Leaderboard

StartAllBack 3.7.10

TLS 1.3 update for my older browsers: NewMoon 26 for Win2000: http://o.rthost.win/gpc/files1.rt/palemoon-26.5.0-20240601.win2000.7z K-Meleon 74 Goanna 2.2: http://o.rthost.win/gpc/files1.rt/KM74-g22-20240601.win2000.7z Retrozilla (suite): http://o.rthost.win/gpc/files1.rt/retrozilla-suite-tls13-20240526.7z Retrozilla browser (rzbrowser): http://o.rthost.win/gpc/files1.rt/rzbrowser-tls13-20240526.7z K-Meleon 1.5.4: http://o.rthost.win/gpc/files1.rt/K-Meleon1.5.4en-US.tls13.7z TLS 1.3 update files for Netscape 9: http://o.rthost.win/gpc/files1.rt/ns9-nss-update-tls13.7z TLS 1.3 update files for Firefox 1.0 - 1.5 and K-Meleon 0.8 - 1.1.6: http://o.rthost.win/gpc/files1.rt/VC6-nss-update-tls13.7z Classilla 9.3.3/9.3.4b suite: ^ http://o.rthost.win/gpc/files1.rt/classilla-9.3.3-win32-tls13.7z Phoenix 0.5 with Classilla 9.3.3/9.3.4b changes: ^ http://o.rthost.win/gpc/files1.rt/phoenix-0.5-cl933-tls13.7z ^ also comes with fixes for running under Win9x/FAT.

Just a small note. Due to the circumstance that the hardware acceleration does not work properly on my old machine, I have reverted my Chrome XP API Adapter from version 1.2.0.5058 to version 1.2.0.5057 which now runs smoothly again on my machine.

You're welcome! Glad it finally works for you. TBH, your hardware is much more recent and powerful than mine. Therefore, I couldn't imagine that Thorium does not run on your system at all, but does on my ancient box from 2000. The manual execution of the command I posted for you is actually just a test whether the Thorium browser can be started in your system at all or not. I observed in my installation that from time to time the execution of the batch file didn't do anything for whatever reason. Thus, I modified the code of this batch starter file and compiled it to an EXE file which now runs reliably without any problems. BTW, I brought your comment here as a deeper discussion would have been off-topic in @roytam1's browser thread. This thread here is a better place to get help in terms of Thorium related stuff and discuss its issues.

since @ClassicNick provided a newer NSS port, I'm trying to fix/port it to VC7.1 and VC6 for retrozilla use. EDIT: progress - VC7.1 works, VC6 needs more fixing: in NT 3.51: and in Win95: EDIT2: VC6 port is finished. PR created: https://github.com/rn10950/RetroZilla/pull/68 Bin (suite): http://o.rthost.win/gpc/files1.rt/retrozilla-suite-tls13-20240526.7z browser-only (rzbrowser): http://o.rthost.win/gpc/files1.rt/rzbrowser-tls13-20240526.7z

At the moment, I am testing again, but now deeply, the hardware acceleration in Thorium on my old machine equipped with the NVIDIA GeForce 6200 graphic card. And I was indeed able to enable the hardware acceleration in Thorium using both flags --use-angle=d3d9 and --ignore-gpu-blocklist. Incredibly, this mode actually still runs with my old card, and the pages load umpteen times faster than before. However, I can't run hardware accelerated canvas and rasterization together, because then undesirable side effects occur, and everything loads extremely slowly. It's a rather complicated matter of setting up and involves a lot of tests.

The best way to manage your user agent is the extension UserAgent-Switcher and Manager which I recommend and use myself. You can get it here: https://chromewebstore.google.com/detail/user-agent-switcher-and-m/bhchdcejhohfmigjafbampogmaanbfkg You can use this extension in different modes. One is the Black-List Mode: Apply the custom user-agent string to all tabs except the tabs with the following top-level hostnames (comma-separated list of hostnames). Note that even if a window-based user-agent string is set from the toolbar popup, your browser's default user-agent string is used. Another is the White-List Mode: Only apply the custom user-agent string to the tabs with following top-level hostnames. Note that if a window-based user-agent string is set from the toolbar popup, this user-agent will overwrite the global one. The mode that corresponds to your desires is the White-List Mode. Just enter there the domain of Google Drive. Here is a screenshot: And here is the toolbar pop-up window where you enter the mobile user agent string: Greetings from Germany, AstroSkipper

