Leaderboard

OT. Last Vista one was probably based on Chrome 85, if I'm not mistaken. XP, most likely from 2016.

... Over the course of several previous years, "we" (UXP-users) have witnessed how the platform, in agony, struggled to cope with whatever new Javascript "feature" Google devs came up with, the most notorious of them being the non-polyfillable "operators" ("?.", "??"; read this excellent write-up by @InterLinked); at the time, I was semi-convinced that JS operators such as those would deal the final blow to UXP; thankfully, some talented UXP coders came up with own implementations of these operators, thus UXP was "saved" - but for how long? I'll have to side with NHTPG and, to a good extent, AstroSkipper that the last nail on UXP's coffin will be a "new" CSS-related "improvement" (sprinkle with a generous dose of sarcasm here ), implemented in a mixture of CSS code inside/alongside JS code (which seems to be the trend recently ); in the case of UXP, MCP's upstream (Mozilla) have CSS features implemented in Servo/Rust, so MCP can't backport easily, if at all, Mozilla code for these and get away with it; CSS-related UXP issues will keep piling up inside their tracker, with very dim prospect ... Case in point is a CSS bug that has been affecting GitHub in particular for many months now, it was acknowledged exactly one year ago, but so far no activity at all towards a resolution ... This particular "feature" was implemented only in Chromium 99, so it does "invalidate" the popular 360EEv13 and/or KafanMiniBrowser variants on XP/Vista x86; only implemented in Firefox 97, thus putting Mypal68 out of the game, too; if it weren't for the new "additions" (Supermium and its semi-fork Thorium) that appeared at a most opportune time , "we" retro-OS-ers (I just devised this term, ) would have run out of options...

You're indeed right! It very well can be, seeing they already call it HTTP/3, I still hope it won't happen very soon because the format is flawed.

Thanks for the new version! Unfortunately, I have to report a new issue on Codeberg.org. When opening a website there (here in Mypal 68.14.1b using single-process mode), the following message is displayed at the top of the site: JavaScript error: e.contentBoxSize[0] is undefined (https://codeberg.org/assets/js/webcomponents.js?v=7.0.0-140-34a85c1536~gitea-1.22.0 @ 10:32190). Open browser console to see more details. Here is a screenshot of the browser console: And when closing the opened Codeberg site by clicking the close button, the complete browser closes immediately which means it crashes without any message. Can you confirm this?

You have other sites opened in Supermium, this obviously adds performance penalty. Msfn, github and reddit. Esp. githun and reddit are heavy.

Probably you hit the size limit. Try to delete your old attachments. It helped me, I can embed since then.

It is not really a necessary feature. You can change that in a very simple way inside your OS. Even by using a third party tool. Anyway! In general, New Moon 28 and Serpent 52 will get what has been developed for Pale Moon and Basilisk.

Hmm... are any of these work on XP? CatsXP doesn't even start on Vista without Kernel.

The last update of the GoodbyeAds-YouTube-AdBlock-Filter filter list seems to be performed 4 months ago. So, for me, not that old at the moment. However, if no more updates are rolled out, I will unfortunately have to remove this list from my mod uBlock Origin Legacy. PS: Not every list is maintained on a regular base. There are many lists maintained by only one person or updated when time is available. So, users have to be patient, unfortunately. But you can disable such a list at any time if you think, it has become obsolete.

https://msfn.org/board/topic/186133-thorium/?do=findComment&comment=1264771

Let me understand this correctly -- you're concerned about security implications of a protocol currently used by the majority of the world, this in the context of Windows XP (an OS that hasn't received security updates for roughly a decade)?

Seems to be switched off by default. At least, in my installation for testing.

Try this - without quotes! Despite what other users/user may tell you. --disable-features=WebUsbDeviceDetection

Knocking on UDP, trying to make connections via QUIC, broadcasting to literally everything and everyone is the default chrome behaviour. The question is, why (if both @win32 and Alexi are trying to implement at least something similar to Ungoogled) didn't cut it out first. Seems like a very logical (and rather simple) first step.

WinNTSetup 5.3.5 - removed redudant code, WimHost options removed - updated wimlib to version 1.14.4 - enforce modded wimlib - HideProductKeys option defaults to 1 - minwim crash fixed - minwim speed improved - minwim extended log files - fixed WinRE GUI bugs - fixed crash installing WinXP

Those they change manually in the end, to target a specific OS version. I think win32 wrote about it somewhere.

AstroSkipper, which sites did you test it with? I have the same issue Dave wrote about, memory leaks (and then crashes, sometimes just hangs) while scrolling instagram. I have 16Gb of RAM currently, can extend to 32.

