Leaderboard

Technically, if we are truly honest here, one of MC's comments is dead-on accurate, whether we like to "admit" it or not. The builds/forks here can and do crash/corrupt at any time. Some have taken MONTHS to discover and isolate, these are RARE but they DO exist. We have all seen them, reports of "last known working version" and we have to go back MONTHS. There is no such thing as a "stable branch" of these builds/forks. We are guinea pigs, whether we want to be or not, whether we think we are or not. We are beta testers, whether we want to be or not, whether we think we are or not. However, what MC fails to see, in my humble opinion, is that we all feel "a part of something" here at MSFN. We cite these findings and discuss them as a community and the developer interacts with those of us that are "a part of something" and fixes are rolled out. How is that any different than Official Pale Moon development channels? Don't they also do the same? But, I submit, to a larger scale (we here at MSFN often thinks we are much larger than we actually are - but that can be our Pro or our Con, arguments exist on "both sides"). Long Live MSFN. Long Live "Browsers working on Older NT-Family OSes".

I'm the same way. Technically, I only use the St52 browser of this thread, but I have a long history with NM27 and NM28 and revisit them every once in a while just to witness "progress" with them. The St52 that I use (daily, btw, I don't own a mobile phone and St52 is my texting/emailing/voicemailing browser, but it's not my "web browser" for actual web-browsing) is 2023-07-31. Though I do get the feeling (can't prove it) that we are the MINORITY in this regard. I have always felt that most here update EVERY WEEKEND - but we need them, they're the one's finding any issues and just where would we (and some times even Official Pale Moon!) "be without them"?

Go to the closest car, touch the engine, is it warm? You betcha! Now calculate the approx. amount of cars in the world.

Will you please finally stop with check disk? It's been explained to you by UCyborg, it's dangerous in such cases. And I agree with him. And that error code is very generic.

Those artificially boosted requrements of that era were mostly related to PVP. https://en.wikipedia.org/wiki/Protected_Video_Path It was in the interests of companies to exclude XP and Vista RTM because "in January 2007 the developer Alex Ionescu announced that he had found a method that allows end users to bypass Vista's Protected Media Path. This would allow digital content to be played on equipment that does not implement DRM restriction measures" https://en.wikipedia.org/wiki/Protected_Media_Path From what I remember, but can't find any articles, those "loopholes" were fixed in XP SP3 and SP1 of Vista in 2008. "These restrictions only apply to DRM-restricted content, such as HD DVD or Blu-ray that are encrypted with AACS, and also apply in Windows XP using supported playback applications."

Most probably nobody will bother to test their ports for Windows XP without SP, without SP fully updated, with SP1, with SP1 fully updated, with SP2, with SP2 fully updated, with SP3, with SP3 fully updated and with SP3 + all POSReady updates, fully updated. Tests under SP3 are logically the most likely.

