Leaderboard

After more in-depth investigations, tests and analyses , I have come to the conclusion that the issue, described in my previous posts, is primarily a timing problem in multiprocess mode and can be easily fixed by inserting a delay in the right position.

I run with 8GB page and 16 RAM, no issues, even with modern apps,

Pagefile is good, Chrome loves it. Many games simply won't start without it. A good example is Titan Quest (XP Era).

You sure? The bottom of your pic.

I have an almost 20 years monitor Dell, 2005 or so. The monitor isn't the issue. I also have no sun in my house, I practice black magic, they don't go well together, you know... or you don't know.

Jody, thank you my friend, you're very helpful as usual. I'll upvote you when the limit is recharged once again. You're full of good ideas, indeed!

I don't think I heard of it before, and I definitely wouldn't miss any good technical site.

Why is it important to write about it here? Was there any archived or somehow important info stored?

The early internet that we knew was a big library where everyone shared info and content. The modern internet is a shopping centre, and browsers money makers machines. Just fire the start page and it is already making money from you....do a search, visit a page, everything.... and remember.,,.... if you computer goes slow or run out of memory... its cause you need to buy new hardware for internet browsing... Adblocking is just the top of the big iceberg...

No I do not. "To each their own." I know how to test for list interference. My lists are fine, but thanks for the concern. You make the MISTAKE of assuming uBO is my only defense.

Personally, I don't expect to be able to surf every website properly with UXP browsers. The bloated, modern and totally over-greedy websites simply have to be accessed with other browsers or devices. Since there's a lot of Googlised junk on them, such sites are thrown at my Android (Google) tablet. That works great. Spoken for me only, I don't like the modern web layout. In my opinion, the entire development of websites is going in the completely wrong direction. I would rather focus on fast page loading and compatibility than overloading pages with all kinds of bells and whistles.

I think a big part of the problem is that Javascript wasn't originally designed to do the kinds of tasks modern Web sites have it doing now. Originally it was just intended to do simple "bells and whistles" tasks that the site could live without - hence you could disable it and still use the site, albeit with less functionality. But since then, it's evolved into a complex programming language that folks even write .PDF viewers in! In a perfect world, perhaps we'd start over with C-script or something; some kind of language easy to JIT compile into efficient machine code. Come to think of it, didn't Micro$oft try to push VBScript as an alternative, many moons ago when IE was the dominant browser? The effort failed because no other browsers supported VBScript, but perhaps we'd be better off now if M$ had succeeded. Or not. Just a thought.

Yeah, that's unusably slow on St 55 (not strictly UXP but close), even with e10s enabled on 64-bit W7 with 8 GB RAM. But I'm not surprised an Android developer site uses Javascript that's only fast on "modern" browsers. After all, everyone who goes there (present company excepted) probably uses Chrome on at least an 8-core processor. Just bloat the site up with as many "cool" features as you can think of, and if someone finds it slow, just quote "system requirements" at them rather than making the slightest effort to optimize the code. It's the modern way of the Internet. My point was only that UXP meets @j7n's definition of "retro" - new functionality with an old UI. I even conceded that it was slow! Not many sites are as bad as that one, though.

One good thing that you may like is when MSFN goes down, we usually post status updates there too, so it helps both ways

Well, I'm sorry. You're right, this feels bad. I've tried to outline a method to my madness; but feedback process still needs improvements. In your case, I've answered about your feedback on different forum and was wrong to hide it here.

I do not get this attitude, seriously. You act like we all collectively p***ed in your porridge or something, when we ask about some features being changed or missing. If the thread is such a garbage dump, whatever that means to you, you can chose not to answer. But instead of you act bitter and make people who actually paid for your software and who ask actual questions in regards to it, feel like they wronged you in some way. Silly attitude, in my humble opinion. The same with deleting posts just because you don't like the questions. Not to mention the double standards, the entire software is about modifying the start menu, taskbar and some other features, but somehow, when some people ask for some features that were present, they are told said feature "breaks functionality" or changed functionality, well doooh, that's the purpose of the software. Yet some other requests are implemented, those apparently do not break functionality and do not modify functionality. You could simply say you do the software as you like, which is fair, in which case, put it in the rules and that's it, no requests, only report bugs. So many ways to make this happen without the p*** poor attitude. Nobody forced you to open this garbage dump thread, nobody forced you to offer support anywhere but on the official email or whatever.

Noticed that too, just today. I couldn't notice earlier because I'm still mending up my eye damage.

What is "Eclipse.cx"?

StartAllBack 3.8.13

@feodor2 To be even more precise, the actual culprit is the command goDoCommand("cmd_copy") if used in multiprocess mode only for fetching selected content on a page without actively copying it via Ctrl+C or using an associated copy feature in the form of a menu item or button. Normal copying content with this command works correctly. BTW, I am only talking here about the native command goDoCommand("cmd_copy") without the help of specialised extensions or scripts. For the purpose of copying selected content automatically to the clipboard, there are a lot of extensions or user scripts. Needless to say, I don't know if it is possible to fix it natively in Mypal 68. However, I personally found different solutions to circumvent this issue or bug in my coding project without the help of specialised extensions or user scripts. Whether the ideas of my solutions can be used for a native fix in Mypal 68 or not, I can't assess. Nevertheless, if you are interested in, please let me know!

