Windows Threshold kernel looks to jump to 6.4, latest build is 9788

Microsoft is deep into development of the next iteration of Windows, currently called Windows Threshold, and today, we are starting to see evidence that this new version of Windows will bump the kernel to version 6.4.

The latest build number to be identified, via tracking mechanisms inside Windows applications, is build 9788 and as you can see in the image above, has a kernel bump to 6.4. The bump in version number is not all that surprising, considering that Windows Threshold is shaping up to be another major release for Microsoft.

We have already uncovered that the modern UI will be disabled by default for desktop users and that on small mobile devices, namely tablets and phones, the desktop will disabled. In addition, modern apps will now be windowed on the desktop and that the start menu will be making a return too, so a bump to 6.4 seems warranted.

Prior to the 6.4 update, there have been quite a few Threshold builds spotted in metadata previously; Threshold is said to have a preview arriving later this year but the final release will not come until the Spring of 2015.

Microsoft is still working an Update 2, which is said to arrive in August but so far, details have been slim on what the update will include. Seeing that major features have not been uncovered yet and the platform has either hit, or is about to hit RTM, we are next expecting much for this update.

Source: @stealth2013

Report a problem with article
Previous Story

Nexus 8 spotted in Indian import logs

Next Story

Japan Airlines will offer charging packs for first-class passengers on domestic flights

52 Comments

Commenting is disabled on this article.

Is the current erratic and somewhat unreliable Windows Update mechanism within Windows-7 going to be fixed? A system that recognizes a problem/missing or corrupt file, but can't identify what is the problem/missing or corrupt file, is hardly complete. It certainly is not worthy of what used to be Microsoft's high standards.
As for the current default Metro UI...it appears that Microsoft has seen the error of their ways. It will be interesting to see what ends with Windows-9. A choice of UI to use???

TsarNikky said,
As for the current default Metro UI...it appears that Microsoft has seen the error of their ways. It will be interesting to see what ends with Windows-9. A choice of UI to use???
One can only hope.

JHBrown said,
One can only hope.

Metro is still very much a part of the OS. Even in with windowed mode, and the live tile based menu, all that is still very much Metro. The Store and the new universal apps aren't going anywhere.

Nope, win9 with current build nos. 97xx will be Metro all the way. If you don't like it, use WinThreshold with Build nos. 9600.2****.

FaiKee said,
Nope, win9 with current build nos. 97xx will be Metro all the way. If you don't like it, use WinThreshold with Build nos. 9600.2****.

How much Metro will be in it?

Dot Matrix said,

How much Metro will be in it?

No idea ATM, I only heard that "win9 is the continuous evolution of Windows" and "Threshold is the compromise to XP/win7 lovers". :)

FaiKee said,
No idea ATM, I only heard that "win9 is the continuous evolution of Windows" and "Threshold is the compromise to XP/win7 lovers". :)

So Win9 is real?

FaiKee said,
Why not? It was mentioned in MS job-listings and in MS guy's CV in linkedin.

Ah, yes, that's right. There's been so much rumor, that it's hard to keep track of what's "real", and what's media invention.

So if iterating to a major version is what caused driver issues between Xp > Vista, why are they still coding drivers the way they were back then?! Its gonna be 2020 and im still going to be on 6.x with some 1990s startmenu still buried in the system for all the hangers on, because just like the people coding drivers they just can't wrap their heads around change

It's just a number. Major revisions, like Windows 8's kernel, are simply named 6.2 instead of 7.0 to keep compatibility. And well, OEMs don't learn, it works now, so why would they change anything? It's wrong to think like that, but that's how it works.

Definitely! While my Win 8.1U desktop experience is already nearly completely Metro-free, it will be nice once it no longer even exists and I won't have to worry about seeing it again.

Klayer said,
What the news: a number was incremented.
Is there really so much little to talk about microsoft?

Yup, that's definitely less interesting than the article about iPhone 6 based around a photo of an unknown purple rectangular object with ruler attached.

agtsmith said,

Yup, that's definitely less interesting than the article about iPhone 6 based around a photo of an unknown purple rectangular object with ruler attached.


And yet when an iPhone from a bar was stolen, the whole world discussed it for weeks...

