Leaderboard

I can't say what's going wrong in your system as I never have used Windows 2008. And I don't know whether New Moon 28 targets this OS at all. Under Windows XP, the "Make New Moon the default browser" button has always worked like a charm. And if this feature didn't work in other browsers, I either edited the respective registry entries manually or I used the tool Default Browser 1.8. Unfortunately, this tool is for Windows XP only but maybe, there is also a tool which does this job under your OS. Here are important registry locations which were set by me for using New Moon 28 as default browser under Windows XP: HKEY_CLASSES_ROOT\Applications\palemoon.exe HKEY_CLASSES_ROOT\http HKEY_CLASSES_ROOT\https HKEY_CLASSES_ROOT\ftp HKEY_CLASSES_ROOT\PaleMoonHTML HKEY_CLASSES_ROOT\PaleMoonURL HKEY_CURRENT_USER\Software\Classes\Applications\palemoon.exe HKEY_CURRENT_USER\Software\Classes\http HKEY_CURRENT_USER\Software\Classes\https HKEY_CURRENT_USER\Software\Classes\ftp HKEY_CURRENT_USER\Software\Classes\PaleMoonHTML HKEY_CURRENT_USER\Software\Classes\PaleMoonURL HKEY_LOCAL_MACHINE\SOFTWARE\Clients\StartMenuInternet\Pale Moon web browser HKEY_LOCAL_MACHINE\SOFTWARE\Clients\StartMenuInternet\PALEMOON.EXE Under the registry path HKEY_LOCAL_MACHINE\SOFTWARE\Clients\StartMenuInternet all registered standard web browsers are listed. So if you want to be offered a standard browser, you have to enter the corresponding one there. BTW, New Moon's native function for setting it as default browser is SetDefaultBrowser(). When checking its source code, you will exactly know what actions the "Make New Moon the default browser" button performs.

... So far, I have resisted selling my soul to Google by registering an account with them (yes, I do use their "services" occasionally, but only as an "anonymous" user ; I do not even own an Android device ), but that would be "the last straw on the camel's back" for me if a Google account becomes now mandatory to grab a local copy of a YT clip (via yt-dl, yt-dlp, etc.) ; not to mention that downloading while being logged-in is a sure recipe for Google's AI to (b)lock your YT account eventually (lots of such reports inside the yt-dl/yt-dlp GH trackers) ...

For yt-dlp, I use no UA parameter and everything works. For youtube-dl, I use: --user-agent "Mozilla/5.0 (Android 14; Mobile; rv:128.0) Gecko/128.0 Firefox/128.0" But it doesn't seem to work any more... Relevant issue: #32905.

@feodor2 Unfortunately, I have to report that the website amazon.de is no longer working in Mypal 68 as before. I tested it in both versions 68.14.4b and 68.14.5b. Here is one link as an example: https://www.amazon.de/Original-Blumax-DMW-BCG10E-Panasonic-DMC-TZ31/dp/B00I04P3ZU?nsdOptOutParam=true&sr=8-3 When I try to open this page, the loading process is a torture compared to how it was months ago. It always stays at 100% CPU utilisation regardless of whether the website has been fully loaded (or appears to be fully loaded) or not. Although my hardware is weak (single core CPU Pentium 4 Northwood 2.8 GHz), amazon.de has worked tolerably well in Mypal 68 in the past. BTW, in New Moon 28, this site is loading much faster and comes to an end without 100% CPU utilisation. As far as I noticed, the issue is related to the loading of images and media. Even the additional use of uBlock Origin couldn't help, unless you neuter the site to the point of informationlessness. P.S.: With the help of additional user scripts and own filters for uBO, I was able to tame a bit the amazon.de website. Now, the loading process finally ends at some point. But all loaded images and media definitely lead to extremely higher processor utilisation than before. The entire loading process still takes some minutes.

I've updated youtube-dl, applying the pending fix for YouTube. See: https://github.com/ytdl-org/youtube-dl/pull/32987. You may have to pass cookies (from logged in account) and the following user agent: --user-agent "Mozilla/5.0 (Android 14; Mobile; rv:128.0) Gecko/128.0 Firefox/128.0" --cookies "C:\Documents and Settings\$USER\Application Data\youtube-dl\cookies.txt"

The air in Canada has become drier this winter. I has to buy a humidifier.

