NotHereToPlayGames

YES! My only conclusion is that these are not "100%" VULKAN files. The UPSTREAM Chromium programmers seem to be using "something" contained within these SO-CALLED "vulkan" files for some other purpose. And forgive the directness, but I lose confidence in those that tell me I do not need these files. I am not a "dumb" computer user, I know my way around computers. I'm no "genius", but I know my way around computers. I need these "vulkan" files on my XP x64.

I have. Regardless of what you and D.Draker think on this matter, I CANNOT LAUNCH 360CHROME ON XP x64 "when I follow your advice" regarding VULKAN FILES.

All I can tell you is that on an XP x64 system, I NEED THESE VULKAN FILES !!! Period!

lol, my Thorium experience is EXTREMELY limited - until when/if it ever becomes "ungoogled".

No matter. Regardless, I am impressed with this flag and so far "it's a keeper". However, I have witnessed these types of flags do great things in Win10 but do NOTHING in XP. I'm kind of waiting to see if @AstroSkipper has success with it on his XP.

That is NOT "my" thread, "for the record".

I wouldn't quite say it that way. I think it was also you (do not remember!) that was suggesting I axe "vulkan" .dll's. THAT suggestion KILLED my Chromium-based on MY computer. So yeah, I started "listening" a tad differently. "Is what it is"

It's still in Ungoogled Chromium v122. I cannot verify Thorium (also at v122!) from here at work.

Wow! I am impressed with --enable-low-end-device-mode !!! Here at work, it's not uncommon for any Chromium-based browser to exceed 4 GIGABYTES OF RAM after a few system hibernate/resume when the Chromium-based browser is running Google Voice + Gmail + Teams Chat + Teams Calendar + YouTube. This will require additional testing, of course. As these four gigabytes of RAM takes an UNKNOWN number of hibernates/resumes. On a computer with 32 GB of RAM, it's never been an issue. But from what I am seeing so far is that the amount of RAM being used by Chromium-based browser is now THE SAME before and after a hibernate/resume. Will monitor. Plans are to KEEP this --enable-low-end-device-mode flag until/if I end up seeing any possible negative side effects.

I can report that here at work, the --enable-low-end-device-mode flag reduces the below from 1.6 GB of RAM to 955 MB !!! 1 Google Voice tab 2 MS Teams tabs (chat + calendar) 1 YouTube tab playing 80s music 1 chrome: // flags tab 13 extensions But this computer (12th Gen Intel Core i7-12850HX with 32 GB RAM running Win10 Enterprise 22H2 certainly does not qualify as "low-end"). The results of this flag most definitely seems to warrant consideration for my default once I am able to try at home.

Did you try --enable-low-end-device-mode? I have not tried this yet.

We need an MSFN Member to step forward that also runs an SSE2 CPU. My ThinkPad T42 is SSE2 but I have no interest in loading Thorium onto it. At least not at the moment.

No offense, but these reports are extremely difficult to follow when the screencap is NOT IN ENGLISH.

Have you parsed "all of" the POSReady updates for what they "do"? ie, maybe "one" of those updates is a sort of "extended kernel" that provides your SSE2 CPU with an SSE3 "workaround"?

Also remember that my era-correct XP single-core Intel Atom N450 is slower performance-wise than your CPU but is SSE3 [I ran SSE2 as requestested for this test] (which I cannot recall offhand, but distinctly recall comparing in the past). SSE3 also works fine for me. Did not notice any gain or loss between the two.

Upgraded to Thorium_SSE2_122.0.6261.171_WINXP_x32 and have been running non-stop for over 15 minutes without any issues. No "gain" in going from .168 to .171. But no "loss" either. Right out of the box. No extensions, no settings changes, all defaults. Browsing works just fine. YouTube will stutter, but YouTube always stutters for me until I add my extensions then YouTube is generally flawless for me also.

I once went to upgrade to v124 but the download's own README cited it as for LINUX. I don't recall seeing a v124 XP version, but I could be mistaken.

I agree. We used to have several XP x86 SP2 forum members. They were "ran off" when they tried to "rep farm".

I'm just thankful that Saxon is on my side for once, LOL.

"no comment"

I have never worked in an IT Department. NEVER. I do use "IT deparment", IN QUOTES, as a REFERENCE to how STUPID our company's "IT department" is!

Fair enough, I can agree with that.

I'm not a fan of "extended kernels". Seems they have been a huge "target for takedowns" at github et alia. Unsure if MSFN moderators "want" projects that github "takes down" to migrate to here. Feel free to ask them directly.

I could do what your followers do - ask for proof of being a mathematician.

Still quite strange! You're a self-proclaimed "mathematician", so as you know, TWO data points do NOT form a hypothesis. Seems to me that the variable that needs changed for this research is no longer POSReady=0 versus POSReady=1 but something else. Can you perform this test between two different CPUs? Both of them with POSReady=0! Both of them with XP + SP3 + post-SP3 thru 2014.