jaclaz

It is rather obvious, but I wanted bphlpt to research on the matter, technically those Zalman (which has BTW gone on the verge of bankrupt in the meantime ;(): http://www.geek.com/chips/zalman-is-not-going-bankrupt-restructuring-under-way-1609996/ are much more sophisticated, dedicated solutions (those and some more listed here): http://reboot.pro/topic/8944-boot-any-iso-image-or-boot-all-iso-images/ I read your current issue more as "forget about the use I will make later of this, how can I make a DVD or a DVD-like image that will be compatible with 1, 2, 3 etc. and booting on *any* machine including BIOS/UEFI?". What I need to be able to try and help you is: how EXACTLY (by EXACTLY I mean which specific xorriso version you are using, on which specific os with which EXACT command lines) you are now creating the .iso hopefully a copy of the (empty, I don't need/want the actual files) pre-made .iso Explanation: I have NO idea how xorriso works and it came out just when I thought that no tool on earth could have a more complex and worse documented set of command options than mkisofs, clearly proving me wrong I am NOT familiar with the whatever APM is and what actually an old Mac would *need*So, the only chances (provided that xorriso cannot definitely create a UDF filesystem or actually *whatever* stupid UEFI Windows *needs*) are some ways (if any) to post-process the .iso modifying it. BEFORE the above, can you try creating a "normal" (or at least "normal according to MS") BIOS/UEFI install cd/dvd using the OSCDIMG tool as detailed here and see if "as is" it works on your test systems?: http://www.msfn.org/board/topic/172122-got-tips-on-how-to-make-a-bootable-cd-from-bootable-usb/ http://www.msfn.org/board/topic/172122-got-tips-on-how-to-make-a-bootable-cd-from-bootable-usb/?p=1080928 jaclaz

As soon as you will find a USB bridge for a hard disk that allows two LUN's of which one (the first) being a CD/DVD like device , I am sure that the info will be useful to port the method/approach on a USB hard disk ... Seriously, some (most) USB stick controllers have the possibility of exposing themselves to the BIOS/OS as a "combo" device with two LUN's (you remember good ol' SCSI? ), the second can only be a "hard disk like" device, the first one can usually be eithe a "hd like" device or a "CD/DVD like" device, see these: http://www.msfn.org/board/topic/121199-formating-a-usb-stick-to-udf-or-cdfs/ http://www.msfn.org/board/topic/150653-booting-windows-on-multi-partition-usb-flash-drive/ http://www.msfn.org/board/topic/121502-install-xp-from-usb-the-easy-way-with-sandisk-cruzer/ jaclaz

Yep, the disk is seen as having 512 bytes/sector, i.e. it is one or the other versions of AF/512e. The syntax (not the most friendly around) of rawcopy you can get with "rawcopy /?". But the principle is simple, that tool (many others will probably do, but I know this behaviour of this particular one) is that programs that operate on block devices can only retrieve a single block or a multiple of it. The attempt for 511 bytes is only useful to make sure that the tool works as expected on your system (as no device will have a block size smaller than 512 bytes, with the exception possibly of a few arcane, ancient and obsolete floppy drives). The result of 512 bytes being actually 512 confirms that your device is seen as 512 bytes/sector. The result of 4095 bytes being actually 3584 makes this a certainty, as 7*512=3584 is the most the tool can get until it "steps" to 4096. The result of 4096 is "normal", you would have the same 4096 result on both a 512 and a 4096 bytes/sector device. The "size on disk" means "nothing" (or "next to nothing", I can now say that your c:\ drive has a cluster size of 4096 bytes , and most probably it is thus using a NTFS filesystem ). The "issues" come on the WMIC output: TotalSectors=3519054315->3519054315*512=1801755809280 Size=1801755809280 TotalTracks=55858005 TotalCylinders=219051 TracksPerCylinder=255 SectorsPerTrack=63 219051*255=55858005*63=3519054315 Somehow the disk is seen as being around 1.8 Tb in size. this may (or may not, it is to be seen) the show-stopper. It's time for you to start (if I were you I would first try in XP, and only later, if successful, go back to the Win2k) experimenting with http://www.rodsbooks.com/gdisk/ jaclaz