... Er, that's just typical ol' Moonchild there , not surprised or impressed... That's why I specifically cited martok (the author of the code involved) as the person to be contacted, NOT MC... In any case, I don't feel guilty myself of proposing something "inappropriate", nor should @UCyborg apologise for his intervention (and his wording, in all of his posts in the linked PMForum thread, was very civil and to the point ) ... In fact, MC was ready to "wontfix" UXP#2452 (based solely on the unwarranted use of NoScript involved), so I just thought, out of genuine consideration in the broader term, that the case of "https://www.theregister.com", which never surfaced among the MCP discussions, be brought to martok's attention, too... Hang me in the town square, won't you? ... A simple yes or no, "thanks for letting us know", from their side would have sufficed... But, perhaps I'm being too naive... Any impartial by-standard will admit that "us" here act, in an indirect way, as α-testers of upstream's UXP master branch, as being the first to be more widely exposed to that code, despite "our" tree's divergence from "official" UXP; this view is, actually, shared/recognised by @dbsoft in this post ; mind you, MC had himself abolished PM's unstable branch/releases some years ago, because very little of his target audience actually used them and reported existing bugs ... And, FTR, I, personally, don't mind being "a guinea pig" ; no-one forces me to update every single weekend (I usually just update once a month or when important webcompat fixes arrive), and before ever updating, I always take a back-up of my current "dirty" profile, in case things break (extremely rarely); going back to a previous build is "a piece of cake" under that scenario ... ... I have always suggested that (even very recently, read this regarding the "dropbox" UXP bug, still "a thing" with official apps, too), but it was NOT feasible in this case because martok's code wasn't present in the latest official "stable" binary releases - I, and I suppose most here (excluding you ), wouldn't be able to compile that "upstream" sourcecode and test that URL on the resulting binaries... Focusing on just "the issue itself", martok responded here that he's unable to reproduce; so, was, in the end, the "www.theregister.com" induced crash in xul.dll (before the fix) particular to "our own" tree? In any case, martok has now prepared UXP#2459 and MC is probably OK to merge it, so all this discussion will become moot... ... Pardon me for asking, but do you think there will ever come a time when some bridges could be built between the two "communities"? We now have @basilisk-dev gracing us with his presence here (but, strictly speaking, not a member of MCP ), plus at least two MCP members are now members here (with different usernames), with at least another one silently following this thread (by his own admission) ... I'm not sure if it was just pretense, but MC seems to still have a beef because: Does this leave a small ray of light come through? Would MC ever accept one "genuine solution" by you for official UXP if you were ever inclined in the future to offer it (rhetorical question, probably...) ? Before anyone here accuses me of old-age-caused dementia , yes, I was here ALL along during the many dramas of the past, but I'd like to see a glass half-full rather than half-empty ; frankly, all this toxicity that existed/still exists between MCP and "us" burdens my soul and if mutual compromises between the two "parties" have to be made at some point, I'd be the first one to welcome them... Just my own humble opinion, though ... Best regards, always extremely thankful PS: December 2023 is well behind us now - any development regarding the "Glory to Hong Kong" song judicial saga (you're still with us here, your repo and blog are still alive, so I suppose things are calmer now?) ?

For checking ansd configuring your hotkeys, you can use Dorando keyconfig. It is very helpful. I agree with the opinion of @j7n and also believe that that the hotkey has been captured by another extension.

RAM limitations are coming from the chipset, not the BIOS.

I totally agree. Mentioning you or your forks is not really a good idea on the Pale Moon forum. I can literally feel and read the hostility, arrogance and aversion there. There is a lack of friendliness and respect in this forum. That's why I haven't posted there for a long time.

Climate change deniers aren't common nowadays, and looks like someone can have real hard currency benefits from such opinions.

I can be mistaken, but that requirement is related to DXVA1, that old version will still open up and work on SP2, but no acceleration will be available, so on that Celeron max 360p, 480p at best. On such old hardware no acceleration is possible anyways, like in my case with the ancient Intel Extreme Graphics on Toshiba.

I found the system requirements for MPC-HC 1.7.13 on the old, abandoned homepage: https://mpc-hc.org/faq/#what-are-the-requirements-for-mpc-hc Due to the fact that MPC-HC 2.1.2.18 is a port for Windows XP, it is completely unclear whether it has ever been tested under SP2 or is supported there at all.

I don't know what the exact system requirements of MPC-HC 2.1.2.18 are with regard to Windows XP, but many programmes require Service Pack SP3 to be installed if they are compatible with XP at all. Thus, stucking at SP2 doesn't really make much sense and is not really a good idea, IMHO.

Comparing to Celeron, Pentium is of a higher tier, especially if we compare desktop to notebook (Toshiba). Besides, my laptop had 512MB of RAM, it was the highest possible amount for the early 2003. Later I upgraded it to 1GB of DDR ram. That was the max. amount the chipset allowed.

