UCyborg

Spamming the forum with new topics won't speed up the development.

It's only supported on Win10 builds up to 18363.752. As long as you stay on 18363 branch, a workaround would be to downgrade just DWM files, as described in the following thread:

I have a working FreeDOS installation on a bootable USB flash drive, so I could've tried Links with the old SMC EZ Card 10/100 (SMC1255TX), for which a packet driver exists, but the card died in a short time frame after being installed. PC went to sleep one day and didn't wake up, didn't respond to reset button, didn't POST after power cycle, no signs of life, just fans spinning. Thought that was it and that motherboard went kaput until I removed the card.

Nothing to do with XP, Good Old YouTube 1.16.0 doesn't work in recent Chrome and Firefox releases anymore neither.

Has anyone run Links in FreeDOS?

I solved the original problem, after removing as much root access traces as possible, it turned out SELinux being in permissive mode rather than enforcing mode was causing problems. I can't keep it in enforcing mode permanently because my OS doesn't have it configured properly. Temporarily putting it in enforcing mode solved the app activation problem. At a later point, another problem showed up, unable to authenticate with the app to access the bank site the normay way, though I can use emergency one-time password generated in the app to get in, but that's not sufficient, eg. if you perform some action on the bank site that requires to be authorized on the mobile app, it wouldn't work. Cause unknown, but apparently there are other users who have the same problem. BlueStacks and other solutions for running Android on PCs have the problem that they don't pass SafetyNet checks, so they're useless for these types of apps.

Lightweight and compatible with modern web is an oxymoron. Moonchild Productions' browsers are good for bells and whistles, but also have problems with memory leaks, though that seems to be a common problem with anything descending from Firefox.

Is the old way of accessing still complaint with the SCA? Customer's knowledge - password, customer's possession - certificate. This one should still render it properly with default configuration at least (just the background color, shouldn't be oversized). Sure you do. I don't only notice it, but it's actually irritating to my eyes. Edit: That's interesting, so only Iron on @jaclaz's machine takes this into account for some reason?

Didn't mention this yet, but at some later point, I've confirmed that the app was only picking up on SELinux status somehow, so removing Magisk for the activation process was unnecessary. Maybe the more accurate description is that permissive SELinux lets the app see something it doesn't like, but I'm unable to confirm that theory. I'd report the issue to Magisk developer to help improve MagiskHide, but unless you actually know the technicalities on how some app is picking that your device is not "clean" (https://github.com/topjohnwu/Magisk/issues/1152), it's pointless. Some banks still send you the SMS with the code to confirm it's really you rather than rely on the app. In my case, the reliability of the particular app is also a problem. You don't get a lot of bad user reviews on Google Play unless your app is really fishy. The title of this topic refers to: https://en.wikipedia.org/wiki/Rooting_(Android) Financial apps don't like rooted phones. Regarding COVID-19 app in this country, they called it #OstaniZdrav (translates to #StayHealthy) and since 14th December, if you cross your municipality's border and your reason is not one of the allowed exceptions for which you have a confirmation like a permit from your employer, you're permitted to move around if you have the activated app on your phone. If a cop pulls you over and you have neither, you're either warned or pay a fine, then sent back to where you came from. There are reported reliability issues with the app, a lot of people don't have the smartphone, others don't have the phone that would support the app (it requires Android 6.0+), it's considered unconstitutional since there's no alternative for those that can't use it for one reason or another etc. etc. The whole crisis has turned into a political game and people have had bad attitude since the beginning. Your typical Slovene's mentality, even when reasonable measures were just recommended rather than enforced: "It's recommended, therefore, it only applies to others, not me.". They've opened up certain shops/services temporarily recently until the holidays and of course people rushed there in big numbers as usual. Third wave is already anticipated for the next year. Slovenia is currently at the top in the deaths per million (last 7 days) criteria, according to this: https://www.statista.com/statistics/1104709/coronavirus-deaths-worldwide-per-million-inhabitants/ App or no app, I don't think it matters at this point.

UMDF drivers are bundled in WMP11 installer and may either be installed automatically along with WMP11 or separately by running two installers packed inside WMP installer file. When I got those drivers ready, it still wouldn't work, device wouldn't be installed (as if drivers were missing). I had to find the device key in registry under HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\USB and modify CompatibleIDs setting by adding USB\MS_COMP_MTP at the top then finally trigger the driver update from Device Manager. Afterwards, the device was installed successfully and showed in My Computer. USB debugging enabled in Android may also prevent MTP from being seen on the computer if ADB drivers aren't installed and working, so my advice to anyone would be to start with USB debugging disabled and worry about that after you get MTP alone working.

Old bugs inherited from the old Firefox codebase that appear to have been fixed in later official Firefox releases. Not so sure about that anymore after a bit more testing. The whole multi-process thing just makes it easier to reclaim memory since you can just terminate the process without having to deallocate explicitly allocated memory. I think browser devs in general should slow down a bit and go back to basics.

Web browser (and/or graphics driver DLL) may also leak memory and those damned multi-process browsers take significant amount of memory before you even open any website. My work laptop (Win10 1809 x64) idles at about 3 GB, but I didn't set it up from scratch and I might have left some less obvious stuff running, which accumulate. Initially, idle RAM usage was about 5 GB and both MS SQL and MySQL servers were running in the background. The same OS build on my home desktop would usually be at 1,3 GB. A screenshot from my personal laptop (Win10 1809 x86): It can get tight, but I can work with that. The most impactful change regarding resource usage that I tend to do is that I disable Windows Defender, though I suppose that's not the most suitable configuration for the common folk.

