jaclaz

Another couple things that you could/should verify may be (now that you mension the "huge" 3 Gb .iso): FiLeNaMe CaSe Isolevel Actual LBA in the .iso Non-iso-hardlinked sources#1 may depend on #2 (original disks were iso-level 1 and thus FILENAMES WERE 8.3 AND ALL CAPS, while later versions were either iso-level 2 or 3, cannot remember exactly but this was a common enough issue in the good ol' days of BartPE) but it is entirely possible that pre-SP2 (or whatever) the CDROM.SYS (or again whatever) cannot read/access properly an isolevel higher than 1 or - and this comes to #3 - access particularly "high" LBA's in the .iso/CD, and the same could happen with hardlinks (because the actual file is beyond a given LBA). Still in the good ol' times there was a file (a script for mkisofs) pe_sort.txt to put "key files" in a given order (which was to make PE's boot faster) but that you might find of use, see: http://www.911cd.net/forums//index.php?showtopic=6869 which later proved to be of good use for sticks too: http://reboot.pro/topic/6041-a-question-or-advice-for-pe-builds/ Just ideas... jaclaz

Well, it greatly depends on the use of a computer. My desktop is almost never booted or rebooted, as an example, and when for some reasons I switch it off, what I normally do next time/next day is to switch it on and go get a coffee while it boots (not that it is anyway slow, but in my use booting time is totally irrelevant). Still my laptop is booted only "on demand" and when I use it I want to have it booted up quickly, in this case it becomes a task, and should it be slow booting it would be very annoying. The statement by Terry Myerson, however, makes no sense anyway. The update is for Windows 10 so it might be of relevance the faster booting when compared to the previous Windows 10 version. I mean, it is entirely possible that 10240 was ALREADY(say) 29% faster than 7 on the same device , which would leave less than 1% bettering for 10586 release (nothing to be actually worth a mention, let alone as first one of the three main points listed) and conversely, provided that any machine actually suited for Windows 7 or later normally boots (or should boot) in under a minute, it would mean *like* half a second in practice. jaclaz

I guess we got the idea by now . Maybe you could ask the moderators to have a thread made sticky containing your "alternate tip" , however, and just for the record: Hello, i'm here now to post my suggestions in relationship of this topic and what i have quoted about my message. I suggest simply of DO NOT USE THIS "TIP" for speeding up the boot of windows, cause it is simply useless, and you will go in my same situation, with this folder that's updating with those useless files that windows 7 DO NOT NEEDS, making only extra hard disk works after the user login. You also WILL NOT find any solution in the web for return back: those damned files will be refreshed every reboot, growing in fragmentation. If you need to speed up your boot process, simply open the Command Prompt and paste this: rundll32.exe advapi32.dll,ProcessIdleTasks Wait some minutes, and Windows wil do EXACTLY the same thing, without the creation of those stupid files. I have now reformatted my system after 3 days of hard work for transfering all my personal settings and software to a brand new pc, and i will NEVER use another time this called ^TIP^. Bye all. jaclaz

Well Gabe Aul referencing Terry Myerson is not like "news". BUT on the actual Terry Myerson blog there is a statement that sounds a bit (too much) bold: http://blogs.windows.com/windowsexperience/2015/11/12/first-major-update-for-windows-10-available-today/?OCID=WIP_r_Body_Blog_LatestPC Queer that you didn't notice such a dramatic speed improvement. jaclaz

Yes and no, meaning that it is likely that there is still some "lying mechanism" about OS versions JFYI: http://blogs.msdn.com/b/cjacks/archive/2009/05/06/why-custom-actions-get-a-windows-vista-version-lie-on-windows-7.aspx and: http://blogs.msdn.com/b/ieinternals/archive/2014/02/19/internet-explorer-and-everybody-else-version-lies-for-compatibility.aspx jaclaz

Well, to be fair OP didn't mention having cleaned the contacts properly. jaclaz

Or you could change the install approach, in the mentioned sub-forum there are quite a bit of alternative ways, namely I would guess that a machine with an USB CD drive is likely to have more than 1 Gb of RAM so this way would do: http://www.msfn.org/board/topic/137714-install-xp-from-a-ram-loaded-iso-image/ but also this way (adapted): http://www.msfn.org/board/topic/121446-install-from-usb-after-booting-with-pe/ could work fine. As a side note - and notwithstanding what you might find here and there - the "right" way (IMNSHO) to install an XP (but actually almost *any* OS) is to do so from a "local" (saved on the machine internal storage) copy because before or later it may happen that you need a file from the original install source (because of WFP or for some other reason) and then the OS will ask you to insert the install CD (unless you installed from local source, keeping a copy) and in your case not only you will have lost or misplaced the actual CD disk but you will also have not handy the USB CD drive . JFYI, there is a thread: http://www.msfn.org/board/topic/121502-install-xp-from-usb-the-easy-way-with-sandisk-cruzer/ about abusing of some USB stick controllers (the ones with two LUNs, 1 CD+1 HD like, often labeled "U3") to install from the CD like device, but it is entirely possible that the tests and reports were relative to SP3 source . jaclaz

