Is the High Price of RAM a Good Opportunity to Promote FreeBSD?

[ruby R53]

What I have not seen is benchmarks themed "what happens when you run out of RAM?". The basics wouldn't be too difficult to do, but there are so many variables to consider once you get down to metal. For example, how does it score if one OS already killed a process and the other one did not? Did that OS do a good job or not? Obviously if the other OS managed to pull through with all processes unkilled it would win this benchmark. But what if it becomes unresponsive for 25 minutes while doing so? Is that still "better"?

and yeah i'd say that's almost like the OS is shooting itself on the foot because if the whole thing becomes unresponsive that means the task it was doing can't keep going well, so it'd be better for the OS to just kill it and the user would have to come up with a way to make it work again

As I said from anecdotical evidence my eyes absorbed in the last decades FreeBSD is really good at pulling through without killing and excessive slowdowns. Just good decision making when paging.

yeah i gotta give FBSD bonus points for that

My FreeBSD runs out of RAM and everything keeps working fine. I just notice now some occasional delays when I change from one app to another.

View attachment 24920

what kinds of apps are you using there? i have 16 gigs of RAM with no swap and it rarely goes that far, even when compiling heavy stuff

 

[AlfredoLlaquet]

what kinds of apps are you using there? i have 16 gigs of RAM with no swap and it rarely goes that far, even when compiling heavy stuff

Two Windows 11 VirtualBox VMs with 4 GB RAM assigned to each one and a lot of other stuff. Normally I don't have so many apps running at the same time. I just wanted to show that my FreeBSD keeps all the apps running almost seamlessly even when I run out of RAM.

 

[rbranco]

zram & zswap is a secret goal. The github repo mentioned in the wiki disappeared:

Making sure you're not a bot!

 

[ruby R53]

Two Windows 11 VirtualBox VMs with 4 GB RAM assigned to each one and a lot of other stuff. Normally I don't have so many apps running at the same time. I just wanted to show that my FreeBSD keeps all the apps running almost seamlessly even when I run out of RAM.

i see, well i think compiling www/chromium alone would already be more stressful to your pc than all of that lol

 

[AlfredoLlaquet]

stressful to your pc than all of that lol

You are assuming things that you don't know and demeaning me. It's not nice.

 

[ruby R53]

it was a joke···

 

[Zare]

Phoronix.com

I found this; https://www.phoronix.com/review/windows-linux-ryzen-ai/9

The synthetic average of results is almost the equal. Some of the Ubuntu tests were compiled with flags - no mention about the Windows counterpart.

Looking through all test pages, they outperform eachother at tests/categories, and at a lot they perform the same. Under no circumstance I would call Ubuntu a clear cut winner here.

 

[cracauer@]

You generally want the gaming benchmarks:

Google Search

 

[Espionage724]

A Windows 10/11 desktop without debloating it extensively? It seems surprising.

Windows seeming;y uses more RAM if its available; I've seen 1.1-1.3GB use 4G RAM minimal debloating (don't have to debloat what doesn't exist LTSC ), and my usual laptop does 1.7-2GB with 16GB (stuff like DWM and Explorer are less than 100MB; Intel's GPU + RST driver stuff adds a little vs computers that don't need those drivers).

GNOME I've seen 1.1-1.3GB min, Plasma 6 slightly higher, and Xfce 600-900MB. On my webserver FreeBSD is about 600MB now, but I've seen 400MB use randomly in the past on Linux relatively same nginx/PHP stuff. Overall FreeBSD seems on-par with reported RAM use vs other OSs.

 

[bgavin]

Windows is very chatty and gobbles up RAM. Excess above the loaded kernel is used for file system cache.

The ideal way to debloat windows is at install time, not by post install bloat removal. This is done with an XML script used by WinPE at install time which configures the windows installer to not use unselected components.