jumper

Claims to support Windows 98, but doesn't. So can we believe any of their claims?

I searched for and was able to find Dxva2_x86.dll : + 5.00 (vista_rtm_wcp.061020-1904) + 6.0.6001.17014 (lh_tools_devdiv_wpf.071009-1043) + 6.0.6002.18084 (vistasp2_gdr.090806-2338) and Evr_X86.dll: + 5.00 (vista_rtm_wcp.061020-1904) + 6.0.6001.17014 (lh_tools_devdiv_wpf.071009-1043) Without needing to add any new Kexstubs definitions, I can load all three versions of Dxva2_x86.dll in any KernelEx compatibility mode. The two Evr_X86.dll versions both load in all 9x modes, but call NtOpenFile (no functioning definition yet) in all NT modes.

If from day one they used Win9x-compatible tools, yes, it would have worked on Window 9x out-of-the-box. But it also would have been a much more difficult project to include all the features that are now built into later versions of the Windows OS and API. I haven't looked at the source code, but they are either writing directly to the Windows NT API or to a framework with API requirements. Either way, changing to a Win9x-compatibile compiler now is not an option. I haven't written a 16-bit Windows app in ages, but I still could. Trying to get it to have the look and feel of a "modern" 32-bit app would be a lot of work, however. Design choices usually involve trade-offs and difficult decisions. Using modern tools for new projects is completely understandable.

Are there any special CPU/GPU or other hardware requirements? Suggested API definitions to try (and what also needs them): >[Advapi32.dll] >GetTraceEnableFlags= GetTraceEnableFlags=z1 ;Chrome.dll >GetTraceEnableLevel= GetTraceEnableLevel=z1 ;Chrome.dll >GetTraceLoggerHandle=t1e GetTraceLoggerHandle=z1e0 >RegisterTraceGuidsW=f8 RegisterTraceGuidsW=z8 ;Chrome.dll >TraceEvent=f3 TraceEvent=z2 ;IE7 >UnregisterTraceGuids=f2 UnregisterTraceGuids=z1 ;Chrome.dll >[NTDLL.DLL] >RtlInitUnicodeString=>Kstub822:IAS RtlInitAnsiString=>Kstub822:IAS RtlInitUnicodeString=>Kstub822:IUS Note: 'f' and 'z' are functionally the same, but I use 'z' for ERROR_SUCCESS instead of 'f' which looks like "fail"

The power supply is dead. I've had several go out over the years. If any case fans are attached directly to the psu, that would confirm it.

In case anyone donated and hasn't heard yet: CNN - 6 days ago: Kickstarter, a popular crowd-funding site, urged users to change their passwords Saturday after announcing that it had been hacked.

> It was my CPU. No, it wasn't. Flash 9.0.47 doesn't need SSE/PentiumIII to work correctly. Now that O9+F9 works, please update from SP6 to SP6a. If that breaks it again, then SP6a is somehow at fault. If it continues to work, then Virtual PC 2007 must have been at fault.

Please try "-v -a6000" ala post #5 or create a batch file to do the same.

Looks good. If you've rebooted, then the original Shell32.dll should be unlocked. You can now replace it and restore the original KnownDlls setting, or leave things as they are. I suggest leaving things as they are.

Try replacing googleupdate.exe with a folder of the same name.

VLC is hiding the actual error which could be almost anything. Try running VLC within a debugger such as FineSSE to see the raw Exception errors.

The sequence should be: - have disaster recovery plan ready - Rename the 2K Shell32 to Shell2k.dll - Add it to the system folder. - Edit KnownDLLs in registry to redirect: Shell32 -> Shell2k.dll (need to be admin to edit registry?) - reboot - If system reboots okay, Shell32.dll will no longer be in use and can be renamed >How do I use NT permissions in NT 4.0? I really don't know. I think you need to be logged in as an administrator.

"CORE.INI [ApiConfigurations] default=2" doesn't disable KernelEx, but sets the default to Win98+enhancements. Set this registry value to disable KernelEx by default on apps not explicitly set to another mode: Kex+.reg------------REGEDIT4[HKEY_LOCAL_MACHINE\Software\KernelEx]"DisableExtensions"=dword:00000001And to enable: Kex-.reg------------REGEDIT4[HKEY_LOCAL_MACHINE\Software\KernelEx]"DisableExtensions"=dword:00000000

In the "Details..."->"Add..." editbox, try: http://ad.doubleclick.net/* or http://*.doubleclick.net/* Or in your HOSTS file add the line: "0 ad.doubleclick.net"

> I gave it a go on Win 98Se with KernelEx and the result is build failure. KernelEx told it you were running it on Windows 2000. Do one of these: 1) uninstall KernelEx or 2) In %windir%\KernelEx\CORE.INI, set [ApiConfigurations] default=2 then reboot and try again.