... and: http://www.matem.unam.mx/~micho/warp.html which may also be of interest to our 9x/Me friends: jaclaz

Yes. http://homepage.ntlworld.com/jonathan.deboynepollard/FGA/questions-with-yes-or-no-answers.html Some info for those not wanting to search half the internet to find references to what you probably are referring (vaguely) to: http://www.techrepublic.com/article/os2-blue-lion-to-be-the-next-distro-of-the-28-year-old-os/ http://www.arcanoae.com/ jaclaz

jaclaz

Sounds like in theory - and hypothetically (of course) - the good AV guys are giving you a full load of bullsh*t . JFYI: https://www.globalsign.com/en/blog/microsoft-announces-updates-sha-1-code-signing-policy/ http://social.technet.microsoft.com/wiki/contents/articles/32288.windows-enforcement-of-authenticode-code-signing-and-timestamping.aspx jaclaz

Maybe you are a son of a lesser God Check nasty surprise #3 here: http://www.forbes.com/sites/gordonkelly/2015/11/14/microsoft-windows-10-threshold-2-problems/ http://www.forbes.com/sites/gordonkelly/2015/11/14/microsoft-windows-10-threshold-2-problems/2/ jaclaz

@JorgeA ... and more or less it amounts to an "open microphone", everyone's conversations within - say - 10 meter radius from an innocent child playing with the doll might be catched and transmitted to the servers... ... and what about this? http://arstechnica.com/tech-policy/2015/11/beware-of-ads-that-use-inaudible-sound-to-link-your-phone-tv-tablet-and-pc/ ... the thingy could well become part of the "cross-device tracking" network. jaclaz

Welcome to the wonderful world of UAC prompts . I mean not really-really "news" Possible way outs: https://www.raymond.cc/blog/task-scheduler-bypass-uac-prompt/ But possibly you might want to try other approaches, using the one or the other version of elevate.exe (or similar) or some other ways, start reading from around here: http://www.msfn.org/board/topic/173265-formatting-an-external-drive-using-different-interfaces/?p=1092535 jaclaz

And it is more common than not (JFYI) it is a cross-OS bug: https://bugs.launchpad.net/unity/+bug/1463112 jaclaz

... meanwhile in Mattel ... http://kernelmag.dailydot.com/issue-sections/features-issue-sections/15018/hello-barbie-security-concerns/ The only senceful use of that thingy might be as a replacement (non-) clay target for skeet practicing.... PULL! jaclaz

Are we supposed to know WHICH 7z file you downloaded from WHERE? jaclaz

Setup /p, the undocumented switch. JFYI : http://www.msfn.org/board/topic/45373-new-ideas-for-the-next-version/?p=321473 Actually documented: https://support.microsoft.com/en-us/kb/186111 You want to try setup /p i ... but maybe (who knows) even j might work in your case jaclaz

As expected Though the new twist to an indirect reference is interesting. And at least the Mark Avrelius fake name shows some more creativity/is less evident than (say) Rodneylex45 or BaldricRace666 jaclaz

No. it's difficult to follow you this way. You should try running the batch I posted and post the whole thing, including the command line you issued. What you posted seems like an error because the system cannot find the file mynicecmd.cmd, it is possible that you need to specify a full path to it, *like*: @dencorso Maybe you could split the topic to a new thread, so that we can go through assisting simonetaddei without taking the original trustedinstaller topic too OT. jaclaz

A lot of time has passed, and I don't think (or cannot remember) if attempts with RTM were made at the time, but the whole stuff began (and was tested) with XP SP1 AFAICR. Anyway the "right" place is here (if you are talking of "installing from"): http://www.msfn.org/board/forum/157-install-windows-from-usb/ and this thread (if you are talking of "install to"): http://www.911cd.net/forums//index.php?showtopic=14181 It is entirely possible that the USB CD-ROM is a "special case" though all experiments/tests/whatever were made with USB sticks or hard disks but -at least on this topic of USB booting - the good MS guys have often published deceiving, incorrect or plainly wrong information, so I wouldn't trust them at face value. The BSOD 0x0000007b is often generated by the "standard" NTDETECT.COM on some systems, so using the "modified" one is always suggested, see also FAQ #3 here: http://www.msfn.org/board/topic/116766-faqs/ jaclaz