Raa said,

And yet when an iPhone from a bar was stolen, the whole world discussed it for weeks...

It was neither "whole world" nor "for weeks". But it's exactly this kind of rhetoric that leads to overhyped articles as I mentioned above.

What is interesting is the build numbers seen before the 6.4.x were not actually Threshold but an update to Windows 8.1. This also indicates that its possible 6.4 will be 8.2 instead of Windows 9.

The problem I see is that Microsoft knows they have had a negative perceptive ( i think thats the correct word ) of Windows 8 so the only question is will Windows 9 be free like 8.1 or paid update.

It would be only three years since Windows 8 was release come Spring 2015. I think its a good possibility that they will basically write off the upgrade fees in the consumer market in order to change the consumer perspective on Windows 8/9.

They would still get the upgrade fees from Windows 7 users. But Widows 8 users have already accepted the platform and if they bring Universal binaries to their platforms they can reward those current users.

TheGhostPhantom said,
What is interesting is the build numbers seen before the 6.4.x were not actually Threshold but an update to Windows 8.1. This also indicates that its possible 6.4 will be 8.2 instead of Windows 9.

The problem I see is that Microsoft knows they have had a negative perceptive ( i think thats the correct word ) of Windows 8 so the only question is will Windows 9 be free like 8.1 or paid update.

It would be only three years since Windows 8 was release come Spring 2015. I think its a good possibility that they will basically write off the upgrade fees in the consumer market in order to change the consumer perspective on Windows 8/9.

They would still get the upgrade fees from Windows 7 users. But Widows 8 users have already accepted the platform and if they bring Universal binaries to their platforms they can reward those current users.

I wouldn't be terribly surprised if it ends up being free for Win7 users too. They want as many as possible on the modern platform and able to participate in the Store ecosystem so developers will be able to make money and thus build apps for Windows.

Brandon Live said,

I wouldn't be terribly surprised if it ends up being free for Win7 users too. They want as many as possible on the modern platform and able to participate in the Store ecosystem so developers will be able to make money and thus build apps for Windows.

That is extremely good news, Brandon! I've used Apple, I've used Linux... But this is why I love the direction that Microsoft is headed toward. You guys are doing an amazing job up there in Redmond. Keep up the amazing work! I'm so excited about the future!!!

Because drivers hate it when the first version number changes, ending up with things like we had in Vista: major compatibility issues.

Lots of software references the major kernel version (for compatibility reasons). Changing it from 6.x would break a lot of legacy software unnecessarily.

CAP-Team said,

Why not kernel 9.0 keep the kernel version in pace with the windows version.

Bad idea ... look what happened when the kernel version was bumped from 5.1 to 6.0 for Vista.

Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

IT did change, but it wasn't designed to break all drivers. Old XP drivers should still work on Vista and higher, the version number however, caused some issues.

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

If there's no truth to it then why hasn't MS changed the major version number since Vista? I'm sure they'd love to do it but still haven't done it 3 OS versions later and we're still on kernel 6.x.

Heck, they change it on Windows Phone without issue.

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

Thats what I thought, an entire new driver model. Where even the display driver in Vista and higher can crash then recover, but so many drivers in XP can crash and bring the entire system down.

George P said,

If there's no truth to it then why hasn't MS changed the major version number since Vista? I'm sure they'd love to do it but still haven't done it 3 OS versions later and we're still on kernel 6.x.

Heck, they change it on Windows Phone without issue.

Maybe they feel that the kernel hasn't changed enough to warrant a major version increase? If they really did have an issue with going from 5 to 6 then why did they even do that? And then why not force drivers to not be reliant on the major version number to receive WHQL certification?

The whole thing just sounds made up but believable enough to catch on.

And what about Windows Phone? It uses the NT kernel, same version number as desktop Windows. Not like that matters since WP is tightly controlled. You can't install it on whatever you want and install your own drivers.

CAP-Team said,
Why not kernel 9.0 keep the kernel version in pace with the windows version.

Because the kernel is not significantly different. v7.0 will probably run a kernel similar to singularity.

McKay said,

Thats what I thought, an entire new driver model. Where even the display driver in Vista and higher can crash then recover, but so many drivers in XP can crash and bring the entire system down.

