Leaderboard

It was available first time, actually, in Pentium 3, but that was pre-9\11 world and this did not fly well and was quickly strike down by EU. https://en.wikipedia.org/wiki/Pentium_III#Controversy_about_privacy_issues After 9\11 right-wingers rig the world and you all should comply and be obedient slave citizen or nine-eleven will happen again. P.S. I long time want to ask, will roytam adapt 29 version of PM? They promise "Google WebComponent" support on they release notes with 29... Maybe this question was asked before, but i never see it. So what will be with 29 for XP?

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210515-963b86a51-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210515-963b86a51-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210515-963b86a51-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20210515-963b86a51-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 1158442 - Remove the "Performance Entry buffer size maximum reached" warning; r=baku (98065e0a3) - Bug 1164552 - create new colors in light/dark themes for devtools specifically for rendering graphs in the performance tool. The highlight-* colors in light theme work well for text, less so for rendered blocks and appear very 'bold'. Most of the dark theme graph colors are the same as their highlight-* equivilents. r=vp (a5f33fb22) - Bug 1158731 - Buffer for Performance APIs (Resource Timing, User Timing) should be separeted. r=baku (87d4be360) - Bug 1166494 - Re-add console timestamp and style markers' metadata. r=vp (24c989240) - Bug 1166494 - part2: correctly add labels to console markers and properly add style markers' restyleHints. r=vp (ecf87cff4) - Bug 1166494 - part3: Correctly handle marker definition fields that are described via function rather than array. r=vp (b71b8575f) - Bug 1162662 - Map JS markers to human readable keys, and hide if platform related via (Gecko). r=vp (7526bec84) - Bug 1167006 - Refactor marker details to not handle stack traces explicitly, and move logic into marker utils. Separate out some source link styles. r=vp (5b69886a7) - remove rough hack, shall be introduced later clean witha round like TFF or proper clamping/jittering as with FF 1443943 (91a202517) - Bug 1155761 - User Timing API in Workers, r=ehsan (0b7d3fe7d) (3517c9d90) - import changes from mozilla upstream: - Bug 1167489 and bug 1153672 - Clamp the resolution of performance.now() calls to 5us, because otherwise we allow various timing attacks that depend on high accuracy timers. r=froydnj, a=abillings (aeab359c2) - Bug 1186489. Apply the performance.now() resolution clamping in workers as well. r=froydnj (e56bc3c9a) and changed to 20us as bug 1427870 suggested. (cbb05a977) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1168411 - Disable test_sharedWorker_performance_user_timing.html on B2G debug for intermittent failures. r=RyanVM (844f2ecd9) - Bug 916893 - Patch 1 - Notification on workers. r=khuey,wchen (058a84ae1) - Bug 916893 - Patch 2 - Deal with onclose. Some grammar fixes. r=wchen (19cf84c5a) - Bug 916893 - Patch 3 - Walk up worker chain to find correct window for WorkerNotificationObserver. r=khuey (ee1f0c3fc) - Bug 1190176 - Make LayerScopeWebSocketManager thread safe. r=:djg (92c5ee233) - Bug 1196682 - DebugDataSender is not thread safe. r=kamidphish (6825e5d9b) - Bug 1196682 - fixup static analysis build bustage on a CLOSED TREE; r=me (eccf7b0cc) (2d7e3ad42) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1175138 P1 Make the dom.caches.testing.enabled pref available in workers. r=ehsan (fe47d0e0a) - Bug 1175138 P2 Expose dom.serviceWorkers.testing.enabled to workers. r=ehsan (efab5d0d3) - Bug 1160458 - Part 1: Use the CSP of the principal passed to CreateServiceWorker. r=nsm (4d0a1d742) - Bug 1172948 - Part 3: Add an explicit test case to ensure that authenticated origins that have a non-authenticated parent cannot register a service worker; r=nsm (78b3087c9) - Bug 803537 - XHR crashes in workers and in debug-builds when blob URLs are used from file scheme documents, r=khuey (aa86f77b7) - Bug 1163900 - crash in mozilla::net::nsHttpChannelCacheKey::GetData(unsigned int*, nsACString_internal&), r=jduell (adb5ddb01) - Bug 1147746 - Null check mInterceptListener in HttpChannelChild::ResetInterception; r=jdm (4c8c4e630) - Bug 1157283 - Recreate IPC redirected HTTP channels as necessary after intercepting the request in the child. r=mayhemer (3b144e45e) - Bug 1172884 P1 Properly decode body when intercepted response redirects. r=jduell (f49c37d4f) - Bug 1172884 P2 Add test for synthesizing a redirect to a compressed resource. r=ehsan (823d2122a) - Bug 1160458 - Part 2: Test. r=nsm (02b9fb3a0) - Bug 1169249 - Unregister service worker registration when uninstalling a service-worker-enabled application. Tests. r=baku (5509a19d6) - Bug 1177621 - SharedWorkers should not be shared between a private and a non-private documents, r=nsm (0836234c7) - Bug 1175138 P3 Expose the devtools SW testing flag on workers. r=ehsan (aade20454) - Bug 1173467 P3 Pass private browsing flag into CacheStorage factory methods. r=ehsan (c4d062a80) - Bug 1173467 P4 Add a test to validate Cache in private browsing window. r=ehsan (dde897e69) - Bug 1162487 - Enable the dom.caches.enabled pref in test_chrome_constructor.html; r=baku (2c73e2929) - Bug 1175138 P4 Enable dom.caches.testing.enabled in existing tests. r=ehsan (c453e03fb) - Bug 1175138 P5 Make CacheStorage reject on untrusted origins. r=ehsan (c85424d4e) - Bug 1175138 P6 Add a simple test to verify CacheStorage rejects in http origin. r=ehsan (5832eb99d) - Bug 1179567 - Make ServiceWorker keep its document and window alive; r=baku (1ae847884) - Bug 1179982 - Fix all compile errors in dom/workers on non-unified build. r=mrbkap (d30bece64) (963b86a51)