... 99.99% of times , when the OS generates that error upon trying to launch an ".exe" file, it means that the "Sub System Version" value inside the EXE's PE Header has been set (by the compiler) to a figure higher than the one corresponding to the current OS (e.g. XP=5.1, XPx64=5.2, Vista=6.0, Win7=6.1, Win8=6.2, Win8.1=6.3, Win10/11=10.0); using special software to modify that field of the PE header to an appropriate value for the current OS will allow for the executable to launch, unless some other kernel dependency isn't being fulfilled (in which case the OS will generate an appropriate ERROR message, different to the initial one ) ... Below screengrab is with the AVX2 Thorium variant: TL;DR; ALL the builds you tried have been configured to require Win10+ out-of-the-box ...

In case you missed, Hybrid Post Quantum Key + QUIC (Kyber draft "X25519Kyber768Draft00"), instructions on bruteforcing that weak "encryption" method. https://msfn.org/board/topic/186176-x25519kyber768draft00-hybrid-post-quantum-secure-key/?do=findComment&comment=1265284

@Dave-H, would you be willing to try out --disable-low-res-tiling? --disable-low-res-tiling reduces cpu usage and saves power https://codereview.chromium.org/196473007/ https://stackoverflow.com/questions/66825943/how-to-reduce-chromedriver-cpu-usage-in-selenium-python

With our without HPET, yes, a bit faster start, like Dave wrote. But I'm not on a single core. Those on a single core would've seen a faster startup.

It's very strange, it's not related to CSS, nowhere near it. And the author of Thorium suggests that flag himself. Unfortunately (or fortunately?), I can't link to his site because he hosts Windows cr*cks, pirated loaders on it. But I can make a quote, those interested may find themselves. "--disable-highres-timer[1] ## Disables high-resolution timer on Windows."

I think their discussion was specifically about DNS over HTTPS, which wasn't a thing until long after XP EOS. @UCyborg brought DNS over HTTPS into play. However, the original enquiry only mentioned DNS. Hence my answer.

Here is a brief report of my testing so far if someone is interested. Thorium can actually be installed on such an old computer under Windows XP. The browser starts up still quite bearable compared to other browsers on my system. However, as always with Chrome browsers on low-performance computers under Windows, the page loading behaviour is problematic. Although I have configured all conceivable flags to increase performance and selected the settings to suit my hardware, loading web pages takes considerably much longer than with New Moon, Serpent or Mypal. Of course, websites can be loaded that no longer function properly in the already mentioned browsers. Surprisingly, RAM consumption is not a problem, it is rather perfect although Thorium is running in multi-process mode. Interestingly, it seems that I have even found a solution for my installation that considerably speeds up the loading of web pages and therefore the browser start, too. It's not a new one, though. One doesn't have to reinvent the wheel, either. But first, this method must still be thoroughly tested.

True, but you can only directly configure Windows versions released in last 4 years or so to use DOH (eg. https://www.howtogeek.com/765940/how-to-enable-dns-over-https-on-windows-11/). Older ones require to use a sort of DNS proxy or special support from the browser side. In Windows XP, you can either directly insert the desired DNS service in the native network configuration window, or you do it more convenient with a very few clicks by using a third-party tool like, for example, ChrisPC DNS Switch. I use the Pro version which has more features and is more convenient.

I had the exact same problem on WinXP SP3. I installed the "Segoe UI" fonts from Win10. Problem solved!

Supermium is the same as usual Chrome, but with tweaks to make it work on old OS. GDI rendering instead of Dwrite for XP, rewritten memory functioning. Supermium understands Chrome flags. Some flags added, some removed. All of the usual connections to Google servers are present.

They say you need to launch from a specially created icon (shortcut), did you do that? https://github.com/win32ss/supermium/issues/531#issue-2275990442

Guess I'm lucky I'm not everyone and Pale Moon still manages all "important" sites. The latter have certainly gotten worse, though I don't care about GitHub, X (Twitter), Facebook, Instagram, DRM encumbered media...pretty much don't care for 99% of the internet. Don't care about Firefox either, it has become a (buggy) monstrosity I don't recognize anymore and I have no time nor interest for seeking workarounds for stuff that recently worked as expected. I've set it up quite some time ago for the family member, assuming it will be "safer" option than Pale Moon...well that was a mistake, Firefox doesn't play videos properly anymore on a humble 5 years old laptop with Intel HD 610, it either lags and the sound is screwed up (with HW acceleration) or it lags and the sound is OK (without HW acceleration). Put Pale Moon on it and it's been smooth sailing ever since.