Well, ms knows that these versions are Simply Win7();Win7.1;Win7.2and upcoming Win7.3

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

Correct, it's not true. Some applications had issue with the second number (6.0) being higher than the second number (5.3). MS addressed this with backcompat hacks.

Either way, apps updated for Vista had to handle the version change and there's absolutely no reason to believe they wouldn't handle another major jump. MS just hasn't done enough since Vista to the kernel to warrant a major jump.

Rosyna said,

Correct, it's not true. Some applications had issue with the second number (6.0) being higher than the second number (5.3). MS addressed this with backcompat hacks.

Either way, apps updated for Vista had to handle the version change and there's absolutely no reason to believe they wouldn't handle another major jump. MS just hasn't done enough since Vista to the kernel to warrant a major jump.

That has absolutely, positively no truth to it whatsoever.

First of all, there isn't really a "kernel version" and a "Windows version". 6.3 is the latest file version for non-kernel files too (like, say, shell32). In Windows 7 we started out with it being bumped to 7.0. Then figured out we could make compatibility way better by making it 6.1.

There are larger changes to the kernel in Win8 than in Vista. But the major version will be staying 6 potentially forever, unless a really major *breaking* change comes along. But the whole idea is that Windows doesn't do those. At the very least, there won't be a major hardware breaking change any time soon.

So in short...
- The file version is largely meaningless and only kept at 6 for compat reasons.
- The philosophy is that it should only change if there's a *major* *breaking* change to the platform.
- There are other compat reasons for keeping it the same and incrementing the minor version: http://brandonlive.com/2009/10...hen-version-checks-go-wrong

Also note that Apple does the same thing. They've kept Mac OS at 10.x for like 14 years. That's not because they haven't made major changes to the kernel...

Brandon Live said,

Also note that Apple does the same thing. They've kept Mac OS at 10.x for like 14 years. That's not because they haven't made major changes to the kernel...

Yup, they've also successfully made people stop associating the 'X' with 10. Sure, it's still officially "oh-es-ten", but damned if the majority of users call it that. You've also got more people referring to versions by animals than by numbers.

Then there's the opposite side, with Chrome's version number skyrocketing whimsically, which inspired a whole new sect of devout believers in the mantra "version numbers don't matter".

So it's weird to see people gripe about Windows versioning.

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.

The best example of that breakage was the move from XPDM to WDDM in Windows Vista which forced many of the hardware vendors to re-write their video card drivers from scratch or at least re-write large portions of it. With that being said, if it weren't for all the driver API change and other improvements bought about in Windows Vista then Windows 7 and successors wouldn't have turned to be as successful as they are now - particularly when it comes to battery life etc.

DJGM said,
Bad idea ... look what happened when the kernel version was bumped from 5.1 to 6.0 for Vista.

All what happened to Vista was due to major OS rewrite. Maintaining compatibility with major version number isn't a problem at all. But basically saying Windows 9 is still Windows 8/9 with just a new additional application framework.

I just think they should not call it Windows 9. They should reconsider the approach and call it Windows 6.4.

coth said,

All what happened to Vista was due to major OS rewrite. Maintaining compatibility with major version number isn't a problem at all. But basically saying Windows 9 is still Windows 8/9 with just a new additional application framework.

I just think they should not call it Windows 9. They should reconsider the approach and call it Windows 6.4.

That would just be silly. We already have Windows 7 and 8 and 8.1. 6.4 would sound like it's an old version to almost everyone.

Windows 9 is a good choice. It gets away from the Windows 8 name which is tainted no matter how you feel about the OS. Even if the next update was the greatest thing ever if it is 8.2 then people will still whine because people love whining. Windows 7 for example, new name for mostly the same OS.

Brandon Live said,
Also note that Apple does the same thing. They've kept Mac OS at 10.x for like 14 years. That's not because they haven't made major changes to the kernel...

Apple bumps the major kernel version every major release. Starting with 5.1 in 10.1.1 to 13.3 in 10.9.4.