New build of Serpent/UXP for XP! Test binary: Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20210515-e29e57e-uxp-e98338935-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20210515-e29e57e-uxp-e98338935-xpmod.7z source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/custom IA32 Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20210515-e29e57e-uxp-e98338935-xpmod-ia32.7z source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/ia32 NM28XP build: Win32 https://o.rthost.win/palemoon/palemoon-28.10.3a1.win32-git-20210515-2ad022d8c-uxp-e98338935-xpmod.7z Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.3a1.win32-git-20210515-2ad022d8c-uxp-e98338935-xpmod-sse.7z Win64 https://o.rthost.win/palemoon/palemoon-28.10.3a1.win64-git-20210515-2ad022d8c-uxp-e98338935-xpmod.7z Official UXP changes since my last build: - Issue #1751 -- Remove cocoa support code from /dom (ca35efb84) - Issue #1751 -- Remove cocoa and uikit widget support code (08da125d9) - Issue #1751 -- Update crashtests (b5ece9f77) - Issue #1751 -- Remove cocoa and uikit gfx and hal support code (1ee35eafa) - Issue #1770 - Fix incorrect reference to event parameter (5c583e905) - Merge pull request 'Fix undefined event error thrown when scrolling on the auto scroll pop-up' (#1771) from FranklinDM/UXP:scroll-wheel-event-work into master (2b7197465) - [network] Fix typo in port blocking range check. (e98338935) No official Basilisk changes since my last build. Official Pale-Moon changes since my last build: - Issue #1865 - Remove mac-specific resources (2e3103d23) - Back-end branch pointer update. (2ad022d8c)