Google "portable projector mp4" e.g. £147.99 for the "Philips PicoPix Pocket Projector with MP4 Playback" + "Large image projection: up to 80" in size" + should be bright enough at 32-40" Or use a surplus Overhead Projector (no bulb needed) with a phone/tablet/laptop lying on the transparency surface. Google "overhead projector" (then add "diy") If you have (or can get) a full-page Fresnel lens, try this: The DIY $5 Laptop, Tablet, or Smartphone Home Theater Projector

> (besides the nice way it is written , it is also a very good introduction to .inf files and driver installation ) -1 - Silly, insulting fluff aside, it's nothing but another dry reference sheet. Look elsewhere for a good intro.

I've been using ImportPatcher.38 for a long time now, so it's about time I post it! The key new feature is that Delay-load processing is now optional and defaults to OFF. For Kexstubs users, this is great for finding the minimal set of missing APIs that need to be added to stubs.ini (or Kstub822.ini). New in ImportPatcher.38: March 29, 2013Delay-load processing made optionalAdded file and data alignment checkingOrdinal import fields reversed in log (to match strings)ImportPatcher.39 is a first attempt at moving towards a dialog interface. The initial MessageBox is replaced by a dialog box and the settings section of the #.ini file has been removed. (In fact, if there are no missing dependencies, the #.ini file will be empty!) The log file output remains the same. New in ImportPatcher.39: July 6, 2013Expanded first MessageBox into fuller DialogBoxAdded export forward patching

In the <GUIConfigs> section of config.xml are: AppPosition "x""y"RememberLastSessionContents either yes or n.

If you don't have a mobile phone and your land line number is not unlisted, I see no reason not to use the land line number. A land line number won't help you regain access to a lost Google account, but then neither will not providing any number. It will get rid of the nags and can be updated if you do get a mobile phone someday. Also, a "Recovery" and/or"Alternate" email address can be set up to (regain) access (to) your account. Since Google is my ISP, I was getting nagged several times a week last year and almost broke down and used my land line number. But then sometime last Fall the nags mysteriously ceased; I don't know why. Maybe when iGoogle went away?

> instead of shell32.vlc, I could type shfolder.dll and that may work. Good idea. But "shfolder.dll" is one character too long so make a local copy and name it shfolder.00 or shfoldr.dll BTW, I think this version of VLC requires SSE. FineSSE might work if you only have a PII. (FineSSE should work on NT, but hasn't been tested yet.)

If you mean the Booting in Safe Mode warning, that is best diabled in MSDOS.SYS with: [Options]BootWarn=0

Image display dimensions are not explicitly set in the html, but in style sheets: <img itemprop="image" src="http://www.msfn.org/board/uploads/profile/photo-301491.png?_r=1382667300" class="ipsUserPhoto ipsUserPhoto_large">If the style sheets are blocked or user-modified, or content from some domains is blocked, Firefox could be defaulting to collapsing the missing or blocked images.Note: I'm not seeing this problem in FF2.0.0.20 even with ABP enabled.

> My system.dat and user.dat file are 14,993 kb and 3,525 kb in size, and I have no policy.pol file. My System.dat is 3629KB, User.dat is 521KB; Install date: 3/26/2003. No policy.pol. Simple and easy Registry things to do: run Scandisk to ensure HDD is okayrun Scanregw and create a good backupwhile (errors) { fix them, reboot, run Scanregw again }run RegClean 4.1awhile (errors) { fix them and run RegClean again }exit (or reboot) to DOSrun "scanreg /fix /opt"reboot into Windowsif (registry is still too large or other problems persist) { uninstall anything you can do without restart these steps at the top!}If you are still experiencing system problems after these simple steps...well, we'll cross that bridge if we come to it.

The manual, D945PWM_TechProdSpec.pdf, mentions a few things about the BIOS setup: Try disabling these (in the Advanced submenu?): - PCI Autoconfiguration - System Management BIOS (SMBIOS) - any unused on-board features There are also 12 BIOS versions available at the Intel D945PWM download site:Latest is "108".