Weird build dates between Longhorn pre and post reset

Discuss Windows Vista/Server 2008 to Windows 10.
Post Reply
lopper
Posts: 154
Joined: Sat Apr 06, 2013 7:05 am

Weird build dates between Longhorn pre and post reset

Post by lopper »

I was looking through the wiki and found these build strings:
Longhorn Build 4093.private/wcp_dev_platform(danmohr).041116-2208
Longhorn Build 4094.private/Lab06_dev_platform(snaptab).041124-0021
Both are built in mid-late Nov 2004, which is strange, considering that 4093 winmain was built on 19 Aug.

Also, the first reset build was compiled on the same day as 4093 winmain. By Nov 2004, Longhorn Build 5012.winmain.041102-1650 was there, showing that the reset was well underway.

Why were those two pre-reset builds compiled so late? Do we even have proof that they exist? And what about the Skatari 4094, what proof do we have that shows that that build exists?

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

coz at the time of 4074 being presented they already discussed
and did what Allchin called "plan B", and that was reset

at the time of 409x builds, new "omega" builds were underway
why were 409x been done, who knows, maybe to test some parts of code that can be
ported over

hounsell

Re: Weird build dates between Longhorn pre and post reset

Post by hounsell »

lopper wrote:Do we even have proof that they exist?
99% of Longhorn build lists are utterly worthless.

The problem is people started taking any file version that began with 6.0 and claiming it was a Longhorn build. The wcp one is almost certainly one such build - it probably came from a WCP/.NET beta.

Definitely good examples are all the ones listed as Milestone 1 or 2 on the wiki. You will actually find they are all driver versions - and that they almost all even exist in Windows Server 2003 RTM, just with the major/minor changed to 5.2.

It's my intention to go through the Longhorn builds and get a more complete, reliable story on BuildFeed. But it's not really there atm, just a few unleaked builds and all the leaked ones.

lopper
Posts: 154
Joined: Sat Apr 06, 2013 7:05 am

Re: Weird build dates between Longhorn pre and post reset

Post by lopper »

I understand how you get the latest build numbers for Windows 10 and Windows 8/8.1 on Buildfeed, but how do you get those for Longhorn? Is it just hearsay?
Offtopic Comment
Also hounsell, I'd like to apologise for being a shithead, posting memes and being a general ass about 2 years ago.

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

I'm interested in that answer too

also if the famous xBetas build that was removed is indeed one among 1st
compiled build, then from that one files should be taken and compared

grabberslasher says

M1 were merely builds after XP... this could only reference to Server jump
M2 prototyping, MS does it in Macromedia Director suite, ("Longhorn Days" flash ????)
M3 36xx builds

then again that makes no sense, why would you start M1 with nothing
same goes with M2

hounsell

Re: Weird build dates between Longhorn pre and post reset

Post by hounsell »

Pre-Win10 on Buildfeed is based on evidence. It's mostly leaked builds at the moment.

There's a few nonsense ones that have snook on there in the early days - I remove them when I see them. I do intend to review them more fully though and flesh them out. I'll also link to the evidence in the Source Details where possible.

I do believe Longhorn was in active development some time before 3683, but it's difficult to say when, exactly. One complicating factor is that some pre-existing projects were merged into the Longhorn project and had been going on independently for years prior. The installer system is one example. The WIM format we first see in 4001 is already carrying an internal version of 7. It was descended from an internal project that was already in development possibly years before Longhorn got into full swing.

lopper
Posts: 154
Joined: Sat Apr 06, 2013 7:05 am

Re: Weird build dates between Longhorn pre and post reset

Post by lopper »

What do you mean by internal version of 7?

ovctvct
Posts: 1058
Joined: Fri Apr 25, 2014 6:19 pm

Re: Weird build dates between Longhorn pre and post reset

Post by ovctvct »

wasabilee wrote:I'm interested in that answer too

also if the famous xBetas build that was removed is indeed one among 1st
compiled build, then from that one files should be taken and compared

grabberslasher says

M1 were merely builds after XP... this could only reference to Server jump
M2 prototyping, MS does it in Macromedia Director suite, ("Longhorn Days" flash ????)
M3 36xx builds

then again that makes no sense, why would you start M1 with nothing
same goes with M2
Actually, M3 begun from 3683, early 36xx is(?) M2, and 35xx is (maybe) M1, but due to the Srv 03 build number, 35xx might be based on an early 35xx net. server build, not XP. Anyway, the weird compile date can be made for MS to "recycle" code, or preserve it for other purposes, you can't tell for sure exactly why. As for file versions, many of them can be LH/Vista post-reset builds, e.g. the version of IE7 B2 preview is 6.0.5296, so it's almost certainly from that build.

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

*cough* to quote someone much much reliable than you
M3: Builds 3663-3718
and again for M1 and M2, same reliable source, is written above

hounsell

Re: Weird build dates between Longhorn pre and post reset

Post by hounsell »

lopper wrote:What do you mean by internal version of 7?
I mean it has a file version field (so different versions of ximage can check quickly if they're compatible) and the 4001 wim is already version 7. It stands to reason there were plenty of versions prior.
wasabilee wrote:*cough* to quote someone much much reliable than you
M3: Builds 3663-3718
and again for M1 and M2, same reliable source, is written above
I wouldn't consider everything grabberslasher says to be incontrovertible. Where, for example, do you fit in the post-3718 37xx builds, since 4000/4001 was the first Milestone 4 build. Obviously Milestone 3, but grabberslasher hasn't accounted for them at all.