@feodor2 Again, same problem as quoted above but a bit more detailed. If you select a word on a page In New Moon or Serpent, the command goDoCommand("cmd_copy") in combination with the gClipboard.read() method fetches this selected word first and only then the last copied content from the clipboard during a second, following execution. In Mypal 68 (running in multiprocess mode), it is unfortunately the other way round. For me, it looks like an issue of execution priorities. So, my request. Can you restore the original execution order when this combination of commands is used in your browser such as it is in New Moon or Serpent? And just as a further information, the combination of the commands goDoCommand("cmd_copy") and goDoCommand("cmd_paste") works correctly in Mypal 68, though. Same applies to the gClipboard.read() method which solely executed also works correctly. @feodor2 After some tests, I can confirm that the issue I reported in my previous posts only happens in multiprocess mode. When running Mypal 68 in single-process mode, the command goDoCommand("cmd_copy") in combination with the gClipboard.read() method works correctly such as it is in New Moon or in Serpent. So, it is definitely a multiprocess mode issue. BTW, I really love to talk to myself.

I have re-enabled the Task Scheduler for Windows Defender once again. I do wonder just how long ago it started working again.

Seems so ... for now. I'll be keeping you posted

Well over a year and a bit later, Windows Defender has started "updating" again. The update button in Windows Defender actually works again! Go figure ???

Now interestingly, Windows Defender has started "updating again. For the last year or so, I've been downloading and manually running definition updates. Now the update button in Windows Defender actually works again!

Wow! I knew it! This new member now posts fake programmes. The account with the nick @lll888. I'd like to notify @Tripredacus. It's already quite a lot of posts that member made with dubious software and batch scripts, now this link to fake Kernel.

Thank you very much for the investigation, I felt like something fishy was going on there. I'll upvote tomorrow (I'm out of likes, they are limited).

Hello @George King! Any news for us? Although I am not a Windows XP 64 bit user, I am interested to hear whether you have made any progress in testing. Did you already install Windows XP 64 bit on your testing machine? Greetings, AstroSkipper

If a user owns a computer equipped with more capable hardware, why should Windows XP be used then? Simple rule: more modern hardware, more modern operation system. I am actually here to support real Windows XP users to inform them about security programmes. And this OS is ideal for old, weak computers. There are many people outside who want to protect their computers and have to use Windows XP due to very old, weak hardware for whatever reasons. People with powerful machines can do what they want to do. Whether it makes sense or not, I leave undecided. I have given my opinion on this often enough here in different threads. And if someone wants or needs to throttle the performance of their computer, then it makes perfect sense to install Avast. In any case, all my articles and investigations are meant to spread information including my experiences whether they are positive or negative. One thing is crystal clear. The mature readers can decide themselves what they want to use as I stated long before.

You are absolutely right! Avast Antivirus 4.8 was a version I liked very much, and it is even today installed on my Windows ME partition which still exists only for historical reasons. And as usual in those days, Avast was lightweight and low on resources. How can a manufacturer get so much wrong, considering versions like 18.8 or the very latest?

I thought that should be said at some point, no matter how long these manufacturers continue to support Windows XP. And as for Avast, frankly, I am not really a fan of it anymore, and certainly not of AVG. I would therefore be able to live with the loss...

Thanks for your efforts! TBH, I think there is something missing in Windows XP 64 bit. But in the end, we have to accept that Panda Antivirus Free is not compatible with this OS due to problems in recognising an already existing .NET Framework 4 installation. Panda Antivirus Free can't be installed out of the box although .NET Framework 4 is installed apparently properly. Personally, I am glad that I use Windows XP Professional SP3 32 bit (including all POSReady updates).

360 Total Security Essential This article is about 360 Total Security Essential and not about 360 Total Security. These versions are different. 360 Total Security Essential is an antivirus for the basic computer protection against malware and online threats. The software uses multiple antivirus engines and intelligent cloud technology to detect infected files and unknown or hidden threats: 360 Cloud Scan Engine, System Repair Engine, QVM II AI Engine, QEX script killing engine and Avira AntiVir Engine. Furthermore, 360 Total Security Essential supports various types of system scans for suspicious files, which can be further analysed to avoid security problems. The software provides protection on the internet by blocking dangerous websites and making secure online purchases. 360 Total Security Essential has a sandbox that allows you to open files or run applications in an isolated environment without the risk of damaging the main system. Its antivirus also uses behavioural analysis to prevent the system infection and protect personal data against ransomware. 360 Total Security Essential is compatible with Windows XP and free of charge. Main features: Additional antivirus engines Privacy protection Internet security Protection of the file system and registry Sandbox Homepage: https://web.archive.org/web/20200203081602/http://www.360totalsecurity.com/en/features/360-total-security-essential/ Version number: 8.8.0.1119 Date of release: 18.12.2019 System requirements: Windows 10/8.1/8/7/Vista/XP/Server 2008 and above version Version history: Review: https://www.accuratereviews.com/best-antivirus-for-pc-software-reviews-list/360-total-security-essential/ Download link: https://web.archive.org/web/20200402234430if_/https://free.360totalsecurity.com/totalsecurity/360TSE_Setup.exe Screenshots: Screenshot of the main interface: Screenshot of the security configuration with settings I made, especially in terms of the Avira Scan Engine: In this screenshot above, you see I changed the protection mode from the default Balanced to Custom, and I enabled the Avira Scan Engine. Alternatively, you can set the protection mode to the highest level Security. This definitely improves the programme's detection rate. More about the used Avira Scan Engine can be read here. In March 2023, I had installed 360 Total Security Essential 8.8.0.1119 on my second system partition and used it for a while. This version has definitely no such bloatware and crappy desktop manager as I had to notice in the 360 Total Security 10.x follow-up versions I tested before. When installing 360 Total Security Essential 8.8.0.1119, there is an option to install additionally the browser Opera but you can untick it. The whole installation process is far more pleasant than it is, for example, in the edition 360 Total Security 10.8.0.1541. 360 Total Security Essential is quite okay as an alternative in Windows XP where the choice of antivirus programmes is very limited. Although I can't say much about the quality of its protection, you have the option of installing the Avira Scan Engine which increases the detection rate of this programme considerably, as AV-Test has shown in the follow-up version 9.2. Here is an image taken from the test: The impact of system resources by 360 Total Security Essential 8.8.0.1119 is very low. One hardly notices the presence of this programme in the background. And that in my very old, weak system! Unfortunately, 360 Total Security Essential in version 8.8.0.1119 was the last release of this edition. It was abandoned at some point in the past, but its virus definitions could still be updated when I tested it in March 2023. Therefore, I think it will probably still work in these days. But as always, the proof is in the pudding. So maybe, someone can test that and report here. Cheers, AstroSkipper