My Windows XP Professional received the SP3 service pack as soon as it was released. My system was already fully updated in 2014 with all updates after SP3. From 2014 to 2019, my system received all POSReady updates. My 32-bit CPU is a Pentium 4 with only 1.5 GB SD-RAM. So very weak hardware. And yet my system runs excellently. By no means worse than when SP2 was still in use , but much more up-to-date.

I'm obviously not an XP expert anymore, if I ever was, but something tells me, those updates are no longer that important anymore. The times and the threats, they were written to deal with, are long gone. When was the last update for XP, a decade ago? I even find the first release of XP64 to be faster than the second one, so I maybe port MPC-HC on it someday.

This Service pack 3 is heavily bloated and slow. It was always the case with my Celeron Toshiba (2003) or any other right era hardware I used. Possibly later, when the first Core Quad came out, it became less apparent. I can relate to those people who prefer to run MPC-HC on SP2.

That's a typical comment from the guys on the Pale Moon forum. Exactly as I put it here: And then the try to row back a little patronisingly: The quoted comment can be found here: https://forum.palemoon.org/memberlist.php?mode=viewprofile&u=17830&sid=c270026a4e9bbd9b3587497dfcb6127a https://forum.palemoon.org/viewtopic.php?p=247714#p247714 What do we learn from such behaviour? One should only post in the Pale Moon forum about the official Pale Moon and Basilisk browsers and only if absolutely necessary. These poor, lost souls there can't cope with anything else.

maybe checking out his latest reply there?

The job is fine, retail just takes a lot of energy. When I am working in the back it's nice though. The job gave me money to get a much needed platform upgrade. I now have a 7800X3D, 32GB DDR5-CL30, and a B650 board. Runs very nice and its an absolute monster of a system now.

I didn't wait for their side's action and I decided to import them for testing now (with another PR) and it seems that it won't crash anymore even if my workaround is reverted.

I love this comment because if I have learned anything during the time I have been trying to make Firefox 3.0 compile targeting the Windows GFX toolkit, Visual C++ 6.0, and now, Windows 95, it is that alpha software is typically stable enough for use, and sometimes, works better than the official "stable" release. Due to that, I/we shouldn't mind running software just because it is "alpha" or "beta". Your second point is going to repeat like a broken record, because I work with source code as old as I do because that is all my hardware is compatible with. I did try compiling official UXP on Linux Mint 19.3 in ~September 2020, but it took 6 hours with 2 CPU cores enabled, and 2x that time with 1 CPU core enabled. I can compile roytam1's fork of Pale Moon 26.5.0 in ~3-4 hours with 2 CPU cores enabled, and 6-7 hours with 1 CPU core enabled, along with ~1.2 GiB RAM usage. With 1 CPU core enabled, I can compile RetroZilla in less than 1 hour, while using ~100 MiB RAM with Visual C++ 6.0 SP5. The Mozilla 1.9.0 codebase does not seem to have increased build times than the Mozilla 1.8.1 codebase, although 1.9.1 and 1.9.2 do come with ~15% increased build times, along with ~150-300 MiB RAM usage on Visual C++ 2003, which uses 40% more RAM than Visual C++ 6.0.

...The "magnifying glass" icons in the column headers open popups that can be used to locate specific filterlists (via their name/description); the "funnel" button on the "software" column can be used to filter based on adblocker used; filtering for "Easylist" and (software) "uBlock Origin" gives me the "Easylist (Optimized)" FL (that I mentioned previously) as the third entry in the second results page:

