jaclaz

What do you mean by "bootmgr failed"? Which error are you having? You do have a bootable USB stick as in the 7-steps set of instructions Ntator posted, right? Let's take those 7 points as reference, what happens when you do each of them? (or where what happens appears different from what is expected to happen?) jaclaz

In my experience very poor people cannot afford more than 512 Mb of RAM anyway, so it looks to me like a non-problem. jaclaz

Maybe it is just about time to bring down the wall between the "we" and the "they" and consider everyone as being "ordinary" slaves under the MS "plaaned obsolence evil plan"? I mean, now that also 2K and XP are out of support, we could maybe manage to rebuild the wall only to keep the NT 6.x people outside? And yes, if we cannot get to it, remember how Win2K is waaaay better than WinMe and how Godzilla can make mincemeat out of King Kong ANYTIME! jaclaz

Yep and it seems like NOONE actually made a compression/decompression tool using MSDelta.dll , so if by any chance you have nothing better to do, a nice project would be using MSDelta AVOIDING the stupid .Net bloat: http://www.msfn.org/board/topic/173071-msdeltadll-usage-help/ jaclaz

I understand and exactly because of this: I asked, because if a driver is *somehow* modified/repacked/compressed/whatever BUT the result is not tested and verified working on the specific hardware it represents IMHO more "wishful thinking" than an useful solution, and you can compress it as much as you want/can, still it will take some space without being of any practical use... Before I forget, we have some news here (possibly useful): http://www.msfn.org/board/topic/171184-microsoft-ipd-intra-package-delta-compression-tool/ http://www.msfn.org/board/topic/171184-microsoft-ipd-intra-package-delta-compression-tool/?p=1091462 jaclaz

Yes and no. Meaning that it may be easier, but not necessarily "new" means "better" and sometimes it doesn't even mean "working", if the given, specific version of imagex.exe - as it has been specifically reported - works as expected in the specific environment/setup/scope there is NO reason to NOT use the specific version linked to and instead attempt using a later version (that may work fine as well or completely fail to). jaclaz

May I ask a question? WHY it is *needed* to "to unpack with any number of available unpackers, change the inf and repack"? I mean would it not make more sense to make the actual drivers be available "as they are" and provide a tool or a set of detailed instructions (or, better, both) to allow the "final user" to modify the drivers (when/where possible) himself/herself (asking nicely to provide back the modified and TESTED driver)? I doubt that a "one-size-fits-all" approach on a huge mass of drivers may lead to anything more than a handful (maybe) of working ones and a large majority of useless crap (additionally modified). jaclaz

Hmmm, little by little we are managing, through applying some (light) torture , to extract from you the actual goal, which now seems to me more *like* "How can I change each and every PATH reference in the Registry to a set of new values?". You should be aware of the generic risk of slipping on chocolate covered bananas, JFYI: http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/put-down-the-chocolate-covered-banana.html Now, the next question is: Are you doing these modifications on an "online" Registry or are you doing it on a set of offline hives temporarily mounted in the online Registry? I mean, this may be of use: http://reboot.pro/topic/11212-offline-registry-library/ http://reboot.pro/topic/11312-offline-registry/ Or, depending on the actual use of the thingy, you may be able to afford paying a few bucks for a Shareware program, like (examples only, not necessarily a valid solution): http://www.funduc.com/registry_toolkit.htm I would also explore/experiment to see how much it is still functional the REGFIND from good ol' W2K Resource Kit: http://www.petri.com/download_free_reskit_tools.htm jaclaz

I see. That file is an update (actually an hotfix) for imagex.exe. It is a self-extracting executable, so you should just download it and then run it, and it should replace your current imagex.exe. If - for any reason - it doesn't work - you can open the 430546_intl_x64_zip.exe in 7-zip, then open inside it the Windows6.1-KB2525084-v2-x64.cab and from it extract the imagex.exe, replacing (overwriting) your current imagex.exe. The full KB insttructions are here: http://support.microsoft.com/kb/2525084/en-us jaclaz

Possibly a line needs to be drawn between "antivirus" and "antivirus with a a real time scanner", just for the record: http://www.clamwin.com/ of course one needs to know what he/she is doing if no real-time antivirus is running. jaclaz

WHICH error? Or same as WHAT? jaclaz

Or if you want to remain strictly inside batch you can use WizApp, the Wizard Apprentice: http://wizapp.sourceforge.net/ jaclaz

Besides the very high probability of being essentially FUD, this has the usual issue: http://www.pcpro.co.uk/operating-systems/1000045/the-moral-mire-of-not-patching-windows-xp? In order to lose "moral high grounds" one needs to have them, as I see it. jaclaz