You're welcome! Thanks to @WSC4 for reporting back! So, Panda really meant what was stated as the system requirements: Therefore, we now definitely know that Panda Antivirus Free is not compatible with Windows XP 64 bit editions due to .NET Framework 4 problems, confirmed by @WSC4 and @UCyborg. Accordingly, I noted that in my article. Cheers, AstroSkipper

Microsoft's strategy is to annoy all users of older operating systems in order to get them to switch to their latest operating system. They "learnt" from the past to avoid such long support as it was in the case of Windows XP.

I saw that yesterday. Thanks for the response.

As of late I have not received automated updated definitions. Here's what I posted on here and Reddit: Hey Folks. I'm on Windows 8, using Antimalware Client 4.8.207.0. Over the past two weeks, it has not updated automatically, or when I press the "Update" button in Windows Defender. In the latter case, It says "Searching", and shows a small amount of the percentage bar; then quits. There is no error message. It just goes back to displaying the time and date of the last time Defender was update (as if there was no update available). I can download the latest definitions and install them myself however. That works. So there's indeed a way to remain updated. Has anyone else experienced this? ... and then in a secondary comment: Upon further investigation, I notice the Event Log there's an Schannel Fatal Alert 40. It only happens when I press the update button in Windows Defender. At the same time in the Windows Defender "Operational" Logs, there's an error under Event ID 2001 at the same time.

I will say that just now I've updated Defender, and I received definitions created at 3:34 am this morning. I'll keep my fingers crossed.

But how are the ESU updates legally obtained? Are they on the Microsoft Catalog site?

@VistaLover Great advices already by Humming Owl and RainyShadow. You can try few things yourself, to check what could be the problem by a process of elimination. If you have 2 RAM sticks, take out 1 and try starting your laptop just with 1 RAM stick, if it still gives black screen try only with the other RAM stick. Here is a video how to change RAM sticks - How To Add Ram Toshiba Satellite Laptop Increase Memory: https://www.youtube.com/watch?v=6vbxHF9w3ZE I only advise not to touch the RAM stick on the "golden" contacts, and hold it on the white labels instead. The advice to try with external monitor to check if the screen is the problem is also good.

It could be. But this "flickering on a frozen frame" coupled with supposed disk activity... ... points more towards a problem somewhere else. @VistaLover Do as they said - try with external display first. You may need to use a key combination to switch displays, and on many systems these won't work during specific points of the boot process. If you're not afraid of opening the thing, disconnecting the internal LCD cable (whichever end is easier to access) normally forces the output to the external monitor. If this doesn't help, try to confirm that the system is actually working - press Caps/Num/Scroll Lock to check that their indicators toggle. Or see if the disk light is blinking as usual. If it boots, you can try to use Remote desktop/Anydesk/whatever to connect and investigate further. If it doesn't boot, try cleaning the RAM contacts. Well, sh!t happens. Last night there was some kind of faint, sweet, fruit-like smell in my room for about a whole hour before i found its source... I lifted my pillow and there was a big hole on the underside, as well as on the electric blanket below, and on the bed itself. When the smoke escaped, the smell suddenly got terrible. The small plastic box in the blanket, where the power cord connects to the heating wire, was completely burned, lol... heh. Now, i gotta buy another electric blanket before the temps go under...

Do you have a spare monitor to connect to the laptop through a VGA cable? In the best case it is only a screen issue. Cheers.

they don't need password to extract, check if you're downloading from my site, and check if you're not using too-outdated archiver.

They're all at default. Let me know if any relevant pref is missing from those screens. This is rather new issue to me since I had browser.privatebrowsing.autostart pref enabled for the longest time. I used to have dom.serviceworkers.enabled turned on just in case, turned it off in recent times, but re-enabling this one doesn't help neither. The console indicates the error supposedly comes from index.php.

