Jump to content

basilisk-dev

Member
 View Profile  See their activity

  • Posts

    53

  • Joined

  • Last visited

  • Days Won

    6

  • Donations

    0.00 USD 

  • Country

    United States

 Content Type 

Everything posted by basilisk-dev

  1. basilisk-dev
    I am wondering if in this PR: https://repo.palemoon.org/MoonchildProductions/UXP/commit/e10d75b7c266c30769ecac71fd8b890c6db7c47b Maybe I should have used return cpuCount instead of return threadCount. It is hard to diagnose this on my end because I am not able to replicate it on Linux or Windows 10 in upstream Basilisk either.
  2. basilisk-dev
    Yes. https://repo.palemoon.org/MoonchildProductions/UXP/pulls/3040
  3. basilisk-dev
    Yeah, they were intended to. However some of them actually hurt performance. I believe that he is just vibe coding with an AI tool for most of his changes and is not actually a developer. I can't verify that but that is my theory.
  4. basilisk-dev
    The Python 3 changes support Python 3.6 and above. I'm not sure what version of mozilla-build you use, upstream we are using mozilla-build 3.4 which ships with Python 2.7 and Python 3.6. I don't know if it will install on Windows XP, but I know for sure it will install on Windows 7. Most of his "performance improvements" code outside of restoring e10s actually make the browser slower according to benchmarks. He did import some newer WebExtensions support so you might want to merge that.
  5. basilisk-dev
    As long as the application runs in the Windows 2000 extended kernel you do not need to directly support Windows 2000. For example, Roytam's browsers work on Windows 2000 with the extended kernel even though he does not target Windows 2000.
  6. basilisk-dev
    Software can be open source software while not meeting the definition of free software. For example Unreal Engine and the unrar CLI tool on Linux match this description. I want to make it very clear I did not support the Pale Moon team's actions at that time and I had absolutely nothing to do with that.
  7. basilisk-dev
    Basilisk supports all of these. We started supporting PowerPC Mac OS X in the last release. We have released GTK2 binaries since I took over the project.
  8. basilisk-dev
    I don't think that anyone involved with Pale Moon or upstream UXP development hates Feodor2. That was mostly Tobin, who is no longer involved with Pale Moon or UXP. I have no opinion on Feodor2. If Feodor2 decided to continue development of Centaury I would not care.
  9. basilisk-dev
    Upstream issue submitted: https://repo.palemoon.org/MoonchildProductions/UXP/issues/3001 It was caused by the second link you posted.
  10. basilisk-dev
    I have seen PowerFox and talked to Jazzzny about it a little. Hadn't seen Machfox until you mentioned it, that is pretty neat. I do worry that I am taking away reasons to use PowerFox by releasing 10.5 PPC binaries of my own, but we have the PowerPC support in the tree so I might as well use it. I had a hard time sleeping last night and thought about it. We don't necessarily have to modify the CSS parser. We just need to de-nest the CSS before the parcer receives it. I have a proof of concept that fixes a bunch of sites using nested CSS, although it isn't 100% compatible with all nested CSS sites yet.
  11. basilisk-dev
    Yes this will get implemented in UXP at some point but I don't know how to implement it yet. It's not like I can snap my fingers and UXP magically has support for it. It takes time to research, implement, test, and verify there are no regressions or crashes as a result of that implementation. That implementation affects the CSS parser so it will literally affect how UXP reads the CSS for every single site on the web. Even if we had an implementation it would take A LOT of testing before we'd feel comfortable merging it in.
  12. basilisk-dev
    I think you misunderstood something here. Web compatibility is my number one priority right now. I already said that earlier in the thread. Basilisk does not, and will never, require AVX for release builds.
  13. basilisk-dev
    That would be a very large amount of work. I understand why users want it but at the moment it isn't anywhere near the list of top priorities. Personally, I view the lack of DRM in the browser as a feature and not a bug, but I could see why other users might not feel this way. Is it something I want to implement? Yes. Am I going to do that? Not any time soon.. If I did do it, the implementation would be Manifest V2 only because Manifest V2 is a stable target and a lot of Firefox WebExtensions still use Manifest V2. For example, the Firefox version of UBlock Origin uses Manifest V2.
  14. basilisk-dev
    I haven't taken a look at how to implement this in our codebase yet, but Mozilla's implementation is written in Rust unfortunately so we will have to do a new implementation from scratch. Happy to work together with you to get an implementation figured out, you need this for your XP builds just as much as we do upstream. I have no idea if Moonchild Productions is just Moonchild or if there are other UXP/Pale Moon contributors involved in Moonchild Productions as well. I've never asked. In my case, Basilisk Development Team is just me because nobody else seems to be interested in contributing to Basilisk outside of the occasional PRs from other UXP developers to ensure breakages don't occur after UXP updates.
  15. basilisk-dev