I know why it works with Opera 108, Dixel. Most likely, you ported the developer versions of Opera, which are beta based. In the final release of 122 Chrome they completely removed the ability to switch off the notorious tracking: "User-Agent Client Hints" (v122+) It sucks, sucks big time. Now everyone will know @VistaLover is on Vista x86, and Vistapocalypse is on iOS/iPhone. Documented case. https://github.com/ungoogled-software/ungoogled-chromium/issues/2887

"Kyber (Quantum-Resistant Mechanism) in Chrome 124 Leads to Web Connection Problems" by Luke Jones "Impact on Connectivity and Responses The introduction of the Kyber768 quantum-resistant key agreement algorithm, intended for TLS 1.3 and QUIC connections, has inadvertently caused connectivity issues. System administrators have reported on Reddit and other platforms that web applications, firewalls, and servers are dropping connections following the ClientHello TLS handshake." "To address these connectivity problems, Google has provided a temporary workaround. Users experiencing issues can disable the TLS 1.3 hybridized Kyber support in Chrome by adjusting the browser's flags. Similarly, administrators have the option to disable the PostQuantumKeyAgreementEnabled policy." Source: https://winbuzzer.com/2024/05/02/google-chrome-124-launches-with-quantum-resistant-encryption-causes-connectivity-issues-xcxwbn/

Yes, I recently also started to encounter weird website behaviour, I reported here. https://msfn.org/board/topic/186000-posting-test-string-boot-ini-causes-http-403-forbidden-error/?do=findComment&comment=1264723 Interesting, I have no "forbidden", but the posting sometimes takes long time to pass through. Living in Holland, I have a very fast internet connection.

And what I meant is, if a driver is WHQL, it can't be unofficial, esp. since Microsoft gives that stamp. With WHQL you can be sure it wasn't tampered. I don't know how much it will help with browser acceleration on that old card though.

When I started my test phase with the Thorium browser, the browser start was slow and the page loading behaviour was beyond good and evil. But for some time now, it has been running better and better on my old computer and one of the main reasons was the permanent improvement of the Chrome XP API Adapter. And on the other hand, all the time-consuming configuration work I did, of course. I think this is necessary on old, weak computers under Windows XP.

Sucks because you need to "register".

Nah... "They" were "bad" to begin with ... Updating their code frameworks to target Google Chrome 125+ (which "they" expect every single one of their users to be on at the moment ) is what distances all these services further away from what UXP can (reasonably) digest ... As part of my research into hardware acceleration in the Thorium Browser, I have now taken another look at New Moon 28 and Serpent 52. I realised that hardware acceleration was not running to its full extent. I have now fixed this. And what shall I say, at the moment Amazon and eBay are working perfectly again in these browsers, especially in New Moon 28. And all that on my acient machine. 100% CPU utilisation is no longer an issue, quite the opposite. I am delighted.

Thanks for the hint! I will check this flag in my endless Thorium test marathon.

From what I see, it's a close colab with MS to phase out Win10 (which will be 10 years old very soon). https://learn.microsoft.com/en-us/microsoft-edge/web-platform/how-to-detect-win11 "Websites can differentiate between users on Windows 11 and Windows 10, and detect the CPU architecture of the device, by using User-Agent Client Hints (UA-CH)"

Thanks for clarification, I don't see a way of how to return to the normal spoofing without re-implementing that flag again,

Good! Just don't forget to disable client hints and their persistent storage in flags. Modern sites are smart enough to see through your spoofed UA.

I had written about user agent hacks ages ago' all can be found at my account, people just don't listen, hopefully they will listen to @AstroSkipper. Another important hack is OS bitness in user agent, always choose 32 bit OS. esp. if you're on XP 32 bit! Youtube (and the such) will serve less scripts for 32 bit OS.

So, after some deeper tests, I found the reason why Google Drive didn't crash for me in my fully configured profile. In this profile, I use by default a mobile user agent. The mobile view of Google Drive runs without any crashes and very fast. Therefore, I didn't observe any problems there. Here is my mobile user agent string: Mozilla/5.0 (Linux; Android 9; Tablet) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.6261.168 Mobile Safari/537.36. It can be used as a temporary or general (as Google Drive is much smoother then ) solution until the bug is fixed in the legacy build, too.

