Jump to content

My Browser Builds (Part 4)


Recommended Posts

9 hours ago, modnar said:

Sadly another crash - Serpent 52.9 - mozjs.dll, I have halved and 256-colored the pic to make it smaller. Either Serpent 52.9 doesn't like the craptor lake refresh news or videocardz.com in general. :)

By the way - it's reproducible. I can go to craptor lake refresh news time and again and it shall crash after loading the site, it crashes just before it shows Disqus system comments that site uses.

Serpentinšek_really_dislikes_videocardz.JPG

Can't reproduce, loads fine here without any crashes. The other articles load fine as well without any crashes, and not just that Intel Raptor Lake Refresh article.

t3m-XPOm-Ztg.png

Edited by mina7601
Link to comment
Share on other sites


On 10/12/2023 at 3:28 AM, AstroSkipper said:

I tested your linked website in the latest release of New Moon 28 (2023-10-06). I couldn't observe any serious problems except the initially cumbersome loading I already mentioned. smilie_denk_24.gif

That's an updated browser. How 'bout a ten years old browser? Wayback Machine has JavaScript those won't understand.

Another problem I always remember to point out, much slower execution of WebGL content in newer browsers, seems performance degradation is worse with maximized than with much smaller window. So the old K-Meleon (KM-Goanna) from 2020 still manages to load Lights:

spacer.png

Frame rate here is fine (number on the very left in the upper-left corner), at least with layers.prefer-opengl enabled in about:config. That trick stopped working with some browser update.

Impossible to get smooth frame-rate with Serpent, New Moon etc. on XP:

spacer.png

Of course the code paths used on newer Windows are faster and there's no problem with D3D and OpenGL (ANGLE disabled scenario) co-existing. Though on XP, even with ANGLE enabled, so you're only supposed to have D3D9 used in the entire browser, it's still slow for some reason.

Edited by UCyborg
Link to comment
Share on other sites

2 hours ago, UCyborg said:

That's an updated browser. How 'bout a ten years old browser? Wayback Machine has JavaScript those won't understand.

TBH, I do not use 10 years old, unsupported browsers at all. :no: And that wouldn't make sense either nimportequoi.gif, at least as long there are more recent ones which is the case under Windows XP. :thumbup Your statement was:

On 10/11/2023 at 6:54 PM, UCyborg said:

Too bad the web site can't be used to view archived web sites in old web browsers.

I replied:

On 10/11/2023 at 10:55 PM, AstroSkipper said:

The website web.archive.org can of course be used to view archived websites in old web browsers. It works, for example, in New Moon 28 or Serpent 52. It's more a question of defining what an old web browser is. :whistle:

New Moon 28 is an old, legacy browser but still updated, thanks to @roytam1. And there is no problem to access your linked website https://web.archive.org/web/20140701122330/http://lights.helloenjoy.com/. As I already stated, it is a problem of defining what an old web browser is. So, if I take you literally, then your statement is not correct. :no: The website web.archive.org works in old browsers to view archived web sites. :yes: The proof is here:

22 hours ago, AstroSkipper said:

I tested your linked website in the latest release of New Moon 28 (2023-10-06). I couldn't observe any serious problems except the initially cumbersome loading I already mentioned. smilie_denk_24.gif Here is a screenshot as a proof:

NM28-lights-helloenjoy-com.png

Kind regards, AstroSkipper albert.gif

Edited by AstroSkipper
Update of content
Link to comment
Share on other sites

I couldn't get anywhere interesting on Lights Hello Enjoy. After waiting for a long time I could click Play, which brought me to a scene of a grey hills. Portions of them illuminated in pulses without interactivity or sound. A 3D scene is a modern invention and an old browser can't be used.

The Archive is always very slow to respond, which is ok if I need to read a text or get a small zip. Feels like they have little elves that pick up a disk, dust it off and insert it.

Link to comment
Share on other sites

17 hours ago, Mehmed said:

Any video on vk.com causes the browser to crash.
https://vk.com/clips-36759319?z=clip-36759319_456241478
https://vk.com/video-101322457_456239576

Serpent 20231007

AppName: basilisk.exe
AppVer: 4.8.7.8657
ModName: mozjs.dll
ModVer: 0.0.0.0
Offset: 000b6ef0

Consol error:
uncaught exception: 5.119190723800904e-246
TypeError: The ReadableStream reader was released.

