Leaderboard

It seems Google is re-considering adding JPEG-XL support in Chromium https://bugs.chromium.org/p/chromium/issues/detail?id=1451807 On this point at least we are a step forward.

you can check if browser is using driver's acceleration in Graphics section in about:support

Hi again ; I did not want to give very detailed explanations on this, because..., well, some MSFN members are "allergic" to long posts of mine ... But indulge yourself: My NM28 tests were done on the 32-bit compile of the default (SSE2+) latest build offered by roytam1; its default UA string (on my system) is: Mozilla/5.0 (Windows NT 6.0; rv:4.8) Goanna/20230731 PaleMoon/28.10.7a1 Without modifying anything in about:config, "userstyles.org" starts to load, but then barfs with a Checking indeed the Web Console, I see the "block" has to do with a Google () related URL, https://accounts.google.com/gsi/client By creating the below SSUAO for "google.com", general.useragent.override.google.com;Mozilla/5.0 (%OS_SLICE% rv:65.0) Gecko/20100101 Firefox/65.0 that "client-side exception" is being mitigated and, consequently, the page does now load! Incidentally, that "google.com" SSUAO will allow you now to be served the most recent incarnation of the https://www.google.com/ search engine result pages, instead of a dumped-down version "legacy" browsers are being served (Fx < 65.0) . My primary UXP browser is St52; this one comes with a native (built-in) SSUAO for google.com: general.useragent.override.google.com;Mozilla/5.0 (%OS_SLICE% rv:71.0) Gecko/20100101 Firefox/71.0 Basilisk/52.9.0 This one is enough to allow the loading of a "basic" "userstyles.org" homepage, but pages for individual userstyles end up as blank ; I had to impersonate Chrome () on Serpent 52 (but NOT on NM28) for the "site" to work properly... general.useragent.override.userstyles.org;Chrome/86.0.4240.198 (in fact, just the word "Chrome" should be enough ...) Kind regards .

... Looks like you have a lot to catch up with ... The official yt-dlp has long ago abandoned support for CPython 3.4 and, thus, the Windows XP OS... My personal records/archives indicate that yt-dlp v2021.06.09 is the last official binary that can launch under XP: [debug] Command-line config: ['-v'] WARNING: Support for Python version 3.4 have been deprecated and will break in future versions of yt-dlp! Update to Python 3.6 or above [debug] Encodings: locale cp1253, fs mbcs, out cp737, pref cp1253 [debug] yt-dlp version 2021.06.09 (exe) [debug] Python version 3.4.4 (CPython 32bit) - Windows-Vista-6.0.6003-SP2 [debug] exe versions: none [debug] Proxy map: {} Usage: yt-dlp_x86 [OPTIONS] URL [URL...] yt-dlp_x86: error: You must provide at least one URL. Type yt-dlp --help to see a list of all options. Up to current times, the 32-bit official binary, yt-dlp_x86.exe, is being compiled with CPython 3.7, which requires Vista SP2 32-bit and higher ... The yt-dlp.exe official binary you mentioned is, in fact, 64-bit, compiled with CPython 3.8 and requires Windows 7 SP1 x64 and higher ; additionally, the yt-dlp devs like/want to use the "latest and greatest" ( ) Python version (not the case, though, with the current youtube-dl maintainer ) and CPython 3.7/3.8 versions (3.7 EoL'ed by the PSF by now) might/will become unsupported in the near future ... The yt-dlp binaries offered here in this thread have been compiled with "community" produced Python versions (patched - 3.7 - or recompiled from modified source - 3.8, 3.9) that have restored WinXP (and Vista , in the case of 3.8/3.9) support ... Best regards .

Many thanks for the explanations, @VistaLover! Quintessence: i'll look in this thread here more often, to get a well maintained yt-dlp version for my classic XP and 7 environments. Best regards!

Not only Home, unfortunately. "Windows Update Turns Itself Back On – How To Fix" "Now, here comes the question, what if Windows Update turns itself back on? This problem occurs now and then and has bothered a lot of users. Noticing this, I decide to provide some practical solutions to stop the Windows Update service." https://www.minitool.com/news/fix-windows-update-turns-itself-back-on.html

Interesting, but I have no extensions that were installed that way. I do have 'Developer Mode' enabled though, perhaps that's the key. Universal Extractor is a wonderful thing, I've been using it for years, and it rarely fails in my experience.

