jaclaz

If the "bad area" includes the MBR there is simply NO address where to look for the actual partitons (JFYI: Windows asks to initialize the disk when the "Magic Bytes" 55AA miss from the end of first sector). But this does not mean that *all* the disk contents are lost (though it is possible). The generic procedure is to create a "clone" of the disk (or an image of it) using a dedicated tool, like the mentioned ddrescue, which basically: attempts to read a sector (and repeats the attempt reading it a given number of times) if it succeeds in reading the sector, it writes the sector to the target if it fails in reading the sector, it writes a 00ed sector to the target then it goes on next sector and loops to #1 the actual algorithm is more complex, but you get the idea. Once ddrescue has finished, you start examining the target (be it clone or image) and see what you can recover from it, in some cases you can get a whole partition/filesystem "sound enough" to be mounted and accessed normally or almost normally, in most you get only "RAW" sectors and when this happens recovering contiguous (i.e. defragmented) files is normally possible, if the files are fragmented (or very large, or both) you can still have some chances through more advanced carving/rebuilding techniques, it's impossible to say what will happen in a specific case, too many factors affect the chances of partial data recovery. Generally speaking, it is a terrible idea to scan the failed disk for fragments or attempting to recover sectors, if you suspect that the disk is failing as it is possible that you have one chance and once only to read a sector, and this chance should be used to copy the sector on a surely working target, but again it depends on which is the actual issue at hand, I have very old disk drives that developed a given "bad area" but that - excluded that area - continued working for years (limited to the non bad sectors) and as well I have seen disks that once started developing bad sectors increased their number very fast. jaclaz

If I get this right, changing the "Flags" value to 59 enables *something*, and resetting it to 58 disables it. So one could open Regedit, change the "AutoRepeatRate" to "20" and the "AutoRepeatDelay" to "440" (or other sensible values) "permanently" and use just the Flags value to flip the thing on and off, leaving the other values changed. It seemingly has not changed much since Windows 7: http://superuser.com/questions/388160/keyboard-repeat-rate-repeat-delay-values-in-win7 and not even XP or earlier. http://msdn.microsoft.com/en-gb/library/windows/desktop/dd318079(v=vs.85).aspx possibly the 59 is hex 3B coming from 1+2+8+10+20 hex? Maybe you could check if this thingy here works on 10: https://geekhack.org/index.php?topic=41881.0 jaclaz

ran, your drive is NOT bricked (or at least it WAS NOT bricked at the time you attempted using the CA-42 cable). It is NOT LBA. It is NOT BSY. Both BIOS and Windows "see" the device. Hard disk die , it is a fact of life . Some die suddenly, some start developing area(s) of "bad sectors" (what you call "a bunch of LBA access errors) and are simply "ill". The illness is called "bad sectors" or "LBA access errors" (or "bad head(s)" or "bad areas") and is NOT "bricking", or it may be anything among a number of other possible maladies, including "bad SA" (whatever it is), "incorrect adaptative" (again whatever it is) or some other ones, but it is more probable that it is simply a "bad area" or a "bad head/translator". It is like, actually it is exactly like, you had a flu and you decided to put your left ankle in a cast (which may be a good cure for a broken ankle but totally irrelevant for the flu). If it is only a "bad area", it is maybe possible to recover part of the sectors using something like ddrescue to copy *whatever* remains accessible to another hard disk, if it is a bad head, a whole side of a platter is simply "lost forever", if it is *anything else* as well the disk is lost forever unless - maybe, and I want to underline maybe - a professional data recovery company can fix it. Sorry for your loss , but really nothing that you can do with a CA-42 cable. jaclaz