I'd only use such limited polyfills on per-site basis to avoid issues. A more capable implementation of structuredClone polyfill: 'use strict'; const VOID = -1; const PRIMITIVE = 0; const ARRAY = 1; const OBJECT = 2; const DATE = 3; const REGEXP = 4; const MAP = 5; const SET = 6; const ERROR = 7; const BIGINT = 8; // const SYMBOL = 9; const EMPTY = ''; const {toString} = {}; const {keys} = Object; const typeOf = value => { const type = typeof value; if (type !== 'object' || !value) return [PRIMITIVE, type]; const asString = toString.call(value).slice(8, -1); switch (asString) { case 'Array': return [ARRAY, EMPTY]; case 'Object': return [OBJECT, EMPTY]; case 'Date': return [DATE, EMPTY]; case 'RegExp': return [REGEXP, EMPTY]; case 'Map': return [MAP, EMPTY]; case 'Set': return [SET, EMPTY]; } if (asString.includes('Array')) return [ARRAY, asString]; if (asString.includes('Error')) return [ERROR, asString]; return [OBJECT, asString]; }; const shouldSkip = ([TYPE, type]) => ( TYPE === PRIMITIVE && (type === 'function' || type === 'symbol') ); const serializer = (strict, json, $, _) => { const as = (out, value) => { const index = _.push(out) - 1; $.set(value, index); return index; }; const pair = value => { if ($.has(value)) return $.get(value); let [TYPE, type] = typeOf(value); switch (TYPE) { case PRIMITIVE: { let entry = value; switch (type) { case 'bigint': TYPE = BIGINT; entry = value.toString(); break; case 'function': case 'symbol': if (strict) throw new TypeError('unable to serialize ' + type); entry = null; break; case 'undefined': return as([VOID], value); } return as([TYPE, entry], value); } case ARRAY: { if (type) return as([type, [...value]], value); const arr = []; const index = as([TYPE, arr], value); for (const entry of value) arr.push(pair(entry)); return index; } case OBJECT: { if (type) { switch (type) { case 'BigInt': return as([type, value.toString()], value); case 'Boolean': case 'Number': case 'String': return as([type, value.valueOf()], value); } } if (json && ('toJSON' in value)) return pair(value.toJSON()); const entries = []; const index = as([TYPE, entries], value); for (const key of keys(value)) { if (strict || !shouldSkip(typeOf(value[key]))) entries.push([pair(key), pair(value[key])]); } return index; } case DATE: return as([TYPE, value.toISOString()], value); case REGEXP: { const {source, flags} = value; return as([TYPE, {source, flags}], value); } case MAP: { const entries = []; const index = as([TYPE, entries], value); for (const [key, entry] of value) { if (strict || !(shouldSkip(typeOf(key)) || shouldSkip(typeOf(entry)))) entries.push([pair(key), pair(entry)]); } return index; } case SET: { const entries = []; const index = as([TYPE, entries], value); for (const entry of value) { if (strict || !shouldSkip(typeOf(entry))) entries.push(pair(entry)); } return index; } } const {message} = value; return as([TYPE, {name: type, message}], value); }; return pair; }; /** * @typedef {Array<string,any>} Record a type representation */ /** * Returns an array of serialized Records. * @param {any} value a serializable value. * @param {{json?: boolean, lossy?: boolean}?} options an object with a `lossy` or `json` property that, * if `true`, will not throw errors on incompatible types, and behave more * like JSON stringify would behave. Symbol and Function will be discarded. * @returns {Record[]} */ const serialize = (value, {json, lossy} = {}) => { const _ = []; return serializer(!(json || lossy), !!json, new Map, _)(value), _; }; const env = typeof self === 'object' ? self : globalThis; const deserializer = ($, _) => { const as = (out, index) => { $.set(index, out); return out; }; const unpair = index => { if ($.has(index)) return $.get(index); const [type, value] = _[index]; switch (type) { case PRIMITIVE: case VOID: return as(value, index); case ARRAY: { const arr = as([], index); for (const index of value) arr.push(unpair(index)); return arr; } case OBJECT: { const object = as({}, index); for (const [key, index] of value) object[unpair(key)] = unpair(index); return object; } case DATE: return as(new Date(value), index); case REGEXP: { const {source, flags} = value; return as(new RegExp(source, flags), index); } case MAP: { const map = as(new Map, index); for (const [key, index] of value) map.set(unpair(key), unpair(index)); return map; } case SET: { const set = as(new Set, index); for (const index of value) set.add(unpair(index)); return set; } case ERROR: { const {name, message} = value; return as(new env[name](message), index); } case BIGINT: return as(BigInt(value), index); case 'BigInt': return as(Object(BigInt(value)), index); } return as(new env[type](value), index); }; return unpair; }; /** * @typedef {Array<string,any>} Record a type representation */ /** * Returns a deserialized value from a serialized array of Records. * @param {Record[]} serialized a previously serialized value. * @returns {any} */ const deserialize = serialized => deserializer(new Map, serialized)(0); if (!("structuredClone" in env)) env.structuredClone = (any, options) => deserialize(serialize(any, options)); Source: https://github.com/ungap/structured-clone