There are a few imprecisions in the definitions you used. A clone is a clone, i.e. something usually NOT DISTINGUISHABLE from the original. The sheer moment that you change a single byte, that will NOT be a "clone" anymore. If you use an actual "cloning" program the result will (or should be) invariably an EXACT copy of the original. What you actually want to do is to make a (working, bootable) copy of the original disk CONTENTS, with some optimizations (i.e. alignment, resizing, etc.). The "cloning" operation needs NOT to be preceded by any form of "cleaning" since what happens during a "real cloning" operation is the integral copy of every single sector from the source to the target, thus overwriting *whatever* was in the target before, while since what you will be doing is not "cloning" - though it is not actually "necessary" - "cleaning" the target (at least partially) before copying the contents of the source to it may be a good idea. Another thing that you may not be fully accurate on is the sector size/alignement. A "normal" hard disk (or other mass storage device) will have INTERNALLY 512 bytes sector and will expose 512 bytes externally. A "native 4Kb hard disk will have internally 4 kb sectors and will esxpose 4 kb externally. A "advanced format" device will have internally 4 Kb sectors BUT expose 512 bytes externally. You can do "nothing" regarding "Advanced Format", a device either uses 512 bytes sector externally ("traditional" or "advanced format") or it uses 4 kb sectors externally ("native 4kb"), *whatever* is used internally is what the manufacturer implemented and you cannot change it. Alignment is ONLY the offset at which the filesystem starts. Traditionally alignment was on the device geometry, i.e. typically with multiples of 63 sectors before, starting with Vista alignment was set to multiples of 4 Kb, which was said (on "normal", "modern" hard disks) to provide some faster operations. A SSD is not much different, it internally uses 4 kb sectors (maybe) but externally exposes 512 bytes sectors, the difference is that here it is strongly recommended to align to 4kb because the operations will definitely be faster. BUT be VERY aware that XP (specifically Disk Manager in XP) should NEVER be used on disks containing 4 Kb aligned Extended partition/Logical volumes, see here why: http://reboot.pro/topic/9897-vistawin7-versus-xp-partitioning-issue/ The Windows NT letter assignment (and conversely a large part of the way the OS boots and accesses files) depends on the starting offset of the partition. Drive letters are encoded in the Registry coupled with Disk Signature and Partition Starting Offset, see (only seemingly unrelated): http://www.911cd.net/forums//index.php?showtopic=19663&hl= This info - if absent - is regenerated when the OS is booted, so if you change alignment (i.e. partition start offset) you need to delete this info from the Registry before attempting to boot from the OS n the new device (or you need to manually coorect this info). Another thing that you need to know is that the Disk Signature is such an important ID for the Windows NT system that you can NEVER have two disks with the same Disk Signature connected to a same Windows OS, as one of the two will be (immediately and silently) changed to avoid a conflict with the other disk. Now that hopefully you have these things more clear, please do take some time to read this thread: http://www.msfn.org/board/topic/157634-hard-disk-cloningimaging-from-inside-windows/ that should provide you with further info. Then we will talk of the details and of possible exact procedures. jaclaz

Review this: http://www.msfn.org/board/topic/119748-howto-integrate-raid-drivers-into-txtsetupsif/ maybe you missed one of the steps. jaclaz

Are you really sure that .reg files do not support them? All this time believing that the REG_EXPAND_SZ type was about using those kind of (environment) variables.... ...did it change with 7? https://msdn.microsoft.com/en-us/library/windows/desktop/ms724884(v=vs.85).aspx Look for any existing key with an environment variable in it's value and export it to a .reg file, you will find that it will come out in .reg as "hex(2)" and the value will be in a set of hex values. and of course there are suitable converters: https://www.raymond.cc/blog/convert-windows-registry-hex-to-text/ jaclaz

Administrator is not TrustedInstaller. http://www.msfn.org/board/topic/155910-taking-back-the-registry-from-trustedinstaller/ Psexec does not provide TrustedInstaller credentials AFAICT, the mentioned devxexec.exe instead can do it,though seemingly with a slightly different set of permissions, at least this is what was reported on the thread related to the precursors of runAsTI64: http://reboot.pro/topic/17501-runassystem-and-runfromtoken/ You can try with the "original" RunasSystem and RunFromToken in the above. It is entirely possble that the good MS guys have changed *something* in Windows 10 and all the tools mentioned need to be upgraded/modified to work on it , historically Error 142 means that some dependency is not fulfilled, but specifically for devx-exec: http://developex.com/custom-software/devx-exec.html it may be caused by failed access to "WinSta0" jaclaz