Darwin Stability-CD.local 13.3.0 Darwin Kernel Version 13.3.0: Tue Jun 3 21:27:35 PDT 2014; root:xnu-2422.110.17~1/RELEASE_X86_64 x86_64

Well if it's true that changing the major version number breaks things, I think this is something that should be addressed. I mean, this is a problem, and now that it is identified work should be done to resolve the issue. At some point Windows should be designed in such a way so that changes in the major revision number don't break things.
Otherwise, what will they do when they reach 6.9, I hope they won't start doing some of the 6.99, 6.999, 6.9999 crap, etc...

coch said,
Well if it's true that changing the major version number breaks things, I think this is something that should be addressed.

There's a great build conference talk over on channel 9 that talks about how it breaks.

It's not just drivers, but many applications break from a major version bump. This is simply because developers (outside MS) do not like to support new OSes until they've been fully tested.

So while something is perfectly able to run on the updated operating system, the developer says "if version <> 6 then don't run". This is what breaks. Trying to convince a world of developers that their code is insulated from OS changes is your first step to resolve the issue.

In a few instances, Microsoft professional services have created "shims" to alter what the OS reports to the application when these sorts of checks are made. This is why being a TAP partner really helps out corporations in the long run.


It's like getting a world of web developers to move from "if browser.useragent == blah" to CSS feature detection. It takes time.

mrp04 said,
Is this whole "driver issue" actually true? It feels like it keeps getting repeated but it doesn't really make sense and I haven't seen it given as the official answer.

Drivers broke in Vista because of the completely new driver model for lots of devices, not because a number changed from 5 to 6.


Yes, MS has mentioned the major version bump as an issue but like you, while it may contribute to problems, I think it's by far overshadowed by the driver model change (WDM) in Vista. THAT is what caused instability at least, and months of work for driver developers.

If Windows 9 would be Kernel 7.0 and use an identical driver model, any issues would be solvable with zero changes to the actual driver, and only (at most) a change to the installer so that it didn't prevent the install when it didn't need to.

Also, the installer issue is present even with the forthcoming Windows 6.4, it's just less likely to happen.

Rosyna said,

Apple bumps the major kernel version every major release. Starting with 5.1 in 10.1.1 to 13.3 in 10.9.4.

Darwin Stability-CD.local 13.3.0 Darwin Kernel Version 13.3.0: Tue Jun 3 21:27:35 PDT 2014; root:xnu-2422.110.17~1/RELEASE_X86_64 x86_64

Well, that wasn't really my point, but sure. Of course, if Apple versions their kernel separately, why can't Microsoft :-)

coch said,
Well if it's true that changing the major version number breaks things, I think this is something that should be addressed. I mean, this is a problem, and now that it is identified work should be done to resolve the issue. At some point Windows should be designed in such a way so that changes in the major revision number don't break things.
Otherwise, what will they do when they reach 6.9, I hope they won't start doing some of the 6.99, 6.999, 6.9999 crap, etc...

Err, work was done. The file/API version was decoupled from the actual product version. You can't do better without inventing:
A) A time machine.
B) A worldwide mind-control device to stop developers from writing bad version checks (and driver writers from writing possibly legitimate "did you break everything" checks).

The current solution is pretty good, and really not something to worry about. By the way, in much of the computing world, 6.9 is followed by 6.10. Apple's Yosemite is OS X 10.10, for example.

Northgrove said,

Yes, MS has mentioned the major version bump as an issue but like you, while it may contribute to problems, I think it's by far overshadowed by the driver model change (WDM) in Vista. THAT is what caused instability at least, and months of work for driver developers.

If Windows 9 would be Kernel 7.0 and use an identical driver model, any issues would be solvable with zero changes to the actual driver, and only (at most) a change to the installer so that it didn't prevent the install when it didn't need to.

Right. Instead of a lot of work vendors won't do for legacy devices, it's a little work vendors won't do for legacy device. Remember, Vista actually supported the same old driver model(s) as XP. Maybe not perfectly, and you missed out on DWM / Aero without a WDDM display driver, but it did support them, even though a huge number refused to install or load.


Also, the installer issue is present even with the forthcoming Windows 6.4, it's just less likely to happen.

Yes, but part of the idea is that developers and vendors *should* be getting *better* at this, not worse.