These definitions are released from time to time alongside the general security updates available during Patch Tuesday. Speaking of Defender, the company has now announced another important change that affects how security updates are delivered to enterprise devices running Windows. According to a recent announcement, Microsoft Defender for Endpoint's endpoint detection and response (EDR) updates will no longer be bundled with the monthly Windows security updates or Patch Tuesdays. Instead the company is shifting delivery of these updates to Microsoft Update, bringing EDR servicing in line with several other Microsoft Defender components. If you recall, Microsoft last year moved PowerShell updates to Microsoft Update (MU) as well since it provides automatic updates for Microsoft products and services. Thus the move is intended to allow Microsoft to deliver EDR improvements and security enhancements independently of the OS's regular monthly update cycle; this should enable faster deployment of protection updates without requiring organizations to wait for the next Patch release. For those unfamiliar, Microsoft Defender for Endpoint's EDR capabilities are designed to help organizations detect, investigate, and respond to advanced threats across managed devices. Keeping these components updated is critical for maintaining protection against evolving attack techniques. The rollout has already began for Windows 10 devices in late May 2026 (last month) and Microsoft says it will gradually expand support to Windows 11 and the remaining supported Windows versions over the coming months. The company expects deployment across Windows 10 and Windows 11 to be completed by fall 2026 or around Q3 of this year. Once the transition is complete, EDR updates will be delivered through Microsoft Update using KB5005292, provided the required prerequisite updates have already been installed. Microsoft is also introducing a new Defender Update Service as part of the change. Following installation of the first update, devices will automatically create a new directory located at %ProgramData%\Microsoft\Microsoft Defender\Defender Update. Microsoft notes that restarts may occasionally be necessary in case of "rare" failure scenarios. For most organizations, the tech giant says no action will be required as long as Microsoft Update is already permitted within their update management strategy. Admins who rely on manually deployed update packages, however, will need to adjust their processes to ensure the new Defender update package is included. Microsoft also recommends reviewing internal documentation and notifying helpdesk and security operations teams about the updated delivery mechanism to avoid confusion during the transition. As a prerequisite, the tech giant notes that systems must be running Sense version 10.8798.25857.1000 or later and have one of the following Windows updates (or later) installed: Win11 24H2 KB5062660 (2025-07 Cumulative Update Preview) Win11 23H2 KB5062663 (2025-07 Cumulative Update Preview) Win11 22H2 KB5062663 (2025-07 Cumulative Update Preview) Win10 22H2 KB5062649 (2025-07 Cumulative Update Preview) Win10 1809 KB5063877 (2025-08 Cumulative Update) Server 2019 KB5063877 (2025-08 Cumulative Update) Server 2022 KB5063880 (2025-08 Cumulative Update) Server 2025 KB5063878 (2025-08 Cumulative Update) As always, organizations should verify that their update policies align with the new servicing approach before the broader rollout reaches all supported Windows platforms later this year. In case of major problems, the EDR update can be rolled back to the inbox version stored in %ProgramFiles%\\Windows Defender Advanced Threat Protection (ATP) using: MpCmdRun.exe -RevertMde -Product Edr -ToVersion Inbox For those who have access to the Microsoft 365 Admin Center portal, you can view the message here under ID MC1381119.
    • My problem with these smart scopes is that IMO they're not really an entrance into astronomy, they're an entrance to looking at pretty(ish) pictures that you take by essentially pressing a button and letting the scope do the work. I still maintain that getting some binoculars or a solid dedicated telescope (which doesn't have to mean expensive) and actually spending time learning the night sky and using a telescope is a much better way to actually learn. But, granted, the learning curve is a bit steeper (as it tends to be).
  • Recent Achievements

    • One Month Later
      DJC50PLUS earned a badge
      One Month Later
    • Week One Done
      DJC50PLUS earned a badge
      Week One Done
    • Proficient
      Eric Biran went up a rank
      Proficient
    • Dedicated
      Conjor earned a badge
      Dedicated
    • Week One Done
      Windows Guy earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      493
    2. 2
      PsYcHoKiLLa
      248
    3. 3
      Steven P.
      73
    4. 4
      +Edouard
      69
    5. 5
      neufuse
      68
  • Tell a friend

    Love Neowin? Tell a friend!