Leaderboard

New build of Serpent/UXP for XP! Test binary: Win32 https://o.rthost.win/basilisk/basilisk52-g4.7.win32-git-20201031-ffb32e0-uxp-6a4c3caa8-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk52-g4.7.win64-git-20201031-ffb32e0-uxp-6a4c3caa8-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-g4.7.win32-git-20201031-ffb32e0-uxp-6a4c3caa8-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-28.10.2a1.win32-git-20201031-fcd19efc9-uxp-6a4c3caa8-xpmod.7z Win64 https://o.rthost.win/palemoon/palemoon-28.10.2a1.win64-git-20201031-fcd19efc9-uxp-6a4c3caa8-xpmod.7z Official UXP changes since my last build: - [netwerk] Make nsIncrementalStreamLoader's GetNumBytesRead threadsafe. (28119b040) - Bump platform version for added features. (d5e5bef3c) - [layout] Avoid negative availSize.BSizes in paginated table reflow. (2cd89d584) - [layout] Re-order rowgroups if reflowing. (6eba9263c) - Update docs for change of repository host. (1950e886b) - Update docs for change of repository host. (0daf842cf) - Issue #1656 - Nuke the remaining vim lines in UXP (6a4c3caa8) Official Basilisk changes since my last build: - Update submodule location (affb1cb) - Nuke vim configuration lines from the front-end (no code changes) (0a95895) - Update back-end branch pointer (ffb32e0) Official Pale-Moon changes since my last build: - [branding] Use about:blank as default newtab for unofficial. (810a8db8c) - Update submodule location (b82a4e75b) - Update 'README.md' (c269e3d0e) - Update 'README.md' (0bd23b462) - Graft updated protected branding assets. (85f7050a7) - Nuke vim configuration lines from the front-end (no code changes) (fcd19efc9) My changes since last build: - skipped branding related commits

Yes, the domain changed recently. Try this: http://o.rthost.win/gpc/files1.rt/palemoon-26.5.0-20180718.win2000.7z

UXP (Unified XUL Platform) is the name of the platform on which several upstream and downstream (forks) projects are built... Organisation: Moonchild Productions (MCP) UXP applications: Pale Moon 28 (+29), Basilisk (52.9.*) Organisation: Hyperbola (unofficial abbreviation: HBL) UXP applications: Iceape-UXP+Icedove-UXP Organisation: Binary Outcast (offical abbreviation: BinOC, unofficial abbreviation: BOC) UXP applications: Borealis Navigator+Interlink Mail & News Thus, Basilisk/UXP -> Serpent 52.9.0 Basilisk/Moebius -> Serpent 55.0.0 (*) Pale Moon 27/Tycho -> New Moon 27.9.x (*) Pale Moon 28[/29]/UXP -> New Moon 28.10.x Borealis Navigator/UXP -> BNavigator Interlink Mail & News/UXP -> MailNews The Hyperbola forks are meant to run only on said Linux distro, but roytam1 has forked them to compile and run on WinXP+... (*): The Tycho and Moebius official platforms have long been deprecated (and also purged completely from official repositories ) , roytam1's forks currently use code from other projects to get updated...

win32k declare own ring3->ring0 service table for user32 and gdi32 see KeAddSystemServiceTable() inside win32k

I apparently missed this announcement, but if you're once again so inclined, @win32 : DST Correction in Windows for the Fiji Islands: October 13, 2020 https://support.microsoft.com/en-us/help/4578623/dst-correction-for-fiji-islands Details: This update makes a daylight saving time (DST) end date correction in Windows for the Fiji Islands whereby at the end of daylight saving time (DST), the Fiji Islands will return to Fiji Standard Time on Sunday, January 17, 2021 at 03:00. (The current erroneous DST end date for the Fiji Islands is Saturday, January 10, 2021 at 03:00.)

Thank you, @hpwamr, for this wonderful reference for sharing files! Might I ask for an additional, and parallel, set of recommendations? That would be specifically for shared images. Some of your above recommendations might be able to fill this request, but different sites might be required. Yes, all of the above would work if you treat the images just as a file, and that's fine, but if you want to share an image that will be displayed in a forum post like you used imgur.com above in the first post of this thread, for example, then different requirements come into play. I can think of a few "requirements" off hand, but there might be others: Direct links to the image must be provided. The image should be maintained as uploaded and not "optimized", degrading its quality. Downloading the image should result in a file with the same file hash as it had when uploaded. The larger the maximum image size allowed, the better, with at least "wallpaper size" as a minimum. Today's availability of 4K and larger screens means that is quite large. A reasonable limit by either dimension or file size is acceptable, but it should be specified. I prefer the image to be displayed as uploaded, others like to use thumbnails. The option for either would be best, with multiple possible thumbnail sizes being ideal. The images should be maintained for as long as possible, preferably indefinitely, and be able to be downloaded and viewed by as many people and as often as desired. The more image formats supported, the better. The site should not require registration, but at the minimum registration should be free. Registration should provide additional features such as file maintenance. Registered free users might have a limit to the number of images allowed, and that's expected as long as the limit is a reasonable one. Extending those limits with a subscription fee should provide additional features in addition to larger capacity. NO spamming, ads, download delays, bandwidth throttling, etc. Forums might impose their own limits to image display, so the image host should have as few of its own as possible so as not to create any bottlenecks. I think it would be great to include this info in this thread as a central reference point for shared files. Using these services helps out all forums regarding the bandwidth and storage space they would otherwise have to support. Many image hosting sites have died over the years, such as TinyPic, so a current set of recommendations would be appreciated. Cheers and Regards

