jaclaz

Sure, here it is : http://reboot.pro/topic/21116-how-to-convert-emmc-hdd-from-removable-disk-to-local-disk-to-install-windows/ though I don't think you are in this case, your issue is seemingly the lack of appropriate Windows 7 drivers for the specific hardware, maybe the "generic" Intel drivers? http://h30434.www3.hp.com/t5/Notebook-Operating-System-and-Recovery/Windows-7-Drivers-for-HP-Stream-Notebook-11-d010nr-ENERGY/td-p/5019111 About the USB, besides the (good as always :)) advice by cdob to enable the drivers: http://reboot.pro/topic/14186-usb-hdd-boot-and-windows-7-sp1/#entry127772 the note was about the fact that if you manage to have USB 3 working the system will be noticeably faster, but for the moment just test the USB 2 booting from external disk. jaclaz

All these years thinking that paying to access a news article was to actually get some news, as opposed to some lame late excuses about how she was wrong in a previous non-news article (BTW likely originally also behind a paywall). She should give back some of the money ;). jaclaz

Yep , but you also failed to read the given link or the comment about the need of 0x08 hex characters (backspaces) in the batch (they cannot be posted), everyone has been (before or later) a newbie at batch scripting (or at something else), but in order to learn one needs to try and solve problems. What would you learn if I fix your time subtraction batch? Maybe you would learn more if you read and understand my time subtraction batch. (it is written in a very "linear" fashion and should be very easy to follow) Anyway the posted snippet has at least one more issue, using TASKLIST to monitor the running program *somehow* makes the line flash AND there is anyway an (intentional) typo in the (just to see if readers are paying attention ) in the :to_hs subroutine. And in the meantime I found a clever way to create the 0x08's "on the fly", so here is an updated little batch. Still you will need to find the typo in the math subroutine. So here it is: I have a couple "better" batches, but I will post them only once you will have had time to test the above and fix the small mis-calculation. jaclaz

@Tsundere Which EXACT model of the HP Stream is it? And is that 32 bit or 64 bit Windows that you are trying to install? You need anyway - as Tripredacus said - a specific driver slipstreamed, but there can be an additional issue, there was recently a case of a (Dell) thingie that needed a BIOS update to have the EMMC be seen as "fixed". BUT wait a minute, is this still the same as this: http://www.sevenforums.com/installation-setup/376325-installing-win-7-new-laptop-emmc.html and you are looking for a "second opinion"? The "external" USB approach should however work fine, still you will probably need USB 3 drivers for Windows 7. jaclaz

There are reports of these issues when one attempts to install the printer AND attempts to share it BUT there is some generic network/firewall/domain misconfiguration. See: https://social.technet.microsoft.com:443/Forums/office/en-US/4e7f2d47-8808-4acc-9c51-cc222acc3f81/when-adding-a-printer-share-within-print-management-receive-there-are-no-more-endpoints-available?forum=winserverprint Particularly see if this is your case: https://asifkhandevadi.wordpress.com/2014/04/24/ws2012-windows-could-not-share-your-printer-there-are-no-more-endpoints/ If you cannot actually start the Add Printer it may be a different issue, still doing an SFC /scannow and rechecking the mentioned services won't do any harm. There also reports of the issue being connected with the spooler service stopping itself and/or some mixup with the RPC service and RPCSS and HTTP services as dependencies, see: http://answers.microsoft.com/en-us/windows/forum/windows8_1-hardware/printer-spooler-wont-start-at-startup-windows-81/207fd102-f5cf-4881-bf2c-be5568a75087?page=3 https://www.reddit.com/r/windows/comments/166t0g/print_spooler_service_broken_wont_start/ jaclaz

I don't know , it may *sometimes* undermine confidence in the perceived uniqueness of one's vision ... jaclaz

Just to confirm that it happens here also with Opera (Presto/2.12.388 Version/12.15), should be reproducible. jaclaz