It's not that simple as it may seem . This should bother you much more . Have you ever heard about IPT ? It's your unique ID that is generated by all Intel (maybe AMD too) CPUs. I'm not sure about the exact year of it's implementation , but a quick searh reveals that it was there in 2014 already. From Intel's wevsite: "Intel Identity Protection Technology is a built-in security token technology that helps a hardware-based proof of a unique user’s PC to websites, financial institutions, and network services; " . Take this one , for example . It's more than a 6 y.o. old basic/cheap mobile CPU from 2015 and it has this "technology" built in. https://ark.intel.com/content/www/us/en/ark/products/89611/intel-pentium-processor-4405u-2m-cache-2-10-ghz.html It was available in earlier models too , starting from 2013-2014 (at least). Found this one article from 2014 : "this token is built right into the device hardware, it is much more convenient " , funny , huh ? "convenient" for them , right. https://www.askwoody.com/forums/topic/intel-identity-protection-technology/ Outright privacy invasion , a very powerful hardware based tracking tool! But no info anywhere ! Amazing.

For people who enjoy testing. I would like to express my thanks for the work. Jumper, I know you've spent endless hours with the further development of KernelEX. I'm able to run Firefox 35 and even 52.9 and see almost the whole contents of the web on Windows Millennium again. Versions 24 up to 35.0.1 are still usable for some websites. However, these versions are no longer up to today's requirements. There is a workaround to run higher versions up to 52.9 as well as RT's Serpent 55/52, NewMoon 28 and others - see below. ftp.mozilla.org/pub/firefox/releases Firefox Portable: sourceforge.net - FirefoxPortable Firefox 24.0 - 35.0.1 Requirements: KernelEx: KernelEx.dll 4.5.2016.25, Kexbasen.dll 4.5.2016.25, Kexbases.dll 25, Core.ini 25h with Kexcom.dll 18, Sheet.dll 19, verify.exe 18 and VKrnlEx.vxd of 18 Set Firefox.exe, Xul.dll and furthermore for portables the FirefoxPortable.exe to Windows XP SP2 or higher. Otherwise you get an error "Couldn't load XPCOM". Explanation: Xul.dll is set to 2k in the Registry (HKEY_LOCAL_MACHINE\SOFTWARE\KernelEx\AppSettings\Configs). This string entry *\XUL.DLL="NT2K" can be deleted to get the file in follower mode (default). Reboot required. 2k mode was set on Xul for older FF versions like 9.0.1. Don't choose Legacy WinXPSP2! Firefox will be very unstable. Recommended Hardware: Pentium 4 or newer processor that supports SSE2 512MB of RAM 200MB of hard drive space After version 48, SSE2 CPU extensions are going to be required on Windows. Improved version 31.8 for ME/98: Details: Download: Firefox 31.8esr portable Extensions: Compatibility PlainOldFavorites 1.3.1 Firefox 3.0 - 56.* QuickJava 2.0.5 Firefox 20.0 - 36.* ClassicThemeRestorer 1.5.3 CTR Firefox 29.0 - 50.0 Newer Versions and Forks: Firefox 42.0 - 48.0.2 Additional requirements: SeaMonkey Thunderbird Cyberfox, GNU IceCat Light Browser NewMoon 27 Firefox 49 - 52.9, Mypal 28/29, Centaury and SeaMonkey 2.46 - 2.49.5 , IceDragon Requirements: Extensions: Classic Add-ons Archive (Firefox 45 - 56). It is a very large collection of extensions and is almost indispensable. FF43 - 47 and 52 ESR: "xpinstall.signatures.required" => false to install unverified add-ons (already set in RT's builds). A workaround is required for versions 48: Override the Firefox Add-on Signing enforcement This is usually also necessary for higher versions, but not in every case. Firefox extensions on archive.org/WaybackMachine Firefox Legacy Collector Serpent/NewMoon 28/IceApe... - Testers are welcome. Short summary to run these browsers: Screenshot with add-ons Plain Old Favorites, Get Archive, Free Memory Button and Classic Add-ons Archive: Note: There is still no known workaround to run FF 36 - 41 (crash on start) and the corresponding forks with a few exceptions like IceCat 38. Versions 12 - 23 are freezing after some seconds and are useless in today's internet anyway. Limitations, issues, solutions, other settings and recommendations: Limitations: WebGL and Direct2D! Blocked for graphics card because of unresolved driver issues. Limited WebGL support with adequate graphics driver and graphics card: WebGL 1.0 based on OpenGL ES 2.0 and requires OpenGL 2.0 graphics driver support. 