 


If you disable these variables, the video works fine.
dom.streams.enabled > false
javascript.options.streams > false

crash confirmed, but still need to wait for upstream fix

Link to comment
Share on other sites

23 hours ago, mina7601 said:

Can't reproduce, loads fine here without any crashes. The other articles load fine as well without any crashes, and not just that Intel Raptor Lake Refresh article...

Interesting - I have been fixing poor Avast implementation of their filter driver (aswSnx) and its interaction with Diskeeper in my XP and looks like that altitude and load order combination wasn't optimal.

I have now (finally) looked at Win7 32bit (thinpc) implementation and replicated the same in XP and set altitudes to completely factory values with load orders fixed and Serpent 52.9 doesn't crash anymore. Jolly good. :-)

Link to comment
Share on other sites

21 hours ago, UCyborg said:

So the old K-Meleon (KM-Goanna) from 2020 still manages to load Lights:

Edited the post, accidentally pasted same picture twice...

23 hours ago, AstroSkipper said:

As I already stated, it is a problem of defining what an old web browser is.

So eg. you compare two programs with common heritage that diverged (say Firefox and any UXP browser), both regularly updated, also in terms supported web standards, but one of them is considered old while other is not. Perhaps old-school (old-fashioned) is more appropriate or distinct term? You'd arrive back at Firefox 1.0 in with both browsers.

This made me think of another example where old could be linked to the history, eg. Oracle Database could be said to be one of the oldest databases around.

It seems old is just too vague for what some of us refer to as "old browsers", perhaps a more distinct "obsolete versions of particular web browser" would be clearer and on-point. But "old" is short and sweet to use.


I'm trying to setup Lights on a local web server, I think I managed to strip the extras Wayback Machine adds, a bit annoying, but seems doable, but now I have to manually download a bunch of files that are fetched during runtime with JavaScript code. So "Save Page As..." really only works fully for static web pages. maybe it still won't run on Chromium 39 in the end due to other factors, but it'll be a learning experience.

Oh, the web server is Apache 2.4.4, packaged inside WAMP 2.4, MySQL version is 5.6.12. Now that's old! There are no updated forks of those that run on XP, aren't there?

Link to comment
Share on other sites

New build of Serpent/UXP for XP!

Test binary:
Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20231014-3219d2d-uxp-d6590573ed-xpmod.7z
Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20231014-3219d2d-uxp-d6590573ed-xpmod.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/custom

IA32 Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20231014-3219d2d-uxp-d6590573ed-xpmod-ia32.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/ia32

NM28XP build:
Win32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20231014-d849524bd-uxp-d6590573ed-xpmod.7z
Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20231014-d849524bd-uxp-d6590573ed-xpmod-ia32.7z
Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20231014-d849524bd-uxp-d6590573ed-xpmod-sse.7z
Win64 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20231014-d849524bd-uxp-d6590573ed-xpmod.7z

Official UXP changes picked since my last build:
- Correct overzealous GMP checks (c3767681a9)
- Issue #2332 - Refactor code that feeds video stream sink when it gets added (1ed778419f)
- Issue #2332 - Use only one listener in MediaPipelineTransmit (580a231706)
- Issue #2332 - Use the same path for passing on missed data to video sink, as during normal operation (a87f8cc7f7)
- Issue #2332 - Add deprecation warnings to writable RTCSessionDescription (f2dbc4c6ce)
- Issue #2332 - Update RTCIceCandidate to spec (ddf29677d9)
- Issue #2332 - Have addIceCandidate take a dictionary (80a3c49ba2)
- Issue #2332 - Have set(Local|Remote)Description take dictionaries (6b713575c9)
- Issue #2332 - Have createOffer/Answer resolve with dictionaries (a1c22cb034)
- Issue #2332 - Update most tests to skip RTCIceCandidate construction (f615587ab7)
- Issue #2332 - Update most tests to skip RTCSessionDescription construction (763cdb00cb)
- Issue #2332 - Fix CamerasChild helper to return by value instead of out-of-scope reference (348f761a9b)

No official Pale-Moon changes picked since my last build.

No official Basilisk changes picked since my last build.

My changes picked since my last build:
- ported from mozilla:
 - Bug 148624 - only show tooltip when document has focus. r=emilio (8ae372dc88d1)
 - Bug 1857513 - Use window activeness rather than document.hasFocus() to display chrome-only tooltips. r=smaug (4cbf370b4e16) (f3deac5f5f)
- Reverted "Correct overzealous GMP checks (c3767681a9)" during merging, since related commits are reverted

Update Notice:
- You may delete file named icudt*.dat inside program folder when updating from old releases.