Yep, I understand your desire to not reveal your script contents, but this (respectable) choice makes it impossible for anyone to help/assist you, so your post can only be a (again perfectly understandable) form of ranting. To me it seems evident (but maybe it is just me ) from what you report that *somehow* what actually "arrives" to decrypt when sent "manually" is different from what "arrives" to it when sent through your script. What I would suspect would be that (still *somehow*) what your script is passing is different because it is (say) in a different encoding or format (think - as an example - to ASCII, Unicode or UTF) and I would use a hex editor to verify (after having dumped both the "manual" and "through script" password to file) if such a difference exists. And same goes if the procedure includes a clipboard copy/paste. At least you could describe the procedure and enumerate WHICH versions of Windows create the issue. To give you an example of what can happen (not necessarily your current issue, of course): http://bavih.blogspot.it/2008/07/notepad-bug.html jaclaz

Sure, like sneakingly removing access to more than (roughly) 3.5 Gb of memory (which happened with SP2) was "fair play" . ... and of course the AARD issue back in the good ol'times was legit ... https://en.wikipedia.org/wiki/AARD_code jaclaz

@submix8c Anyway the getTimeInAdvanceFromCrystalBall function is not available in the API unless user is "Magician" . Just in case the needed XKCD: https://xkcd.com/612/ @Kali Hmmm, the batch you posted seemingly mixes liberally HH , MM, SS and hs, you will have a number of incorrect results in many cases. And - if I get this right now - what you want is that the thingy shows time elapsed since start of the program updating it in "real time", i.e. something more like a "progress bar". This could be done in batch of course, BUT it will hog the CPU. AND you will have the whole display flashing, as the only simple way to update a command window from batch is to clear screen and re-output the new info. BUT there is a neat trick using SET /P and the backspace character (0x08) see: http://stackoverflow.com/questions/368041/how-to-code-a-spinner-for-waiting-processes-in-a-batch-file so that we can have *something* like: For the example above I used the Wait.exe available here: http://www.softtreetech.com/24x7/archive/58.htm jaclaz

I wonder why he used "months" instead of the more accurate "decades" . jaclaz

... and on the BBC too: http://www.bbc.co.uk/news/technology-36367221 jaclaz

Well, here, once saved the posted batch as mytymer.cmd, when run in a command prompt it shows: batch Program by jaclaz for Start Time and End Time of a Program execution Parameter is Missing Usage: mytimer.cmd <executable> PING -n 6 127.0.0.1 Program Start 18.32.43,74 Program End 18.32.49,05 Program Time 00.00.05,31 How different is what happens there? jaclaz

Maybe you want to change the text "jnz" to "jz" or "jmp" instead .. Seriously , what you posted is the disassembly view, you need to change the OPCODE(s) in the binary, that is the idea of the given commands: First one search and replaces a number of hex bytes: gsar -o -s:xE8:x02:xBA:x02:x00:x85:xC0:x75:x41 -r:xE8:x02:xBA:x02:x00:x31:xC0:xEB:x41 update.exe Second one re-validates the PE checksum: pechecksum -c update.exe In the -s hex the 85C0 is "test eax,eax", and 75 is "jnz", in the -r hex this becomes 31C0 for "xor eax. eax" and (you never know) the EB is "jmp" (short, unconditional), see (example): http://www.mathemainzel.info/files/x86asmref.html You will need to change the other bytes before and after in the gsar command line to match the different binary you have (or use an hex editor, maybe easier). In any case make sure and double sure that gsar finds the right occurrence/area before adding the -r vaues. jaclaz