When it comes to pre-3683, everyone I know of is pretty hazy. When it comes to pre-3663, I've never head anything remotely solid, with the possible exception of the fact that there were builds before it.

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

well I'd take his word over ovc... as he was around circles years ago
some precious builds leaked and obviously he still has some

as for 3663 and below, I guess xBetas is the only community or... was
that could know, from what I saw they only round up when new OS is comming
and no LH question was answered on their boards

and if LH was indeed started from gathering pieces of technologies
(as sidebar, .net, wim...etc, was all separate projects)
then it can as easy be it started with 5.2.35xx

its just bit sad that NOBODY from MS wants to give us some info

lopper
Posts: 154
Joined: Sat Apr 06, 2013 7:05 am

Re: Weird build dates between Longhorn pre and post reset

Post by lopper »

xBetas appears to be dead. Their forum is pretty much screwed with spammers right now, and it is impossible to sign up atm (weird captcha)

DiskingRound
User avatar
Posts: 1534
Joined: Thu May 01, 2014 10:26 pm
Location: Inside the space between . and I

Re: Weird build dates between Longhorn pre and post reset

Post by DiskingRound »

Has anyone tried contacting Microsoft, or people who were involved in the Longhorn build process, about this? If they even respond at all, they may be able to tell us something about early Longhorns.

Windows OS
User avatar
Posts: 455
Joined: Tue Jul 08, 2014 9:43 pm
Location: DLL Hell, United States
Contact:

Re: Weird build dates between Longhorn pre and post reset

Post by Windows OS »

DiskingRound wrote:Has anyone tried contacting Microsoft, or people who were involved in the Longhorn build process, about this? If they even respond at all, they may be able to tell us something about early Longhorns.
It would be pretty interesting what their response would be. Would they tell everything, or will they go all Vince Perri on us?
Offtopic Comment
For those who don't know, Vince ran the company who developed Action 52, and when asked about the development via email, he said this:
Vince Perri wrote:WHY DO YOU WANT TO KNOW THIS?
Do Not Make Illegal Copies Of This Signature.
YouTube | Twitter | BA Wiki | BetaWiki

Melcher
User avatar
Donator
Posts: 695
Joined: Tue Apr 26, 2011 1:36 pm
Contact:

Re: Weird build dates between Longhorn pre and post reset

Post by Melcher »

hounsell wrote: I do believe Longhorn was in active development some time before 3683, but it's difficult to say when, exactly. One complicating factor is that some pre-existing projects were merged into the Longhorn project and had been going on independently for years prior. The installer system is one example. The WIM format we first see in 4001 is already carrying an internal version of 7. It was descended from an internal project that was already in development possibly years before Longhorn got into full swing.
For the people that would like to know more about this very early WIM format, have a look here: https://docs.google.com/document/d/1J6z ... sp=sharing. Furthermore, I can say without doubt that Longhorn developement was already in progress at least in December 2001. I found the following snippet in some Microsoft documents dating from that month and year:
Introducing the ability to search for files into the places where people need it. An obvious example of this is the File Open dialog. This is implemented today in longhorn but there’s no UI. Our team could help with the UI/code needed to make this accessible (...)

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

now finally things gets interesting
which brings me to just 1 question

does anyone know if LH builds were compiled EVERY day ?
like blinds example, monday 4000, tuesday 4001

hounsell

Re: Weird build dates between Longhorn pre and post reset

Post by hounsell »

Not really. Plus Longhorn was far more prone to periods where things wouldn't build at all.

Evidently, because 4001 and 4093 are 9 months apart, which averages out about 1 every 3 days.

Melcher
User avatar
Donator
Posts: 695
Joined: Tue Apr 26, 2011 1:36 pm
Contact:

Re: Weird build dates between Longhorn pre and post reset

Post by Melcher »

Though main was supposed to build every day, right (excluding holidays etc)?

rthdribl
User avatar
Posts: 171
Joined: Sun Jul 15, 2012 2:05 am

Re: Weird build dates between Longhorn pre and post reset

Post by rthdribl »

Melcher wrote:Though main was supposed to build every day, right (excluding holidays etc)?
Yes, but the build might simply be unsuccessful. I read somewhere, probably on SuperSite, that there were sometimes days between any Longhorn builds, because the damn thing didn't want to work at all.
MINI NT EXECUTIVE Version 1.0
rthdribl

Overdoze
User avatar
Posts: 1762
Joined: Mon Feb 24, 2014 10:28 am
Location: Slovenia

Re: Weird build dates between Longhorn pre and post reset

Post by Overdoze »

Well that just shows how badly broken it was.
All roads lead to Neptune™

KRNL386 - my site about retro computing | My site about Windows 1.0 | My blog | 86Box Manager | LeakDB - list of PC OS warez leaks

wasabilee
Permanently Banned
Posts: 491
Joined: Tue Jun 10, 2014 5:09 pm

Re: Weird build dates between Longhorn pre and post reset

Post by wasabilee »

it was same thing with whistler
if compile comes to an error, build stops compiling
the next day they find the faults and fix, then you get new build

Post Reply