The UBCD contains a (relatively complex) menu system that basically loads a floppy image + a number of chosen tools. It is much easier to make a new floppy image with a DOS and RPM than to "extract" it from the UBCD, I suggested because you reported failing to get RPM by itself. You can try again, getting Ranish Partition Manager from here: http://www.ranish.com/part/ I just tested the doanload of version 2.40 and had no issues in getting the file. You were not paying attention (or were not in class) when the teacher explained MBR partitions. Quick recap: A MBR's partition table has 4 available entries. These can be filled maximum with 4 primary partitions or with 3 primary + 1 Extended (containing *any* number of logical volumes). There cannot be more than one Extended partition on a disk. There are of course no limits to the number of partitions formatted with a "same" filesystem. The only real "need" is to have the DOS be residing on the first hard disk, on an active partition, formatted as FAT16 (for DOS up to 6.22) or as FAT16 or FAT32 (for DOS 7.x/8.x, please read as Win9x/Me). You have to put things into it's historical perspective. When Windows NT 3.1 and soon after 3.51 and 4.00 came out, people were running DOS 5.00 (or so), and later 6.00 and 6.22. DOS up to 6.22 wants for "itself" an active primary partition and it wants it formatted as FAT16. Additionally the built-in FDISK will allow only one primary partition and one Extended partition. So (remember that we are talking of disks that were typically 300-500 Mb in size and lated grew to 1.0 Gb, 2.1 Gb or 4.3 Gb) NT was designed to take advantage of the possibilities offered by the extended partition. The DOS would go to first partition (primary and active) together with a few "needed for booting files" from NT, i.e. NTLDR, NTDETECT.COM, BOOT.INI and in some cases a SCSI driver NTBOOTDD.SYS, while the rest (large part) of the NT system would go on a logical volume inside extended. BUT still, NT 4.00 had some limits to the size of a volume/partition, typically only the first 7.8 Gb of the disk would be accessible at boot time and the windows NT setup would not work if the first, active partition (what contains NTLDR and that MS calls, reversed "system") was not FAT16 and within 4Gb of size. Consider how the size of files that need to go on the "system" partition is just a few kilobytes and that the "whole" NT 4.00 installation (on the volume that MS calls "boot", i.e. the one that contains \WINNT\System32) is around 100 Mb or so. So, typically an "average" at the time 500 Mb hard disk dual booting DOS and NT would usually be partitioned in: a first, primary, active, partition 100 Mb or less with DOS 5.00-6.22 a second partition extended, containing a logical volume inside extended sized 200 Mb or so for the NT 4.00 OS and another logical volume around 200 Mb for "data"Since there were - since day 1 - issues (depending on the order on which what was installed and/or when some other tools were used) bootpart was developed (it dates back to Windows NT 3.x times) to allow to "switch" between the NT loader and the DOS IO.SYS booting or repair "botched" installs. Only later came third party tools (like Partition Magic) capable of making more advanced settings/changes to the partitioning scheme. Then came Windows 9x. A typical Windows 95 system with a same "average" 500 Mb hard disk with Windows 95 (OSR2+) installed would come with a single active primary partition, usually formatted as FAT32. Most people that already had a DOS system and that already was dual-booting with NT 4.00 would of course want to try the new Windows 95, and so every kind of tool and approach were developed. In such a scenario(s) there were not that much difficulties, systems with more than one hard disk were rare, you had only one disk, you always had a primary partition on it (that always got C:\ as drive letter on *any* OS and it was always formatted as FAT16, the very first version of Windows 95 did NOT have FAT32 support, NT was limited to either FAT16 or NTFS, if the volumes in the extended partition were FAT16 they got a drive letter in all OSes, if they were NTFS only got additional drive letters and were accessibly under NT, the DOS 6.20 or 6.22 and the Windows 95/DOS7.0 were resident on the first partition and the NT stayed on it's logical volume. But soon came the new version of Windows 95 with FAT32 support (and that attempted to remove the dual boot with "previous DOS versions") and everyone wanted to test this new filesystem and even more tools were developed to make these OS easier to manage. MS intention was clear: DOS was the "only" OS available used by both the businesses and consumers. Windows 3.x was it's graphical version. NT was the New Technology for businesses (and for businesses only) Windows 95 was the new GUI OS for consumers (and for consumers only) Later: Windows 98 was the enhanced new GUI OS for comsumers (and for consumers only) Windows ME was the enhanced (only worsened) new GUI OS for consumers (and for consumers only) Windows 2000 was the New Technology enhanced for businesses (and for businesses only) Then they changed their mind an forced down the throat of everyone a "business OS" (XP is actually very like 2000 with some added bells and whistles besides a few actual enhancements). But in any case, the idea was that you had only one OS, at the most (and limited to the few people that already had DOS and wanted to try a NT system) a dual boot was "allowed", in their mind, no machine should have ever dual booted a Windows 9x/Me with a NT 4.00, and the very day Windows 2000 was available, everyone should have ditched NT 4.00 replacing it with Windows 2000. This is part of the reasons why making a Windows 9x/Me live together with BOTH a NT 4.00 AND a Windows 2000 needs some planning before and has a few limits that one must know when attempting to put these OS all together on the same machine. To know how is currently (or at any time) your system partitioned (including CHS data) you can get PartIn9x.zip (for 9x use) and PartInNT.zip (for NT/2K use) from Symantec they are "Partition INFO" tools, they are "safe" because they are "Read Only": ftp://ftp.symantec.com/public/english_us_canada/tools/pq/utilities/ And you may want to review the Ranish "Partitioning primer" which for some reasons is unavailable online, but that can be retrieved through the Wayback Machine, like here: https://web.archive.org/web/20050830030859/http://www.ranish.com/part/primer.htm At the light of the above info, try thinking of how you think you would like to setup your system, and post your idea, and we will check together if it is doable, if it "hits" again one of the known limitations (or only it is likely to cause issues) and needs to be changed, and how exactly to make it. jaclaz