I will mirror the new versions in the next minutes/hours... It seems you forgot to replace the copy & paste with the right link for the Win64 version, also on your blog. Edit: Latest NM27 causes reproducible crashes when opening the important German IT portal www.heise.de (see attached screenshot) and when playing a video on YouTube (i.e. Édith Piaf - Non, Je ne regrette rien) - build 20201024 from last week works. -> Finally I reverted back to NM27 SSE2 202010324, for now...hopefully this issue can be fixed. Additional information: NM27 normal (SSE2) version, JS allowed, tested on two WinXPSP3 POSReady 2009 "hacked" machines. Edit2: I also want to mirror the other builds, but I'm very confused by their names. What does "BOC", "HBL" or "UXP" mean, for example? original name -> fork name Pale Moon -> New Moon Basilisk 52 -> Serpent (UXP) Basilisk 55 -> Serpent (moebius) MailNews -> BOC??? BNavigator -> BOC??? IceDove -> HBL??? IceApe -> HBL??? Is the name schema above correct? Did I forget something? Edit3: Added the latest builds of New Moon 27 / 28 to soggi.org - tools. kind regards soggi

New build of post-deprecated Serpent/moebius for XP! * Notice: This repo will not be built on regular schedule, and changes are experimental as usual. ** Current moebius patch level should be on par with 52.9, but some security patches can not be applied/ported due to source milestone differences between versions. Test binary: Win32 http://o.rthost.win/basilisk/basilisk55-win32-git-20201031-9db2e14ff-xpmod.7z Win64 http://o.rthost.win/basilisk/basilisk55-win64-git-20201031-9db2e14ff-xpmod.7z repo: https://github.com/roytam1/basilisk55 Repo changes: - remove camellia-gcm code from repo as they should be in separated patch until it is working (747093d89) - partly import changes from tenfourfox: - #627: M1631583 M1631597 M1636771 M1637222 M1649316 M1651520 (2373458b5) - #627: M1631573 (0abd0fc5d) (6ea417b15) - import tzdata, DNS, pins, HSTS from tenfourfox (as of rev 16c85bb3) (f72df46c2) - follow-up of rev 747093d89ec14002e39883d97941b50014a191e0 in platform's security manager part (d1a62298d) - import change from tenfourfox: - #622: M1646140 (665369699) - closes #626: M1440677 (81acd2b99) - #612: M253143 M1643126 (9d3a83602) - #628: M1663642 M1663439 (bc184b4c4) - #628: 2020b timezone update plus LTDs + HSTS (1387da73f) (9db2e14ff)