* Notice: From now on, UXP rev will point to `custom` branch of my UXP repo instead of MCP UXP repo, while "official UXP changes" shows only `tracking` branch changes.

Link to comment
Share on other sites

New build of BOC/UXP for XP!

Test binary:
MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20231014-d0fd16ed-uxp-d6590573ed-xpmod.7z
BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20231014-d0fd16ed-uxp-d6590573ed-xpmod.7z

source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom

* Notice: the profile prefix (i.e. parent folder names) are also changed since 2020-08-15 build, you may rename their names before using new binaries when updating from builds before 2020-08-15.

--

New build of HBL-UXP for XP!

Test binary:
IceDove-UXP(mail) https://o.rthost.win/hbl-uxp/icedove.win32-20231014-id-656ea98-uxp-d6590573ed-xpmod.7z
IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20231014-id-656ea98-ia-93af9a0-uxp-d6590573ed-xpmod.7z

source repo (excluding UXP):
https://github.com/roytam1/icedove-uxp/tree/winbuild
https://github.com/roytam1/iceape-uxp/tree/winbuild

for UXP changes please see above.

Link to comment
Share on other sites

New build of post-deprecated Serpent/moebius for XP!
* Notice: This repo will not be built on regular schedule, and changes are experimental as usual.
** Current moebius patch level should be on par with 52.9, but some security patches can not be applied/ported due to source milestone differences between versions.

Test binary:
Win32 http://o.rthost.win/basilisk/basilisk55-win32-git-20231014-f3d14e640-xpmod.7z
Win64 http://o.rthost.win/basilisk/basilisk55-win64-git-20231014-f3d14e640-xpmod.7z

repo: https://github.com/roytam1/basilisk55

Repo changes:
- import from UXP: [Basilisk] Don't build hunspell shared on SunOS (ebf28e41) (d93dd0161)
- ported from UXP: [Basilisk] Remove stubs and references. Remove some telemetry leftovers (32e41150) (14d0a8725)
- ported from UXP: [Basilisk] Pref changes for old reports that no longer exist (6b8a4e9b) (ca064beea)
- Basilisk: remove DirectoryLinksProvider, and also mobilePromo code in real (2911d3ef7)
- ported from UXP: [Basilisk] Remove some mobile leftovers (77bc3513) (5516d75d6)
- import from UXP: [Basilisk] Remove some android leftovers (c4f84bd3) (a01e26101)
- import from UXP: [Basilisk] Remove ios leftovers (3d339bae) (84ca0e6ef)
- ported from UXP: [Basilisk] remove more mozilla spy garbage (6c886245) (c734f12fd)
- ported from UXP: [Basilisk] Clean up an unused pref. Remove link to mozilla site from crash reporter (23ef0c81) (6b3f5eaa9)
- ported from UXP: [Basilisk] Remove last bits of telemetry (b5a13a2a) (638d4255e)
- ported from UXP: [Basilisk] Remove obsolete documentation (a11b107f) (d1673432b)
- ported from UXP: [Basilisk] Remove unused functionality for reporting crashes to an external server (e73c73a4) (764c20341)
- ported from UXP: [Basilisk] Remove unused prefs for download of tiles on new tab page (3fa770bc) (f945431ea)
- ported from UXP: [Basilisk] Remove some leftover crashreporter stuff (264cd17e) (34e38de02)
- ported from UXP: [Basilisk] Remove browser.tabs.crashReporting.sendReport prefs (36ca1ce4) (de43839dd)
- ported from UXP: [Basilisk] Remove unsubmitted crash reporter from crash handlers (0ae50366) (5c55144af)
- ported from UXP: [Basilisk] Remove some unused crash report strings from localization data (04e9b40d) (b00e49f34)
- ported from UXP: [Basilisk] Report GTK version in about dialog for Linux builds (a0328304) (8f6acb5f8)
- ported from UXP: [Basilisk] Issue #67 - begin porting over changes to stop using appconstants (e1e2d7e8) (6b1df8f46)
- ported from mozilla:
 - Bug 148624 - only show tooltip when document has focus. r=emilio (8ae372dc88d1)
 - Bug 1857513 - Use window activeness rather than document.hasFocus() to display chrome-only tooltips. r=smaug (4cbf370b4e16) (f3d14e640)

Link to comment
Share on other sites

New NewMoon 27 Build!