1. Nvidia Geforce 7 (full support OpenGL 2.0) 2. Drivers NV81.85 and higher 3. Settings: webgl.force-enabled => true and webgl.disable-angle => true Browser crashes or freezings are possible and are probably caused by graphics drivers infirmities. So turn it on only when necessary. WebGL-Renderer ANGLE: 1. It needs at least hardware providing DirectX VertexShader 2.0 and PixelShader 3.0 (vs2.0 ps3.0), DirectX 9c. (graphics cards Nvidia Geforce 6 and 7 series - Shader Model 3.0) 2.Setting: webgl.force-enabled => true It doesn't cause any crashes or freezings. It can be switched on permanently. D3D11 compositing: There is simply neither OS/Software support nor a DirectX11 capable graphics driver for ME/98. First Nvidia graphics cards with DX11 support come with GeForce 400 series. Direct2D requires Direct3D 11 compositing. There is no workaround. Common issues: * It is becoming more and more common that some web pages no longer open. For this reason an user agent switcher addon like User-Agent Switcher 0.7.3.1 (Firefox 1.0 - 56*) is increasingly important. Certificates may also have expired. The cert file is nssckbi.dll (NSS Builtin Trusted Root CAs). Replace the file by a newer one of Firefox 52 or higher. * Sound/audio doesn't work with vxd sound drivers (see topic). Use a WDM driver. The support for vxd sound drivers has been removed with version 15. Issues all versions 24 - 52 and forks on ME/98SE: * Adding bookmarks doesn't work (also the popup - dialog window doesn't occur). Recently visited addresses aren't saved. Partial solutions: The History and the Bookmarks created on XP+ can be used on ME/98. Copy the file places.sqlite of your Firefox profile on XP+ into the profile folder of Firefox on ME/98. And there is the bookmark extension PlainOldFavorites. * Firefox crashes with Adobe Flash plug-in activated. Make sure that this plugin is disabled. If it is still needed, deactivate the plugin container: dom.ipc.plugins.enabled;false and dom.ipc.plugins.timeoutSecs;-1. The boolean "dom.ipc.plugins.enabled" has been removed with version 43. * Own fonts of pages can be blurred or unreadable. Recommended: Uncheck "Allow pages to choose their own fonts". Tools > Options > Content > Advanced * The popups for installing addons, saving passwords and such have big black lines around them. * The button on the download popup "open containing folder" has no function. * Printing doesn't work. * Squares and boxes instead of icons: Install suitable fonts - Awesomefont (version 4.7 tested) for this Site and and see Workaround Possible setting (You can even see the "search magnifier" and the "Rate stars"): browser.display.use_document_fonts;1 gfx.downloadable_fonts.enabled;false Note, older versions, like 4.7, have recently been removed from GitHub, but are available on archive.org or fontawesome.com/versions. * Last character of the URL in address bar is missing - The search bar can also be affected: Different solutions are possible. * Possible: Second process "Explorer" remains after close. Open the Taskmanager and close the lower (second) "Explorer" process. * Cookies may not be saved correctly and aren't recognised after browser restart. There is no workaround. * Download list in Library is empty after browser restart. * Set Firefox as default browser: Start=>Run > ...