Usually they dim when old, or become orange/yellowish.

But then again, not worth if it's very old, better find the same model on the used market. As it could be costly to buy the good caps.

A very unfortunate event indeed! If you already checked it with another PC (not only the card), then I'd start with more in-depth testings, like replacing the caps and measuring their capacity/ripples.

My French brother, please accept my deepest condolences. That loss is so tragic, no words can describe. And yes, be on your guard with scammers, they usually know our weak places, where to push. Prenez soin de vous ! / Take care.

Forgot to tell, It's good to have clear thinking people on MSFN, (I just made you a compliment). So, please go on, keep up.

I agree with you, at this point we still know nothing, I too want to see more proof! System kernel is way too important to install from unknown sources! I'd prefer the developer to have a verified MSFN account, I don't trust github after reading their TOS.

Yes! The flags have updated themselves to reflect the real location. So now we all know Tommy is from Canada. How far from me is still unknown though.

No mystery, sorry! And I wrote about it 1.5 years ago. https://msfn.org/board/profile/439498-karla-sleutel/?tab=field_core_pfield_18

Would be about the level of nVidia G210, at best,

... The separate standalone mpeg-dash streams are in a media format/container (iso6dash) that IS NOT compatible with old media players (such as mpv from back in 2015 ); when you download via yt-dl and a recent-ish build of ffmpeg is used, the separate media streams are muxed together into a standard iso MPEG4 container that has been supported for at least two decades... The 360p encode is itself by default packaged (and streamed) inside a "proper" MPEG4 container (with its mov atom placed in the beginning of the MP4 file) ...

I believe that updating Chrome components in the background on Windows XP may now be more of a problem than a benefit. It's been so many years since support for Chrome on Windows XP was discontinued that components for this operating system are no longer being developed, let alone checked for compatibility. Thus, using the flag --disable-component-update seems to me a logical consequence.

As I already wrote, no components were installed in my Thorium installation. The list is simply empty. It was just an idea. Personally, I use the flag --disable-component-update.

@Dave-H I have just seen your recently opened issue "Side by Side Error Recorded in Windows XP #85"on GitHub. I then checked my event logs but can't see any corresponding entries so far although the sub folder screen_ai also exists in my USER_DATA folder including this ominous chrome_screen_ai.dll file. Furthermore, I checked my installed components under chrome://components/ but there are none. Maybe, you simply delete or rename the complete screen_ai sub folder and use the command line flag --disable-component-update for the next browser start.. It might be necessary to de-register the chrome_screen_ai.dll file, too. I can't say whether this will help you, but you could give it a try.

I know these command line flags. Personally, I use the extension Animation Policy in Thorium. By using this extension, you are able to exactly adjust via a toolbar popup how the animations are supposed to be handled by the browser. Thus, the user has the possibility to let an animation play or stop from inside Thorium at any time without editing the flags and restarting the browser.

After some tests, I would even say that the loading behaviour of different websites also has been improved by the updated Chrome XP API Adapter. More test are necessary, of course.

As already reported in the Supermium thread by @rereser, a new progwrp.dll file was released for Supermium. This file is the Chrome XP API Adapter and is supposed to be compatible with Thorium. Therefore, I replaced the old version 1.1.0.5002 by the new version 1.2.0.5035. The file size from version 1.1.0.5002 to 1.2.0.5035 has been significantly reduced by an incredible 82%. In any case, I can confirm that now the browser start is much faster and smoother than before. I think this is a great improvement.

The image problem you're discussing looks like a layout issue. Try the following CSS: @-moz-document url-prefix("https://www.tunwalai.com/story/") { .story-image { z-index: unset !important; } } Interesting that unsetting filter on the image (which sets contrast in this case) also makes it display, but then it obviously doesn't display with intended contrast.