I see. But AFP is not more something like a network protocol? Maybe it is APM: http://www.zytor.com/pipermail/syslinux/2014-June/022287.html but anyway doesn't matter, " a rose by any other name ..." . jaclaz

Well, WHY EXACTLY do you need that? And BTW (though the dual LUN CD/HD like devices is a very nice approach at compatible hardware, i.e. sticks for which you have the "right" manufacturer tool and it actually works as expected) allow me to doubt that it is actually needed. As a side note, it is considered not very polite to bump an existing thread: http://www.msfn.org/board/topic/172610-acronis-iso-boot-on-uefi-pc/?p=1095786 only to ask for attention on a new one. jaclaz

Well, this is "normal", in the sense that the PCB is disconnected exactly to avoid the 'LED:000000CC FAddr:0024A051' "loop". It seemed to me from your previous posts like you had that loop both when the PCB was connected and disconnected, which was "queer". The CTRL+Z not working might mean that the issue is in either the connection of the Tx (as seen from the converter side) or of a non working/non suitable converter, everything is grounded together, right? I seem to remember people having timing issues (like issueing the CTRL+Z too fast or too slow) but you mentioned trying with different delays, disn't you? What remains if the above are nto the cases might be a (partially) damaged PCB, but it would be a rare case. Your next attempt should be to procure a similar hard disk and verify that your tools work on it. jaclaz

The point is to learn what the disk exposes (and also what Windows "sees") when it comes to "physical sector size". The "enhanced" FSUTIL provides (or should provide this kind of info), which your posted output does not. But we have a "poor man's way" alright. Get rawcopy from here: http://www.ltr-data.se/opencode.html/ You want to run, once found N which is the PhysicalDrive number, i.e. the Disk number in Disk manager: and then look at the exact file sizes of the result, the idea is that rawcopy operates on "blocks" so that it cannot retrieve a number of bytes less than the block size. Try also on XP (possibly it is available also on 2K, but cannot say) what WMI sees. You want to run: which would provide a number of interesting/useful info. jaclaz

Good. Yes it is . Though I am not sure what specific command triggers it. The only one that may cause it, could be the MOUNTVOL command , because we actually (when the switch is needed) mount to the first drive letter the volume with the "other geometry" bootsector. Try the attached, it is only a (hopefully working) quick fix, in which I moved the mounting to first drive letter to after the switch (if needed) is performed. If that is the issue, then I will rewrite the routine less half-@§§edly. jaclaz SwitcherQEVmod2.zip

It is possible that the issue lies in there , you don't have the possibility to run hyperterminal or Putty (which are both know to be working)? jaclaz

How EXACTLY did you try to reach the prompt? jaclaz

Good . this should be it: Try the attached, it's the same as above, but I added the condition for the "RAW" in the diskpart output: jaclaz SwitcherQEVmod.zip

Yep, it seems like the good guys at HSTG do not put these specifications, not even in the spec sheet : http://www.hgst.com/hard-drives/internal-drive-kits/nas-desktop-drive-kit Newish versions of windows allow to determine easily if a disk "exposes" a 512 or a 4096 byte/sector interface, see this only seemingly unrelated thread: http://www.msfn.org/board/topic/173265-formatting-an-external-drive-using-different-interfaces/ Try running the FSUTIL tool (on a more modern Windows). The 2.2 Tb limit that the good Intel and MS (and EFI/UEFI guys) like so much to attribute to MBR partitioning is FALSE. The MBR limit is 232 -1 sectors, i.e. 4,294,967,295 sectors, which translate, IF the sector is 512 bytes, to the known 2.2 Tb limit: http://support.microsoft.com/kb/2581408/en-us BUT with modern disks with a 4096 bytes sector size the same amount of sectors translate to 8 times that, i.e. more than 17 Tb. The NTFS filesystem, on the other hand, has not even this file limit, so, in theory there are no issues, as long as you do not want to boot from that drive (because there is no support for booting from 4096 bytes/sector drives in most - possibly all - BIOSes). BUT, while it seems (see the other thread) that XP has NO issues whatsoever with a 4kb sectored disk, it is possible that 2K may have some limitations, it has AFAIK never been tested. However, GPT is supported in Windows 2000 (for a non-boot/system volume): https://msdn.microsoft.com/en-us/library/windows/hardware/dn653580(v=vs.85).aspx In a nutshell: if the disk exposes a 4096 bytes/sector interface, it is more likely that a way to access it fully is found (with either MBR or GPT) if the disk exposes a 512 bytes/sector interface it is more likely that this will happen (with GPT ONLY) Still, both MBR and GPT schemes may find some other issue in a 2K system, possibly also connected to the specific drive model, the only thing is to experiment with it. jaclaz