So, you don't actually want a "stopwatch", you want a "timer" instead. This latter can easily be made in batch. The following is slightly obfuscated (so that readers may have an occasion to exercise their little gray cells): @ECHO OFF SETLOCAL ENABLEEXTENSIONS ENABLEDELAYEDEXPANSION REM Time format expected HH.MM.SS.hs 24h format SET N=batch Program by jaclaz for Start Time and End Time of a Program execution CALL :SET ECHO 1,6,11,3 CALL SET Var=42 %§%%P%=%* CLS IF NOT DEFINED PROGRAM ( ECHO %N% ECHO Parameter is Missing ECHO Usage: %~nx0 ^<executable^> SET PROGRAM=PING -n 6 127.0.0.1 ECHO !PROGRAM! ECHO. ) %§%%S%=!%T%! %?% %P% %S% !%S%! %_9% :M %S% %S% /W "" !%P%! %§%N=!%T%! %?% %P% %E% %N% %_9% :M N %§%/A D=%N%-!%S%! %?%. %§%/A F=%D%/%.%&%§%F=0!F! %§%/A A=(%D%-%F%*%.%)/%...%&%§%A=0!A! %§%/A J=(%D%-%F%*%.%-%A%*%...%)/%..%&%§%J=0!J! %§%/A H=(%D%-%F%*%.%-%A%*%...%-%J%*%..%)&%§%H=0!H! %?%. ECHO %P% %Y% %F:~-2%.%A:~-2%.%J:~-2%,%H:~-2% GOTO :EOF :M N O P Q SET /A %1=(1!%1:~0,2!-%..%)*%.%+(1!%1:~3,2!-%..%)*%...%+(1!%1:~6,2!-%..%)*%..%+1!%1:~9,2!-%..% GOTO :EOF :SET Answer=42 SET §=%0 &&SET §=!§:~1!&!§!#=%2-%4&!§!?=%1&!§!_9=%6 FOR /F "tokens=%#% delims= " %%N IN ('%?% %N%') DO ( %§%P=%%O&%§%!§:~0,1!=%%S&%§%!§:~1,1!=%%V&%§%!§:~2,1!=%%T&%§%Y=%%W %§%..=%200&%§%...=%3000&%§%.=%5%30000 ) jaclaz

And now for the usual OFF TOPIC ... ... meanwhile in Maine ... http://www.macrumors.com/2016/05/23/apple-replace-ipads-macbooks-classrooms/ I personally read in the above "tablets" (in the meaning of underpowered touch enabled devices) instead of iPads, of course. jaclaz

Two semi-random thoughts: 1) Some laptops have a keyboard combo (usually FN+*something*) that switches Audio (speakers) off. 2) It is not uncommon that the actual jack/plug connector switch does not make good contact when you remove the headphone jack jaclaz

That is NOT a "C/C++ script", it is a small C++ program SOURCE CODE, which you got from here: http://www.thecrazyprogrammer.com/2013/06/cc-program-to-create-digital-stopwatch.html Normally such a Source Code needs to be COMPILED (though there are C and possibly C++ interpreters, very probably, like 99.9999% you DO NOT actually want that). Also it seems like targeted to a "pure DOS" (not Windows). Try asking what you *need* or would like to obtain (FINAL GOAL, as opposed to the way you believe you should reach that goal), right now you are slipping on a chiocolate covered banana : http://homepage.ntlworld.com/jonathan.deboynepollard/FGA/put-down-the-chocolate-covered-banana.html jaclaz

Translated: makebt\mkisofs.exe -v <- be verbose (i.e. output more information while creating the .iso) -iso-level 4 -l -joliet-long <- a few parameters to make the inner format of the .iso correct -volid "PE3" <- the name of the .iso volume -b boot/etfsboot.com <- name of the bootsector to use -no-emul-boot <- specify that above is a NO-EMULATION bootsector image -hide boot.catalog <- optional, there is no need to have the boot.catalog visible (or hidden for that matters) -duplicates-once <- this - in case of two or more "same" files in two different paths will copy only first instance and make the rest links, think of hardlinks on NTFS -force-uppercase <- this will make all directory and file names in the .iso UPPER CASE, no matter ThE OrIgInAl CaSe On HaRdDiSk -o win7pe_x86\PE3.ISO <- this is the output .iso win7pe_x86\ISO zzz_add\x86 <- these are the folders which contents will be added to the .iso The actual PE files are inside the win7pe_x86\ISO folder and will go in the root of the .iso. The drivers are in the zzz_add\x86\Drivers and since the contents of the zzz_add\x86 directory are added they will go in \Drivers in the .iso. The folder on hard disk is called zzz_add because this way the contents will be added to the .iso AFTER the PE booting files (there might be some limitations in the booting files if they are saved into the .iso at a "too large" address). jaclaz