\uninstall\helper /SetAsDefaultAppGlobal Install version: C:\Program Files\Mozilla Firefox\uninstall\helper.exe /SetAsDefaultAppGlobal Portable version e.g.: H:\Programme\Firefox\52.9.1\App\Firefox\uninstall\helper.exe /SetAsDefaultAppGlobal Note, that the used profile isn't the portable one initially. Usually "C:\WINDOWS\Application Data\Mozilla\Firefox\Profiles", if not changed. * NewTabPage doesn't work. You can set an url with an extension like ClassicThemeRestorer or newtab.url or others. FF49 - 52.9 and forks: * Slipped letters. Open about:memory and press "Minimise memory usage". The addon FreeMemoryButton allows to minimise memory with one click only. Mypal 28/29 and NewMoon 27/28: * Input in address bar doesn't work on second browser start anymore and Search Engine List is empty. Problem solving: With the first run you can remove search engines you don't want from the Search Engine Manager list and then set the search.json file in the profile folder quickly to read only. After that add your favorite engines and set a Home Page. Restart FF and check the functionality. If the first attempt fails, delete "search.json" and "searchplugins", restore the original state (Restore Default Search Engines ...) and try it once more. If you despair, delete (or rename) search.json , create an empty text file, rename it to search.json and set it to read-only before you start the browser. Mypal 28/29: * Low GDI resources - Possible problems with programs starting/running at the same time. Open about:memory and press "Minimise memory usage". The addon FreeMemoryButton allows to minimise memory with one click only. * Sometimes I can't close Tabs after a while. Restart Mypal when it annoys. * Sometimes - Error in Xul (browser restart required) Serpent/Newmoon/Iceape 20250621: * Additional issues due to Sqlite lib update 3.50.1. A Release up to 20250531 might be the better choice. Spoiler - issues FF 24 - 51 on ME/98: All significant issues should be ascertained. Other settings: * FF24 - 26: "security.tls.version.max" should be set to 3 (TLS 1.2 enabled). TLS 1.2 is enabled since version 27 by default. * Youtube: Note - The button "Remind Me Later" on request page for Browser update doesn't work anymore. This means that youtube no longer works. Now also version 52 and Mypal 29.3 are affected. One solution is to change the useragent (Firefox 68 or higher). Or you can use RT's forks like Serpent or Newmoon anyway. Another possibility seems to be to set in options "Limit Cache to MB of space" to 0. This disables disk caching. FF33 - 35: Set "media.mediasource.enabled" to true. (probably obsolete) It's possible to enable "media.mediasource.enabled" on lower versions, but it's quite likely useless. FF42: Set "media.mediasource.webm.enabled" to true. (probably obsolete) It should run on FF43 and higher by default. In case the player doesn't run on FF43+, set "media.mediasource.webm.enabled" to true, too. (probably obsolete) Settings not needed, but interesting - matter of taste: * "browser.tabs.onTop" => false to move the Tabs toolbar below other toolbars (FF24/27). FF 29 and higher: You can move the browser tabs with the extension ClassicThemeRestorer. * "browser.search.showOneOffButtons" => false to restore Firefox's classic search where you could see the favicon for the current search engine (FF33 - 42). However, the new search should be a bit faster. * "browser.urlbar.trimURLs" set to false to see protocol / scheme in URL (http/https...) * "browser.tabs.closeWindowWithLastTab" => false avoids browser exit and can show the homepage Recommendation for testing: Use portable versions or create different profiles for the installed. In the second case call the Profile Manager (Start > Run > ..../firefox.exe -P) or open "about:profiles" (Firefox 46+ only) and start the Profile Wizard. A note about my test environment: All my observations, tests, and findings were on real machines and not on a VM. All processors (Intel) meet SSE2 or higher and memory sizes are 2 and 3.5 GB RAM. The graphics cards are NV GF7800GT, NV GF7900GS, NV GF6600GT(AGP) and NV GF5900XT(AGP). And my OS is Windows Millennium. Tests with graphics cards e.g. Nvidia GF 2/3/4 series or from other manufacturers are unknown. Recommended versions for use on ME/98: Update cert file nssckbi.dll by a newer one of Firefox 52.9 or higher For troubleshooting see issues KernelEx Core Updates only: Version 27.0.1 (Tab bar glitch easy to solve) or better Improved 31.8esr CPU Instruction Set SSE (Pentium 3 / AthlonXP) only systems: KernelEx Core Updates and some dll files required Version 45.9esr or 48.0.2 CPU Instruction Set SSE 2 and higher (Pentium 4 / Athlon64 and newer) systems: Note additional requirements Version 52.9