New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.4.5-Goanna-20220101.7z Changelog: Out-of-tree changes: * update Goanna3 to git 50c54215f...cbefc77b3: - imported changes from mozilla NSS: - Bug 1737470 - Ensure DER encoded signatures are within size limits. r=jschanck,mt,bbeurdouche,rrelyea - Bug 1735028 - check for missing signedData field r=keeler and bump patch version. (c0d865e51) - partly import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1172080 - Part 2: Use ^ instead of ! to delimit originAttributes from the URI in nsIPrincipal.origin, r=bholley (929ef38df) (23896f6de) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 967494 - "Preference Composition/Spelling/Language is ignored, and changing spellcheck language in one composition window affects all open and new compositions" [r=ehsan] Bug 1157421 - Fix typo "suggesteed". r=ehsan (d48b61df9) - namespaces (c9e3edbf1) - Bug 1167409 - 3/4 - Change ScriptLoadRequeest::mLoading to mProgress. r=jandem (34377ba15) - Bug 1104732 - having deferred scripts shouldn't cause async scripts to delay domcontentloaded, r?smaug (f6b3a5ad9) - Bug 1138395 - Optimize nsDocument::mExpandoAndGeneration.expando out from the cc graphs when possible, r=mccr8 (d944130ab) - Bug 874838 - Make CreateElem return Element. r=khuey (ac65a35cf) - Bug 1194619 - fix comment r=dholbert (017a488a2) - Bug 1137494 - Change the type given to type validation check. r=jgilbert (05885cc7c) - Bug 1106138 - Remove the early unpremultiply in WebGLContext::SurfaceFromElementResultToImageSurface, and let the texel conversion code handle it instead. r=jgilbert (b8010b16b) - Bug 1185815 - Hoist generation increment. r=jgilbert (f6a276b5e) - Bug 1175931 - TexImageFromVideoElement uses GL_HALF_FLOAT if it does not support GL_HALF_FLOAT_OES which would be the case on non-ANGLE systems. Using GL_HALF_FLOAT_OES on a non OES system would result in an error when using TexImage2D. r=jgilbert (d692281f1) - Bug 1184534 - Add support for target size retrieval to GLX backend. r=jgilbert (0e5ba1f8e) - bug 1174705 - add GLContext::GetDefaultFramebuffer. r=jgilbert (99c0e70aa) - Bug 1179935, introduce complex viewport projections to Compositor, remove PrepareViewport; r=mstange (1753d65d3) - Bug 1184534 - Fetch viewport size from context in CompositorOGL, discard if changed during composition. r=nical (4f57bc4ed) - Bug 1187440 - Implement GLX shared surfaces on the OpenGL compositor. r=jgilbert,nical (4844e96ce) - Bug 1033375 - Nudge simple linear gradients with hard stops to half-pixel gradient. r=nical (331ddd4fa) - Bug 1185636 - Remove hard stop workaround for Cairo due to regressions. r=jrmuizel (ccefe7abc) - Bug 1177807 - Mark cairo surface dirty in ReleaseBits. r=jrmuizel (ae9d508b9) - Bug 1170390 - Detect 16bpp cairo xlib surface format. r=jrmuizel (25857ae30) - Bug 1019063 - Check for ::CreateDCW failing when printing. r=dvander (7f54ba8d2) - Bug 1170390 - Add gfxASurface::GetSurfaceFormat for retrieving precise surface format where necessary. r=jrmuizel (f70d11b29) - Bug 1155626 - Don't assume that Factory::GetD2D1Device returns a non-null device and add some gfxCriticalLog. r=Bas (0c896a368) - Bug 1182209 - Additional info with some critical errors. r=mchang CLOSED TREE (f4841baec) (e8f234939) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1185636 - Part 2 - Add some utility methods to Point and Size. r=jmuizelaar (67d447379) - Bug 1185636 - Part 4 - Implement CSS border corners by splitting geometry instead of gradients with hard stops. r=mstange (e8e35023e) - Bug 1185636 - add test for border-radius splits. r=jmuizelaar, r=mstange (091f0dae3) - Bug 1185636 - Part 5 - fuzz some reftests to compensate for new border rendering approach. r=mstange (10412c6b0) - Bug 1156536 - "BaseMargin<T>::Side casts enum SideT to T and adds to pointer". r=jmuizelaar (d9e7b2538) - Bug 1172964 - Fix iteration bug in CopyBGRXSurfaceDataToPackedBGRArray. r=jrmuizel (27199b35c) - Bug 1144147 - Warning fix in DrawTargetCairo.cpp. r=jrmuizel (a98227330) - Bug 1171454 - Workaround for Cairo's source operator with alpha handling. r=jrmuizel (827bdd5f0) - Bug 1169370 - Mark tree Cairo surface dirty after borrow. r=jrmuizel (69002f555) - Bug 1151145 - Add assertions to detect unbalanced calls to CGContextSaveGState / CGContextRestoreGState. r=jrmuizel (9f7d5d6e0) - Bug 1160335. Make DrawTargetCG's transform handling better. r=mstange (30429334d) - Bug 1169126 - Part 1: Make DrawTargetCG::CopySurface able to handle all SourceSurface types. r=gw280 (afbbd895f) - Bug 1169126 - Part 2: Add webgl reftest for captureStream(). r=jgilbert (2b980e81e) - Bug 1169126 - Part 3: Add canvas 2d reftest for captureStream(). r=gw280 (a6be999ce) - Bug 1175286. Remove unnecessary CGContextSave/Restore pair from ClearRect. (2f07cd4b0) - Bug 1151145 - Make sure we don't call