Leaderboard

I made an unofficial update to BlackWingCat's Extended Core because the official Extended Core is not on BlackWingCat's new website, and I wanted to make an English version of Extended Core v16d because I cannot read Japanese. Most of the files in this update are from Extended Core v16a because they are the same as the ones in Extended Core v16d except for the language. Update: Made installer with SFXCAB utility. The new installer can be slipstreamed with nLite. ┌───────┐ │ Download │ └───────┘ Files not from Extended Core v16a: acpi.sys from http://blog.livedoor.jp/blackwingcat/archives/1974336.html ndis.sys from Extended Kernel v16d videoprt.sys from KB829884-v9 ntoskrnl.exe* ntkrnlmp.exe* ntkrnlpa.exe* ntkrpamp.exe* *All of these files are from Extended Core v16d but modified (see below) Modifications made to NT Kernel & System (the 4 exe files with '* next to them): All Japanese resources replaced with resources from Extended Core v16a. Exports for ExVerifySuite and RtlIntegerToUnicode added because the code for the functions was present but the exports were not added to the export table. New boot screen for Windows 2000 Professional:

You can review the code of the script here: https://greasyfork.org/en/scripts/375264-youtube-to-invidious/code When installed, you can even change a few options in the script via Greasemonkey (Userscripts ---> Preferences): About your edit: I don't know. Maybe someone else here knows?

Actually, each post is numbered, (your last post above is comment #1184077), but they are not numbered consecutively. One way to find the post number is to simply hover over the post header, either on the far right where it says "report post" or on the left where it shows the time/date that the post occurred, and then look down in your browser's status bar, or wherever it shows the link address that you are hovering over, to see the post number as the last part of the address of the link. Cheers and Regards

Microsoft bashing is... Sometimes warranted. Ignoring their own "desktop consistency guidelines" to what, try to make it look "different"? Just bad policy. Show me someone who honestly thinks desktop usability is "new and improved". I love the system Windows is based on. I've been a Windows afficionado since the time of NT. Second to none the kernel is. Dave Cutler's design was so many decades ahead of its time and is still better than any version of Unix, IMHO. But it is not open and it is starting to look like we have already seen the best it could become. Today Windows development appears to have become about hanging all kinds of things on that solid kernel and calling them operating system improvements, because "perception is reality". Truth be told, even as a software developer I don't need cloud-integration. But Windows is no longer a system for developers. Microsoft envies Apple, and that is a very, very bad thing. -Noel

The problem is in the wpsmain.dll file. If you replace it with the old version or patch xompie, then the office will start. The ksomisc.exe error still appears, but it does not affect the work, it can be deleted, without it there is no error and everything seems to work without problems.

Update about FASTEST Windows 10 builds for normal hardware / Fanless devices / NUCS / Lowend hardware (max speed and the less LAG you can get without a dedicated graphic card; yes, running Intel Graphics ) : What I say about 1507 and 1607 it's not 100% true. Best version (after 10074) it's probably 1511 (10586), it's an improved version of the previous w10 version (10240). It doesn't update WDDM to 2.1, instead of it improves wddm 2.0 and memory management. At the end you get better disk usage and smooth GUI with less lag. How can you choose the perfect W10 version for lowend hardware? Use the XP era thinking when Windows 2003 appears. You could say "it has no sense to use one build or another if the Windows Server version has the same kernel version nowadays" This is the starting point In other words, fastest w10 Workstation editions shares the kernel version with the Windows Server edition. So now I can really understand why 10074 and 10586 are faster builds than 10240 (and also 1607 (14393), but it's the first one I started to hate because it has metro more integrated and lot of new things). Kernel version 10.0.X and Server build equivalence 10074 = Server 2016 Technical Preview 2 ( ¿Windows 8.2? ) 10240 = NO SERVER VERSION ONLY WORKSTATION RTM ( slow + lot of hdd usage in RTM ) Threshold 1 10514 = Server 2016 Technical Preview 3 ( I didn't tried this build but as you can see this server build is really close to 10586 TP4 ) 10586 = Server 2016 Technical Preview 4 ( Windows 10 November update Workstation kernel version ¿Windows 8.2 SP1: Service Pack 1? ) Threshold 2 14393 = Server 2016 RTM build ( After this one, all builds are pure Windows 10. You won't get Windows 7 / 8 / 8.1 experience anymore. In other words, starting with 1607 14939 builds are a little heavy for HDDs. 1607 it's not bad but some classic windows shell is removed, metro is more integrated, windows updates can't be controled as easy as in RTM and November update (10240-10586), WDDM version changed to 2.1 in this build so.. more things introduced ) WARNING To get the best speed of your CPU, make sure you have Disabled Meltdown and Spectre (these cpu protections are intended for servers running windows) _ttps://www.grc.com/inspectre.htm AND ALSO: If you want to get fast speed on every build starting with 1607 make sure you have Control Flow Guard disabled too (kernel 14393) ___ Another interesting script I recommend which really make a difference: Remove-Windows10-Bloat.bat by matthewjberger https://gist.github.com/matthewjberger/2f4295887d6cb5738fa34e597f457b7f ___ And also, what REALLY makes a difference it's DWS (Destroy Windows Spying) utility ___ + Windows Defender & Cortana disabled of course. ___ Get rid of Metro UI: rename c:\windows\system32\windows.immersiveshell.serviceprovider.dll to anything else and restart explorer.exe NOTE: All this information is based on a personal appreciation after I tried Windows 10 different versions on 4 different hardware configurations and compared the experience to the same hw running w7,w8 and w8.1