From my experience , the results may vary with different hardware . You can just place your three modules into that PC and test how fast (or slow) it performs. There are plenty of benchmarking tools. And then you can try placing them in another slots , yes , it can perform quite different . Then try to add 2GB to the rest of them (3x4GB + 2GB). Try NOT to leave corsair in the first slot though , since Corsair tends to develop faulty sectors right at the start of a given module , which will prevent you from booting at all.

NUSB includes english systems files replacing whatever os got. I had same issue on finnish Windows 98SE. One option is to manually replace files with right ones from extracted NUSB

I thought it a bit strange about the Corsair 'hate' here as my experience with my primary desktop (4 x 2GB XMS3 DR3 1600MHz) has been very good, as I mentioned earlier working without problems since 2016. But prompted by those comments I did some online research on RAM and I did find a significant number of negative reviews from users on both other forums and retailers web sites. That does have to be tempered with the knowledge that Corsair RAM is a top seller, probably the top seller, which necessarily means there are going to be more people complaining if they have a problem. However there does seem to me to be a concerning disproportionate trend over recent years reporting modules failing after comparatively little use than for any other branded RAM. Going back to the main topic of the thread; now I have the alternative new 2 x 8GB modules I also have to decide what to do with, now 3 X 4GB good Corsair XMS 4GB 1600MHz DDR they've replaced. Is it worth replacing the matched 4 x 2GB XMS3 set in the other PC with them ie. will 8GB of RAM dual channel 2GB be usefully out performed by 12GB made up of 2 x 4GB XMS dual channel plus a single as near as is possible matched 4GB one? I suspect I know the answer to the idea of mixing in one 2GB module (14GB) or omitting the single 4GB and using 2 x 2GB matched ones (12GB) instead but its an option so I thought I might as well ask.

roytam1, please remove the limitation 4K texture for hardware acceleration. This can return hardware acceleration for Radeon 9500 - X1250 and Intel GMA900 - 3150 graphics cards https://bugzilla.mozilla.org/show_bug.cgi?id=637089 And add opengl32.dll file for SSE2 + , that it would be possible to enable software WebGL on computers with a bad video card https://bugzilla.mozilla.org/show_bug.cgi?id=1193695 https://download.qt.io/development_releases/prebuilt/llvmpipe/windows/ (25-Sep-2014) and rename opengl32sw.dll

Personally, if it helps, I never search for drivers by "name". I would start by going to your Device Manager and selecting your Network adapters in question and uninstalling whatever you have installed (Vista default or otherwise). Highlight, right-click, uninstall. After you have uninstalled, now highlight, right-click, and select Properties. Go to the Details tab. Select Hardware Ids. You should see a list of numbers with VEN, DEV, and SUBSYS. Search for a driver using those identifiers and skip trying to find a driver by "name". This is from my WinXP x64 Device Manager -> Network adapters -> right-click Properties -> Details but you should be able to find something similar in Vista.

qemu

There is an old thread about this on the Pale Moon forum: https://forum.palemoon.org/viewtopic.php?t=14617 The technology involved a NPAPI plugin that could be set to "Never activate".

Is it normal that the Palemoon 28 interface becomes unresponsive while executing high CPU JavaScript on single core? Often I encounter a heavy webpage and want to press the stop button or close the tab, but cannot do that until the page has finished running its code. Then sometimes I close multiple tabs or the entire window once it starts accepting input. I also generally feel that the browser milks every bit of system power to itself with other programs feeling sluggish during a full speed file download over SSL. Lately I started running it at /belownornal priority, but that doesn't address the unresponsive GUI. I "upgraded" from Opera and Firefox 27 which felt better, compatibility aside.