CGContextSaveGState without calling CGContextRestoreGState in DrawTargetCG::StrokeRect. r=jrmuizel (c301b1834) - Bug 1153609: Push layers for cleartype when trying to draw cleartype glyphs to a transparent layer. r=jrmuizel (8c7d7aead) - Bug 1178816 - Fix a crash when DrawTargetD2D1 fails to initialize. r=milan (5e08c929a) - Bug 1169039 - Check for NULL in every function that uses CreateEffect(). r=mstange (791a77faa) - Bug 1168189 - Force D2D1 CopySurface to use DrawBitmap like D2D. r=bas (b14307c05) - Bug 1161642 - Utilize primitive blends where available for faster blending on D2D 1.1. r=bas (26dbe0886) - Bug 1130195: Don't immediately crash if we can't allocate image. Carry r=bas (a4f6dd644) - Bug 1160070 - Used BitmapBrush instead of ImageBrush when no sampling bounds. r=bas (fc0bb0683) - Bug 1166585 - Check for CreateFilter() call returning NULL. r=mstange (35fadc74f) - Bug 1168427 - Add a PostScale method to Moz2D's Matrix class. r=Bas (e40fe2f6b) - add missing namespace (c696ec9e3) - Bug 1097464 - Part 1 - Remove Z value from 3D transforms for D3D9. r=roc,matt.woodrow (10a7691bd) - Bug 1097464 - Part 2 - Snap translation for 3d transforms. r=roc (0c0078960) - Bug 1151617 - Add some basic reftests for async scrollbar positioning on B2G and Linux w/APZ. r=tn,botond (d17187621) (3ab112ba1) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1151617 - Add async-zooming tests for async scrollbar positioning. r=tn,botond (b05201e85) - Bug 1161206 - Add a scrolling test with native mousewheel event synthesization. r=mstange (1f4ee0121) - Bug 1161634 - Enable the test_wheel_scroll on Linux as well. r=mstange (64a23bacb) - Bug 1164218 - Enable some guinea pig tests to run in chaos mode. r=botond (b05caafa5) - Bug 1186004 - Split the asyncPanZoom reftest sandbox condition into asyncPan and asyncZoom. r=mstange (3ec461116) - Bug 1097464 - Part 3 - Fix and add reftests. r=roc (6d67df3fb) - Bug 1097464 - Part 4: Fix preserve3d wording for layer flags. r=roc (d18108671) - Bug 1172239. Expand height change hint to its components. r=heycam (04454fa9c) - Bug 1172239. Make vertically-resizing scrollframes reflow their percent-height descendants if necessary, and remove nsChangeHint_NeedDirtyReflow for height changes. r=bz (fe20fb8aa) - Bug 1172239. Use nsChangeHint_UpdateComputedBSize to only dirty intrinsic sizes when necessary. r=bz (7d6adafb1) - Bug 1179078 - Recover from parse errors inside image values in border-image properly. r=dholbert (874c593d8) - Bug 1190635. Don't early-return for an mHeight change, since width changes can add extra change hints. r=heycam (2dd5b9271) - Bug 1179288. Make position:fixed induce a stacking context. r=heycam (99d01e332) - Bug 1172239. Add NS_FRAME_DESCENDANT_INTRINSIC_ISIZE_DEPENDS_ON_BSIZE#. r=bz (22fb22820) - Bug 1172239. Ensure nested overflow:auto elements are tested. r=bz (049347625) - namespaces (737dff870) - Bug 1174510 - Fix fallback from cursor formats that we don't support or that fail to decode to other cursor files. r=seth (34fb64864) - Bug 1202029 - Use the containing block for determining perspective for transformed elements. r=dbaron (3bd60a8a9) - Bug 1097464 - Part 5: Fix preserve3d wording for nsIFrame. r=roc (5e5457f02) - cleanup (6306774fc) - Bug 1138895 - Use proper origin for CSP fast path cache. r=geekboy (394c64963) - Bug 1175122 - Add more assertions to the in-tree content policy implementations to ensure that they receive external content policy types; r=baku (37021992d) - Bug 1177651 - Add areas covered by MozMousePixelScroll listeners to the dispatch-to-content region. r=dvander (6f68653fd) - Bug 1205466, make HasApzAwareListeners check faster, r=masayuki (8ccb3cfe2) - Bug 1191539 - DisplayPort should ForceInside frameRect to match compositor DisplayPort. r=kats (fb95b17f4) - attempt to fix misspatch (31dc4dcb4) - Bug 1182551 - Don't set STATE_IS_BROKEN on HTTP pages when mixed content is allowed by default. r=smaug (7e0a5e6fa) (7b4edb5ee) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1181920. Persist XUL attributes off a scriptrunner. r=peterv (4b29b8efa) - Bug 1156100 - Mark XULDocument::CachedChromeStreamListener::mDocument as an nsRefPtr; r=baku (0a5b7bd5b) - Bug 1168823 - Repair XML syntax highlighting. r=jst (83fd915e3) - Bug 1157059 - Avoid calling AddRef on the pointer enclosed in an nsCOMPtr in nsBindingManager::DoProcessAttachedQueue(); r=baku (6e13d9beb) - Bug 1177816 - Dump a message when the service worker for test_request_context receives an unexpected context value; r=baku (b44d39184) - Bug 1168903 part 1: Give nsSMILValue a move constructor & move reassignment operator. r=birtles (ba7ee0bed) - Bug 1168903 part 2: Use Move() instead of nsSMILValue::Swap() to populate outparams from temp variables in SMIL functions. r=birtles (a4858c958) - Bug 1180048 - Switch warning about having