jaclaz

Sure, but whether Microsoft supports (or fails to support) any given OS does not imply that another OS can run on it. While, as said, it is very unlikely that 7 can be installed on a Surface Go for technical reasons, the fact that MS does not support it is not particularly relevant. jaclaz

AND it doesn't mention XP64. Here is a Wikipedia page about scope: https://en.wikipedia.org/wiki/Wikipedia:Out_of_scope It will help the OP as much as your post (not at all ) but you might find it interesting . jaclaz

@Comos Then that is likely the issue. The Hitachi microdrive filter is (was) specifically made at the time for CF cards/Microdrives and only "flipped" the "removable" bit, as Andalu suggested, the DiskMod is a newer (and more complete/more featured) filter driver that can "flip" *something else* and make the (queer) drive suitable to host the pagefile. jaclaz

I know, it was a rhetorical question to quickly check the reliability of the suggestion. jaclaz

Check this, though it is improbable: https://learn.microsoft.com/en-us/archive/blogs/astebner/registry-key-to-force-windows-to-use-short-filenames Maybe it is *something else* that *somehow* prevents the proper setting of the secondary entry, you could check the actual file/folder entry on the filesystem: https://www.ntfs.com/fat-filenames.htm AFAIK the long and short filenames are two separate entries (the short filename one is automatically generated at file creation), which explains why you have your long file name shortcuts still working, so it could be *something* that only shows the short filename. Does this happen with *any* copy utility/method or just with a given one? As well, do you see the short file name with *any* tool (dir from command line, explorer, other file manager, etc.) or just in a given one? Are you sure that the issue is with filenames (and not with long PATH)? jaclaz

Can you explain? Maybe that is the issue ("wrong" type of device). jaclaz

Only for the record the limit for a file in $MFT is around 750 bytes on 512 bytes sector ($MFT record sized 1Kb) and around 3800 on 4K sectors ($MFT entry sized 4Kb), exact values here: https://www.forensicfocus.com/forums/general/mft-resident-data/paged/2/ jaclaz

Does "CSM bios mode" exist on a Surface Go? jaclaz

I could not understand anything in your post , but - generally speaking - it is not multiplication where changing the order of factors does not change the product. jaclaz

The right video link, as the video in itself is "wrong", the first thing they say is that they are not going to actually install windows 7, but rather that they are going to install the Windows 7 inside a VM (Virtual Machine) running on Windows 8 or 8.1. Anyway the OP has seen people installing 7 on modern hardware, so possibly he could ask those people how they installed it, though I have to remind that modern hardware is not a synonym of a Microsoft Surface, and particularly not of a Surface Go. I don't think it is possible at all, due to the peculiarities of Surface's (generally) hardware (namely the UEFI only and - if I recall correctly - GOP only video[1]), but additionally the Surface GO is a relatively recent model and I doubt that - even if *somehow* you can get past the two mentioned roadblocks - appropriate drivers can be found for it. jaclaz [1] there are actually workarounds for these - though complex and risky - that may apply to some "modern" hardware, but I doubt specifically to the Surface

Might be? That seems like a link to a search results page with *anything* related to the Surface Pro and or the Surface 7, if you can find and post a link to one of the results on that page that actually documents that windows 7 can actually be installed on a Surface Pro, that would count as a good answer (BTW to another question, since OP asked about a Surface Go, not a Pro), right now it seems like a generic "search for it". jaclaz

You are already out of date. Why all this interest in my (exact) seniority? I am already feeling old enough , I don't need a timer to remind me how much. jaclaz

I think (but do not quote me on this) that it may work, though I don't see the actual scope of this replacing of the processor, the notebook has probably a Core i3 or i5, there shouldn't be much difference in speed/performance replacing the original processor with a Pentium 6100, they seem very similar, at least on this comparison: https://technical.city/en/cpu/Core-i3-380M-vs-Pentium-P6100 Maybe if it has a Celeron, there would be some improvement, but I doubt it will be much, if any: https://technical.city/en/cpu/Celeron-P4600-vs-Pentium-P6100 Also the Fujitsu BIOS might be locked to the default processor installed in factory, clock speed of the P6100 seem to be same as the Celeron, but different from i3 and i5. IMHO all in all such an upgrade (which seems more like a same-level-grade) could be risky/problematic (if possible at all) without bearing any noticeable advantage. jaclaz

Maybe the issue is connected with the different way USB "Mass Storage" devices are handled in Windows 10 (as compared to previous Windows, i.e. multiple partitions visible as opposed to single partition)? Is that stick partitioned or not (direct volume or "superfloppy")? And if it is partitioned, is it MBR or GPT "style"? AFAIK USB 3.0 devices are (transparently) compatible with USB 2.0 ports, they will only be slower in data transfer. jaclaz

And - just for the record - at each and every update/new version the good guys managed to bork either links or contents of CODE tags (or both). jaclaz