Yes, Palefill is a community effort, it's pretty much about coding site specific fixes using some known workarounds where they may work by injecting certain polyfills (JavaScript functions not natively supported by the browser) or performing some smaller alterations to the original JavaScript code on a specific site. Some can be fixed while others rely too heavily on specific features, those are hopeless. https://www.tunwalai.com/story/651554 The text in the bracket says "Omegaverse". You might be missing the font the site is using. I'd say what it is, but I'm honestly not sure after trying to identify it if I'm doing it right... If possible, having Office 2010 installed is one way to add some fonts Windows XP normally doesn't have. Generally, if a used font is known, that font alone can be installed on the system and then sites using it will display correctly. May not apply for the poster above, but if anyone is unaware, there are extra fonts on Windows XP CD covering letters of specific languages that aren't installed by default:

@Taiga-chan AFAIK, you need to install Palefill extension to be able to use that website. I'm not a registered user there so can't tell if any additional issues have appeared since the compatibility problem due to lack of support for Web Components in the browser has been addressed with that extension.

fixed

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221008-0a5fd83fe9-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221008-0a5fd83fe9-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221008-0a5fd83fe9-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20221008-0a5fd83fe9-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1149728. Move CreateXForOffscreen functions. r=jrmuizel (87d69e4b7c) - missing part of Bug 1179280. Update ANGLE (ee6228fa90) - missing bit of Bug 1191042 - Use CreateOffscreen for WebGL instead of CreateHeadless. - r=jrmuizel (87c26edf2c) - add back XP bits (f198a909bd) - put back SQLITE_ENABLE_DBSTAT_VTAB checks (a06fd5b7a9) - Bug 1197387 - Use glXGetProcAddress to link against GLX_ARB_create_context_robustness. r=jgilbert (651b85207d) - add back feature report (a0a33bb8e5) - Bug 1209022 - Fix AA handling in EGL's CreateOffscreen. - r=jrmuizel (64d4e0f65a) - Bug 1171228 - Expose WEBGL_debug_renderer_info to web content on non-RELEASE_BUILDs. - r=kamidphish, sr=jst (e7405ef944) - Bug 1175257 - Remove the fails-if(B2G) annotation on capturestream.html. (80bf24c623) - Bug 1177726 - Inflate the bounds when stroking the canvas text. r=jmuizelaar (15969cb828) - Bug 1183363 - Make EnsureWritablePath and EnsureUserSpacePath always imply EnsureTarget. r=Basimply EnsureTarget. r=Bas (12d825cf30) - Bug 1183363 - Add canvas crashtest. r=Bas (69b6099607) - Bug 1186689 - check for OOM on destination array before mapping source to avoid needing unmap. r=bas (7a3ed105d6) - Bug 1206161 - Make sure the DrawTarget is available when creating a CanvasLayer so that we can check for SkiaGL. r=Bas (fdf15fd552) - Bug 232227 - Do not expose system colors to CSS or canvas. r=jmuizelaar r=enndeakin tor-r=arthuredelstein (dfd6f07fd8) - Bug 1209206 (part 1) - Remove mgfx namespace synonym. (5916056863) - Bug 1209206 (part 2) - Rename and re-comment HasNonOpaqueColor(). r=jdaggett. (768a0d549b) - Bug 1209206 (part 3) - Use sinf() and cosf() instead of sin() and cos() in a couple of places. r=Bas. (7af9ebaa7f) - No bug - Remove outdated comment and additional comment questioning its relevance. r=me DONTBUILD (6f7eca8452) - Bug 1145934 - Update obsolete glyph-run assertion. r=jdaggett (ce2442e64b) - Bug 1182361 p1 - move old generic lookup methods into gfxPangoFontGroup. r=heycam (ef39e6fd52) - Bug 1182361 p2 - count generic lookups. r=heycam (53050ad6e8) - bug 1180010 make some methods protected and non-virtual r=jdaggett (815e503481) - Bug 1165179 - use all style matched faces within a font family. r=heycam (d2711db48b) - Bug 1182361 p3 - move generic lookup methods to platform fontlist. r=heycam (63b974a000) - Bug 1182361 p4 - move pref font util methods to platform fontlist. r=heycam (23d7352442) - Bug 1182361 p5 - cache pref fonts per langGroup. r=heycam (c4244432cb) - Bug 1182361 p6 - eliminate old pref font caching. r=heycam (c5293f0b8d) - Bug 1182361 p7 - fixups based on review comments. r=heycam (51a2204179) - Bug 1170688 - Move initialization of text-run parameters until *after* the draw target has been updated; r=jfkthame (e523ff6e96) - Bug 1157629 - Fix uninitialized member warning in BufferAlphaColor. r=jdaggett (a564effd24) - Bug 1184282 - Remove break suggested inside cluster warning. r=roc (54da46a09f) - Bug 1197650 - remove duplicate validity check for newly created fonts. r=m_kato (5ffb28a313) - Bug 1171357 - log font matching for textruns. r=m_kato (2354cd36f6) - Bug 1185812 - rejigger the ordering of fonts within a family to avoid obscure faces. r=heycam (0518f4ef64) - Bug 1167697 - Mark refs to gfxFontEntry in UserFontCache as MOZ_NON_OWNING_REF. r=jtd (1fdcccdac2) - bits of Bug 1073209 - Eliminate usage of CreateSamplingRestrictedDrawable (9058557527) - and sync gfx/gl to same state as af-frontend branch (d041f9bde0) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1204192 - IonMonkey: MIPS: Split shareable code to mips-shared in BaselineIC-mips32. r=nbp (eeae38da4e) - Bug 1204193 - IonMonkey: MIPS: Split shareable code to mips-shared in Bailouts-mips32. r=nbp (f24f496235) - Bug 1204194 - IonMonkey: MIPS: Split shareable code to mips-shared in MoveEmitter-mips32. r=nbp (5c3bdc2fc2) - Bug 1204214 - IonMonkey: MIPS: Split shareable code to mips-shared in BaselineCompiler-mips32. r=nbp (01903406d1) - Bug 1205229 - IonMonkey: MIPS32: Make more CodeGenerator functions can be shared. r=nbp (d437fe618e) - Bug 1209528 - IonMonkey: MIPS32: Add suffix 'f' for constant float32. r=arai (0f6600c083) - Bug 1205232 - IonMonkey: MIPS32: Fix rounding of big negative float32 values in Ion. r=bbouvier (4652bb9b7a) - Bug 1203044 - IonMonkey: MIPS32: Atomics operations should throw on oob access. r=lth (e898b3af95) - Bug 1205135 - IonMonkey: MIPS: Split shareable code to mips-shared in CodeGenerator-mips32. r=nbp (b6d81b922d) - readd some missing bits (2aa8c9fafd) - Bug 1167189: Remove unnecessary checks after infallible allocations. r=bholley (fa7d8b9f7b) - Bug 1209398 - Make AC_SUBST_SET emit a list of unique items instead of an actual set. r=gps (83e5877882) - Bug 904572 - Add support for generating clang compilation database; r=glandium,r=gps (ab3e6e62fa) - Bug 1209398 - Enable the FasterMake backend by default for desktop Firefox builds. r=gps (29f6fb3883) - Bug 1204719 - Don't create interfaces.manifest files from the build backend. r=gps (5a63e36259) - Bug 786520 - Install things to $(DIST)/branding from moz.build instead of manual rules in Makefile.ins. r=mshal (652552b547) - Bug 1160563 - Part 1: Make ANDROID_RES_DIRS a moz.build variable. r=gps (e0719524f8) - Bug 1159390 - Set sharedUserId in robocop.apk. r=gbrown (dde6716e2d) - Bug 1160030 - Use TEST_HARNESS_FILES to install Robocop test files. r=froydnj (00cd01c3f2) - Bug 1160030 - Use TEST_HARNESS_FILES to install Robocop ini files. r=froydnj (66d52518e7) - Bug 938659 - Part 2: build system changes. r=mfinkle (0833d22ba3) - Bug 1160563 - Pre: Allow ANDROID_RES_DIRS that are not relative to srcdir. r=gps (0d8a3ce66e) - Bug 1160563 - Part 2: Make ANDROID_ASSETS_DIRS a moz.build variable. r=gps (9f4c132e2a) - Bug 1195388 - Part 1: Make ANDROID_APK_{NAME,PACKAGE} moz.build variables. r=gps (4a2fa3fc09) - Bug 1195388 - Part 2: Add ANDROID_EXTRA_{PACKAGES,RES_DIRS} moz.buildvariables. r=gps (7ad49e355e) - Bug 1207893 - Change how we track object consumption from the build backend. r=gps (20ccc5b67a) - Bug 1184405 - Add annotations for tags, file patterns, and test flavos to moz.build to specify