a zero length axis to a LAYOUT_WARNING. r=dholbert (19d405908) - Bug 1175289 - Part 1: Add LAYOUT_WARNING. r=mats (fd4d1bdd9) - Bug 1175289 - Part 2: Disable "Subdocument container has no frame" warning by default. r=mats (8ce582bf5) (de52ef406) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1178664 - Part 1 - Make Animation interface EventTarget inheritance. r=smaug (0ea50ab93) - Bug 1178664 - Part 2 - webidl for AnimationPlaybackEvent. r=smaug (3a5d86f40) - Bug 1150808 - Implement Animation.reverse(). r=smaug r=birtles (9ce9fc4db) - Bug 1178664 - Part 3 - Implement Animation.onfinish event. r=bbirtles, r=smaug (9139c4227) - Bug 1178664 - Part 4 -Implement Animation.oncancel event. r=bbirtles, r=smaug (415b068e6) - Bug 1175751 - Apply playback rate to compositor animations. r=bbirtles (0ffd9cb90) - Bug 1181905 - Animation::IsPlaying should check playbackRate != 0 to stop playing on compositor animation. r=bbirtles (e3a2fd2f4) - Bug 1208385 part 0 - Fix up some references to Web Animations spec; r=heycam (fca9626d4) - Bug 1203009 part 1 - Rename sequence number to animation index; r=heycam (cd7dc513f) - Bug 1203009 part 2 - Remove {CSSAnimation,CSSTransition}::OwningElement() getter; r=heycam (4f49c63b1) - Bug 1203009 part 3 - Add mNeedsNewAnimationIndexWhenRun flag to CSSAnimation and CSSTransition; r=heycam (73a45b5a7) - Bug 1195523: Use type-safe LinkedList instead of PRCList to manage AnimationCollection objects. r=birtles (35b233981) - Bug 1194037 part 1 - Make nsAutoAnimationMutationBatch batch multiple elements at once; r=smaug (83f808043) - Bug 1194037 part 2 - Make WillRefresh no longer call FlushAnimations; r=dholbert (449b0fbd5) - Bug 1194037 part 3 - Add Animation::HasEndEventToQueue(); r=dholbert (b1ddc33d3) - Bug 1203009 part 4 - Implement new composite ordering; r=heycam (4c571e608) - Bug 1203009 part 5 - Remove IsUsingCustomCompositeOrder; r=heycam (242d0142c) - Bug 1203009 part 6 - Add tests for new composite order; r=heycam (5f8711177) (cbefc77b3) * 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 NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220101-cbefc77b3-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220101-cbefc77b3-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220101-cbefc77b3-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20220101-cbefc77b3-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - imported changes from mozilla NSS: - Bug 1737470 - Ensure DER encoded signatures are within size limits. r=jschanck,mt,bbeurdouche,rrelyea - Bug 1735028 - check for missing signedData field r=keeler and bump patch version. (c0d865e51) - partly import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1172080 - Part 2: Use ^ instead of ! to delimit originAttributes from the URI in nsIPrincipal.origin, r=bholley (929ef38df) (23896f6de) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 967494 - "Preference Composition/Spelling/Language is ignored, and changing spellcheck language in one composition window affects all open and new compositions" [r=ehsan] Bug 1157421 - Fix typo "suggesteed". r=ehsan (d48b61df9) - namespaces (c9e3edbf1) - Bug 1167409 - 3/4 - Change ScriptLoadRequeest::mLoading to mProgress. r=jandem (34377ba15) - Bug 1104732 - having deferred scripts shouldn't cause async scripts to delay domcontentloaded, r?smaug (f6b3a5ad9) - Bug 1138395 - Optimize nsDocument::mExpandoAndGeneration.expando out from the cc graphs when possible, r=mccr8 (d944130ab) - Bug 874838 - Make CreateElem return Element. r=khuey (ac65a35cf) - Bug 1194619 - fix comment r=dholbert (017a488a2) - Bug 1137494 - Change the type given to type validation check. r=jgilbert (05885cc7c) - Bug 1106138 - Remove the early unpremultiply in WebGLContext::SurfaceFromElementResultToImageSurface, and let the texel conversion code handle it instead. r=jgilbert (b8010b16b) - Bug 1185815 - Hoist generation increment. r=jgilbert (f6a276b5e) - Bug 1175931 - TexImageFromVideoElement uses GL_HALF_FLOAT if it does not support GL_HALF_FLOAT_OES which would be the case on non-ANGLE systems. Using GL_HALF_FLOAT_OES on a non OES system would result in an error when using TexImage2D. r=jgilbert (d692281f1) - Bug 1184534 - Add support for target size retrieval to GLX backend. r=jgilbert (0e5ba1f8e) - bug 1174705 - add GLContext::GetDefaultFramebuffer. r=jgilbert (99c0e70aa) - Bug 1179935, introduce complex viewport projections to Compositor, remove PrepareViewport; r=mstange (1753d65d3) - Bug 1184534 - Fetch viewport size from context in CompositorOGL, discard if changed during composition. r=nical (4f57bc4ed) - Bug 1187440 - Implement GLX shared surfaces on the OpenGL compositor. r=jgilbert,nical (4844e96ce) - Bug 1033375 - Nudge simple linear gradients with hard stops to half-pixel gradient. r=nical (331ddd4fa) - Bug 1185636 - Remove hard