I'm not up to arguing (especially when there's no need to). "Microsoft ends Windows 7 security updates | TechCrunch" Windows 7 is a legacy OS. https://techcrunch.com/2023/01/10/microsoft-ends-windows-7-security-updates/

I am very satisfied with the efficiency of uBlock Origin Legacy when it comes to filtering and blocking unwanted content. Especially when you consider how old the original extension actually is. However, the decisive factor is the correct configuration and sensible selection of effective filter lists that are still compatible with uBlock Origin Legacy. And of course, the filters and rules you create yourself are also crucial, as they make everything even more interesting and effective. If the user proceeds in this way, then the combination of PBCF filter list and user script is the icing on the cake.

On my weak machine, these unnecessary connections do have an impact to my CPU and are counterproductive for smoother browsing. So, I am glad to have got rid of them.

No need to add any firewall rules, it's not a routable address. I did not add any specific rules. In Windows 10 Firewall Control, Thorium is restricted to the standard web browser zone as all other browsers, too. The connection to the posted address was not allowed in this zone. Therefore, it was blocked automatically. No more, no less.

Wow! Is this thread here about hardware? And performance comparisons between Win XP and Win 10? The answer is crystal clear: Definitely not! This thread should actually provide information about the Thorium browser. Thus, the last 7 posts are therefore completely off-topic. The reader will not learn anything worth knowing about Thorium here. No matter how you feel about this browser.

You will always find the latest version of New Moon 28 in this thread: https://msfn.org/board/topic/185966-my-browser-builds-part-5/?do=findComment&comment=1264369 The link points to the latest post with the most recent version. In general, you can find all releases from @roytam1 on his server: https://o.rthost.win/

Yes, it forces the connection to that specific IPv4 IP address. Just the same as forcing a block of a connection to a site by: 0.0.0.0 evilsite.com For example, I have www.facebook.com like that in my hosts file and when I try to get DNS info about it with the `dig` command in Linux, this is the answer: dig www.facebook.com ; <<>> DiG 9.16.48-Ubuntu <<>> www.facebook.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 23139 ;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1 ;; OPT PSEUDOSECTION: ; EDNS: version: 0, flags:; udp: 65494 ;; QUESTION SECTION: ;www.facebook.com. IN A ;; Query time: 0 msec ;; SERVER: 127.0.0.53#53(127.0.0.53) ;; WHEN: Tue May 07 16:21:57 CEST 2024 ;; MSG SIZE rcvd: 45 ANSWER: 0 ; Query time: 0 msec You're only in trouble when the IP address of void.gr changes. That's up to the maintainer of uBO Legacy. What you could also do, is tell the maintainer of the list (@kargig) about his website issue on GitHub.

:OT Unfortunately, there are really no forks of browsers which bring newer browser versions to older Android versions. Thus, my old, beloved Samsung Galaxy S3 with the old Android 4.3 stucks on very old browser versions. When setting up my new Galaxy smartphone, I will probably root the S3 device to upgrade it to Android 7. Just for fun, of course! The good is my tablet is on Android 9 and still gets the most recent browser versions from the Chrome world. :END OF OT

Oh, not again! Kyber has been obsolete and outdated for ages, I warned you, you don't listen, I honestly don't understand why. D.Draker never gives bad advice. On the link below, you can enjoy the precise instructions on bruteforcing that weak "encryption" method. "Challenge 1: Breaking baby Kyber." "After a couple of hours (or way less depending on your configurations), we obtain the following solution..." Decrypting the communications: "From this point, having recovered the four server private keys (one for each communication), the rest of the challenge is straightforward. Having access to the ciphertext from the pcap file, we can use the OQS_KEM_kyber_NS_decaps function to recover the 32-byte shared secret. Here is the example code for the case NS5" https://blog.quarkslab.com/sandboxaq-ctf-2024.html

Yes, looks very weird it's still not there, I mean for a fifteen years old totally out-of-support buggy OS.

Yeah when I tested 24H2 briefly on my main machine control center was overhauled but borked. Leave it to the world class engineers at Microsoft. I wish they'd get back to the way they were in the 90s and 2000s.

Thorium actually starts relatively fast compared to other Chrome browsers under Windows XP, even on low-performance hardware like mine. Of course, it has first to be configured well.

Thanks! I will test this flag.

--disable-highres-timer speeds up considerably.

I think you posted in the wrong thread. This is about @roytam1's browsers and not Supermium. There are no Chrome processes here, fortunately.

@win32

Strange, it doesn't mention whether these fixes were applied. https://chromereleases.googleblog.com/2024/04/stable-channel-update-for-desktop_24.html

Usually, I do not use any Chrome or Chromium browsers under Windows XP. My daily drivers under Windows XP are New Moon 28, Mypal 68 and Serpent 52. And there is no such behaviour.

When uBlock Origin is installed, it controls the preloading of websites. You do not have to disable uBlock Origin to control this feature again, you can simply enable the option "Disable pre-fetching" in uBO's dashboard page "Settings" as already mentioned by @Dixel.

Zeal