tests potentially impacted by source files. r=gps (2db17131cc) - Bug 1171105 - Ability to aggregate metadata from Files instances; r=glandium (59c8759e61) - Bug 1184405 - Add a container type to mozbuild with a namedtuple-likeinterface and typed, mutable fields. r=gps (3062dbae71) - Bug 1138283 - Fix bad documentation around wildcard patterns; r=glandium (5785d3284d) - Bug 1155816 - part 2 - move EXTRA_*COMPONENTS manifest check to buildbackend time; r=mshal (c539616b45) - Bug 1161270 - Add source manifest to reftest manifest representation in moz.build.;r=gps,roc (376b080084) - Bug 1180813 - Use a cache for BuildConfig to improve mach speed. r=gps (d431b1ec23) - Bug 1207882 - Associate an install target with every ContextDerived object. r=gps (62f63487a7) - Bug 1126228 - Directory (--directory) argument for |mach warnings-list| and |mach warnings-summary|. r=gps (37a85d904f) - Bug 1210642 - Allow to pass defines overrides when processing install manifests. r=gps (5eb483dc8d) - Bug 1210642 - Add support for --silence-missing-directive-warnings for preprocessing within install manifests. r=gps (3a07d30b36) - Bug 1176642 - Use absolute_import in mozpack; r=glandium (863ab20c4a) - Bug 1162826 - Properly display full moz.build processing errors when empty KeyError or ValueError are thrown. r=mshal (57e8367680) - Bug 1184405 - Add a mach command to expose test-deps file info. r=gps (7bcc10679c) - Bug 1195735, r=zer0 (7fe14224e3) - Bug 1205166 - IonMonkey: MIPS: Import MIPS64 support into Architecture-mips-shared. r=nbp f=rankov (84e1f9ec5d) - Bug 1205166 - IonMonkey: MIPS64: Import Architecture-mips64. r=nbp f=rankov (fe173dd5af) - Bug 1210322 - IonMonkey: MIPS: Rename BaseFloatRegister/s to FloatRegister/sMIPSShared. r=nbp (ac262cbb5d) - Bug 1205167 - IonMonkey: MIPS64: Import Assembler-mips64. r=nbp f=rankov (079e84ea49) - Bug 1215555 - Improve simulated OOM testing of ARM assembler buffer and fix bugs r=jolesen (253b976cec) - Bug 1205167 - IonMonkey: MIPS: Import MIPS64 support into Assembler-mips-shared. r=froydnj f=rankov (94b23ac585) - Bug 1213146 - IonMonkey: MIPS: Fix build failure caused by bug 1194139. r=nbp (12f7d5d572) - Bug 1205566 - IonMonkey: MIPS: bailoutCmp32/Ptr optimization. r=nbp (9a40ee80d0) - Bug 1209873 - IonMonkey; MIPS: Implement memory barrier. r=lth (964535cb6b) - Bug 1209572 - IonMonkey: MIPS32: Goto done by short jump in convertUInt32ToFloat32. r=nbp (0b388199e7) - pointer style (312fc6fe4f) - missing bit of Bug 1140890 - Make sure the first argument cannot bail in between negative zero removal and creating result in substraction, r=nbp (afcc9bfdc7) - and sync more build scripts with frontend branch (fc52b62763) - partly import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1152538 - Removing RTCIdentityAssertion interface code, r=jib (b89f1a9ebc) - Bug 1158868 - PeerConnection should not depend on navigator.onLine r=jesup (77c66505ff) - Bug 1167922 - Handle broken entries in media.peerconnection.default_iceservers more gracefully. r=jib (b24406479f) - Bug 1189030 - Add pref "media.peerconnection.ice.relay_only" (default=false). r=jesup (eafbcf9412) - Bug 1187206 - Adding checks for null SDP, r=jib (8de79e683e) - Bug 1189060 - add CreateOfferRequest.webidl interface for add-ons r=florian,peterv (e901784f32) - Bug 1191180 - Wait for certificate before generating identity assertion, r=jib (ced0fc6aaf) - Bug 1207784 - skip permission hooks in createOffer when called from privileged code (add-ons). r=mt (0052b94f8a) - Bug 1098015 - Make createAnswer tolerate (no-op) options argument. r=bwc, r=bz (9dd5250811) - Bug 1064223 - Remove support for constraint-like RTCOfferOptions predecessor. r=mt (7fad1b56d5) (34929c3a27) - import from `custom` branch of UXP: D3D9 initialization fix by XPRTM@MSFN, Thanks! (a16549dd) (6aa41bc113) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1199171 part 0 - Add spewInst function for