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201031-16988569b-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201031-16988569b-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201031-16988569b-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.9.7.win64-git-20201031-16988569b-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 1177892 part 1 - Remove BOOLEAN_TO_JSVAL and STRING_TO_JSVAL. r=evilpie (1fac17ff3) - Bug 1177892 part 2 - Remove PRIVATE_TO_JSVAL. r=evilpie (4d7e80200) - pointer style (cd27bd41a) - Bug 1177892 part 3 - Remove OBJECT_TO_JSVAL. r=evilpie (ff66aebe5) - Bug 1155342 - Disallow flagging a [NewObject] method with a [DependsOn] value that implies it might return the same value when called twice. r=bzbarsky (7ba1bb806) - Bug 1129239. Don't require 'optional' keyword on trailing dictionary arguments if the dictionary has a required member. r=smaug (3ee24f57b) - Bug 1152902 part 1. Add a way to flag a method in webidl as being identity-testable, so we'll expose such an identity test from bindings to other C++ code. r=peterv (43e038677) - Bug 1152902 part 2. Add a fast path for the case when a Promise is resolved with another Promise. r=nsm (59f56603e) - Bug 1172785 - Adding StaticClassOverride routing for JS implemented WebIDL, r=peterv (6ae76e198) - Bug 1172785 - Using RTCCertificate for WebRTC, r=ekr (a69ee6167) - fix build, file built twice (04f7e3d75) - Bug 1155942. Treat USVString and ByteString as serializable values. r=bkelly (33e635737) - Bug 913053. Split up IDLTypedef and IDLTypedefType, and likewise for IDLCallback vs IDLCallbackType, so it's clearer whether we're operating on types or objects represented by those types. r=khuey (661f45319) - Bug 1168471 - Implement support for SharedArrayBuffers and SharedArrayViews in WebIDL. r=bz, r=lth, r=luke (77d63babc) - Bug 1151269 - Fix the test used by Ion ICs for whether values can definitely be written to an unboxed object, r=jandem. (0463d6b53) - Bug 1139474 - Watch for unboxed object properties when attaching Ion SETPROP ICs, r=jandem. (67bc9884d) - Bug 1162199 - Use unboxed objects by default, r=jandem. (4dba9da90) - Bug 1166678 - Inline first ObjectGroup::maybeSweep test, r=jandem. (ff179acb3) - Bug 1166709 - After converting unboxed objects created by some initializer to natives, create native objects at that allocation site in the future, r=jandem. (14a5c92b5) - Bug 1166709 - Mark definite properties when replacing the unboxed group used for literals at some pc, r=jandem. (487017105) - Bug 1170372 - Use unboxed arrays for Array() and other functions keyed to allocation sites, r=jandem. (9153a5313) - Bug 1175535 - Don't require objects embedded in MIR nodes to always be tenured, r=jandem. (75399d353) - Bug 1190272 - Improve type checks when storing values into unboxed objects in Ion code, r=jandem. (0a91a08d5) - Bug 1216130 - Fix extra checks for unboxed objects in PropertyWriteNeedsTypeBarrier. r=bhackett (4862b91ca) - Bug 1166700 - Tolerate null failures targets when storing to an unboxed object must fail, r=jandem. (87a07aa3d) - backport of Bug 1389436 - Explicitly instantiate gfxFont::GetShapedWord<uint8_t> for its use in gfxTextRun.cpp. r=jfkthame (5f6fc9f18) - Bug 1164374 - Use StaticMutex in BrowserProcessSubThread. r=froydnj. (b0ad93d17) - Bug 1167771 - Simplify the pre-barrier verifier's tracer use; r=sfink (0fc926804) - Bug 1166037 - Part 2 - Modify common Baseline code for ARM64. r=djvj (739b36d83) - Bug 1168864 followup - Remove unnecessary addProperty check from CheckHasNoSuchProperty. r=bhackett (3a464d07e) - Bug 1166037 - Follow-up - Revert SimulatorType to Simulator. no_r=me (613b095d3) - Bug 1166944 - Inline the only user of TraceObjectSlots; r=jonco (d830d12f0) - pointer style (d567d3be2) - Bug 1167291 - Generalize the marking tracer's cross-compartment check; r=jonco (b5f753147) - Bug 1167318 - Remove GetGCThingTraceKind in favor of the typed Cell variants; r=jonco (f698f3a64) - Bug 1167323 - Use internal interfaces to implement MarkCycleCollectorChildren; r=jonco (4acf85bdb) - Bug 1167433 - AccessorShape is an AllocKind but not a TraceKind; r=jonco (ea92cd26a) - pointer style (403e83b64) - remove namespace hack (e3663d39b) - Bug 1167453 - Rename JSGCTraceKind and make it a C++11 enum class; r=jonco (f7f435a16) - Bug 1165966 - Update destructor assertions to handle failed initalization r=terrence (48237ed24) - pointer style (f637042fc) - Bug 774364 - Part 1: Inline Math.random() in Ion on x86_64. r=sstangl (8379ab0b2) - Bug 1167677 - Try harder to find scratch registers for memory->memory MoveGroup moves, r=sunfish. (14d4b9962) - pointer style (93bb53345) - Bug 1185653 - Fix enumerate hook on unboxed objects to skip non-enume#able properties. r=jorendorff (4ecf41a10) - Bug 1125624, part 1 - Implement ValidateAndApplyPropertyDescriptor steps 3-4, so that (once the corresponding code in StandardDefineProperty is deleted) freezing an already-frozen object with an addProperty class hook will not call the hook. r=Waldo. (f67f98231) - Bug 1125624, part 2 - Change js::StandardDefineProperty to forward to s::DefineProperty. r=Waldo. (3ac5d9e30) - Bug 1148568 - In JSON.parse with a reviver callback, ignore failure when defining properties. r=Waldo. (30c35a758) - pointer style (826d092cf) - Bug 1166950 - Only give constructor functions a prototype. r=efaust (d1b909603) - Bug 1140482 - Add JSPROP_RESOLVING. Give NativeDefineProperty standard behavior in cases where a non-resolving define needs to trigger a resolve hook. r=Waldo. (b2d650abd) - Bug 1140482 followup: Update ResolveInterpretedFunctionPrototype() failure-cases to return false instead of nullptr, now that return type is bool. rs=jorendorff (38eef1812) - Bug 1148188 - part1: defaultShims. r=billm (9eb98a8fa) - pointer style (ec74889f1) - Bug 1101182 - One interpose call for one property access. r=bholley (111afff12) - pointer style (b8dd593e9) - Bug 1148188 - part2: interposeCall. r=billm (6fce5829e) - Bug 1125624, part 3 - Remove js::StandardDefineProperty and js::DefineOwnProperty. r=Waldo. (a8a228f6e) - Bug 1167244 - Handle nullptr return from maybeGetProperty(). r=bhackett (91958bdfc) - pointer style (ebfc629a6) - Bug 1170355 - Watch for indexes that don't fit in a jsid in Array.shift, r=jandem. (6c2f0a8a2) - pointer style (a31367643) - Bug 1165348 - Move Scalar Replacement after GVN. r=jandem (9296335d1) - Bug 1161584 - Add TrackedStrategy::SetProp_InlineCache. r=shu (612ea0b32) - Bug 923717 - Add IC fuzzing mode. r=efaust (a6a6a7460) - Bug 1166711 part 2.1 - Check Scalar Replacement with both unboxed object and without. r=bhackett (e839f034a) - Bug 1172943 - Use unboxed arrays for JSON and script literal arrays, r=jandem. (964d5a42f) - Bug 1162986 - Relax type requirements for using baseline cache information when compiling GETPROP, r=jandem. (68e234a08) - Bug 1168500 - Replace the operator, by variadic templates. r=Waldo (ccb173cb7) - Bug 1166711 part 0.1 - Use JitSpewPrinter instead of stderr. r=bhackett (239022b72) - Bug 1166711 part 0.2 - JitSpew add scope-base indentation level. r=bhackett (89288621e) - Bug 1129313 - Scalar Replacement: Remove PostWriteBarrier at the same time as the stores. r=h4writer (f9df0503e) (16988569b)