32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20231014-b780764f57-xpmod.7z
32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20231014-b780764f57-xpmod-sse.7z
32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20231014-b780764f57-xpmod-ia32.7z

64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20231014-b780764f57-xpmod.7z

source repo: https://github.com/roytam1/palemoon27

repo changes since my last build:
- bits of Bug 1229519: Fix toolkit/components/satchel to pass eslint checks. r=dolske (c57ebb477f)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 937810 - disable application reputation check on b2g r=mmc,mossop (403f8e0353)
 - Bug 1216537 - Check and request storage permission if file download is started. r=nalexander,paolo (9233e7193d)
 - Bug 1163937 - Added forceSave function to DownloadIntegration and ensured that downloads removed in Sanitizer do not persist. r=margaret (9610017b5d)
 - Bug 989960 - Unhandled rejections in DOM Promises should cause xpcshell tests to fail. r=Yoric (be260d55d2)
 - Bug 1177237 - Implement OS query parsing and user search choice. r=jaws (86c807e606)
 - Bug 1177443 - Add 'system' purpose for searches coming from outside Firefox, r=MattN. (f6ac86ec2a)
 - Bug 1072037, part 2 - Tests for the effect of setting CSS animation's AnimationPlayer.currentTime. r=birtles (79639d2ea4)
 - bits of Bug 1145246, part 2 (849519f25d)
 - bits of Bug 1145246, part 7 (3ec5500e08)
 - bits of Bug 1109390 part 5 (6d849b9b3c)
 - Bug 1074630, part 2 - CSS animations tests for Web Animations finishing behavior. r=birtles (bd0aa575bd)
 - Bug 1073379, part 6 - Tests for the effect of setting CSS animation's AnimationPlayer.startTime. r=dholbert (9d8aad8d28)
 - Bug 1073379, part 6b - Re-enable the player.playState check in checkStateAtActiveIntervalEndTime, but disable the first checkStateAtActiveIntervalEndTime call in the 'Skipping backwards through animation' tests. r=orange (3a9ea58cef)
 - Bug 1141710, part 1 - Update to the new version of addDiv() in testcommon.js. r=dholbert (262b9dc018)
 - Bug 1141710, part 2 - Update comments to be non-Mozilla specific. r=dholbert (177e8818f8)
 - Bug 1141710, part 3 - Stop using ECMAScript 6 features in test_animation-player-starttime.html, since other browsers don't support them. r=dholbert (6f0c6ab687)
 - Bug 1141710, part 4 - Avoid race condition when taking timestamps by reusing the original timestamp. r=dholbert (bc6ff2f3b1)
 - Bug 1141710, part 5 - Change from assert_true() to the new assert_unreached(). r=dholbert (0c1608c4c4)
 - Bug 1141710, part 7 - Create helpers to get the startTime corresponding to various points through the active duration. r=dholbert (568cf9a054)
 - Bug 1141710, part 7 - Store the generated 'animation' property string in a global constant and reuse that constant. r=dholbert (ce85c0d7a9)
 - Bug 1141710, part 8 - Create helpers to get the startTime corresponding to various points through the active duration. r=dholbert (6d6b1625ff)
 - Bug 1141710, part 9 - Get the timeline from the player instead of from the document. r=dholbert (9660c00f1d)
 - Bug 1141710, part 10 - Update some assertion text and comments. r=birtles (d238a482b5)
 - Bug 1141710, part 11 - Assert that seeking over the before and active phases worked. r=birtles (4aedf388f9)
 - Bug 1141710, part 12 - Check that the hold time is updated when the startTime is set to null. r=birtles (0473baeabd)
 - Bug 1141710, part 13 - Wrap all lines at the 80 column mark. r=birtles (7d3ed6b453)
 - Bug 1141710, part 14 - Get rid of the checks at 90% through the active duration. r=birtles (0f3c4cb7ca)
 - part of Bug 1145246, part 2 - (ec09bf3fdd)
 - half of Bug 1109390 part 1 - Add tests for getting the startTime; r=jwatt (a682851ba6)
 - Bug 1149832 - Replace the Web Animations test helper waitForTwoAnimationFrames() with a helper that takes a frame count. r=birtles (da312eae7d)
 - Bug 1235286 - Part 1: Add an argument to waitForAnimationFrames to run a task in each requestAnimationFrame callback. r=birtles (199a789546)
 - Bug 1235286 - Part 2: Tests for animation optimizations. r=birtles (e510c68a7c)
 - Bug 1235286 - Part 3: Comment out some compositor animations tests. r=birtles (61791d5b69)
 - Bug 1235345 - Remove services/metrics. r=gfritzsche (29ac4a5895)
 - Bug 1204846 - Modify the NetworkStatsDB to allow getSamples returns expired data at first sample and modify the test case. r=ethan (3e9d1f9a77)
 - Bug 1237227 - Check the return of context->GetDisplayRootPresContext() for validity. r=roc (2ad8dcf545)
 - sync parts of AppConstants.jsm for isPlatformAndVersion* functions (30f7e10db1)
- ported from mozilla:
 - Bug 148624 - only show tooltip when document has focus. r=emilio (8ae372dc88d1)
 - Bug 1857513 - Use window activeness rather than document.hasFocus() to display chrome-only tooltips. r=smaug (4cbf370b4e16) (b780764f57)