debugging Jit code. r=lth (12cd32209d) - Bug 1192786 - support an installed printer, for profiling. r=nbp (59fe21c4a5) - Bug 1199171 part 1 - Assembler::GetCF32Target: Decode instructions as needed. r=h4writer (b120e75c4f) - Bug 1192786 - add forgotten ifdeffery on CLOSED TREE. r=me (7ccf2e792c) - Bug 1195263 - ARM: Instruction::extractCond asserts that the instruction has a condition flag. r=h4writer (fe30c9a6ae) - Bug 1207449: Conservatively mark MFilterTypeSet with phis as not float32-compatible; r=h4writer (9062bae786) - revert 1130679 that was never commited (a9c6fcc984) - Bug 1201793 - correct scratch register scope r=nbp CLOSED TREE (6df1d46059) - Bug 1202643 - bailout from udiv on nonzero remainder. r=nbp (b001fc7c42) - Bug 1206650 - Part 1/2 - Clean up load8ZeroExtend(). r=nbp (288a451461) - Bug 1206650 - Part 2/2 - Use the dest register for scratch in alu_dbl(). r=nbp (4e601cff7f) - Bug 1201793 - ScratchRegisterScope: ma_alu expect to have scratch register owned by the caller. r=sstangl (6783084545) - pointer style (04d2b491cc) - Bug 1202757 - disassemble more instructions. r=sstangl (39e29fbc98) - Bug 1211832 - Disable functions that can easily cause artificial OOMs. r=jonco (3264e2dcb6) - Bug 1188390 - Fail when Sprinter in disassemble ooms. r=h4writer (f59e3b00ca) - Bug 1191635 - Fix stack limit for evalInWorker runtime. r=bhackett (b4b8f37dc1) - Bug 1209873 - IonMonkey: MIPS: Add AssemblerMIPSShared::as_sync. r=arai (80dbd653b7) - Bug 1212469 - Fix some OOM handling issues shown up by the previous patch r=jandem (ca7e25d243) - Bug 1215058 - Fix various OOM handling issues related to off-thread compilation r=jandem (6d95767f75) - Bug 1207827 - Copy pool data into buffer immediately. r=nbp (d3fcd69e1e) - Bug 1207827 - Stop perforating AssemblerBuffers. r=nbp (8ac7947355) - Bug 1207827 - Use a Vector for poolInfo_. r=nbp (d97f814916) - Bug 1207827 - Switch jit::Pool to a LifoAllocPolicy. r=nbp (caf547349e) - Bug 1207827 - Eliminate poolSizeBefore(). r=nbp (adad1dfa68) - Bug 1181612: Hoist codeLabels_ and associated common functions into Assembler-shared; r=luke (5086762e53) - Bug 1207827 - Remove ARM temporary offset buffers. r=nbp (7bc80ab795) - Bug 1207827 - Delete Assembler::actualOffset() and transitive closure. r=nbp (34141fde56) - Bug 1213146 - IonMonkey: MIPS: Remove AssemblerMIPSShared::asAsm. r=arai (8d71e648c9) - const-var (f860ce4da7) - Bug 944164 - set the shell variable scriptPath to communicate the currently running script to the debugger, r=jorendorff (5ca9508912) - align to firefox (f895d43a07) - Bug 1213146 - IonMonkey: MIPS: Move Assembler::PatchDataWithValueCheck to architecture specific. r=lth (c8a7b478f6) - Bug 944164 - make the shell actually quit when you call quit(), r=jorendorff (8e737a7450) (b4a9437711) - import from `custom` branch of UXP: follow-up D3D9 initialization fix by XPRTM@MSFN, stop creating pure device. (51bb5d5c) (df1c937cce) - import changes from `dev' branch of rmottola/Arctic-Fox: - bit of Bug 1202902 - Fix the world. (b7b307b5b1) - Bug 1179102 - The async version of loadSubscript doesn't get the right JS version. r=bholley (d390d929f5) - Bug 1206149 - Use channel->AsyncOpen2() js/xpconnect/loader/mozJSSubScriptLoader.cpp (r=sicking) (0c5b42221d) - Bug 1185634 - Part 1: SandboxPrivate shouldn't be public. r=gabor (3b63251f58) - Bug 1185634 - Part 2: Remove obsolete comment. r=gabor (48359ea679) - some crash reporter and space style (c72b1291a7) - Bug 1178274 - Ensure the BrowserTabsRemoteAutostart function is first called on the main thread during xpcshell testing. r=billm (3e81414e4a) - Bug 1186993 - Add MOZ_CHAOSMODE to XPCShell. r=roc (a296f46add) - Bug 1187410 - Use GetNameShared() in XPCNativeInterface::NewInstance() to avoid a leak. r=mrbkap (54f1496a03) - var-let (48ada4e8ba) - Bug 1209077 - handle cached empty string in StyleSheetActor._getText; r=bgrins,pbrosset (1940c70e2e) (0a5fd83fe9)

