Leaderboard

Wow! Is this thread here about hardware? And performance comparisons between Win XP and Win 10? The answer is crystal clear: Definitely not! This thread should actually provide information about the Thorium browser. Thus, the last 7 posts are therefore completely off-topic. The reader will not learn anything worth knowing about Thorium here. No matter how you feel about this browser.

... Over the course of several previous years, "we" (UXP-users) have witnessed how the platform, in agony, struggled to cope with whatever new Javascript "feature" Google devs came up with, the most notorious of them being the non-polyfillable "operators" ("?.", "??"; read this excellent write-up by @InterLinked); at the time, I was semi-convinced that JS operators such as those would deal the final blow to UXP; thankfully, some talented UXP coders came up with own implementations of these operators, thus UXP was "saved" - but for how long? I'll have to side with NHTPG and, to a good extent, AstroSkipper that the last nail on UXP's coffin will be a "new" CSS-related "improvement" (sprinkle with a generous dose of sarcasm here ), implemented in a mixture of CSS code inside/alongside JS code (which seems to be the trend recently ); in the case of UXP, MCP's upstream (Mozilla) have CSS features implemented in Servo/Rust, so MCP can't backport easily, if at all, Mozilla code for these and get away with it; CSS-related UXP issues will keep piling up inside their tracker, with very dim prospect ... Case in point is a CSS bug that has been affecting GitHub in particular for many months now, it was acknowledged exactly one year ago, but so far no activity at all towards a resolution ... This particular "feature" was implemented only in Chromium 99, so it does "invalidate" the popular 360EEv13 and/or KafanMiniBrowser variants on XP/Vista x86; only implemented in Firefox 97, thus putting Mypal68 out of the game, too; if it weren't for the new "additions" (Supermium and its semi-fork Thorium) that appeared at a most opportune time , "we" retro-OS-ers (I just devised this term, ) would have run out of options...

Knocking on UDP, trying to make connections via QUIC, broadcasting to literally everything and everyone is the default chrome behaviour. The question is, why (if both @win32 and Alexi are trying to implement at least something similar to Ungoogled) didn't cut it out first. Seems like a very logical (and rather simple) first step.

@dimo70 You can't disable defender while the system is still running, it does protect it self.

DarkMagic 3.7.9 fixes

No sure if you still can disable defender online. This should work from WinPE/WinRE. Another Windows would need to run it with TrustedInstaller rights. Defender.cmd reg load HKLM\SOFT C:\Windows\System32\config\software reg load HKLM\SYS C:\Windows\System32\config\system reg import Defender.reg reg unload HKLM\SOFT reg unload HKLM\SYS Defender.reg Windows Registry Editor Version 5.00 [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows\CurrentVersion\Run] "SecurityHealth"=- [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows\CurrentVersion\Explorer] "SmartScreenEnabled"="Off" [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows Defender] "DisableAntiSpyware"=dword:00000001 "DisableAntiVirus"=dword:00000001 [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows Defender\Features] "TamperProtection"=dword:00000000 [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows Defender\Real-Time Protection] "DisableRealtimeMonitoring"=dword:00000001 "DisableAntiSpywareRealtimeProtection"=dword:00000001 "DpaDisabled"=dword:00000001 [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows NT\CurrentVersion\Image File Execution Options\MpCmdRun.exe] "Debugger"="NUL" [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows NT\CurrentVersion\Image File Execution Options\MsMpEng.exe] "Debugger"="NUL" [HKEY_LOCAL_MACHINE\SOFT\Microsoft\Windows NT\CurrentVersion\Image File Execution Options\smartscreen.exe] "Debugger"="NUL" [HKEY_LOCAL_MACHINE\SOFT\Policies\Microsoft\Windows\System] "EnableSmartScreen"=dword:00000000 [HKEY_LOCAL_MACHINE\SOFT\Policies\Microsoft\Windows Defender] "DisableAntiSpyware"=dword:00000001 "DisableAntiVirus"=dword:00000001 [HKEY_LOCAL_MACHINE\SOFT\Policies\Microsoft\Windows Defender\SmartScreen] "ConfigureAppInstallControlEnabled"=dword:00000001 "ConfigureAppInstallControl"=dword:00000001 [HKEY_LOCAL_MACHINE\SYS\ControlSet001\Control\CI\Policy] "VerifiedAndReputablePolicyState"=dword:00000000 [HKEY_LOCAL_MACHINE\SYS\ControlSet001\Services\SecurityHealthService] "Start"=dword:00000000 [HKEY_LOCAL_MACHINE\SYS\ControlSet001\Services\WinDefend] "Start"=dword:00000004 [HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\MRT] "DontOfferThroughWUAU"=dword:00000001 P.S: Not sure why I set SecurityHealthService to autostart, either there was a problem with disabling or it's a typo.