    Eventually I will look into it if no one else on the UXP dev team steps up to try to look into it. I did read last week on the Pale Moon forum that someone was writing a XUL extension that un-nests nested CSS, but I can't find the post right now. Right now my priorities are: * Full ES2023 compliance (eventually 2024/2025 but now ES2023 is priority) * Make Tailwind v4 playground render correctly (this will fix a lot of sites not just sites using Tailwind) * Performance improvements * Make WebRTC actually work with modern services * Fix any other sites that I find that are broken If I am being completely honest with you, I have been a bit frustrated because I have felt that the other UXP developers have prioritized things other than site compatibility, and that those other things they are prioritizing are the wrong things to prioritize. I frequently see forum posts where users are told that the CSS or JS feature a site uses is bad and that the user should email the site asking them not to use those features because they are bad. Users don't and shouldn't have to do that. People don't care about what CSS or JS a site uses, all they care about is that the site they want to use is broken. They'll just go use another browser instead. Due to that I've been learning C++, learning how our JS and CSS engines work, and I'm doing what I can to change that in order to keep UXP/Basilisk/Pale Moon/roytam's forks relevant.
  16. basilisk-dev
    Set security.tls.hello_downgrade_check in about:config to false. This affects upstream as well.
  17. basilisk-dev
    Which operating system are you asking about? Windows or Linux? I have not looked into this for Windows, as I am not a modern Windows user(I do use roytam1's Serpent builds on a Ryzen 3600X running Windows 2000 though). I was already looking into doing this for Linux last week. I had to patch some code but I was able to get a successful build that works on a Pentium MMX and newer. The oldest hardware I have is a Pentium 3, it ran fine on that machine. http://dl.basilisk-browser.org/basilisk-20260305163630.linux-i586-gtk3.tar.xz I want to start making our x86 releases support 10.5+, right now we are stuck at 10.7. The contributor who does the Mac builds has an issue with his build environment where he was not able to do a build targeting OS X 10.5 or 10.6. It was producing SSE related errors which is strange because even the oldest CPU that can run OS X 10.5 has SSE support. We did introduce PowerPC builds in the last release if that's something that interests you. You could run them in Rosetta, you might get Google.com to load after 1-2 years of waiting for Rosetta to run the browser.
  18. basilisk-dev
    All three of these services work fine in Basilisk. They might not work in Pale Moon because I ship an internal polyfill loader that Pale Moon does not have. I am not able to replicate your performance issues in Gemini.
  19. basilisk-dev
    Implemented "that support". https://repo.palemoon.org/MoonchildProductions/UXP/pulls/2991 Maybe you should hold your breath next time.
  20. basilisk-dev
    This is incorrect. Pale Moon was always intended to be a browser that was utilized newer CPU optimizations to increase performance. Your claim that it was done with the explicit intention of running on older CPUs is false, although that is an unfortunate side effect of enabling the AVX instruction set. There is absolutely a noticeable performance improvement when comparing the AVX builds to the pre-AVX builds. I disagree with the requirement to have AVX instruction support to run Pale Moon and I think it was a bad move on Moonchild's part, but basically every claim here that you've made is either false or misguided.
  21. basilisk-dev
    The site works fine for me on both ARM MacOS and aarch64 Linux.
  22. basilisk-dev
    I am working on a fix right now. You are welcome to contact me here or on the Basilisk GitHub mirror if you are unable to contact me on the Pale Moon Forum or Pale Moon repo. There is also an official subreddit at https://old.reddit.com/r/basiliskbrowser
  23. basilisk-dev
    I am not doing this, I considered doing it to prevent buggy code being pushed to roytam1's users, but I never decided to actually go forward with it. What's sad is that roytam1 continues to release code to his users that has not been verified as production ready. Code in the master branch of upstream UXP or upstream Basilisk is not always production ready. I have seen issues reported on MSFN for roytam1's builds because he was pulling code from master branches before they were ready. He is doing a disservice to his users by giving them potentially buggy code.
  24. basilisk-dev
    Basilisk releases use the date for the version, for example "2023.11.05" is the most recent release of Basilisk. Basilisk release builds are compiled with "export BASILISK_VERSION=1" in the .mozconfig. Internally Basilisk releases present themselves to addons as "52.9.date" rather than just "date". This ensures that Basilisk works with extensions that were developed for Firefox 52. The 52.9.date internal version as well as the date version number comes from the confvars.sh shell script: https://repo.palemoon.org/Basilisk-Dev/Basilisk/src/branch/master/basilisk/confvars.sh You can also set BASILISK_VERSION to a date if you want to specify a date. For example, if you do "export BASILISK_VERSION=2023.12.01" then the Basilisk version number will be 2023.12.01. This functionality is in the above shell script as well. If "export BASILISK_VERSION=1" is not present in the .mozconfig then the version defaults to 52.9.0. This is the reason that Serpent uses 52.9.0 as the release number. 52.9.0 comes from the txt files in the basilisk/config directory:https://repo.palemoon.org/Basilisk-Dev/Basilisk/src/branch/master/basilisk/config
  25. basilisk-dev
    This happens for me even in Basilisk. I was never able to figure out why, sometimes archive.org works for me and sometimes it doesn't.
×
×
  • Create New...