Link to comment
Share on other sites

New regular/weekly KM-Goanna release:
https://o.rthost.win/kmeleon/KM76.4.9-Goanna-20231014.7z

Changelog:

Out-of-tree changes:
* update Goanna3 to git e9dee823ef...b780764f57:
- bits of Bug 1229519: Fix toolkit/components/satchel to pass eslint checks. r=dolske (c57ebb477f)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 937810 - disable application reputation check on b2g r=mmc,mossop (403f8e0353)
 - Bug 1216537 - Check and request storage permission if file download is started. r=nalexander,paolo (9233e7193d)
 - Bug 1163937 - Added forceSave function to DownloadIntegration and ensured that downloads removed in Sanitizer do not persist. r=margaret (9610017b5d)
 - Bug 989960 - Unhandled rejections in DOM Promises should cause xpcshell tests to fail. r=Yoric (be260d55d2)
 - Bug 1177237 - Implement OS query parsing and user search choice. r=jaws (86c807e606)
 - Bug 1177443 - Add 'system' purpose for searches coming from outside Firefox, r=MattN. (f6ac86ec2a)
 - Bug 1072037, part 2 - Tests for the effect of setting CSS animation's AnimationPlayer.currentTime. r=birtles (79639d2ea4)
 - bits of Bug 1145246, part 2 (849519f25d)
 - bits of Bug 1145246, part 7 (3ec5500e08)
 - bits of Bug 1109390 part 5 (6d849b9b3c)
 - Bug 1074630, part 2 - CSS animations tests for Web Animations finishing behavior. r=birtles (bd0aa575bd)
 - Bug 1073379, part 6 - Tests for the effect of setting CSS animation's AnimationPlayer.startTime. r=dholbert (9d8aad8d28)
 - Bug 1073379, part 6b - Re-enable the player.playState check in checkStateAtActiveIntervalEndTime, but disable the first checkStateAtActiveIntervalEndTime call in the 'Skipping backwards through animation' tests. r=orange (3a9ea58cef)
 - Bug 1141710, part 1 - Update to the new version of addDiv() in testcommon.js. r=dholbert (262b9dc018)
 - Bug 1141710, part 2 - Update comments to be non-Mozilla specific. r=dholbert (177e8818f8)
 - Bug 1141710, part 3 - Stop using ECMAScript 6 features in test_animation-player-starttime.html, since other browsers don't support them. r=dholbert (6f0c6ab687)
 - Bug 1141710, part 4 - Avoid race condition when taking timestamps by reusing the original timestamp. r=dholbert (bc6ff2f3b1)
 - Bug 1141710, part 5 - Change from assert_true() to the new assert_unreached(). r=dholbert (0c1608c4c4)
 - Bug 1141710, part 7 - Create helpers to get the startTime corresponding to various points through the active duration. r=dholbert (568cf9a054)
 - Bug 1141710, part 7 - Store the generated 'animation' property string in a global constant and reuse that constant. r=dholbert (ce85c0d7a9)
 - Bug 1141710, part 8 - Create helpers to get the startTime corresponding to various points through the active duration. r=dholbert (6d6b1625ff)
 - Bug 1141710, part 9 - Get the timeline from the player instead of from the document. r=dholbert (9660c00f1d)
 - Bug 1141710, part 10 - Update some assertion text and comments. r=birtles (d238a482b5)
 - Bug 1141710, part 11 - Assert that seeking over the before and active phases worked. r=birtles (4aedf388f9)
 - Bug 1141710, part 12 - Check that the hold time is updated when the startTime is set to null. r=birtles (0473baeabd)
 - Bug 1141710, part 13 - Wrap all lines at the 80 column mark. r=birtles (7d3ed6b453)
 - Bug 1141710, part 14 - Get rid of the checks at 90% through the active duration. r=birtles (0f3c4cb7ca)
 - part of Bug 1145246, part 2 - (ec09bf3fdd)
 - half of Bug 1109390 part 1 - Add tests for getting the startTime; r=jwatt (a682851ba6)
 - Bug 1149832 - Replace the Web Animations test helper waitForTwoAnimationFrames() with a helper that takes a frame count. r=birtles (da312eae7d)
 - Bug 1235286 - Part 1: Add an argument to waitForAnimationFrames to run a task in each requestAnimationFrame callback. r=birtles (199a789546)
 - Bug 1235286 - Part 2: Tests for animation optimizations. r=birtles (e510c68a7c)
 - Bug 1235286 - Part 3: Comment out some compositor animations tests. r=birtles (61791d5b69)
 - Bug 1235345 - Remove services/metrics. r=gfritzsche (29ac4a5895)
 - Bug 1204846 - Modify the NetworkStatsDB to allow getSamples returns expired data at first sample and modify the test case. r=ethan (3e9d1f9a77)
 - Bug 1237227 - Check the return of context->GetDisplayRootPresContext() for validity. r=roc (2ad8dcf545)
 - sync parts of AppConstants.jsm for isPlatformAndVersion* functions (30f7e10db1)
- ported from mozilla:
 - Bug 148624 - only show tooltip when document has focus. r=emilio (8ae372dc88d1)
 - Bug 1857513 - Use window activeness rather than document.hasFocus() to display chrome-only tooltips. r=smaug (4cbf370b4e16) (b780764f57)