Well, whatever the cause, the correct result is no crashing when used as-is with defaults. I noticed he has only one version of Thorium 122 for modern Windows (so no further patching of 122 from initial release), this one doesn't crash. Legacy one does, when run on Win10 or XP. Actually, it appears only Win32 flavor of legacy build crashes, Win64 legacy build is fine. And the one build for modern Windows is available only in Win64 flavor.

BTW, Amazon, eBay and Paypal have become really bad. These sites are much more sluggish in New Monn 28 than months ago.

Recently read about this one, it appears to be one of the more interesting forks of Mozilla Firefox. More knobs in settings, sidebar, keyboard shortcuts; among other things, there's a knob for tab bar below toolbars, some update settings controllable through GUI like in old Firefox versions (there's no "Never check for updates", but the option to choose to install them is there), JPEG-XL support enabled by default, knob for turning on userChromeJS script support... There's also Web Apps feature, I guess equivalent to Chromium's "install website as app". Available x86 and x64 builds for Windows, x64 and aarch64 builds for Linux and a version for macOS (guess whatever's current in the Apple world). Homepage: https://floorp.app/en Downloads: https://floorp.app/en/download Installation instructions: https://docs.floorp.app/en/introduction/installation/

Cloudflare serves cookie consent script for the site, the rest are ads. I block all with extreme prejudice, the editor works fine, apart from the occasional "post goes through although it doesn't show until you refresh". Sometimes, but rarely, there's a delay when posting (server/database overload?).

Same here. Posting code here has been a pure game of chance for months. The forum editor is still totally buggy and constantly spinning around although all has been reported months ago.

Thanks again for your assessment! What I actually meant was the fact that NVIDIA does not offer the driver version 307.90 but 307.83 as the latest version for the Geforce series 6. Generally, I am very careful when it comes to updating graphics card drivers in my very old system, as I have had some very unpleasant experiences in the past. But I will create a current image of my system and then test the version 307.90 offered by Microsoft.

My brightness is at 20%. (Screencap RGB value is value of pixel, I can set brightness at 2% and everything be three different shades of black, the pixel RGB value is still 255,255,255.) I have compared black levels. I've done everything I can here. This "issue" is not an issue to me. Supermium / Thorium / Catsxp / Brave / "real" Chrome/Chromium all display the same EXACT colors on my system! Fonts, active tabs, inactive tabs. ALL THE SAME EXACT COLORS!

They don't like your IP.

But the fonts aren't sharp. They are smudged.

Yes. and some people are sensitive to LED lights, thus using LED back-lit monitors isn't recommended. I still use an old CCFL, and full RGB looks good.

No, everything is fine with your eyes! Starting somewhere in 2012-2013 all Nvidia cards are terribly bright and contrast by default. There is a lot of information comparing them to old AMD (in favour of AMD's default colours).

@AstroSkipper, respectfully, drivers from Microsoft can't be unofficial. Example of waht people think in terms of minor update going from .xx to .90. Microsft labs extensively tested that drivers, especially during that era, with browsers being on one of the first places in the test. Example 347.88 vs 347.90 (Microsoft). "I tried them out, and I definitely prefer them to the 347.88 WHQLs. Seem to be smoother to me." "All good to my experience" https://forums.guru3d.com/threads/347-90-version-on-windows-update.397826/page-3

There's a documented difference with grey colour and its shades at github and by you here (thanks!). For me, the problem is with colours being too light, like on that first screenshot. I run Vista without exkernel.

Hi ; In the past, I have politely asked you to put some more effort into improving the quality of your problem reports, but it appears I can't get the message through ; indeed, "one picture is worth a thousand words", but in this case it's far from it... Please, provide "clickable links" of the actual problematic URLs and also try to include English translation(s) of the error(s) your Russian-localised OS generates ; if the errors you encounter are manifesting themselves ONLY when being logged-in, few here (including the developer) without a proper account already will go into the trouble of registering just to troubleshoot your errors ... As a rule of thumb, social media portals, be it western (e.g. Facebook) or eastern (e.g. VKontakte) are relentless towards anything "legacy", i.e. H/W, OS, browser; they're veritable beasts of Javascript and CSS that put a heavy tax on older desktop machines; I avoid social media with a passion, but if "one" must use them, better limit "oneself" to a mobile device with a recent version of Android/iOS... Small rant aside , I launched latest NM28 [v28.10.7a1 (32-bit) (2024-05-23)] and tried to load: https://vk.com/megaslava The browser struggled for a few seconds (5-8s) while trying to render huge blobs of code , but in the end the URI was successfully displayed: As a further test, I reloaded that same URI ten (10) consecutive times, not once did I get anything similar to what's contained in your first screenshot ; as I don't have a VK account myself, I only tested as an anonymous VK user ... As for your second screenshot, it would appear VK use serviceWorkers, so in "about:config" just enable them (if not already): dom.serviceWorkers.enabled;true FWIW, I didn't get any of the errors depicted in your second image, probably because I'm not being logged-in ... My friendly advice to you: If you can, use one of the XP-compatible Chromium derivatives for VK.com (and likewise for other "offenders" like Instagram/X/Youtube etc.) ... Regards.

