jaclaz

Isn't it the "normal" way for a range? I mean in the given example: -range:8808448-9236127 should mean get the bytes from offset 8808448 to offset 9236127 in file <given long url>.iso, this would make the Windows 8.1 bootmgr 9236127-8808448=427679+1=427680 bytes in size, which sounds just right. @JFX Seemingly the -out parameter works fine (i.e. the file is downloaded to the given directory and file name, but the "final" message in the GUI is misleading as it uses as path the "current directory" (where GWT.exe is). If using the GUI and asking to download the whole 8.1 Waik files, there is an error message (with no specification) "Failed to download the necessary files.". As a matter of fact some of the files are actually downloaded, cannot say if *all* of them. Maybe if you could add something a -list option, so that one knows which files are included with a given switch, it would be useful, like: GWT -Win8.1 -list \ADK5\ \ADK5\x86\ \ADK\x86\BCDBoot\bcdboot.exe ... As well (IMHO) adding a -x86 and a -x64 to limit the download to only one architecture would save a lot of bytes/time. jaclaz

Nice , is there anywhere a way to retrieve the currently used URL's (and optionally change them)? Just in case an URL becomes invalid and/or should another source become available, etc. As a side note, ideally there could be some sort of .ini file (.ini just to say "human readable") with the pre-configured URL's and ranges and the possibility to add a button like "Custom" ( with a drop-down set of checkboxes fetching the "title" from the .ini entry). jaclaz

Three years? Possibly developing two os's at a time is more terrifying than terrific. jaclaz

Yep, my bad, didn't recheck and forgot to post the actual direct link. There are seemingly 4 days in total on Wayback Machine: https://web.archive.org/web/*/http://download.microsoft.com/download/A/6/E/A6EFFC03-F035-4604-9FB0-3B8169ED6BB6/WindowsXP-KB955704-x86-ENU.exe but only the Feb 13 2013 and the August 1 2013 are good ones. @bizzibody The speeds you report are "terrible", I still suspect that there is something "wrong" with the actual stick. If you have time and will of course, it would be interesting if you could run on the stick all three of these utilities and post the results: ChipGenius, Flash drive information extractor and Chipeasy: http://www.usbdev.ru/files/chipgenius/ http://www.usbdev.ru/files/usbflashinfo/ http://www.usbdev.ru/files/chipeasy/ jaclaz

exFAT is already included among the filesystem accessible in Windows 7, (and on Vista SP1/2), the one that needs the driver to be installed is just XP: https://kb.sandisk.com/app/answers/detail/a_id/3389/~/operating-systems-that-support-the-exfat-file-system Anyway, courtesy Wayback Machine: https://web.archive.org/web/20140623201218/http://www.microsoft.com:80/en-us/download/details.aspx?id=19364 jaclaz

Which - if it exists (which I doubt) - will probably be a CardBus and not a PCMCIA (Yes, I am picky): If more modern, it will be ExpressCard: https://it.wikipedia.org/wiki/ExpressCard jaclaz

Yes, this could be a possibility, at least you will have a "known base". Method one (easier): Leave the MBR alone and (after making a copy of the files) try reformatting as NTFS the partition under XP and let's see how it goes. Method 2 (IMHO better): Ideally you should however remove the current partitioning by clearing the disk (use diskpart or disk manager of Windows 7+) and recreating it so that the volume is aligned to 4Kb (which should provide some increase in speed), but do not format it (i.e. create the NTFS filesystem) it under 7+, but rather under XP. @dencorso Looking for troubles? (at bizzibody's expenses , BTW) Come on , apart the need to install the EXT2 FS drivers everywhere (thus losing the portability factor between different Windows machines) which EXT2 FS driver(s) do you suggest compatible with BOTH XP (which is easy) and Windows 7/8/8.1/10 (that may be trickier)? Both UDF and exFaT may be better suited, but while there is an exFAT driver for XP (and we will be back to the need to install driver, at least n XP), I am not too sure if there is a UDF driver for XP that is compatible with the implementation in 7/8/8.1/10 . jaclaz