* 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

Link to comment
Share on other sites

2 hours ago, roytam1 said:

 

Hi roytam1, on NM27 build 20230812-6e095708be when open the new Private Tab, the position of the Tabs is very balanced.

But since newer build when open the new Private Tab, tabs are created in a downward direction causing the bottom part of the Tab to be covered by the taskbar. Is there a way to restore the old behavior?

Thank you.

Old.png

New.png

Link to comment
Share on other sites

On 10/12/2023 at 12:32 AM, AstroSkipper said:

TBH, I do not use 10 years old, unsupported browsers at all. :no: And that wouldn't make sense either nimportequoi.gif, at least as long there are more recent ones which is the case under Windows XP.

It can make sense for historical reasons for instance and period-correct web browsers are fitting. Pockets of old web are still out there.

I cleaned up the Lights, loads and runs on Chromium 39 fine now, also works on Firefox 6.0.2.

spacer.png

spacer.png

Although it's interesting that it's still not as smooth in old Firefox as one might expect (like some old K-Meleon builds that were better in that regard), not 60 FPS, but feels it's above 30 FPS, so improvement over updated UXP forks. No frame counter in that browser and I don't have any software on XP that could report it though.

Chromium didn't work with --use-gl=desktop here (to use native OpenGL rather than ANGLE's D3D9), just hung on startup, no noticeable performance difference between the two on old Firefox. Updated UXP forks however, they fail to get some shaders compiled with ANGLE, so native OpenGL is the only option.

spacer.png

Download link for the curious, needs to be extracted in the folder of an actual web server as browsers' security protocols prevent access to certain resources through file:///. I also had to add the music in Vorbis format, Wayback Machine only has MP3 version, despite the existing reference to OGG...old browsers can't do with MP3 it seems.

On 10/13/2023 at 3:58 AM, j7n said:

A 3D scene is a modern invention and an old browser can't be used.

Ha, you can even add it in an old Internet Explorer. But the content must be programmed with it in mind and it can be very limiting.

spacer.png

https://github.com/mengyou658/iewebgl

Also an old website of the project - http://iewebgl.com/, the person above just compiled it since no binaries are available from original authors.

iewebgl.exe under Bin is an installer, it install the add-on for the current user. I've put the whole WebSite2.0 folder content on the web server, then could see some samples that don't require HTML5 features.

Edited by UCyborg
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...