There's a trick not everyone knows, all unpacked extensions that were then installed in "developer mode", will never update themselves in any Chrome. This extractor will do the job, I use UniExtract 2 to extract .crx extensions. https://github.com/Bioruebe/UniExtract2 Universal Extractor 2 v2.0.0 RC 3 The method is proven to work 100%.

I read it gets back on its own, I don't have links at hand now, sorry. But it was the reason people started to simply remove this service from their ISOs completely, like he did, before the install. And I think delaying doesn't stop from downloading them, so they still accumulate on HDD.

For starters, this GPU model is not even properly detected, judging by the screenshots. Memory size: N/A and so on.

If GPU-Z shows the GPU Load on Windows 7 but not on Windows XP, it is most probably a driver problem. Additionally, you can try other tools as for example Process Hacker or Process Explorer which can also show the GPU load. If they all fail, it is definitely a driver problem. @we3fan! Maybe, it will comfort you a little, in my Windows XP system the GPU Load is not displayed, either. This is probably due to hardware and drivers. Presumably, there is simply no sensor for measuring the GPU Load. I have a very old graphics card installed there, an NVIDIA GeForce 6200.

If GPU-Z shows the GPU Load on Windows 7 but not on Windows XP, it is most probably a driver problem. Additionally, you can try other tools as for example Process Hacker or Process Explorer which can also show the GPU load. If they all fail, it is definitely a driver or missing sensor problem.

What errors exactly? There is indeed some issue at the moment (YT changed something again...😡), as described here. RegexNotFoundError: Unable to extract yt initial data Maintainer @dirkf is aware of it and I'm confident he will eventually fix it. I think it's related to this yt-dlp issue, which is fixed now. Just rename `yt-dlp.exe` to `youtube-dl.exe`.

Officially it doesn't, I think the question was about the official support. For 116 to start, it needs to be done like this. I also saw some reports of video decoding/playback troubles. The installer of 116 is blocked for anything below Win 10. About the video and text problems, I can't confirm or deny, I don't use Firefox myself. https://www.reddit.com/r/windows7/comments/15i9dfd/firefox_116_on_windows_7/

Thanks guys. @AstroSkipper Same happens in Process Hacker and Process Explorer, shows GPU Load on Win 7, doesn't show it on XP (I guess we are both missing some resource ). I know the chance is slim to show the GPU Load sensor for me on XP, should I try new Chipset Driver and new Video Driver? @Dixel Yes my GPU is pretty weak, but GPU Load sensor works properly on Win 7: 0% when Idle, 21% when playing video. Since it works on Win 7, I thought to try make it work on XP also.

Ah, I see. A non-SSE2 version. I thought you meant the source was compatible with 3.4.

Latest yt-dlp for XP is also uploaded to my GitHub repo. Direct link: https://github.com/nicolaasjan/yt-dlp/releases/latest/download/yt-dlp_x86_Windows-XP.zip For Windows 7 you can also still use one from the official Release files, or one from the nightly builds. The 64bit version (yt-dlp.exe) is built with Python 3.8.10 and the 32bit version (yt-dlp_x86.exe) is built with Python 3.7.9, so that one will work on Vista as well. However support for 3.7 wil probably be dropped soon...

How did you do that? On a Windows 7 VM with Python 3.4.4 32bit and the latest source, I get: python "yt_dlp\__main__.py" -v "https://www.youtube.com/watch?v=8kl6q_9qZOs" Traceback (most recent call last): File "yt_dlp\__main__.py", line 14, in <module> import yt_dlp File "C:\Users\nico\Desktop\yt-dlp-master\yt_dlp\__init__.py", line 5 f'You are using an unsupported version of Python. Only Python versions 3.7 and above are supported by yt-dlp') # noqa: F541 ^ SyntaxError: invalid syntax

It depends on whether you have to use these codecs for your specific encoding/decoding. libsvtav1: So for being able to play AV1 video, libdav1d is enough for me. libdav1d: However, I prefer to save only h264 video (avc codec; ≤1080p) from YouTube, because that is the only format that plays hardware accelerated (via iGPU) on my PC. yt-dlp command: yt-dlp -f bestvideo[height<=1080][ext=mp4][vcodec^=avc]+bestaudio[ext=m4a]/best[ext=mp4]/best "URL" libvmaf: No idea why I should need that.