F11 works in Serpent 52. Maybe the hotkey is being captured by another application. Although I have disabled most animations in New Moon 28. Full screen entry still does a gradual hiding of the bar at the top, which is laggy on heavy sites.

https://forum.palemoon.org/viewtopic.php?f=62&t=30769&p=247631 Many thanks! but it seems that no positive feedback there. as I said before, better test with theirs before creating something like this in their side. mentioning my things there is not a good idea as always.

Any service pack past SP3 can't be official.

Double posting! The editor is out of control.

I was born in the seventies (exactly the time you say it's started to get warmer), I remember those already heated up seventies very good. But even comparing to the already hot late seventies, what we have have here today is bloody hell, mind you - it exceeds the mentioned 30 years average weather period by almost 80%. So yeah, it's good enough to make clear observations, and not trust those 'official' sources.

Yes indeed. We usually do it when new builds are issued, so if you could create a new thread for your next builds @roytam1 that would be good, and I will then lock this thread. Thanks, Dave.

I liked the covid times, it was much quieter in my town, I live in a town with lots of tourists, obviously covid was a man-made crap, but it works for de-pop, and that is good for us and the climate. Less people, more oxygen.

Yes, and it happened again to @DanR20! What I'm saying, though, is that if any extension crashes the browser, that has to be the browser's fault! Unlike Flash and other plug-ins, extensions don't have their own machine code. (Well, a tiny few do, but those "helpers" have to be downloaded and installed separately from the extension itself, so at least you're aware of them.) Extensions are just CSS, JavaScript, HTML, etc., and it's the browser's job to read and interpret those things, and handle any errors therein. Of course, as a practical matter, if an extension is crashing your browser, the best course of action may well be to remove and replace the extension, rather than waiting for the browser developer to fix the crash. After all, even if the crash is fixed, the extension still may not work properly. We're lucky (and a bit spoiled) to have @roytam1 able to deal with issues like this so quickly.

Agree with this part. This nightmare is also here. The lobby is very strong worldwide. "...who have doubts about their gender identity and may be considering sex change surgery". https://www.alliancevita.org/en/2021/12/gender-french-senate-adopts-controversial-bill-on-sex-change/ Huge, huge amounts of money in this industry.

Results : 391.35 works fine, but I didn't like the way it represents the colours and its contrast is way too high. Otherwise - fine ! 399.24 is one hell of a glitchy piece of crap. Interferes with my good ol' sound card. It booted up fine (without the NVCPL, of course) . Played a DX11 game for 1 hour fine. Then started to watch Octopussy 1983 using MPC-HC and look at the crash errors. (TDR had happened) Happens at random intervals. Also 399 doesn't like Chrome.

This orb problem had happened once for me also, then I quickly understood it was due to the fact the NVCPL couldn't start because of the missing 345 (0x0159) "N/A" in COMCTL32.DLL. If I install only the driver (without the CPL), all is fine.

I see, thanks , I think I wrote I already tried the iCafe 390.xx (77 ?) and it's too old already. Regarding games, nothing major between 390-391. Could you elaborate on hit and miss ? Apart from the non-working panel, no troubles for me . But the panel doesn't start and we know why. Maybe wait for the new kernel then ? Or you want me to mod the hell out of it so we could use it without the kernel ? I'm afraid the new tutorial would be too complicated for some folks then.