UCyborg

Good to know, at least users of poor legacy versions can be aware they're not missing anything in that regard compared to stock up-to-date version. I rarely have YouTube tab opened for such long periods, wonder if would've got it in the past under similar circumstances. Interestingly, AdNauseam is forked off uBO. Another extension that will never be ported to UXP.

So this is what some YouTube users have been seeing, second time I witnessed it, previous was yesterday when I didn't think to make a screenshot and this tab has been opened since Wednesday evening, with computer asleep when I was at work or sleeping. uBO's counter disappeared with detaching the tab to its own window to make a screenshot with less baggage, it counted over 4000 blocked requests. I wonder if upstream, much more sophisticated version of uBO can prevent it and what part of it (or filter) does. This popup pauses playback, you can close it and resume. On an unrelated note, bug (or feature?) of the browser, it closes dev tools window when you detach the tab, so you lose whatever temporary data was logged, set in there.

Aye, old low-end computers emphasize the slowness even more. Slower response is common on better hardware as well, even with it comes tab switching. St52 can be made slightly more responsive with multi-process mode, which has its own issues. My slowest computer has a dual-core AMD E1-6010 APU running at 1,35 GHz, just 2 GB of RAM (with 256 MB used as video memory) and a 32-bit Win10 1809. It's fun, use Pale Moon for too long and it hangs, even with some free RAM. Could upgrade it at some point, get more RAM and replace Win10 with 64-bit version, but don't feel it to be urgent as long as I have another faster computer available. I guess the weight of discontentment with Google and Mozilla's ways of developing a web browser is greater for me than the weight of drawbacks of an UXP based web browser.

It was just an observation that it currently happens to still launch on that particular Windows version. I did user older versions of VS more often in the past, but given programming is not for me and that I abandoned the older projects I messed with, it's not a serious need at this time.

If I would've assumed, I'd be wrong, because I'd think DWM would only ever put on screen what was specifically rendered by the application. I tried on Windows 8.1 and it does appear to be influenced by DWM. You can force it off there with some hackery, when it's on, empty window does appear white when background color is not specified, but it does switch to selected color if specified without flashing white. Though we should assume application may not have the need to specify background color, so the whiteness will remain until the client area is filled with content. With DWM forced off, the client area is see-through if the color is not specified right after ShowWindow call. It's not something that DWM should be causing, it's fine on Vista, didn't get around to trying Windows 7. IDK, it's not something I personally witnessed on any Windows version in the past. If nothing else runs and I just watch GPU utilization, it's 0%.

Probably same as Firefox/Basilisk. But that applies to classic extensions, of which very few may run in Mypal, I doubt anyone that maintains them checks Mypal. Firefox needs some hacks to load specially adapted bootstrapped extensions, one and only remaining kind of classic extensions that may still be loaded. https://github.com/xiaoxiaoflood/firefox-scripts Mostly, you're looking for Web Extensions, which have manifest.json rather than install.rdf.

I was curious about something at work, so installed VS 2022 Community on that laptop I got. It went through, with warning about installing on unsupported OS at the beginning. Only installed components related to C# development. I could build a simple Windows Forms application linked to .NET 8.0. Although with such massive application like VS 2022, covering the array of languages to develop in platforms that can be targeted and all specifics that go into it, who knows where things could break.

Testing in my Win10 20H2, there seems to be something at fundamental level. Creating bare bones app with a window, I made it use red color for background, hints here. So register a class, create a (initially hidden) window, then show it. I made a breakpoint before ShowWindow and right afterwards. After it went in ShowWindow, window was briefly shown in white and then it switched to red. Without color specified, it would stay white. On Windows XP, when it finished with ShowWindow, the window's content (technically called client area) was see-through if background color isn't specified, but if it is, the selected color is seen right away. I'm sure Microsoft could optimize it if they wanted or go with the right theme dependent color right away, but it's probably low on priority list, if it's there at all!

I wasn't thinking much in terms of cores back then when I still used that single core computer. But I guess I wanted that one core to be faster than it was, hence resorting to overclocking. Quad core is still adequate for me these days, just I don't do much demanding stuff. Onto another subject, WIn10 (version 1809), I don't recall exactly how long it took to install on the slowest computer I have (year 2014, 1,35 GHz dual-core APU, 5400 RPM HDD). Heck, I really don't remember exact time length of any Windows install I did, but I'd recall if any took much over 1 hour, so that laptop wasn't an edge case. Not sure about usability after install if it took too long to even get there.