... and: http://reboot.pro/topic/21538-booting-2-windows-partitions-on-single-hdd-one-via-uefi-vs-one-via-csm-bios-possible/ jaclaz

Good , that is what we call in jargon the "Bloubul's maneuvre" Happy of (yet another) happy bunny in the basket: jaclaz

I had a look at it, but besides (quite a few) strange settings, I found nothing hinting what the issue may be. Let's call them "anomalies": 1) the partitioning is "up to XP style" i.e. with 63 sectors before the partition 2) the bootsector code is instead that of a Windows 8+ (possibly also 10) 3) the MBR code - at first sight - is not a "common" one 4) the $MFT is where XP and earlier wouold put it (@ cluster 786432) 5) the $MFTMirr is where I haven't ever seen it (@ cluster 2) It seems like the thingy has been partitioned (or re-partitioned) and later formatted or re-formatted using various different tools or under different OS's, there is simply not a "coherent" set of values/locations. I'll have a look if I can find any hint of what the MBR code is, maybe that will explain how it was partitioned. jaclaz

You are not the only one (and this should show the shortsightedness of using mediafire or similar to host files if not temporarily). For some reasons Wim_b didn't upload it where it should have been, he uploaded it to mediafire and to datafilehost, and now both show "no file". Temporary upload/repository: http://www77.zippyshare.com/v/48Nzbc66/file.html jaclaz

The first thing that should be done on any install is to protect the built-in Administrator access with a password, bur you are right those that know and do this will also store the password somewhere (and thus not need any procedure) while the vast majority will leave the built-in Administrator (disabled but accessible via Safe Mode) with a blank password. jaclaz

Don't say that aloud, someone may decide to make a simple web api interface to a kubernetes container leveraging on a cloud saas third party library to connect a phpadmin instance to a raw registry editor (and use Rust and python as parsers and interface to the cryptographic engine) . jaclaz

Naah, the device is partitioned after all (and it is partitioned along the "old" standard of respecting the cylinder boundary, i.e. with 63 sectors before, which in itself it is not particularly smart on a flash device NTFS formatted). So I need the first 16 sectors of the volume. When the drive is accessible is not a problem to use the HDhacker (as the volume will be a logicaldrive in it) but when it is not accessible that may be a problem. Switch to dsfo. Get the dsfok package: http://members.ozemail.com.au/~nulifetv/freezip/freeware/ You want to extract the first 63+16=79 sectors, let's make it 100, i.e. 100*512=51200 bytes That will be on the XP: dsfo \\.\physicaldriven 0 51200 c:\good51200.dat dsfo \\.\physicaldriven 0 51200 c:\bad51200.dat where n is the disk number (as seen in Disk Manager) http://members.ozemail.com.au/~nulifetv/freezip/freeware/ It would also be interesting to know what is the opinion of 7 on the stick, i.e.: Stick working on 10 -> Stick working on 7? Stick working on 10 -> Stick not working on XP -> Srick working (or not working, needing repair) on 7? AND, I have to ask you this, what is the source of the stick? (like known local dealer vs. el-cheapo on e-bay or similar) jaclaz

Very, very strange. I take it that it is not partitioned. Make a copy of the first 16 sectors (at a time when the device is seen "normally"), then make another copy of the same first 16 sectors at a time when it shoes the RAW behaviour. To make this copy you will need something like dd (any windows port will do on XP, on 10 I am not sure you can get normally direct access). Try HDhacker (it will work fine under XP): http://dimio.altervista.org/eng/ you want first 16 sectors of the PhysicalDrive. jaclaz