stop workaround for Cairo due to regressions. r=jrmuizel (ccefe7abc) - Bug 1177807 - Mark cairo surface dirty in ReleaseBits. r=jrmuizel (ae9d508b9) - Bug 1170390 - Detect 16bpp cairo xlib surface format. r=jrmuizel (25857ae30) - Bug 1019063 - Check for ::CreateDCW failing when printing. r=dvander (7f54ba8d2) - Bug 1170390 - Add gfxASurface::GetSurfaceFormat for retrieving precise surface format where necessary. r=jrmuizel (f70d11b29) - Bug 1155626 - Don't assume that Factory::GetD2D1Device returns a non-null device and add some gfxCriticalLog. r=Bas (0c896a368) - Bug 1182209 - Additional info with some critical errors. r=mchang CLOSED TREE (f4841baec) (e8f234939) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1185636 - Part 2 - Add some utility methods to Point and Size. r=jmuizelaar (67d447379) - Bug 1185636 - Part 4 - Implement CSS border corners by splitting geometry instead of gradients with hard stops. r=mstange (e8e35023e) - Bug 1185636 - add test for border-radius splits. r=jmuizelaar, r=mstange (091f0dae3) - Bug 1185636 - Part 5 - fuzz some reftests to compensate for new border rendering approach. r=mstange (10412c6b0) - Bug 1156536 - "BaseMargin<T>::Side casts enum SideT to T and adds to pointer". r=jmuizelaar (d9e7b2538) - Bug 1172964 - Fix iteration bug in CopyBGRXSurfaceDataToPackedBGRArray. r=jrmuizel (27199b35c) - Bug 1144147 - Warning fix in DrawTargetCairo.cpp. r=jrmuizel (a98227330) - Bug 1171454 - Workaround for Cairo's source operator with alpha handling. r=jrmuizel (827bdd5f0) - Bug 1169370 - Mark tree Cairo surface dirty after borrow. r=jrmuizel (69002f555) - Bug 1151145 - Add assertions to detect unbalanced calls to CGContextSaveGState / CGContextRestoreGState. r=jrmuizel (9f7d5d6e0) - Bug 1160335. Make DrawTargetCG's transform handling better. r=mstange (30429334d) - Bug 1169126 - Part 1: Make DrawTargetCG::CopySurface able to handle all SourceSurface types. r=gw280 (afbbd895f) - Bug 1169126 - Part 2: Add webgl reftest for captureStream(). r=jgilbert (2b980e81e) - Bug 1169126 - Part 3: Add canvas 2d reftest for captureStream(). r=gw280 (a6be999ce) - Bug 1175286. Remove unnecessary CGContextSave/Restore pair from ClearRect. (2f07cd4b0) - Bug 1151145 - Make sure we don't call CGContextSaveGState without calling CGContextRestoreGState in DrawTargetCG::StrokeRect. r=jrmuizel (c301b1834) - Bug 1153609: Push layers for cleartype when trying to draw cleartype glyphs to a transparent layer. r=jrmuizel (8c7d7aead) - Bug 1178816 - Fix a crash when DrawTargetD2D1 fails to initialize. r=milan (5e08c929a) - Bug 1169039 - Check for NULL in every function that uses CreateEffect(). r=mstange (791a77faa) - Bug 1168189 - Force D2D1 CopySurface to use DrawBitmap like D2D. r=bas (b14307c05) - Bug 1161642 - Utilize primitive blends where available for faster blending on D2D 1.1. r=bas (26dbe0886) - Bug 1130195: Don't immediately crash if we can't allocate image. Carry r=bas (a4f6dd644) - Bug 1160070 - Used BitmapBrush instead of ImageBrush when no sampling bounds. r=bas (fc0bb0683) - Bug 1166585 - Check for CreateFilter() call returning NULL. r=mstange (35fadc74f) - Bug 1168427 - Add a PostScale method to Moz2D's Matrix class. r=Bas (e40fe2f6b) - add missing namespace (c696ec9e3) - Bug 1097464 - Part 1 - Remove Z value from 3D transforms for D3D9. r=roc,matt.woodrow (10a7691bd) - Bug 1097464 - Part 2 - Snap translation for 3d transforms. r=roc (0c0078960) - Bug 1151617 - Add some basic reftests for async scrollbar positioning on B2G and Linux w/APZ. r=tn,botond (d17187621) (3ab112ba1) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1151617 - Add async-zooming tests for async scrollbar positioning. r=tn,botond (b05201e85) - Bug 1161206 - Add a scrolling test with native mousewheel event synthesization. r=mstange (1f4ee0121) - Bug 1161634 - Enable the test_wheel_scroll on Linux as well. r=mstange (64a23bacb) - Bug 1164218 - Enable some guinea pig tests to run in chaos mode. r=botond (b05caafa5) - Bug 1186004 - Split the asyncPanZoom reftest sandbox condition into asyncPan and asyncZoom. r=mstange (3ec461116) - Bug 1097464 - Part 3 - Fix and add reftests. r=roc (6d67df3fb) - Bug 1097464 - Part 4: Fix preserve3d wording for layer flags. r=roc (d18108671) - Bug 1172239. Expand height change hint to its components. r=heycam (04454fa9c) - Bug 1172239. Make vertically-resizing scrollframes reflow their percent-height descendants if necessary, and remove nsChangeHint_NeedDirtyReflow for height changes. r=bz (fe20fb8aa) - Bug 1172239. Use nsChangeHint_UpdateComputedBSize to only dirty intrinsic sizes when necessary. r=bz (7d6adafb1) - Bug 1179078 - Recover from parse errors inside image values in border-image properly. r=dholbert (874c593d8) - Bug 1190635. Don't early-return for an mHeight change, since width changes can add extra change hints. r=heycam (2dd5b9271) - Bug 1179288. Make position:fixed induce a stacking context. r=heycam (99d01e332) - Bug 1172239. Add NS_FRAME_DESCENDANT_INTRINSIC_ISIZE_DEPENDS_ON_BSIZE#. r=bz (22fb22820) - Bug 1172239. Ensure nested overflow:auto elements are tested. r=bz (049347625) - namespaces (737dff870) - Bug 1174510 - Fix fallback from cursor formats that we don't support or that fail to decode to other cursor files. r=seth (34fb64864) - Bug 1202029 - Use the containing block for determining perspective for transformed elements. r=dbaron (3bd60a8a9) - Bug 1097464 - Part 5: Fix preserve3d wording for nsIFrame. r=roc (5e5457f02) - cleanup (6306774fc) - Bug 1138895 - Use proper origin for CSP fast path cache. r=geekboy (394c64963) - Bug 1175122 - Add more assertions to the in-tree content policy implementations to ensure that they receive external content policy types; r=baku (37021992d) - Bug 1177651 - Add areas covered by MozMousePixelScroll listeners to the dispatch-to-content region. r=dvander (6f68653fd) - Bug 1205466, make HasApzAwareListeners check faster, r=masayuki (8ccb3cfe2) - Bug 1191539 - DisplayPort should ForceInside frameRect to match compositor DisplayPort. r=kats (fb95b17f4) - attempt to fix misspatch (31dc4dcb4) - Bug 1182551 - Don't set STATE_IS_BROKEN on HTTP pages when mixed content is allowed by default. r=smaug (7e0a5e6fa) (7b4edb5ee) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1181920. Persist XUL attributes off a scriptrunner. r=peterv (4b29b8efa) - Bug 1156100 - Mark XULDocument::CachedChromeStreamListener::mDocument as an nsRefPtr; r=baku (0a5b7bd5b) - Bug 1168823 - Repair XML syntax highlighting. r=jst (83fd915e3) - Bug 1157059 - Avoid calling AddRef on the pointer enclosed in an nsCOMPtr in nsBindingManager::DoProcessAttachedQueue(); r=baku (6e13d9beb) - Bug 1177816 - Dump a message when the service worker for test_request_context receives an unexpected context value; r=baku (b44d39184) - Bug 1168903 part 1: Give nsSMILValue a move constructor & move reassignment operator. r=birtles (ba7ee0bed) - Bug 1168903 part 2: Use Move() instead of nsSMILValue::Swap() to populate outparams from temp variables in SMIL functions. r=birtles (a4858c958) - Bug 1180048 - Switch warning about having a zero length axis to a LAYOUT_WARNING. r=dholbert (19d405908) - Bug 1175289 - Part 1: Add LAYOUT_WARNING. r=mats (fd4d1bdd9) - Bug 1175289 - Part 2: Disable "Subdocument container has no frame" warning by default. r=mats (8ce582bf5) (de52ef406) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1178664 - Part 1 - Make Animation interface EventTarget inheritance. r=smaug (0ea50ab93) - Bug 1178664 - Part 2 - webidl for AnimationPlaybackEvent. r=smaug (3a5d86f40) - Bug 1150808 - Implement Animation.reverse(). r=smaug r=birtles (9ce9fc4db) - Bug 1178664 - Part 3 - Implement Animation.onfinish event. r=bbirtles, r=smaug (9139c4227) - Bug 1178664 - Part 4 -Implement Animation.oncancel event. r=bbirtles, r=smaug (415b068e6) - Bug 1175751 - Apply playback rate to compositor animations. r=bbirtles (0ffd9cb90) - Bug 1181905 - Animation::IsPlaying should check playbackRate != 0 to stop playing on compositor animation. r=bbirtles (e3a2fd2f4) - Bug 1208385 part 0 - Fix up some references to Web Animations spec; r=heycam (fca9626d4) - Bug 1203009 part 1 - Rename sequence number to animation index; r=heycam (cd7dc513f) - Bug 1203009 part 2 - Remove {CSSAnimation,CSSTransition}::OwningElement() getter; r=heycam (4f49c63b1) - Bug 1203009 part 3 - Add mNeedsNewAnimationIndexWhenRun flag to CSSAnimation and CSSTransition; r=heycam (73a45b5a7) - Bug 1195523: Use type-safe LinkedList instead of PRCList to manage AnimationCollection objects. r=birtles (35b233981) - Bug 1194037 part 1 - Make nsAutoAnimationMutationBatch batch multiple elements at once; r=smaug (83f808043) - Bug 1194037 part 2 - Make WillRefresh no longer call FlushAnimations; r=dholbert (449b0fbd5) - Bug 1194037 part 3 - Add Animation::HasEndEventToQueue(); r=dholbert (b1ddc33d3) - Bug 1203009 part 4 - Implement new composite ordering; r=heycam (4c571e608) - Bug 1203009 part 5 - Remove IsUsingCustomCompositeOrder; r=heycam (242d0142c) - Bug 1203009 part 6 - Add tests for new composite order; r=heycam (5f8711177) (cbefc77b3)

OT: I made some updates on both libchardet and GreenPad, you may have a look :)

If you mean restore site-specific user-agent overrides to FF 52.9.1, this old post explains the trick:

It's bug 1737470. From Mozilla: Emphasis in original. MailNews and IceDove (and Navigator and IceApe, which include them) are likely to be affected, but not pure Web browsers like Mozilla's Firefox, MCP's Basilisk and Pale Moon, or @roytam1's builds of them. anyway they're merged from https://hg.mozilla.org/releases/mozilla-esr91/rev/b6173cd78c9d522f2e3257c408933db4f9cedc13 and will be there in next build