I remember, when I actively used 360explorer in 2021, my uBlock (that was not installed in developer mode), I simply dragged and dropped the .crx file, got itself updated to a newer version, which then acted slowly. Also, "normally" installed (crx drag and drop) extensions' update behaviour pattern is heavily dependent on the used UA.

I never tried them myself, though I am under the impression any somewhat modern browser should do video accelerating for decades now. Esp. on Nvidia and Intel GPU.

Thanks for the link! Never heard of this driver before. I use the GeForce Release 307.83 WHQL which was actually the last official driver for my NVIDIA GeForce 6200. But we don't want to go into that here, as the whole topic is off-topic. This thread is still about @roytam1's browser releases.

This thread is now locked. I'm sure I don't need to explain why.

GPU usage is not relevant in this case, you need to look at Video Engine load.

It will then not install, says manifest is damaged, better replace with zeroes. Like: https://0.0.0.0.org/board/topic/

GPU-Z does seem to work in XP x64 on my GeForce GT 635. But it's pretty much always at 0% with a 1% blip very rarely. YouTube 1080p videos would bring it up to between 6 and 8%. Nothing seemed to bring it any higher than 8%.

Hello @Reino! Allow the website userstyles.org completely in eMatrix! Then, check if you have done that correctly: For google.com, I use @VistaLover's SSUAO Mozilla/5.0 (%OS_SLICE% rv:65.0) Gecko/20100101 Firefox/65.0. Here is a proof that it works as already reported: This is the link of the website shown in my screenshot above: https://userstyles.org/styles/180927/disable-css-animations-and-transitions-safely If you still have problems accessing this page in New Moon 28, simply delete the browser cache and reload it! I had the same problem. Lately, this website has really become crappy for UXP browsers, unfortunately. Cheers, AstroSkipper

For me the problem is not one of whether Win7 will support the latest version of FF it is how long the FF versions it can run are viable for at least general browsing. I still use a much earlier version for that purpose and another even earlier version which works for a WinXP VM on the same PC. But increasingly web page functionality, including posting here, does not work correctly. That is only going to get worse but if FF Quantum v115ESR follows a similar ageing pattern you're still going to be able to use it for a lot of online purposes, like browsing forums, 5+ years or more from now which means on any Windows version that can run it.

The website https://developer.mozilla.org/en-US/docs/Web/CSS/animation causes 100% utilisation of my CPU on my old computer. Animations of any kind are now killers of old, underpowered computers. I have already deactivated various animations in different ways in my browsers. Now, I have also installed the extension Stylem and installed the following CSS style via copy and save from userstyles.org: https://userstyles.org/styles/180927/disable-css-animations-and-transitions-safely This actually stops any CSS animation on the site above and brings my processor back to normal.

Userstyles.org now works in NM 28 and loads the site with a specific style. There is a button Install but clicking onto it doesn't install anything and shows a message I have to install the Chrome extension. However, I use the extension Stylem and thought it works for that site. I can only copy the provided CSS code and create an own file. Or do I something wrong?

Ok, I have managed it. You must only change the UA for general.useragent.override.google.com and not the one for userstyles.org. Thanks you very much for your help, @VistaLover!

Setting the UA to Mozilla/5.0 (%OS_SLICE% rv:102.0) Gecko/20100101 Firefox/102.0 fixes the loading of the main page, but clicking onto any theme or style leads to an empty window as it is the case in Serpent 52. @VistaLover Same effect when creating general.useragent.override.google.com with your UA Mozilla/5.0 (%OS_SLICE% rv:65.0) Gecko/20100101 Firefox/65.0 .

In Serpent 52 (32-bit) (2023-08-10), the main site of userstyles.org can be loaded, but clicking onto any theme or style leads to an empty window.

Sure, but those businesses should (in theory) have experts capable of mitigating them, no need to (further) scare to death common users about vulnerabiliities they won't ever experience (simply because they don't use the affected services/programs). If we want to scare them, as general advice, we have the (evergreen) opinion by Armand Gracious : https://www.dedoimedo.com/computers/experts.html jaclaz

@win32, as I don't have a github acc., answering the question "Why does Geforce Now claim this Browser is unsupported?" here. Try to completely disable ClientHints, not to spoof them, as I was able to use Geforce Now on an old Chrome 106, spoofed as 111 via my app. I got no unsupported warnings.