When uBlock Origin is installed, it controls the preloading of websites. You do not have to disable uBlock Origin to control this feature again, you can simply enable the option "Disable pre-fetching" in uBO's dashboard page "Settings" as already mentioned by @Dixel.

Forgot say, the laptop is from 2013, based on Haswell Pentium mobile 3550M. https://ark.intel.com/products/77404

So I got some time today, I tried the driver, I had to add my device inf., it installed fine, but gives error 43 code.

It doesn't work that way, here we respect intellectual property, even though we understand most of it belongs to another man (win32), despite all of the above, there needs to be a clear statement with Open Source code published by the author, none of that happened, so far.

Why not release a better optimised product together, then? One million releases with constant fixes only confuses people. And what does he mean by "legacy" Thorium repo? It's also confusing since Windows 7,8,8.1 are all legacy, too.

Please provide with a link where the author clearly states his softwaie is Open Source, and he (she?they?, if they work as a team) doesn't/don't mind it gets redistributed/modified.

They claim this issue is now fixed, could you confirm? And it's again fixed by @win32, not "Thorium" author. https://github.com/Alex313031/thorium-legacy/releases/tag/M122.0.6261.169 "GDI text rendering alignment is fixed This bug caused text to be askew, either too high or too low, or too much to the left. The reason previous releases didn't have this fixed, is that win32ss fixed it in Skia, which is a third_party submodule of Chromium."

lead

dear

Walking stick

Nvidia RTX 4080 Super

ListDisk This tool provides detailed information about connected physical disks, including their partitions and volumes. Download https://github.com/GeorgeK1ng/ListDisk/releases Features Mounting/unmounting partitionsSetting partition labelsModifying MBR partition attributes Available Commands /mount - Mounts a partition to a specified drive letter./unmount - Unmounts a specified partition or drive letter./setlabel - Sets a new label for a specified partition./setactive - Sets or clears the bootable flag of a partition./settype - Changes the MBR type of a partition. Usage Examples /mount 0 1 U - Mounts Partition 1 on Disk 0 as U: drive./unmount 0 1 - Unmounts Partition 1 on Disk 0./unmount U - Unmounts U: drive./setLabel 0 1 Lbl - Sets 'Lbl' as the label for Partition 1 on Disk 0./setactive 0 1 - Marks Partition 1 on Disk 0 as bootable./setactive 0 1 80 - Sets the boot flag for Partition 1 on Disk 0./setactive 0 1 00 - Clears the boot flag for Partition 1 on Disk 0./settype 0 1 0x17 - Sets Partition 1 on Disk 0 to type NTFS Hidden (0x17)./settype 0 1 0x07 - Sets Partition 1 on Disk 0 to type NTFS (0x07). Requirements Operating System: Windows XP and newer. Note for Windows XP Users: This tool requires the installation of the Microsoft Visual C++ Redistributable. Please install it from this link.Compatible with Windows Preinstallation Environment (WinPE).Place ListDisk in %WinDir%\system32 to make it easy accessible from commandline! Note: Use this utility with caution. Incorrect usage may affect data integrity.

Try this - without quotes! Despite what other users/user may tell you. --disable-features=WebUsbDeviceDetection

Many thanks for taking this "upstream" ; indeed, I now fully recollect that martok was/is responsible for creating that pref and the underlying code that is behind it (UXP #2030) ; actually, I had that pref toggled (to false) in my St52 "dirty" profile, but not in my "minimal" NM28 profile... Toggling that pref was necessary for me (in St52) for GitHub's "sake"; prior to Dec 2023, when a GH repo was being displayed in "Commit List View" (e.g. like that), a button is provided to the right of each commit entry to easily copy (with one click) the commit's HASH: Read more in a now closed Palefill issue ... However, after the mid-December 2023 "massacre" , GitHub have replaced their previous code (in many places, including the "Commit List View" mode) with "legacy-web-engines-excluding", "heavily-CPU+GPU+RAM-taxing" ReactJS scripts and, under the new status quo, toggling the referenced pref is no longer required for GH's re-iteration of the "Copy-SHA-button" ...

None of these frequently discussed (on MSFN) Chromium clones are worth losing a sleep over. Most of their code is copy-paste off Chromium, they don't have their own legs to stand on, without Google LLC, they are nothing. I can relate. I think I should be making better use of my time here. How, I haven't a clue.

