So if I read this right. At Present 22000.1 seems to be the foundation build that all other builds are built on (But has a heap of bugs) 22000.9 is the build that is the first that has enough stability to be rolled out to Manufacturers (OEM/RTM) 22000.176 is the current build with minor bug fixes/features (and wont be the last) Would that be right
Yeah I installed the current latest beta CU on my AMD system and the L3 cache issue is 100% there... I hope they fix it this week... I havent done tons of high performance things on my desktop lately so its been fine for light gaming/browsing but I want it fixed 100% soon. I dont feel like wiping it again and installing 10 if I can avoid it so I am going to wait it out a small bit and see if they fix it.
Do Not count on a New Build on Thursday maybe Not Friday. Microsoft was closed on Monday for Labor Day in USA. So stuff has been pushed back.
yeah I popped it in a vm with internet disabled and it showed 22000.9 still on winver and cmd ver. It was likely the oobe update download. This cumulative is tiny. There's no way it updates anywhere near the .176 mark.
I showed exactly what happened connected and disconnected. Considering that the updating to the latest build when connected during oobe was observed and reported weeks ago my question was a rhetorical one.
For Dev users yes. But for beta they already have two builds compiled 270 and 271. I think we are getting something this week
The problem with making predictions like that so close to release is that they almost certainly have already been working on that cumulative build on ms corp only for a while now. As they notice problems in builds like .176, those issues could also exist in .270 so they might need to recompile with code changes and start re-testing. For example, say that issue where they were trying to insert ads in the taskbar and it was causing taskbars to completely fail requiring people to delete registry keys. Say that coding still existed in 270, well they might need to fix that and make a new build. Repeat this for every major error they discover and you could be way off on predictions by the time the thing actually gets signed off on.