New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.4.3-Goanna-20210515.7z Changelog: Out-of-tree changes: * update Goanna3 to git 32a8a4473...963b86a51: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1158442 - Remove the "Performance Entry buffer size maximum reached" warning; r=baku (98065e0a3) - Bug 1164552 - create new colors in light/dark themes for devtools specifically for rendering graphs in the performance tool. The highlight-* colors in light theme work well for text, less so for rendered blocks and appear very 'bold'. Most of the dark theme graph colors are the same as their highlight-* equivilents. r=vp (a5f33fb22) - Bug 1158731 - Buffer for Performance APIs (Resource Timing, User Timing) should be separeted. r=baku (87d4be360) - Bug 1166494 - Re-add console timestamp and style markers' metadata. r=vp (24c989240) - Bug 1166494 - part2: correctly add labels to console markers and properly add style markers' restyleHints. r=vp (ecf87cff4) - Bug 1166494 - part3: Correctly handle marker definition fields that are described via function rather than array. r=vp (b71b8575f) - Bug 1162662 - Map JS markers to human readable keys, and hide if platform related via (Gecko). r=vp (7526bec84) - Bug 1167006 - Refactor marker details to not handle stack traces explicitly, and move logic into marker utils. Separate out some source link styles. r=vp (5b69886a7) - remove rough hack, shall be introduced later clean witha round like TFF or proper clamping/jittering as with FF 1443943 (91a202517) - Bug 1155761 - User Timing API in Workers, r=ehsan (0b7d3fe7d) (3517c9d90) - import changes from mozilla upstream: - Bug 1167489 and bug 1153672 - Clamp the resolution of performance.now() calls to 5us, because otherwise we allow various timing attacks that depend on high accuracy timers. r=froydnj, a=abillings (aeab359c2) - Bug 1186489. Apply the performance.now() resolution clamping in workers as well. r=froydnj (e56bc3c9a) and changed to 20us as bug 1427870 suggested. (cbb05a977) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1168411 - Disable test_sharedWorker_performance_user_timing.html on B2G debug for intermittent failures. r=RyanVM (844f2ecd9) - Bug 916893 - Patch 1 - Notification on workers. r=khuey,wchen (058a84ae1) - Bug 916893 - Patch 2 - Deal with onclose. Some grammar fixes. r=wchen (19cf84c5a) - Bug 916893 - Patch 3 - Walk up worker chain to find correct window for WorkerNotificationObserver. r=khuey (ee1f0c3fc) - Bug 1190176 - Make LayerScopeWebSocketManager thread safe. r=:djg (92c5ee233) - Bug 1196682 - DebugDataSender is not thread safe. r=kamidphish (6825e5d9b) - Bug 1196682 - fixup static analysis build bustage on a CLOSED TREE; r=me (eccf7b0cc) (2d7e3ad42) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1175138 P1 Make the dom.caches.testing.enabled pref available in workers. r=ehsan (fe47d0e0a) - Bug 1175138 P2 Expose dom.serviceWorkers.testing.enabled to workers. r=ehsan (efab5d0d3) - Bug 1160458 - Part 1: Use the CSP of the principal passed to CreateServiceWorker. r=nsm (4d0a1d742) - Bug 1172948 - Part 3: Add an explicit test case to ensure that authenticated origins that have a non-authenticated parent cannot register a service worker; r=nsm (78b3087c9) - Bug 803537 - XHR crashes in workers and in debug-builds when blob URLs are used from file scheme documents, r=khuey (aa86f77b7) - Bug 1163900 - crash in mozilla::net::nsHttpChannelCacheKey::GetData(unsigned int*, nsACString_internal&), r=jduell (adb5ddb01) - Bug 1147746 - Null check mInterceptListener in HttpChannelChild::ResetInterception; r=jdm (4c8c4e630) - Bug 1157283 - Recreate IPC redirected HTTP channels as necessary after intercepting the request in the child. r=mayhemer (3b144e45e) - Bug 1172884 P1 Properly decode body when intercepted response redirects. r=jduell (f49c37d4f) - Bug 1172884 P2 Add test for synthesizing a redirect to a compressed resource. r=ehsan (823d2122a) - Bug 1160458 - Part 2: Test. r=nsm (02b9fb3a0) - Bug 1169249 - Unregister service worker registration when uninstalling a service-worker-enabled application. Tests. r=baku (5509a19d6) - Bug 1177621 - SharedWorkers should not be shared between a private and a non-private documents, r=nsm (0836234c7) - Bug 1175138 P3 Expose the devtools SW testing flag on workers. r=ehsan (aade20454) - Bug 1173467 P3 Pass private browsing flag into CacheStorage factory methods. r=ehsan (c4d062a80) - Bug 1173467 P4 Add a test to validate Cache in private browsing window. r=ehsan (dde897e69) - Bug 1162487 - Enable the dom.caches.enabled pref in test_chrome_constructor.html; r=baku (2c73e2929) - Bug 1175138 P4 Enable dom.caches.testing.enabled in existing tests. r=ehsan (c453e03fb) - Bug 1175138 P5 Make CacheStorage reject on untrusted origins. r=ehsan (c85424d4e) - Bug 1175138 P6 Add a simple test to verify CacheStorage rejects in http origin. r=ehsan (5832eb99d) - Bug 1179567 - Make ServiceWorker keep its document and window alive; r=baku (1ae847884) - Bug 1179982 - Fix all compile errors in dom/workers on non-unified build. r=mrbkap (d30bece64) (963b86a51) * Notice: the changelog above may not always applicable to XULRunner code which K-Meleon uses. A goanna3 source tree that has kmeleon adaption patch applied is available here: https://github.com/roytam1/palemoon27/tree/kmeleon76