sure, even posting with pastebin will be OK if you want. EDIT: got a modified source file from @XPRTM and I compiled new xul.dll on top of basilisk52-g4.8.win32-git-20221001-3219d2d-uxp-de51d3878-xpmod-ia32.7z. to test, download basilisk52-g4.8.win32-git-20221001-3219d2d-uxp-de51d3878-xpmod-ia32.7z and replace xul.dll from http://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20221005-xpmod-ia32-d3d9test-xul.7z

Off-topic: Thanks for consumption voucher from HK government, and a special offer of HKD 2020 of AMD 5800x, I got a almost free upgrade from 3700x. So there may be about 20% faster for compiling browser's source code.

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221001-1406b5eef9-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221001-1406b5eef9-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20221001-1406b5eef9-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20221001-1406b5eef9-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1182147: Ensure ImageHost does not attempt to call DrawQuad with an effect unsupported by the compositor. r=nical (5624e9134f) - Bug 1151749 - "Win-Only: Disabled Hardware Acceleration + CSS 3D transform on canvas -> black artefacts on hover". r=nical (54c49071b9) - Bug 1184224 - Remove gfxPrefs.h include from gfxPlatformGtk.h. r=karlt (eb907daa58) - Bug 1183820 - Expose if cairo uses XRender to GetAzureBackendInfo. r=mattwoodrow (10044ecfb0) - Bug 1128934 - Call XFlush after compositing to prevent jank. r=karlt (f63b9dc1b5) - Bug 1151663 - New mochitest for this bug. r=kats (9344d22ed6) - Bug 1176402 - Have synthesizeNativeWheel() and synthesizeNativeMouseMove() accept coordinates relative to the element's bounding rect. r=kats (cbdffe3869) - Bug 1177018 - Work around the iframe in which mochitests are run not being scrollable. r=kats (d24291080c) - Bug 1177018 - Disable smooth scrolling in the APZ layerization test. r=kats (e1af545da1) - Bug 1181678 - Improve the check for skipping test_layerization if APZ is disabled. r=botond (91e5ead610) - Bug 1177018 - Enable chaos mode for the APZ layerization test. r=kats (f07fdeb936) - Bug 1177018 - Only enable APZ test logging for tests that actually use it. r=kats (9f24ea01c4) - Bug 1196294 - Remove the mCause from TimelineMarkers, r=tromey (5abd1c057d) - Bug 1195232 - Stop using TracingMetadata from GeckoProfiler.h, r=tromey (e736038873) - Bug 1200118 - Create a barebones TimelineMarker abstract class, r=tromey (9930c83c2f) - Bug 1200252 - Add marker for when the IPC request for a composite is sent to the compositor thread, r=jsantell,matt (becc11f58b) - align some minor thing to gecko (5bb64e322a) - Bug 1204581 - Add a deprecation warning for the usage of AppCache when service worker fetch interception is enabled; r=mcmanus,baku (69949c8d96) - Bug 1210941 P2 Use LOAD_BYPASS_SERVICE_WORKER in HttpBaseChannel instead of mForceNoIntercept. r=jduell (fdb7afc0be) - Bug 1173171 - Disable external jar: via preference. r=jduell (704cfb0b9c) - Bug 1210941 P3 Make jar channels use LOAD_BYPASS_SERVICE_WORKER internally. r=jduell (d871dc4837) - Bug 1210941 P4 Use LOAD_BYPASS_SERVICE_WORKER instead of ForceNoIntercept in nsDocShell. r=ehsan (ea5ed76717) - Bug 1210941 P5 Use LOAD_BYPASS_SERVICE_WORKER in nsObjectLoadingContent instead of ForceNoIntercept(). r=ehsan (9e821adfaa) - Bug 1210941 P6 Use LOAD_BYPASS_SERVICE_WORKER instead of ForceNoIntercept in FetchDriver. r=ehsan (02685be882) - Bug 1210941 P7 Use LOAD_BYPASS_SERVICE_WORKER in ServiceWorkerScriptCache. r=ehsan (45fa163dd2) - Bug 1182112 - XML documents should not be intercepted by ServiceWorkers. r=ehsan (a9dfeffcb3) - Bug 1210941 P8 Use LOAD_BYPASS_SERVICE_WORKER in xslt txURIUtils. r=ehsan (d7686d572b) - Bug 1210941 P9 Use LOAD_BYPASS_SERVICE_WORKER in nsCORSListenerProxy. r=ehsan (653f50de45) - Bug 1205302 - Disallow intercepting OCSP requests; r=jdm (b4650e78d9) - Bug 1210941 P10 Use LOAD_BYPASS_SERVICE_WORKER in nsNSSCallbacks. r=ehsan (b4b8b3f503) - Bug 1210941 P11 Use LOAD_BYPASS_SERVICE_WORKER in worker ScriptLoader. r=ehsan (ac07404832) - Bug 1210941 P12 Remove http channel's ForceNoIntercept. r=jduell IGNORE IDL (a7592a83c4) - Bug 1210941 P13 Remove ForceNoIntercept from jar channel. r=jduell (617544fcee) - Bug 1169613 - Use content type of synthesized response for JAR channel requests if available. Part 1: Set Content-Type of synthesized reponses for JAR channel requests. r=jdm (e32061a26a) - Bug 1169613 - Use content type of synthesized response for JAR channel requests if available. Part 2: Set Content-Type for JAR Channel requests on Fetch API responses. r=nsm (da43e29583) - Bug 1207556 - Part 1: Stop reusing the loadinfo in StartCORSPreflight; r=sicking (4724bfa8cd) - Bug 1207556 - Part 2: Fix the beacon CORS preflight tests; r=sicking (c61a699f9f) - Bug 1111834 - CORS preflight of navigator.sendBeacon() should not follow 30x redirect - tests. r=sicking (2871ad22e7) - fix build because of previous backports (e3096e6f9a) - Bug 1161677 - Expose dev mode state read-only through the navigator.hasFeature() api r=ehsan (013399847b) - Bug 1203680 P8 nsCORSListenerProxy::OnStartRequest must always call outer OnStartRequest. r=ehsan (9a67709770) - spacing (d4511b37c4) - missing bit of 920017 and some telemetry (2bf267ce72) (938961e76b) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1146888 - Build more files in ipc/chromium in unified mode; r=bent (1143418e0f) - Bug 1206558 (part 1) - Combine handling of BSDs in ipc/chromium/moz.build. r=mshal. (2bb24997a2) - Bug 1206558 (part 2) - Move Linux-specific code ipc/chromium/moz.build. r=mshal. (94bf4c550e) - Bug 1206558 (part 3) - Move Android-specific code ipc/chromium/moz.build. r=mshal. (43ca8e06dd) - Bug 1206558 (part 4) - Factor out include handling in ipc/chromium/moz.build. r=mshal. (4ae55262ca) - Bug 1206558 (part 5) - Give libevent its own moz.build file. r=mshal. (1919309e8e) - Bug 1206558 (part 6) - Factor out common libevent moz.build stuff. r=mshal. (734e9a4b71) - actually use arc4random_bug check so allow compilation on recent libc (851de42a37) - var-let (8c910312d3) (1406b5eef9)

yeah some other websites affected https://forum.palemoon.org/viewtopic.php?f=70&t=28881 while we don't have users visiting such site.

and by the way I pushed win2000 branch with changes ported from pm26-win2000. and result binary runs on neptune. (I didn't further test the build)