Something *like* http://sourceforge.net/projects/sacddecoder/ or more like a DSD converter: http://en.wikipedia.org/wiki/Direct_Stream_Digital and/or "ripper": http://code.google.com/p/sacd-ripper/ Warning: the latter NEEDS a compatible PS3 and the use of it may (or may not) be allowed in your country. jaclaz

@gunsmokingman The point is (was) IMHO to find a "fast" (as opposed to or at least not necessarily "elegant") way to find that stupid file, in my (BTW VERY 1984ish ) experience WMI is not that fast (please read as usually slow) and it may actually not be running (or running properly) on the target machines. Could you try "timing" your WMI approach? @jumper Another interesting idea , that should also be timed to have a comparison with the other tests. jaclaz

Are you sure that actual WRITABLE (i.e. that can be "burned" and not "pressed") SACD media exist? http://en.wikipedia.org/wiki/Super_Audio_CD jaclaz

If you read the actual .script, it shouldn't be that much difficult to find out which files and which Registry changes are needed. If you need to extract the files encoded, you can use, besides Winbuilder itself, this: http://reboot.pro/topic/10783-release-unwbzip/ jaclaz

The syswow64 support has been added in this project: http://www.msfn.org/board/topic/170546-win81se/ If you cannot/do not want to use it, you can still look in it to see how/which files, etc. are added for it: http://win81se.cwcodes.net/projectindex.php http://win81se.cwcodes.net/Projects/Win8.1SE/Build/5-Wow64.script jaclaz

No, a much higher resolution image is needed, but we can play the game reversed , this is how one will look like, you find it/them on the board : http://www.littelfuse.com/products/tvs-diodes/surface-mount.aspx See this for an actual example on your board: http://forum.hddguru.com/viewtopic.php?t=18870&start=41 http://img827.imageshack.us/img827/9905/hds721010cla332.jpg However they are rectangular, usually but not *always* black, and they will be very near the connector, with one side connected to 0/ground and the other connected to 5v or 12V. jaclaz

Let's not confuse things. If you are going to a Data Recovery firm, you are going to spend anything from 300 to 3,000 US$ to get the data recovered, this includes a new disk drive holding the recovered data, and of course you will need to send them the WHOLE drive (and normally the failed drive is NOT returned, as they will need to find a replacement PCB and they will keep it for further uses) If you are going to a PCB replacement firm, you are going to simply buy a functional PCB recovered/salvaged from another (working) disk drive. A number of firms, if you send them the "old" PCB ONLY, will - as a service - transfer the old firmware to the salvaged PCB they will send you, typically this will cost, including the supply of the PCB, anything between 40 and 80 US$. Obviously in this latter case, all the data recovery (if needed) is your own problem, and as well IF other parts of the disk drive were fried (let's say one or more heads) the risk of having spent vainly 50 bucks or so is on you. jaclaz

