Chrome, Firefox rein in memory-hogging websites Because even 16GB of RAM can feel cramped on a laptop these days. July 28, 2018 https://www.cnet.com/news/chrome-firefox-rein-in-memory-hogging-websites/
To reduce RAM and CPU usage, I use the Tab Suspender extension, which suspends tabs after being inactive for ten minutes. It has a whitelist so you can specify websites you don't want to get suspended. I'm using Cent Browser, which is still using Chromium 67.
There shouldn't be much difference. Its main benefit, is that it is much more configurable than Chrome.
Any suspender is fake. Either it swap to disk (means heavy disk usage) or need reload from web. Less memory? Insert more memory! Same as 50 years ago when computing starts
It does free up memory, which is important to me, as I've only got 8GB and it can also reduce CPU use as well. I often have a number of tabs suspended (sometimes for days).
What's the issue with downloading it again from the web? Chrome doesn't do that natively, so we have to use that extension. It loads in a second back, too.
I did not deny that it free memory. But data just dont disappear, it is stored elsewhere and in common on a drive. And reloaded on demand. That causing higher drive activity. On ssd its fail and reduces its life time. And it raises data activity. Anyhow it has no gain. Its an endless story about a myth like alu helmets... Btw firefox owns such routines itself, but its timing is different.
I'd rather have more free memory and CPU time, than worry about disk activity. I'm sure my SSD will still last for a good many years anyway. Sometimes it is quite some time before I go back to a suspended tab.
I have never seen Firefox using excessive amount of ram. It usually takes between 300-500MB of RAM but I don't load more than 20-30 sites at once. uBlock Origin and uMatrix could help in this regard. user_pref("browser.cache.disk.capacity", 0); user_pref("browser.cache.disk.enable", false); user_pref("browser.cache.disk_cache_ssl", false); user_pref("browser.cache.memory.capacity", 0); user_pref("browser.cache.memory.enable", false); user_pref("browser.cache.offline.enable", false);
Now they come up with this? Why not 10 years ago? I believe Firefox Quantum has got the right approach by limiting the amount of child processes being launched to a maximum of 4. Because at the end of the day, the zillions of spawned browser processes is causing the problem.
In opinion the browsers started to use so much RAM in order to speed things up. Then they needed to isolate sites in their own processes for security reasons. Of course every site needs the same underlying framework, which I guess, lies in its own RAM pages, multiplied over the amount of child processes. So we ended up with todays situation: It is safe, extremely fast, but uses much RAM to achieve this - what, of course, never was a big issue because everyone usually has enough RAM for work- or game-related tasks. At first, I suspect that nobody expected that you would want to have the browser running while you do work or games. So when we all changed our mindset to more and more multithreading, things got a little bit cramped - although I never had an issue with too few RAM personally. I don't tend to have the browser on when I do other things. (Mainly because my PC is crappy) It is generally not a problem to have your RAM 90% full. Speed would be identical. The only problem is having the pagefile enabled when that happens. And the only other problem is having it disabled when that happens. Two problems that are inversely correlated to each other identify the culprit for me: The pagefile-mechanism/Memory mangement. In German the windows pagefile is called: "Auslagerungsdatei", which could be translated to: "outplacement file". That is a good explanation to what it does: It is an extension of your RAM in which the least used pages from the RAM are stored in order to free some space. The space in RAM that remains doesn't even have to be filled with any data, as programs can secure themselves this precious commodity, up to a limit. That's how those RAM cleaner work: They demand all space in RAM from the operation system with highest priority, and the OS is throwing out all the older stuff and pages the other programs wanted for themselves (I forgot haw that is called xD private stuff...) are given to the cleaner - which of course sets it all free afterwards. You can program one for yourself easily. Does all this sound very logical to you? I'd say it shouldn't be possible for a RAM-cleaner to work and there should be no reason for a pagefile - cause that's limiting the actual usable RAM to less than what is physically available, just by the mechanics of managing free space in it! On the other hand, not having this function will make all programs hang and wait for available RAM if there is none. The way we have it now; there is always guaranteed that some RAM is free and that any new process can run. But that is a compromise: It is done to keep everything working even if it does not work optimally. So we ended up with a mechanism that usually works fine until there is not enough RAM left, then everything slows down gradually because more and more programs need to use the pagefile instead and the SSD, rendering it busy writing and reading nonsense - until it all breaks down and nothing works anymore. It does literally pull everything in the abyss if one process wants too much RAM, even with enabled pagefile. (If it is a little aggressive about it) The solution is: Buy enough RAM. It's stupid, but you can't bet on Microsoft to fully change their RAM management because that would most definitely break all programs. (I guess this is also how we ended up with it: To keep backwards-compatibility.) And you can't bet on other developers to introduces big changes in their own software, because that could induce security issues or incompatibility with windows, leading to other slow-downs. Afaik the JAVA-Runtime Environment did it way better. And it is kind of an emulator because of this. Unfortunately it has way too many other problems. I literally have no special knowledge on this field. So please tell me if i'm wrong.
I can still remember when 8 MB RAM (yes, MB not GB) was huge amount of memory. And also the bad days when you had to tweak autoexec.bat and config.sys so that the damn games could get the needed 640 KB of conventional RAM that they needed to run even tought the machine had whopping 8 MB of it (thanks a bunch IBM, DOS and Mr. Gates!) Anyway, coders are just darn lazy nowadays and don't care because "RAM is cheap" (yeah...tell that to someone living in developing country...). And even if there still was few alive that cared about the art of memory optimization, their bosses most probably don't care anyway.... On linux you can do some compression tricks like zram and zswap https://en.wikipedia.org/wiki/Zram https://en.wikipedia.org/wiki/Zswap And on some (not all) embedded devices could maybe use XIP: https://en.wikipedia.org/wiki/Execute_in_place And that's about it ... Wish that others started to optimize memory usage of their apps too. I really hate to think that I probably need to double my 16 GB in the future and max out my machine specs....
Auto Tab Discard as well on Firefox. No more memory problem even when having 150+ Tabs on my 8 GB laptop.
firefox is suspending itself - running temporarly 200+ tabs with 2gig and no problem. ridcoulous when installed more ram - either content is ditched or went elsewhere to be reloaded. you should use about:memory or about: performance (Taskmanager) if you experience memory hogs. about:memory is showing up for each process (Web Content, pid xxx) and website in that process.
Microsoft Could Help Reduce Google Chrome Memory Usage Microsoft already contributing to Chromium development May 8, 2019 https://news.softpedia.com/news/microsoft-could-help-reduce-google-chrome-memory-usage-525948.shtml
Google Chrome’s memory use could drop significantly after adopting a new Windows 10 feature June 18, 2020 https://www.xda-developers.com/google-chrome-memory-use-drop-adopting-windows-10-feature/
thats a development from microsoft for edge https://blogs.windows.com/msedgedev/2020/06/17/improving-memory-usage/
Chrome’s RAM problem fix for Windows 10 lands in Canary update June 26, 2020 https://www.windowslatest.com/2020/06/26/google-chrome-canary-segmentheap-windows-10/
Google Chrome ends Windows 10 memory optimization test after CPU hit https://www.bleepingcomputer.com/ne...ws-10-memory-optimization-test-after-cpu-hit/