UCyborg

Auto Screen Capture supposedly needs at least Windows 8. It turned out that application doesn't depend on anything that only newer systems have. Here's the fixed version that will happily run on XP/Vista/7 (.NET Framework 4.0 required). These were required code changes.

Syncing through Date and Time in Control Panel starts the Windows Time service and keeps it running until you fully shutdown or restart, so it should sync at regular intervals according to its registry settings. Might explain the events that occur when it can't reach the NTP server. The tasks in Task Scheduler start the service, sync time and then stop the service. If sync fails, it will keep running until it succeeds. We have 2 tasks that virtually accomplish the same thing. Editing ForceSynchronizeTime and trying to save changes shows an error unless the custom trigger is deleted (invalid trigger?). Unless there's a catch, the only difference between 2 tasks is the way the time service is invoked. The funny part is having 2 tasks defined for 1 useful purpose in Task Scheduler that aren't actually being scheduled. And there's a third way to have the time synced, by keeping time service running all the time, like how it is on Windows XP. Though technically, it's just waiting majority of time consuming little memory. They said they changed it from auto-start to manual in Windows 7 to improve bootup performance, but then added significantly heavier background stuff in newer versions, especially Windows 10.Microsoft logic. I do use pool.ntp.org. Just wrapping my head around the automation part, which the good guys at MS have disabled. Late edit: not really...

OK, good point. You could also let the user select any option, but make an exclamation mark icon appear when less than optimal option is selected with an explanation why your selection might not be the best choice.

It hasn't. But why bother with built-in tools for any kind of work with partitions when alternatives are much richer in functionality?

Another one of those undocumented changes they didn't tell anyone about. They didn't even update the group policy setting Requirements (still says "At least Windows Server 2003" on Windows 10, even though it's ineffective on Windows 8+ systems and their server counterparts). Read this conversation for details. I might add, another one of those changes that involve nothing more than deleting existing code.

In Windows 7, they changed Windows Time service to be started manually or by the event of computer joining a domain. For non-domain computers, there's a scheduled task setup called SynchronizeTime that syncs time every Sunday at 1:00:00 or as soon as possible if it can't run then. But starting with Windows 8, SynchronizeTime task lacks the trigger, so it never runs. In addition, ForceSynchronizeTime task is added with custom action and trigger defined, so cannot be inspected in Task Scheduler. Does it ever run by itself? Running it manually does sync time. There are some clues when exporting ForceSynchronizeTime to XML: <Actions Context="LocalService"> <ComHandler> <ClassId>{A31AD6C2-FF4C-43D4-8E90-7101023096F9}</ClassId> <Data>TimeSyncTask</Data> </ComHandler> </Actions> According to registry entries, this invokes code in TimeSyncTask.dll. But the following part is still a mystery: <Triggers> <WnfStateChangeTrigger> <Enabled>true</Enabled> <Delay>PT1M</Delay> <StateName>7510BCA32F018915</StateName> </WnfStateChangeTrigger> </Triggers> Any ideas?

I thought deleting Manage Wireless Networks GUI in Windows 8 was going to be the most radical idea they'd come up with. Nope, here came the new way to manage file associations. To associate a program with all file types it can open, you're supposed to click through all file types and associate them one by one instead of having a button to associate them all with a single click. The Explorer interface also has ability to sort, but who needs that, right? /s Let's just remove the old interface and make it redirect to Settings app, also crash Explorer while at it. Some other bugs/oddities I've encountered in the current supposedly stable build that come to mind: Error message about not enough memory resources being available instead of copyright message in the Command Prompt in WinPE. Various Explorer/shell glitches: Trying to delete a file from the root of plain removable storage device presents a file confirmation dialog always shows file size as 0 bytes and file modification date is missing. Accent color change makes the file view scroll up (forcing usage of the old ListView control fixes that). IFileCopyHook interface is broken. Aero Peek is buggy with partial rendering of window frames and offsetting their positions (Aero Glass fixes that). Inconsistent behavior when a network adapter connected to internet has *NdisDeviceType set to 1 in registry; login/lock screen indicates internet connectivity, tray icon in the taskbar doesn't (the popup when you click it says Connected). Programs' thumbnails used in alt-tab interface and taskbar program buttons' popups are lost when a fullscreen game switches to non-desktop resolution. At various places throughout the OS, where text must be shortened to fit, an ellipsis isn't always appended. Also, shouldn't the notifications/notification center appear on the monitor with the main taskbar?

Version 1.5.9 has been officially released yesterday! Has anyone noticed that Aero Glass actually fixes broken Aero Peek behavior where under certain circumstances window frames are shifted and not rendered entirely? It usually happens if you don't have any window maximized or placed in the way that it covers upper-left portion of the desktop.

To summarize, prevalent thing across both threads are complaints about debug windows and watermark. There is very little about supposed technical issues with current experimental version. Someone mentioned memory leak, but didn't specify if the version of AG used was current at the time. And then the Windows updates and things shuffle around in DWM, requiring new debug symbols, which aren't downloaded automatically by AG unless the user has donation key and require manual intervention. So either the current experimental version already works well or if there are crashing or similar critical bugs still lurking in there, they're edge cases and nobody ran into them. There's also a fact that Big Muscle doesn't share his thoughts on release-ready status of the experimental version, so one can only speculate. It already happened that a lot of time passed between when experimental version was released and when it was followed by release version and changes between the experimental and release version were minimal at best, not worth mentioning in the changelog.