Dencorso posted a picture on how a TVS diode looks like, unfortunately it is not a "circular black piece". SATA power connection: http://pinouts.ru/Power/sata-power_pinout.shtml 0V=Ground jaclaz

Wait a minute. The idea is that you send just the board and they will provide a replacement board, taking the firmware from the "old" board and transferring it to the "new" one. This makes sense only to attempt recovering the data. There is no way to know if the *whatever* fried the board affected other parts of the disk drive, so it makes no sense whatever to attempt "repairing" a disk drive, once you hopefully get the data, you buy a new disk drive (or two). jaclaz

Bascially measure the continuity (resistance) between the +5V and the 0 and from +12V and 0 with *any* multimeter, if you have a direct short (0 Ohms) it is likely that the TVS was triggered. A TVS is more like a switch that can only be flipped once, it is normally "open" and when triggered becomes "closed", shorting the power line. If this is the case it is only needed to desolder it/them, the disk won't be anymore "protected" of course. If you can take a picture of the board, usually it is easy to identify components. For a PCB transplant there are two possible cases, there is a separate chip with the "adaptive data" (which is relatively easy to "transplant" as usually it is an 8 pin chip) or there is the need to transplant the whole processor (and that may be really tricky). IF the first case, it can be done even DIY, if the second there are no or little chances without a professional lab. Some of the people that can provide replacement boards will also make the ROM (actually NVRAM) swap or transfer the firmware. I don't want to seem like I am "sponsoring" a given firm, but I know one that can provide the PCB and do the firmware swap, and they are nice, competent guys in my experience, in case of need PM me. jaclaz

Check if, by any chance, it is not a TVS diode. Some (most) hard disks have one, or in some cases 2, one for the 5 V and one on the 12 V one. They are there to protect disks from "queer" things happenings to the power supply (and thus sending wrong voltage and the like to the power line of the device). If thisis the case, it is just a matter of removing the component (by design when it is triggered it shortens the V+ to ground). BUT while you are shopping, do invest a few bucks and buy a PSU tester, a hard disk PCB may fry for *whatever* reasons, but two in a row seem to me like too much a coincidence to have no connection with the Power Supply Unit. Which EXACT make/model are those hard disks? jaclaz

Essentially: 74EB means the OPPOSITE of 75EB one is "jump if" and the other one is "jump if not" (or viceversa, cannot remember)9090 means "do nothing" jaclaz

Yep, they are working on it: http://www.911cd.net/forums//index.php?showtopic=24464 jaclaz

Find a site/page that gives you this issue and post a link to it, until then what you'll get would be probably random attempts to explain something that is yet "undefined". jaclaz

Well, "a" browser might not, but my Opera (most probably "recklessly" ) got it fine . jaclaz

DP, what do you mean? I mean is the link the issue (you cannot resolve/get to it) or is it the batch file (i.e. you got the batch file "schedule.bat" from the link BUT the batch does not work for you)? The file is attached, in case. jaclaz schedule.zip

And, only to add to the list, a lesser known tool by McAfee called Bintext: http://www.mcafee.com/us/downloads/free-tools/bintext.aspx is very convenient/easy to use. jaclaz

Oww, come on, I was kidding, like "Thou shalt not take the name of ..." jaclaz

While still using FAT16 as underlying filesystem for the "boot" volume, personally I would still attempt to run Windows 3.11 on 7.x rather than on 6.2x, just in case: http://www.msfn.org/board/topic/97945-windows-311-and-ms-dos-71/ Working link to 3xStart: https://web.archive.org/web/20081221094417/http://oldfiles.org.uk/powerload/win3x.htm Opera 3.62 (16 bit) used to work fine on 3.11: http://www.vogons.org/viewtopic.php?f=24&t=12102 http://arc.opera.com/pub/opera/win/362/english/ jaclaz

Just in case, whenever "service management" is involved, a good resource/reference is Black Viper: http://www.blackviper.com/service-configurations/black-vipers-windows-xp-x86-32-bit-service-pack-3-service-configurations/ jaclaz

Sure , but if you are subject to this slight mistyping, you'd better call them "retired people" as it provides somewhat less possibilities of a misunderstanding . jaclaz