Which EXACT model is it? Does the disk expose 512 or 4096 bytes/sector? Do you have access to a more recent Windows System? (like a Windows 7) How are you going to connect the disk? (SATA, USB bridge, etc.) jaclaz

I tend to see the half full glass , the bottom part of it (the XP part ) is full, let's see if we can fill also the top part with Vista or later. Test the attached in 8.1, at first in an open command prompt with Admin privileges, I just added to the previous version a few PAUSEs and some added "feedback" in order to catch where the error occurs, and post the output, notwithstanding what it may seem, it is "better" that the issue is "symmetric", i.e. it is not related to one geometry or the other. jaclaz SwitcherQEV.zip

Really , then I guess that all the people taking part to this forum that reported success in doing exactly that must be liars. jaclaz

Very interesting MagicAndre, thanks. jaclaz

I would say a little more than that, it is so common and so common on Dell's that it is FAQ #3 http://www.msfn.org/board/topic/116766-faqs/ (which is what you should have checked first, now it would be the right time to check and bookmark the above , they are handy, should you happen to encounter any of the other very common issues ) jaclaz

I belived that windows (and other OSes) use non-paged pool memory for critical things like ISRs and drivers to prevent be paged out. What do you exactly mean by "disable the PagingExecutive"? You mean disable page file in control panels? No, he means "DisablePagingExecutive": https://technet.microsoft.com/en-us/library/cc959492.aspx https://technet.microsoft.com/en-us/library/cc757875.aspx jaclaz

Yep , it is very possible that *something else* is missing or mis-configured (accidentally or by design by the good HP guys) or already ruined by some previous user action or even botched (accidentally or by design by any function of the running OS or by the parts of the XP setup that were already run) but let's go step by step , let's see first what is in the MBR (as opposed to what might be in it or be missing from it or in/from any of the partitions bootsectors or anywhere else) and let's make next move starting from that actual data. jaclaz

Maybe doing a full boot time log with Procmon would help. Hmmm. jaclaz

Well, besides being myself (sucking), you stated and re-stated how ALL MS OS suck or sucked in their time, exception made for (maybe) Windows 7 and now Windows 8/8.x (but only your specially tuned version doesn't suck). I am of the opposite opinion, i.e. not only NONE of MS OS sucked in their time (well, wait, no , exception made for Vista as released) and - as a matter of fact - they don't even suck (much) nowadays (which does not mean that many betterings were not made at each new OS release and not even that some - few - betterings were not made lately). I have been happily and contentedly computing before DOS, in DOS times, in Windows 3.1x times, in Windows NT times, in 2K times (sorry 9x/Me friends I actually never used those) and in XP times, and I am pretty sure that - when I will find an actual reason/need to switch to 7 or to a later OS, I will be able - like I managed to do all these years - to continue happily and contentedly computing alright. While reportedly all these years you lived and worked unsatisfied of the operating systems and hardware you had, dreaming of better ones until Windows 7 came: but still you were not happy or contented about them and it took you years or months to tweak them in such a way to satisfy you. I fully understand how after living so many years working in environments that didn't fully satisfy you, and dreaming about better computing experiences once you finally managed, through some really hard work, to reach your dream you are very proud of the result, and you wish to tell everyone how your lifetime dream became reality , still, maybe you overdo it a bit when you regularly and repeatedly downplay other people's (more modest) experience, used Operating System or hardware, underlining the uniqueness of your achievements, boasting how powerful and what not is your extremely top-end system, how smooth goes each and every operation you perform on them and how all other people are having a worse computing experience than you have. Now: jaclaz