Found a bug, nothing wrong with browsers: However, under certain circumstances, eg. when browsing with Firefox in private mode, the offending script will be automatically blocked as part of its tracking protection and the bug will not occur: Basically, the advertising on the site interferes with its basic functionality.

Well in my case, with the current debug version, I do get the watermark on the second monitor.

It works correctly on my PC with Firefox 51.0.1, but broken with latest Chrome. It also works with stock browser on Android 4.4.4 and again, broken with Chrome. Edit: seems loading circle is absent on older browsers.

Given the choice of running debug version of Aero Glass or no Aero Glass at all, I personally prefer the latter option. It's just not the same as the release build, which is the real deal. What bugs me more than debug window and watermark are choppier window animations, which isn't really surprising outcome for debug build compiled with all the logging enabled and lack of compiler optimizations.

Windows 95 says when the subsystem version in executable file's header is higher than OS version that this program expects a newer Windows version, but NT systems say that it's not a valid application. I wonder why is that. Maybe due to the fact that NT could/can historically run on different CPU architectures and they just put the same error regardless whether it's just version mismatch or CPU for which the EXE was compiled for.

I guess it depends whether you find documented changes useful. Cumulative updates are great from the ease of maintenance perspective, but there's a problem as soon something you don't want is included, in this case Meltdown/Spectre patches. I'm currently on April 2018 patch level. Windows 8.1 is a bit shaky on my hardware, I've had occasional DRIVER_POWER_STATE_FAILURE BSOD when shutting down in the past. The problem hasn't surfaced once since January 2018 update. So whatever changes they did in kernel obviously have positive impact on my system's stability. Probably an isolated case though, potentially depending on uncontrollable factors. Edit: Never mind, those BSODs were caused by buggy driver for my network card. I forgot I've turned off every checkbox in its properties on Power Management tab. Guess I got some things mixed up in my head as I remember reading about some past unrelated update dealing with certain DRIVER_POWER_STATE_FAILURE BSOD, but it's unrelated to the problem I had.

When I tried to be smarter than Microsoft when it comes to updating strategy, I ended up with Explorer that continuously leaked memory. Then I risked re-installing Windows on top of existing installation, aka. repair install, something I haven't done since XP days. Results were catastrophic back then, turned my computer into a slow turtle. Fortunately, the original problem was fixed and I haven't encountered any major problems that would involve more than re-installing/re-applying an odd program/registry tweak or two.

If just K32* functions are needed for Pale Moon to launch on Vista, then PSAPI_VERSION=1 should be added to Preprocessor Definitions and the Minimum Required Version setting value changed to 6.00 or just deleted since Vista should still be the Visual Studio's default target.

There are some extra dependencies and I'm no expert on Windows API. I noticed that for Vista, those 2 missing functions in kernel32.dll were originally implemented in psapi.dll. Microsoft moved them to kernel32.dll in Windows 7 and since that OS onwards, the versions in psapi.dll just redirect to real functions in kernel32.dll which Rufus tries to use, but they don't exist on Vista, so we can use the ones in psapi.dll.

Those are usually triggered when some UWP app is opened or something else triggers UWP window framework. They shouldn't keep popping up if you don't close them. Nah, Aero Glass doesn't spy on its users. Strange. Previous experimental version did that on my system as well, but the latest one is fine.

Here's the patch that can be applied on top of latest revision of Rufus source code to restore Windows Vista support and the executable (version 3.1.1321). I experienced strange bug where the code that is supposed to prevent DLL sideloading causes UI elements to lose Aero look and reverts to classic, so the patch removes that. The mitigation doesn't have much value IMO, if you're worried that the program will load malicious DLL from its folder, then you've already lost by getting such DLL in the first place. Still, the relation between UI look and the DLL search directory is odd. At first I thought older version of comctl32.dll was getting loaded, that's not the case though. Calling SetDefaultDllDirectories does it. It's one of those functions that were added with an update and the issue appears to be exclusive to Vista. Unless there's a separate update addressing that.

The exact cause of the issue remains unknown at the moment. I was able to make 3 test posts in Windows 10 section, but the problem returned quickly. We know there's something strange going on with the site and that it's not related to user permissions.

No biggie. The name is very similar to how they name their sound cards. You're maintaining quite a list there, GJ!

Tried with a different browser, still no dice. Good point, I tried posting in a different topic in Windows 10 section and still the same error. Strange, I didn't put any links in my post.

Sound Blaster X-Fi MB3 is not hardware, it's pure software solution. The download link is on this page. Normally, when you run it and click Next, activation prompt requiring activation key will open, except on Vista, where you get an error that Windows operating system running on this computer is not supported. But, if you attach debugger to the installer and do what I said before clicking Next, the activation prompt will open. Of course, you can't proceed from this point onwards without license key.

@WinClient5270 You misunderstood, Wolfenstein: The Old Blood works great without any workarounds. But downloading it and running it on Vista machine could become the problem when Steam drops Vista support. On an unrelated note, I could play Wolfenstein: The New Order on Windows XP x64 some time ago. This game was compiled with an older compiler and id Software's engines were historically minimalistic in Windows API usage, though recent games use more functions. Those instructions are for unrelated commercial piece of software, Creative Sound Blaster X-Fi MB3. It does use InstallShield installer. If there's a known method for those like with Windows Installers (I'm personally not aware of it), it oughtta be easier than attaching a debugger and hunting for version check routines. I know some small pieces of the puzzle when it comes to programming for Windows, so I used that knowledge to get that particular installer to cooperate.