I recently start using as main browser New Moon 28.10.2a1 (32-bit) (2020-10-02) and finally stop using 28.3. I am very satisfied with (2020-10-02). It actually play YouTube videos just as good as old 28.3 - no micro-lags, no subfreezing. Now in versions after 28.3.0a1 (2018-12-21) there was (in my experience) for long period of time "choppy video playback". It was in versions of 2019 year and in early 2020 as i remember (again, in my experience). But in New Moon 28.10.2a1 (32-bit) (2020-10-02) video plays without micro-lags. So, from my experience as user, it looks like video-playback have now higher priority then other processes in browser or something like that. I think it was can be noticeable on single core machines also many users here on msfn say that they do not watch videos in browser and because of that nobody notice that. I don't know what actually was changed, but kudos to Roytam1 - (2020-10-02) version works just as fast and smooth as old good versions. And yes i know that chat don't work in that version because of google sabotage. But for me it's fine - YouTube chat usually filled with inadequate users, unfunny trolls and crazy fans of current "leader" who garbage spam every channel chat in the universe with incoherent rambling. Like you watch channel about astronomy and then some douchebag starts yelling with all caps in chat: maga!!!111.

Indeed, the September and October cumulative updates didn't affect the DWM files. Fingers crossed that further updates don't break Aero Glass. Man....if only Microsoft left us the choice to run Aero/Fluent transparency in-built instead of forcing users to rely on 3rd party solutions.

Only the updroots.sst file has the changed content, the other files have not changed.

Thank you for the kind words about my father. Rudy has two sons. Jason loew is his blood son and we have the same mothers and I am Tony Sotomayor. Rudy was my step dad. I miss him so much.And Im glad to see how he impressed his fellow programmers on this thread and how much they respect his work. I knew as a child Rudy was very intelligent and a gifted programmer. If anyone would like to download a picture of Rudy when he was in his last year in College at University of Oklahoma 1970(Sooner 1970). I have a pic of him that was taken of him at the Engineers Club Engineers week party pic. Maybe someone can put it on this thread or something? He was young fresh leaving college and healthy. Picture is how I like to remember him. You can email me direct for that pic or anything else you want to know about my dad at jasongeo2@aol.com

In case anyone's interested, I've managed to make a standalone executable of the latest youtube-dl on Python 2.7.17 (OpenSSL 1.02t). Requirements and dependencies: KernelEx and msvcr90.dll. http://s000.tinyupload.com/index.php?file_id=84433364913891885743