No they still can't figure out which team fixes compat.

FWIW Alex tells me that uBlock Origin is bundled with Thorium. 🙂

No. It is indeed Thorium trying to connect to this address. To be more precise it is the process thorium.exe. Disabling the flag #media-route-dial-provider seems to be the solution I found by checking all available Thorium flags.

I ported recently somehow AMD Radeon HD 8750m driver to Windows Vista😄 Photo: https://ibb.co/BzrKYgt

Can anyone tell me why Thorium is trying to connect via UDP to the address 239.255.255.250 using the port 1900 almost every minute? My Windows 10 Firewall Control is blocking and logging this connection, fortunately. I performed a WHOIS check and got this information: # # ARIN WHOIS data and services are subject to the Terms of Use # available at: https://www.arin.net/resources/registry/whois/tou/ # # If you see inaccuracies in the results, please report at # https://www.arin.net/resources/registry/whois/inaccuracy_reporting/ # # Copyright 1997-2024, American Registry for Internet Numbers, Ltd. # NetRange: 224.0.0.0 - 239.255.255.255 CIDR: 224.0.0.0/4 NetName: MCAST-NET NetHandle: NET-224-0-0-0-1 Parent: () NetType: IANA Special Use OriginAS: Organization: Internet Assigned Numbers Authority (IANA) RegDate: 1991-05-22 Updated: 2013-08-30 Comment: Addresses starting with a number between 224 and 239 are used for IP multicast. IP multicast is a technology for efficiently sending the same content to multiple destinations. It is commonly used for distributing financial information and video streams, among other things. Comment: Comment: A full list of IPv4 multicast assignments can be found at: Comment: Comment: http://www.iana.org/assignments/multicast-addresses Comment: Comment: A document describing the policies for assigning multicast addresses can be found at: Comment: http://datatracker.ietf.org/doc/rfc5771 Ref: https://rdap.arin.net/registry/ip/224.0.0.0 OrgName: Internet Assigned Numbers Authority OrgId: IANA Address: 12025 Waterfront Drive Address: Suite 300 City: Los Angeles StateProv: CA PostalCode: 90292 Country: US RegDate: Updated: 2012-08-31 Ref: https://rdap.arin.net/registry/entity/IANA OrgAbuseHandle: IANA-IP-ARIN OrgAbuseName: ICANN OrgAbusePhone: +1-310-301-5820 OrgAbuseEmail: abuse@iana.org OrgAbuseRef: https://rdap.arin.net/registry/entity/IANA-IP-ARIN OrgTechHandle: IANA-IP-ARIN OrgTechName: ICANN OrgTechPhone: +1-310-301-5820 OrgTechEmail: abuse@iana.org OrgTechRef: https://rdap.arin.net/registry/entity/IANA-IP-ARIN # # ARIN WHOIS data and services are subject to the Terms of Use # available at: https://www.arin.net/resources/registry/whois/tou/ # # If you see inaccuracies in the results, please report at # https://www.arin.net/resources/registry/whois/inaccuracy_reporting/ # # Copyright 1997-2024, American Registry for Internet Numbers, Ltd. # It seems to be somehow related to multicast. Can this be turned off in Thorium?

Ah, I somehow missed that. I've only tried Supermium (and Thorium) on a single-core ThinkPad T42, WinXP x86 SP3, Pentium M 745. It's the only real (bare metal) XP that I've not upgraged.

Fired up my Satellite A100 just for you (core solo t1350, 32-bit, single core). Supermium runs, but is unusable (nearly freezes with 100% CPU usage -- as advertised, win32 mentioned it needs 2 cores min.), Thorium (latest, all versions) is not recognized as a valid Win32 app (errors out, doesn't run). Might be my fault, though other browsers (Kafan Minibrowser, 360Chrome) run fine[ish. pretty slow]. Disregard, wrong windows version. Downloading/installing now, will update in a few mins Specs: XP 32 SP3, Core Solo T1350@1.86GHz, 4GB installed RAM.

You might find some info on DCI in Video for Windows SDK for Windows 3.1. Also, check the docs and code examples from early edition of Windows 95 SDK or beta SDKs. For example, build 484: https://archive.org/search?query=build+484

https://youtu.be/wJ46Y6xt2pU https://youtu.be/SFYRVVpatqo https://youtu.be/-7QhRdhtnzE

I noticed a huge delay, before it finally got through, when posting inf. in a post. Like it checks something or so.

It's meaningless, since many people simply don't report crimes, don't want to get involved with the police, etc, and the issue is rather serous (remember the scandal you mentioned), Supermium browser, which this one is based on, is paid.