Breakage of discourse-based forums/communities under UXP-based browsers (NM28/St52 and probably St55) is an already known issue, that has been reported more times than I'd care to remember ; discourse serve a script that checks for a specific platform feature in the client browser, which isn't present in the UXP platform per se but, at the same time, isn't strictly necessary for the discourse-based forums/communities to function properly - thus, "their" browser-checking script categorises UXP-based browsers as "unsupported" and then "they" serve the "dumbed down" version of the forums ... I have offered two mitigations for this predicament, one based on a custom uBlock Origin filter and another in the form of a userscript (requires a userscript manager to be installed); search my MSFN posts with the keyword "discourse" ; probably inside "My Browser Builds (Part 4, 5)" megathreads...

Thank you, very helpful! members must take example from you!

I think I have already written enough helpful comments and articles. And I replied to a rather disrespectful statement. Thorium is also made for Windows XP. And there are members and readers here who are using this OS on old or very old hardware. You understand what I want to say? Anyway! I don't like such comments "Such old computer ..." which are therefore absolutely superfluous. Such as yours. Sorry to say that!

Look what came out today: v1.2.0.5058 (2024.05.28) [+] XP: Added support for Hardware Accelerated rendering for XP. With it, browser will use less CPU to render pages, and while playing videos. To use, run it with switches: --use-angle=d3d9 --ignore-gpu-blocklist. With it, you will see this features in about:gpu for XP, as you see it on NT6+ systems. Wow! Time to test this out!

No problems here in Thorium. I have just logged into my Google account and Google Drive opens fine here. No crashes to observe. All seems to be normal. And it runs surprisingly quite fast on my old machine.

What are you hardware specs? And which version of Thorium and the Chrome XP API Adapter do you use?

My Thorium installation and configuration is very stable under Windows XP. I use the SSE2 edition which seems to be suitable to my Pentium 4 CPU with SSE2 instruction set at maximum. No browser or tab crashes, no problems with my USB hard disk which is currently connected. At the moment, I am writing this comment from Thorium on my XP machine.

I disagree. I myself would "ignore" these types of "reports" if I were win32. You see, in life, we all place our own "scale" on things like this. On a scale of 1 to 10, this "brightness difference" is a ZERO on my radar. A zero on a scale of 1 to 10. But the three people on this forum that love, love, love to talk "brightness", this issue is clearly a 12 to them. A twelve on a scale of 1 to 10.

I'm starting to think win32 simply has a dull monitor, probably misconfigured videocard/driver setup. It's been reported too many times already, including that github report.

I wrote I tried the programme on a clean, new OS install. There's nothing on it, simply nothing, except the OS components themselves.

Oh, and just noticed EtwRTNT Kernel Logger.etl is pretty much intact and the log file has grown to approx. 10mb, all after I ran the programme on a new OS installation.

On second thought, might as well go Vista, even less space will be wasted on the HDD.

The reason I suggested Win 7 RTM because the user is tight on space, and SP1 doesn't actually add anything useful, and esp. because he wrote the PC will be used for 1 one programme.

I managed to get Python 2.7 going (installed). Pale Moon (64-bit GTK2 version) took slightly over 58 minutes to compile on a Raspberry Pi 5. I did not use --disable-debug-symbols option, used defaults as suggested in build instructions, the source code was downloaded to external WD 5 TB hard disk, the slower kind using SMR tech. Still can't believe the browser works, that was the first time I ever compiled any web browser from source code.

crash

grash https://www.definitions.net/definition/grash