Can't load the linked image. Welp, it was fruitless endeavor anyway as broken capacitors weren't the biggest problem.

We didn't have any such fancy multi-CPU computer at home back then. I heard about such boards at some point in the context of servers. One CPU just jumps from one task to another, multiple CPUs / cores allow true parallelism. You do get whole lot for money these days compared to earlier days. The computer I'm typing from was actually just about 500€ in 2009, excluding the dedicated graphics, which I bought later, integrated GPU could be used in the meantime. Besides another GPU swap 5 years later, majority of other investments were just expanded storage. Probably the slowest thing on Win10 is cold boot, then loading user session. Quite a contrast to earlier versions. Fast storage makes incredible job of hiding it. I don't mind slower disks, got spoiled by storage capacity and just didn't get the urge to replace spinning disks with solid state disks, I realize latter are much more economical today than they used to be.

V3 - The Fantastic Nostalgic YouTube Client Not entirely self-contained, non-free software, so proceed with caution.

I don't normally use any kind of filtering software mentioned above either, no firewall, no anti-malware, no anti-virus, nothing. FWIW, I'm behind NAT. Proxomitron - only for some small functional tweaks of one specific piece of software at work. And to stop extensions auto-update on Chromium browsers, but I rarely launch those these days. No long-term experience with roytam1's stuff, but I'm still carrying on with the session from 2019 at work. Even managed to migrate it from Basilisk to Pale Moon few years back, somehow even tab groups survived (needs an extension). No session related extensions used (Tab Mix Plus offers some session management, but I don't use that feature).

I don't have the patience to use single-core CPU computer in this age. I remember using it in the days of simpler WWW, it wasn't anything to be thrilled about.

I haven't confused anything. I even showed a screencap of Supermium on XP showing the used WebGL renderer. But I won't go out of my way to screenshot every small detail as those who genuinely care and don't spend enormous amount of time here arguing to prove something can simply test and see for themselves. Well, duh, because nothing else works on XP. Me too, life's too short to waste on cyclical arguments. This topic of inability to use GPU for graphics as intended on XP has been done to death. And the solution to have these functioning as intended is very simple, use supported hardware running a supported OS. Everything else is just pointless politics and reinventing the wheel that takes a lot of time that would be better off spend on other endeavors IMO.

That was with defaults such as stock progwrp.dll as it comes with Supermium and no special flags, especially not the ones dealing with hardware acceleration.

I always forget (read my mind doesn't register it properly as it's not that picky on some details) it's first launch after hibernating that @NotHereToPlayGames is emphasizing. I can try with rebooting. Besides that, I only have standby working, but not hibernation; space constraints of my XP x64 partition. Though in any case, I still think it's preferable to not intentionally cause error conditions. Edit: OK, no first launch crash. Went through a bad wave of depression. It hit me that another decade is behind me and I don't know what to think of the future. The past still haunts me.

vulkan-1.dll is still used to load software-backed Vulkan implementation for running WebGL content. Inside the GPU process: Getting rid of vulkan-1.dll will produce errors and break WebGL. Screenshots are from Supermium, but I doubt Thorium is any different in that regard. So Vulkan components that come with these browsers are functional on XP, but that doesn't change anything regarding the ability to use GPU for it since no XP drivers provide the working implementation AFAIK. Deleting DLL doesn't break the browser here, but generally such interventions appear pointless to me. Why look for even more trouble? Experience of running these browsers on archaic operating systems already seems a lot like experience of living a life with mutilated genitals as it is.