New build of BOC/UXP for XP! Test binary: MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20210515-355db4de-uxp-e98338935-xpmod.7z BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20210515-355db4de-uxp-e98338935-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom-older * Notice: the profile prefix (i.e. parent folder names) are also changed since 2020-08-15 build, you may rename their names before using new binaries when updating from builds before 2020-08-15. -- New build of HBL-UXP for XP! Test binary: IceDove-UXP(mail) https://o.rthost.win/hbl-uxp/icedove.win32-20210515-id-f0915f3-uxp-e98338935-xpmod.7z IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20210515-id-f0915f3-ia-c642e3c-uxp-e98338935-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/icedove-uxp/tree/winbuild https://github.com/roytam1/iceape-uxp/tree/winbuild for UXP changes please see above.

Thanks, that's fascinating stuff! I bought my first PC in 1993 in the Windows 3.1 days, but I never attempted dual booting until I dual booted Windows 98 and Windows 2000, in 2000! I thought that having an NT option would be better for video editing, which I'd just got into, faster and more stable than Windows 98. I waited ages for Windows 2000 to come out because I wanted to wait for FAT32 capability on the NT system, as Windows 98 couldn't use NTFS of course.

No, there is not any such limitation for Vista+. There is the need of a "boot" (what the MS guys call "system") volume, i.e. an active, primary partition (which normally would get a C: drive letter) where the boot files must reside (i.e. namely) NTLDR, NTDETECT.COM and BOOT.INI for XP and BOOTMGR and \boot\BCD for Vista, then you can have (say) XP on another volume, let's say D: and Vista in yet another volume, let's say E: (or viceversa, XP on E: and Vista on D:) these other volumes can be either primary partitions or .logical voluems inside extended As an example, I have (or have had) machines with DOS on C: (this is non-negotiable) Windows 98 on D: (and yes, it is possible, though complex to install it this way), NT or 2K or XP on H: and Vista or 7 on G:. jaclaz

that's why i don't use ubuntu anymore they only care about what they want not you

weird (that corona wont fly away)

You're welcome!