Agreed! (case in point - LINUX !!!)

Since WebGPU ("unsafe" GPU, how the devs still call it) now works on Vista+, it'd make sense for @XPerceniol to switch it off, too. (since he disables webgl, I thought it'd be of interest) https://github.com/Alex313031/thorium-legacy/releases/tag/M122.0.6261.169

64-bit OS. Even my XP is 64-bit. I thougth the PAE patch was only for those trying to get XP x86 to see more then 4 GB RAM ???

Mear

There's an older driver exist, probably more luck with that one? Sorry, I have no link to 64bit. Driver 14.51.10.64.5436 for Windows XP https://www.techspot.com/drivers/driver/file/information/16818/

Maybe it works for this too? Thank you very much! Setting dom.forms.button.standards_compliant to false did the trick. For the very first time, I was able to upload a test file from New Moon 28 to my main folder on Codeberg. I am really happy. Great find! Thanks again!

Moonchild mentioned a workaround - set dom.forms.button.standards_compliant to false on about:config page and restart the browser (page reload is not enough). Maybe it works for this too?

Oh, I pretty much guarantee that the answer to that, worded that way, is "nothing". Rather, v114 *plus* v115 brought a SO-CALLED "security fix" that Chase just "has to have".

Fish

Without the need to install yet another extension, changing the font can be done with uBlock Origin as well. 🙂 I like the Noto Sans font (hinted). Apply this rule in uBO to apply to all elements in the page: forums.digitalspy.com##*:style(font-family: "Noto Sans" !important; src: local(Noto Sans) !important;) Another example. Have Google search results display in bold: google.com##H3:style(font-weight:bold !important;)

You're right. My Windows XP Professional SP3 32-bit is fully updated with all patches/updates after SP3 including all POSReady 2009 ones. But this was never a problem. And uBlock Origin 1.57.2 works fine here in single-process mode. Thus, I don't think that the problem is related to POSReady 2009 patches. But thanks for your opinion!

Ok! I checked my system once again. In my Windows XP main partition, I unfortunately discovered that one of four registry keys relating to the region and language options had disappeared. No idea why. However, the time and time zone display was still intact. Fortunately, I have a second Windows XP partition from which I was able to export the missing key HKEY_CURRENT_USER\Control Panel\International and import it into my main partition. Now, all is working again. This fixing has actually led to a change in Firefox ESR 52.9.0. Now I also get the correct time and time zone displayed there:

All is correct. Do you really think that I wouldn't have checked that first of all? So, the question is what did you and @UCyborg do that makes Firefox ESR 52..9.0 show the correct time and timezone?

On the other hand, the following JavaScript command returns the correct time zone including DST in Firefox ESR 52.9.0: new Date().toTimeString().slice(9); "GMT+0200" or new Date().toString(); "Mon Apr 22 2024 20:45:37 GMT+0200" or new Date().toDateString(); "Mon Apr 22 2024"

I found a further JavaScript command to get the correct, local time including DST in the web console of New Moon 28, Firefox ESR 52.9.0 and Serpent 52 probably too: new Date().toTimeString().slice(0,8); returns the correct,, currently local time at the moment: "19:41:01"

The JavaScript command Intl.DateTimeFormat().resolvedOptions().timeZone; leads in New Moon 28, Serpent 52 and Firefox ESR 52.9 to the same, wrong output (or should I say the same, currently wrong output? ) under Windows XP: "Etc/GMT-1" which means UTC/GMT +1 hours and is the correct winter time. In my location, however, it should actually be UTC/GMT +2 hours due to the currently valid summer time (DST). But this JavaScript command new Date().getTimezoneOffset(); returns the correct difference (in minutes) between UTC time and my local time: -120 My conclusion is that the JavaScript commands new Date().toLocaleString(); new Date().toLocaleString('de-DE'); only return the general time zone (here winter time, UTC/GMT +1) without any DST which has been set in Windows XP whereas. The JavaScript command new Date().toLocaleTimeString('de-DE', {timeZone: 'Europe/Berlin'}); can do the correct output in my case, of course. Anyway! There is a problem fetching the correct DST. Cheers, AstroSkipper

You couldn't run it on XP or even Vista. It happened only recently. And Thorium from that era was based on 109 Chrome, it simply couldn't be "unstable" on officially supported OS. We are interested only in recent tests.

I have a similar Xeon Board from 2009. It was sold with the first edition of Server 2008. Server 2008 was issued to OEM sales on January 2008. Server 2003 and XP64 had some odd behaviour with the C-states of Xeon CPUs. I had to force C1 state only, and the issues were gone.