Anything planned for NT 4.0 and Windows 95? jaclaz

Maybe a bit "niche" as hardware? jaclaz

Sure, and you may also want to notice how none (or next to none) are about "working", they are about learning, about communicating, about social networks, about blogging, about playing, but not a single one about someone in an actual office (or factory) at a desk actually doing his/her job through a PC. Even the very few that are work related, are invariably about "niche" activities carried out from home or from some (clearly artificial and improbable) "studios", here are a few: https://www.ispot.tv/search?term="windows+10"&qtype=ads&limit=24 This one particularly intrigues me: https://www.ispot.tv/ad/wI1F/microsoft-windows-10-angela-makes-clothes-that-make-a-difference My mom operated for years in the field and Rule #1 was you DO NOT store fabrics touching the floor, Rule #2 was you DO NOT lean them against a wall (and Rule #3 was BE VERY, VERY aware of Rules #1 and #2). It is also a miracle how Angela manages to move the apparently very heavy table/desk around the room and how there are nowhere scissors to be seen nor (everywhere) paper models ... jaclaz

@98SE You seemingly missed the base point . XP SP1 can access the memory just fine (without any patch). The limitation is an artificial one introduced in XP SP2 (in theory to prevent badly written drivers to crash the system). Some more info (and working links to Geoff Chapell's page on Vista that touches anyway the XP topic ) here: http://reboot.pro/topic/19933-accessing-3gb-to-8gb-ram-in-winxp-32-bit/ jaclaz

@cc333 That post you quoted is a leftover from the split that was made. At a given point 98SE and RLoew started talking here of Ramdisk related stuff for Windows 98 and the topic was split, keeping here the XP (or NT 5.x if you prefer) and 2TB MBR limit stuff, moving all the rest to this new topic: Everything is cool and dandy, now. jaclaz

@98SE With all due respect , but really, you DO NOT *need* a swapfile (at all) and surely you DO NOT *want* such a large swapfile (having ramdisk for your loading of an iso "as is" or similar may have some merits, but NOT a swapfile, which is nothing but an extension to available memory that you won't need[1]), I am all for "thought experiments" but maybe here you are taking it to the realm of pure speculation, just for the sake of it. Why don't you put together a suitable machine and try? Very likely, even if each and every theory expressed here is valid, you will find in real testing a number of roadblocks (be it BIOS, hardware, software, whatever) that may (please read as "will") likely prevent you from actually achieving what in theory is possible with a sufficient stability and you'll have to settle for "lower standards". jaclaz [1] In the real world at the time when Windows 98 was fully supported (and programs were actually written for it), a "typical" machine would have had 64 Mb, sometimes 128 Mb, exceptionally 256 Mb of RAM and respectively 128 Mb, 256Mb and 512 Mb swapfile at the very most. In other words, no machine at the time ever had more than 256+512=768 Mb of "virtual" RAM (i.e. physical RAM+swapfile) available. Still they worked fine. Now, with some tricks, you can have 1 Gb or more of "real" RAM available, that will be more than enough to run *each and every* program written for 98 (+1), unless (and until) some programmer will write an actually useful/working piece of modern bloatware that will need more RAM.

Is this to be intended as a road-map or preliminary timetable for the project? jaclaz

Ironically? Try guessing WHO EXACTLY asked the Wayback Machine to archive it? Maybe you meant "thankfully" , and now you also know who to thank for it . jaclaz

You are welcome I am sure , let us know how it goes. jaclaz

Ok, I have to ask. Is there any reason why on Windows 9x/Me it makes more sense to have a swapfile on ramdisk (when compared with NT and later)? Some past references, just in case: http://www.overclock.net/t/1193401/why-it-is-bad-to-store-the-page-file-on-a-ram-disk http://www.msfn.org/board/topic/156944-delete-not-clear-pagefile-at-shutdown/ http://www.msfn.org/board/topic/174002-windows-7-possible-advisable-to-disable-the-page-file/ http://www.msfn.org/board/topic/174520-optimal-page-file-setting/ jaclaz