You can try to rule out the CD out of the equation using a local source (copied to the hard disk) and WINNT.EXE from DOS (beware the approach is complex). Reference: the 911CD board is long death, you will need to use the Wayback Machine to check the contents. Or you could try the WinsetupfromUSB on a USB stick: https://msfn.org/board/topic/120444-how-to-install-windows-from-usb-winsetupfromusb-with-gui/ jaclaz

That laptop model is declared compatible with XP: https://www.levnapc.cz/ProductsFiles/FS-Lifebook-S710-technicke-specifikace-en.pdf Is the question only related to the new processor? In itself the processor seems compatible with XP.: https://www.game-debate.com/hardware/index.php?gid=719&graphics=HD Pentium P6100 So the question is if specifically you can replace that specific processor on the specific laptop? jaclaz

I am so sad. He was a great person. I knew him since my first days on MSFN, many, many years ago, and besides the public interactions on the board we had a lot of correspondence via PM's, both on computing and on a whole range of different topics, from cinema and TV to art and linguistics (he could read and understand very well Italian and I can understand a very little Portuguese, we had a common interest in Latin - as root of both languages - and we often exchanged opinions on these and many other things), I considered him a friend. I will greatly miss him and our exchanges. Besides his great work in helping and advising other people on technical matters, he was always polite and just (as a moderator), while often managing to keep a veil of humour in his posts. He told me about his illness, describing it as a "a quite mean cancer", from the tone of his latest messages he was fighting fiercely against it, and was hoping to defeat it only one month before, at the end of september. May he R.I.P. jaclaz

Is that the Max_Path issue? You should be able to access the file via the \\?\ syntax, but it has to be tried, as any tool may (or may not) support this alternate notation: https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation But Long Path Fixer should be able to see those files, maybe there is some other issue involved? The "traditional" way was to use subst to assign temporarily a drive letter to a "deep" folder: https://support.code42.com/Incydr/Agent/Troubleshooting/Windows_file_paths_longer_than_255_characters jaclaz

Yes, the way "cloning" is used varies a lot, it is a years long debate on how cloning is often used - if not improperly - at least not strictly enough (IMHO), JFYI . https://msfn.org/board/topic/157634-hard-disk-cloningimaging-from-inside-windows/ jaclaz

Never happened to me. Maybe there is some setting that gets "stuck" to automatic. If you have the guts for it , you can try to replicate this: https://msfn.org/board/topic/156944-delete-not-clear-pagefile-at-shutdown/?do=findComment&comment=1000282 jaclaz

No, that is not what I meant. Of course WinHex can copy (clone) disks or make images of them. The point is about having two exact copies (clones) disks connected at the same time to a same Windows NT based OS. Up to a certain Windows version (I think Vista or 7, but I may well be wrong), when this happened the Windows OS would silently and automatically change the disk signature of the second mounted copy. From what I read here and there more recent Windows (possibly from 8 onwards, but as above I may be well wrong) the collision is detected and the second disk is put offline to prevent this change. As well from what I read here and there some specialized "cloning tools" have special provisions to change (in the few places where it is hardcoded, namely the Registry DosDevices and the BCD on Vista and later) settings to reflect the newly changed disk signature. Such a clone won't be an actual clone anymore but will boot and work like the original, and can even co-exist with the original still connected as they will have different disk signatures. jaclaz

Does WinHex have such a feature? I was thinking of dedicated tools like Acronis. Macrium or similar, WinHex (at least the very old version I used many years ago) is AFAIK a generic (and very good AFAIK) hex/disk editor with some advanced features, didn't thought it had this provision automated, but in any case it would have not worked in j7n's case, since he had the two disks mounted at the same time (to create the third partition). On older systems you needed a copy of the MBR (or of just the disk signature) and replace it on the (modified) clone through another system (or via DOS or grub4dos, linux bootdisk etc.), on newer systems the clone is - I believe - put offline to safeguard the disk Signature, but if you put it back online on the same system the collision will happen again. jaclaz

That is the Disk Signature in the MBR: In more recent windows a signature collision is detected and the user is warned (and usually the second disk is taken "offline"), on older systems it is changed automatically and silently to a new value, as you cannot have two disks mounted at the same time with the same disk signature, when you remove the original disk, the new disk is not anymore a "clone". Commercial tools to "move/upgrade" systems to new disks have mechanism to avoid this collision, but since you did a straight direct copy the disk signature was duplicated. @D.Draker JFYI, the bootrec /fixmbr will only check and correct the MBR boot code, it won't change or set the Disk Signature, which was needed in this case. jaclaz

And - should it not work - you can try the similar solution using grub4dos (already linked to in a comment on that xp.win.at page): https://msfn.org/board/topic/154071-f6-without-a-floppy-drive/ though most of surrounding info is about XP, so it may or may not work with 2000. jaclaz