FX 52.9 does support WebGL.

Aye. And guess what's the alternative...another app...HID Approve, exists as Android 5+ app and as Windows 10 UWP app. And they suggested it without sending me the user name and activation code. When I asked them about those, they just sent me the link to the same instructions saying I need to input the user name I'm supposed get in the email and the code which I'm supposed to get in the SMS. I guess the user name is the same as the one used to login to the bank site, but no sign of the SMS neither. People are insufferable.

Asked about it on PM forum as the issue is inherited from upstream.

The problematic code is in one of the core Vista's system components, not VLC...

There is indeed a difference between VLC versions due to what kind of icons developers chose to put on the buttons. But this topic was started as a discussion for the solution about the technical problem that happens under specific circumstances and causes the application to render GUI in the old fashion, like how it was always done before Microsoft added visual style functionality in Windows XP. The topic was also initially placed in Vista sub-forum, which I think was appropriate, considering it is related to Vista specific issue. The workaround ensures theming works for the application so the buttons and other GUI elements get background texture from Microsoft's Aero theme. That's all there is to it. It doesn't seem like this discussion is going anywhere, so I believe that unless there is an interest in fixing the issue at the OS level, we can just call it a day. BTW, I have tested the following code snippet: SetDefaultDllDirectories(LOAD_LIBRARY_SEARCH_SYSTEM32); HMODULE hUxThemeLib = LoadLibraryExW(L"C:\Windows\System32\UxTheme.dll", NULL, LOAD_WITH_ALTERED_SEARCH_PATH); It works as expected on Vista. So if the code that loads UxTheme.dll used the absolute path, there wouldn't be a problem. It's also done like that on Win7.

Application should follow OS setting for displaying date and time, especially if it doesn't provide the ability to specify custom format.

My computer went to sleep yesterday and then didn't wake up, just fans were spinning, it didn't respond to reset button. Turning it off and on again, no change. I took the card out and then it started working. Guess I'm better of buying a new ethernet card. It isn't as much of an emergency as long as the phone works. There's no cable setup in this house, but sometimes I plugged the PC to my laptop, which has ethernet and wireless adapters bridged so that way I could connect the PC directly to the home network without having a NAT point in the way. The backstory is that I've apparently damaged onboard ethernet card by accidentally plugging in PS/2 keyboard while the computer was on. It's still detected by the OS, but AI NET 2 in BIOS reports funky status for pairs 4-5 (alternates between "Open" and "Failed") and 7-8 (constant "Open"). The normal status is "Normal". I don't suppose it can be fixed and even if it was taken somewhere for the repair, I imagine the cost would go way beyond the price of a new ethernet card. Oh well...

The question in the title. There are some subjects that don't pertain to any specific Windows version. Edit: just found someone made a suggestion over a decade ago. So if one has to choose, I guess Software Hangout is the most generic forum for this sort of thing, huh?

Thanks, I'll give the "backup the first sector and zero it" approach a spin. I can use dd from Linux for the task, I'm quite familiar with it. And I was going to make it complicated by looking at what bytes to change in the first sector.

When I thought this was solved...over a week ago, I got some vague error on the web page after confirming logon credentials, the push notification didn't arrive on the phone. It arrived on second try, but after confirming it on the phone, I get another vague error on the phone and this error persists. I can still get in with emergency one-time password, but it's only meant as a means to reset a forgotten password. If some action will have to be confirmed on the phone, it's going to be a problem. Absolutely nothing has changed on my device, I even restored backup of data partition from the time it worked, but all that did is that now even the push notification doesn't arrive anymore when trying to get in the normal way. Maybe because I reactivated the app some time after the backup because I wanted a different PIN that can't be changed otherwise. It still doesn't arrive when restoring the current state. I think the time has come to go somewhere else. I might have just found a suitable alternative, even better at some other aspects besides online banking.

This is quite easily reproducible. I've known about the existence of this issue on Vista for a while, though never looked into it deeply. Your screenshots make it really hard to see the problem due to whatever customizations you applied. I forgot about that DLL and the fact that even old programs that use older comctl32.dll still have skinned scrollbars and menus, so that's another clue. Bingo! LOAD_WITH_ALTERED_SEARCH_PATH flag may only be used with absolute path to the DLL, otherwise the behavior is undefined.

You're right, I didn't have the list opened so was going from my memory... It was definitely mounted in the past. I forgot to mention, if I have already booted Windows on partition 1, before I hibernate it to boot OS on partition 2, I have to give that partition a drive letter (let's assign a letter H) then run the following: mountvol H: /P Otherwise the OS on partition 2 will be upset. Has to be done once per clean boot of OS on partition 1. It's like the action done by mountvol doesn't persist across reboots. I'll have to check what mountvol reports for partition without drive letters that I have run the previous command on after reboot, right after the running the command it says: *** NOT MOUNTABLE UNTIL A VOLUME MOUNT POINT IS CREATED *** I figured I can do that to save session in one OS, then I can go try something in another OS and then be back right back where I left. And since this doesn't reset uptime timer and everything else to reload from scratch, it doesn't disrupt seeing how the OS fares with longer periods without rebooting. But when two Win NT 6.x+ OSes are involved, this little problem occurs. So first 512 bytes of partition I want to hide (I've verified that sectors are 512 bytes on my disk)?