https://www.reddit.com/r/sysadmin/comments/1990jvf/new_teams_client_on_windows_10_ltsc/ On Win10 2004+, it should install without jumping through most below hoops. For older versions such as popular 1809, I'm summarizing the relevant post: Install the latest Windows-Updates (2024-01). Install Edge WebView2 Runtime (or Edge itself, which contains WebView2): https://developer.microsoft.com/en-us/microsoft-edge/webview2?form=MA13LH Allow msix sideloading (run Command Prompt as administrator): reg add HKLM\SOFTWARE\Policies\Microsoft\Windows\Appx /v AllowAllTrustedApps /t REG_DWORD /d 1 /f Enabling “Developer Mode” manually or via Group Policy should also work. Download Teams MSIX: https://learn.microsoft.com/en-us/microsoftteams/new-teams-bulk-install-client#option-1b-download-and-install-new-teams-using-an-offline-installer Install MSTeams via MSIX (run PowerShell as administrator): Add-AppProvisionedPackage -Online -PackagePath "$PSScriptRoot\MSTeams-x64.msix" -SkipLicense It may work with older builds of Win10 1809 as well. Regarding Edge and Edge WebView2, it's probably the best to have both formally installed for best compatibility with stuff requiring WebView2. When both are correctly installed, there's technically just one copy of files for specific Edge version on disk, they're hardlinked together. I had Developer Mode enabled from before, so 3rd step was skipped. I wonder what it would say otherwise. In the last step, I just specified absolute path to where .msix file was downloaded, which was in Downloads folder.

The text for that website on search engines such as Bing, DuckDuckGo and Yahoo appears to be in Chinese. Google has it updated to currently used English. 99% of users won't have problems with their computers rendering such texts. Support for Chinese, Korean and Japanese was optional in Windows XP, but could be installed by ticking a single checkbox. These render out-of-the-box in subsequent versions. Cyrillic fonts are available out-of-the-box on XP (and in subsequent versions). Certain alternative operating systems may still require user to do some extra installation steps for certain texts to be correctly rendered.

I only visited videocardz.com on 32-bit Serpent 52 on XP x64 back when it was first mentioned for the short time, couldn't make it crash. Same today on Win10 on 64-bit Pale Moon, except this time I opened 28 tabs instead of just checking few articles in a single tab. Though uBO blocks scripts from everywhere except videocardz.com, disqus.com and disquscdn.com. Can't say I ever had to look into GroupOrderList and driver services' tags. BTW, about that Kingston's RAM stick of mine, it's probably fine, seems to be just Memtest86+ 5.01 glitching out. I probably ran it in single thread mode for the first time when I got the stick while I enabled multi threading for recent run, to give CPU more work. Read that mode was still buggy in that version. Repeated testing (2 passes) with latest and greatest x64 version of Memtest86+ 7.00, no errors. It looks like existing tests are more thorough compared to the old version. It's not the most optimal match for the other two Mushkins, I missed the train when DDR2 was still in common use and I usually don't buy used stuff, just skeptical about it in general.

Aye, those components don't have to be compatible with other components from different builds. Though Windows 8.x is unusually buggy in some places in my experience, couple of years ago when I still used it, I could use Win10's (one of then actual builds) quartz.dll on Win8.1, that fixed otherwise broken quick preview functionality of QTTabBar when it came to certain audio and video files.

Looks like it won't be necessary for now. What about 63/37 variant? Some swear by it.

Yeah, can't say I noticed any. BTW, I missed some when counting for first time, there's 31 T-on-top-shaped ones and 11 ones with s red stripe, and possibly 6 smaller ones and 8 even smaller ones. I didn't get around to blowing dust out for at least 2 years now, so finally did it, seems stable now, even with 7% overclock, CPU heats to 50 °C now rather than 54 °C when under heavy load. There was a lot of dust in a quite massive heatsink of Akasa AK-968, the aftermarket CPU cooler I have. That Kingston's RAM stick still doesn't work correctly, one of the tests just make everything freeze, so stuck with 4 GB of RAM again. If it was at least DDR3...oh well, I'm used to it, but those extra 2 GB could really make a big difference in general.

Yeah, I suspect it's capacitors too, they look fine, but as far as I know, as they age, capacitance goes down and equivalent series resistance goes up. I'm shocked by the sheer number of them, there's more than 26 of them on the board! 26 with familiar top with a T shape, then at least 6 with a red stripe on top and possibly few much smaller ones (assuming I didn't confuse them with something else). With my poor (de)soldering skills and tools, I'm not confident about touching this board right now. I guess the heat could be a smaller factor, it's pretty hot these days and the room it's in is about 27 °C. Blowing the dust off harder accessible spots is still on my TODO.