@submix8c Yes , I also noticed the 11602 MB free on a 1 Mb partition queerness when retyping the contents of the screenshot, first time I thought I had made a typo, but it is actually in the image, though I would not give to it too much relevance, it is entirely possible that it is a "glitch in the matrix". No MBR CODE may be "peculiar" as you wish (or be missing altogether) but the DATA (i.e. the partition table entries) should remain however "standard" (though it would be well possible that the Windows XP setup parses *something* strangely) @oldtiger64 No matter what setup/solution/approach you will choose, you next step should be to make a backup copy of the MBR and post a copy of it for examination. To make this backup, get HDhacker (don't be fooled by the name, it is a commonly used tool useful to simply backup some relevant sectors of the hard disk): http://dimio.altervista.org/eng/ You want to run it, then, in "Drive to operate": Select "Physical Drive (MBR)" Choose Physical drive "0" (since you have only one disk that will be it) in Select sector, leave the default of "First Sector (MBR)" selcted Click on "Read sector from disk" Click on "Save sector to file", saving the file as (say) C:\HPoriginal.mbr Then compress the C:\HPoriginal.mbr into a .zip archive and attach to your next post the .zip archive file (NOT the .mbr one). jaclaz

Hmmm , given that: and that: and that you seemingly spend what little remains of that time by bragging about how good you are at doing your demanding stuff on your top class hardware, perfectly tweaked and set up (only thanks to your unique and superior to anyone else experience and knowledge): you are most probably right. I seem however to better enjoy my free time, and I have to remark how should being me actually suck, it would anyway suck in a modest, minimal way, requiring much less experience and computing power. _______________________________________________________________________________________________ JFYI, the above represents a line, which is the thing you just crossed, or - to be more exact - the thing beyond which your ego expanded. jaclaz

Well, the story is nearing the end, once also the SwitcherQE will have been tested (hopefully with success) all that remains is to double check everything and "RTM". Being the smart guy I am , I left however open the possibility of a sequel, for which your opinion/ideas/etc. would be appreciated, here is the rough storyboard (second part of the post): http://www.msfn.org/board/topic/173265-formatting-an-external-drive-using-different-interfaces/?p=1095074 jaclaz P.S.: As a bonus for the (affectionate) readers of this saga, find attached the (hopefully) final version of the accompanying Excel worksheet, as always password protected, but the password is empty PriLogxls.zip

But anyway all this has nothing to do with my use of WIN31FAHR as unit of measure and it's accuracy. Was NT 4.00 8 to 16 time better than WIN311? You can bet on it. Was 2K 16 to to 64 times better than WIN311? Hmmm, surely 16 to 32 times better, maybe not 64 times so. Was XP 64 to 128 times better than WIN311? Hmmm, hardly 64 to 128 times better, was it 2 to 4 times better than 2K? . ... We can use however MISNT4 as an alternate unit of measure . MISNT4 equates to the minimal install size of a NT 4.00, which is a little less than 200 Mb, 175 or 180 if I recall correctly. Windows 2000 at around 700 Mb scores a MISNT4 of around 3 and actually reflects how better 2K was when compared to NT 400. Windows XP at around 1500 Mb scores a MISNT4 of around 8, i.e. more than double, almost triple the 2k, but without corresponding "enhancements". Vista and it's Service Pack 3 rectius Windows 7 grew to some 12 Gb, if I recall correctly, i.e a score of 67 MISNT4 or a x8 growth when compared to XP, almost 20 (twenty) times when compared to 2K, and in this case the lack of corresponding enhancements seems to me self-evident. As said, not only it is sad, but when analyzed through these units of measure, sadder. jaclaz