Correction : Mozilla, aping Google to the letter, but with a delay of a few years, put Vista SP2 on the same boat as XP SP3; FxESR 52.9.x was/is the last version of Firefox to run on (plain) Vista SP2; release channel v53.0.x. could be "hacked" to launch there, too, but with serious shortcomings in the media playback department... TL:DR: Fx 53.0+ requires Win7 SP1 and higher (obviously, I'm not talking here about Vista's 64-bit Extended Kernel implementations ) ...

Nightly no longer launches due to a missing entry point. The last one working after editing through CFF Explorer is 2023-07-18-09-09-25-38.

Latest Thunderbird upgrade supports Windows 7: https://www.thunderbird.net/en-US/thunderbird/115.0/releasenotes/ However, it seems version 116 beta will require Windows 10.

It is currently still trivial to revert the patch from (https://hg.mozilla.org/mozilla-central/rev/f8db4b5e5d95) to the installer and then compile the installer using ./mach package from a Mozilla build environment (https://firefox-source-docs.mozilla.org/setup/windows_build.html). I now have a working nightly installer again that works on Windows 7.

Then we need to turn off the bloody DirectWrite for Chrome on Windows 7 completely. I know they removed that flag a long time ago, but in some chinese forks - the flag still works, the problem is, no chinese forks on v.111 exist yet. I myself use a chinese fork I re-butchered for my needs, it's based on Chrome 102 and it has the flag to switch the DirectWrite off completely. What do you think of this idea ? We can compare and see the portion of that code in the fork. Also, don't forget - starting from 105 or so, everything is only soft, no hard rendering for windows 7 anymore. Surely it would be slower due to this also. EDIT: @win32see this comment and try to test the browser with krenel from October, locally. I just tried it and the 110 version loads up even faster and the responsiveness is amazing. https://msfn.org/board/topic/181612-wip-windows-vista-extended-kernel/?do=findComment&comment=1236240

I assume you talk about the 111 version ? Did you try to apply the huge list of flags I gave you ? A few seconds is still slow, on my 14 years old PC (2009) with mechanical HDD Chrome 110 starts in 0.5 sec or smth like that. That is with that flags and the Dixel's launcher. I f you still use the old launcher, it needs to be re-butchered because starting from 92 (or so) chrome had changed the user data dir path.

This method is also based on my discovery (from my ported Opera 96 on Chrome 110), and I'm against sharing it in public, I already told @win32 about it. Why ? Simply because you all could easily end up with nothing when the evil developers read this topic. Just wait for new kernel to be ready, you will all get it then ! Be patient . Besides, you ask about win 8.1, which is not being discussed here.

Isn't this version an early stage alpha ? No wonder it crashes then, like another user wrote.

Fake news. Proved by several members, including me. For example here. There is a screenshot that with "nosandbox" applied, doesn't help. https://msfn.org/board/topic/184046-future-of-chrome-on-windows-7/?do=findComment&comment=1230307

Chrome 110 doesn't even start on Windows 7, there are plenty of posts confirming this in this stopic, just scroll back.

It supposed to ! This topic tells not working on Windows 7.

Busy on holidays, maybe ?

"Clever" Indian spammer ?

You already know my point of view in terms of Russia or unwanted, hidden connectoins to whatever. I avoid connections to many countries if possible. Especially, if I need privacy, anonymity and security! OpenVPN has a huge impact on the CPU, but in the case you are equipped with a powerful CPU, it's no problem at all. In my Windows XP system works a single core CPU from 2000, and OpenVPN is not really a good choice. But it works! The native protocols of Winfows XP work much better and have a very small impact on the CPU. Simply great if still supported by a VPN service.

I have no idea what you have with Russia all the time. We are talking about VPN. You have to be able to separate one from the other. Objectivity has top priority. A VPN connection if established works for all browsers and programs. It's a system-wide connection to servers or, simply put, to the internet. Much more convenient. And many proxy servers are dubious, slow and unsafe. The only good proxy servers I used in the past were paid ones. You have to find good ones, and these are very rare if existent at all. But, as I already said, and this will definitely be the last time, proxy servers is not the topic here.