Well, this is "by design" in the sense that a protected mode OS does need a driver for the .iso (which came a little bit later in the game through Firadisk or WinvBlock and for some uses Imdisk besides the *need* of a booloader such as grub4dos and/or Syslinux/Memdisk, if not using the good ol' Server 2003 SP1 Ramdisk), and rest assured with Easy2boot (based on grub4dos) you can make almost *anything* boot from USB stick in a multi-boot fashion (or directly with grub4dos if you know where your towel is). Using a .iso on hard-disk-like device (such as an internal or external disk or a USB stick) is a choice which is suited to have a same object (the .iso) that can be used indifferently on the device or - if really needed - burned to CD/DVD, the "original" approach to install windows from USB used not the .iso approach, it came later as a "convenience": http://www.msfn.org/board/topic/61384-how-to-bootinstall-from-usb-key/ http://www.msfn.org/board/topic/111406-how-to-install-xp-from-usb/ The use of .iso's, particularly for the install of XP became prevalent much later in the game, see this as a reference: http://reboot.pro/topic/8944-boot-any-iso-image-or-boot-all-iso-images/ exactly because of the lack of "proper" NT drivers, whilst the "newer" OS's of the NT families provided through the use of a "real" PE as a setup environment and .wim access/deploying a much more "portable" approach to installing from "any" media. but nothing prevents you from creating a "flat" install build manually (as the mentioned USB_multiboot set of batches do), the issues you had with WinSetupFromUSB are expected when using a modified source because obviously the tool has been tested only with unmodified or very little modified sources. Still, on a multiboot stick it would make sense to boot to another capable OS (usually a PE 3.x or 4.x which you would have anyway on such multiboot device) and install the XP from hard disk after having copied the install files to the internal disk (which remains IMHO the best approach, used since the good ol'times of Winnt.exe and Winnt32.exe, as it is usually faster and the install remains with actual install files "on board" so that you don't happen to have "please insert install CD" messages if running SFC or needing an additional driver for a device). And nowadays this can be done - besides manually - through the WinNTSetup: So, all in all it is just a matter of preferences and personal likings, but we now have a lot of possibilities . jaclaz

So, basically we need to imagine your script, then debug it? Anyway, check line 42. jaclaz

Well, the issue here is that some (intelligent or otherwise) checking/modifying of the .iso (or the exact way to mount it) is vital for these setups. JFYI, there are two specific threads for Ram disk (and also non-ramdisk) XP installing from .iso: However it is EXTREMELY queer that WinSetupFromUsb did not work (at least for the 32 bit version which as said has been really widely tested) that motherboard/UEFI must have something really "queer". The HAL issue might be due to *something* that misrepresented the needed HAL OR it could be the (anticipated) possible issue with disk numbering (which might also be connected with the tons of files missing. Was the actual file missing/unable to be found? If the latter, it might be due to an incorrect configuration in BOOT.INI that should be easy to fix. jaclaz

Yep, and - to be picky (as I am) - EasyBCD is IMNSHO also "border line" (if not downright over it) as anyway re-distributes a few MS files, beside mis- or under- documenting the grub4dos and ultradefrag that is underneath (not complying with their GPL license, i.e. not providing source, etc.), and "forgetting" the PLoP altogether. Just for the record at the time there was some bickering with the Author of the tool that completely failed to make the thingy compliant : http://reboot.pro/topic/17655-boot-into-3rd-party-efi-application-via-bcd/ nothing blatant - of course - still not the kind of quality/carefulness I would expect from a (largely Commercial) initiative. jaclaz

Yep, as said it is not easy , a lot of files went through the stupid doanload.aspx or thankyou.aspx and then possibly some through to conxion.com and they were not cached by the WaybacK Machine or - more probably - they were cached but we can't find them, most of the files for which we do have an actual download link are usually cached, but in the meantime a lot of places where such direct links were posted went down, it's a mess. Additional resource (just in case) sporting many such "direct links": http://erpman1.tripod.com/archive/wnt4upd.html jaclaz

Well, to be fair, you take something totally UNdocumented (the BIOS) that empirically and through years of experience works fine (though with a few limitations) and decide to create a completely NEW "standard" (the UEFI) senselessly complicated by having it developed by a committeee and OVER-document it, and you couple it with another NEW "standard" (the GPT) also developed by a committee, but this time UNDER-documented and then use anyway a non fully compliant implementation of both and let most motherboard manufacturers have each their own (different and not fully compliant - but differently) implementation, something is about to happen before or later. Nothing can be fully compliant with more than 2,000 - two thousands - pages of (BTW unclear and often contrasting) documentation, particularly if even each member of the committee do not respect the standard. Besides the MS or the Asus guys, the people to blame are IMNSHO the ones on UEFI.org, particularly the INTEL peeps, that pushed and helped this madness. jaclaz