Of course I am gonna approach it from the side you did consider but then swiftly discarded. The problem can be "generalized", as I see it, it amounts to two three questions: How can I get a .reg file containing the export of a SINGLE key with binary data?How can I convert such a .reg file representing a SINGLE key with binary data into a "bastardized" format that I find useful for search and replace?How can I convert back, once I have performed the needed search and replace, the data from this "no separator" format into a .reg file, ready to be re-imported in the Registry?Answers:#1 Use REG.EXE export [ROOT\]RegKey FileName.reg <- and this is pretty much all you can do with "built-in tools" For #2 and #3 external tools are needed, and some pretty much complex fiddling I have jumped over the "once I have performed the needed search and replace" and tried putting together two small batches, one is intended to convert the .reg file to the specified "RAW" format, and the other one is intended to recreate the .reg file from it. I have used what I had handy (and that I often use), i.e. gsar and SFK (SwissFileKnife): http://stahlworks.com/dev/swiss-file-knife.html These batches have NOT the ambition of being well written or "good", they are as usually half-@§§ed, quickly put together, batches with very little (if not none) "error control", making use of a stupid amount of temporary files and not even cleaning the mess after having been run, but maybe they could be of inspiration to make something "better". regtoraw.cmd: @ECHO OFFSETLOCAL ENABLEEXTENSIONS ENABLEDELAYEDEXPANSIONSET SourceReg=testbinary.regSET /A Counter=0IF EXIST RegHeader.txt del RegHeader.txtFOR /F "tokens=1 delims=:" %%A IN ('TYPE %SourceReg%') DO (SET /A Counter+=1SET Line=%%AIF "!Line:~-3,3!"=="hex" SET Line=!Line!:&ECHO !Line!>FirstLine.txt&GOTO :Out_of_ForECHO !Line!>>RegHeader.txtIF "!Line:~-4,4!"=="5.00" ECHO.>>RegHeader.txt):Out_of_ForIF EXIST myraw.txt DEL myraw.txtFOR /F "skip=3 tokens=1,2 delims=: " %%A IN ('TYPE %SourceReg%') DO (SET Line=%%AIF "!Line:~-3,3!"=="hex" (ECHO %%B>>myraw.txt) ELSE (ECHO %%A>>myraw.txt))gsar -s:x5C:x0D:x0A -r -o myraw.txtgsar -s:x2C -r -o myraw.txtrawtoreg.cmd @ECHO OFFSETLOCAL ENABLEEXTENSIONS ENABLEDELAYEDEXPANSIONgsar -s:x0D:x0A -r -o FirstLine.txtCALL :get_size FirstLine.txtSET /A offlen=%FirstLine.txt%-2SET /A offset=%FirstLine.txt%-2sfk filter myraw.txt +hextobin myraw2.txtsfk hexdump -hexsrc -recsize 25 -nofile -offlen 0 %offlen% myraw2.txt>myraw3.txtsfk hexdump -hexsrc -recsize 25 -nofile -offset %offset% myraw2.txt>>myraw3.txtgsar -s0x -r -o myraw3.txtgsar -s:x0D:x0A -r:x5C:x0D:x0A:x20:x20 -o myraw3.txtsfk replace -case myraw3.txt "/A/a/" "/B/b/" "/C/c/" "/D/d/" "/E/e/" "/F/f/" -yesCOPY RegHeader.txt + FirstLine.txt + myraw3.txt mytempreg.txtsfk partcopy mytempreg.txt -fromto 0 -7 myreg.txt -yesECHO.>Unicode_myreg.txtCOPY myreg.txt + Unicode_myreg.txt + Unicode_myreg.txt myreg.txtgsar -s:x0D:x0A -r:xFF:xFE -o Unicode_myreg.txtCMD /U /C Type myreg.txt >> Unicode_myreg.txtGOTO :EOF:get_sizeSET %1=%~z1GOTO :EOFSFK is a very comprehensive tool and I know only part of it's usage, so it is very likely that there are "better" ways using it or that "more suited" tools exist.I used gsar only because I am more familiar with it for the simple replacement, most probably it is not *needed* at all and everything can be done within SFK. Most of the complication is in re-creating a file identical to the original source, very likely the REG.EXE IMPORT will not be so "picky" on line length of the comma separated values, or with the CaSe of the letters corresponding to hex values. Have a Merry Christmas. jaclaz

Well, anything beyond any of the two mentioned "limits" (roughly 4 Gb ad 8 Gb) are very likely to be mis-recognized or *whatever*. Which SP do you have integrated in the NT 4.00? You should try to integrate SP6a: http://reboot.pro/topic/2383-windows-nt-service-pack-integration-slipstreaming-toolkit/ the mscab is still available through Wayback Machine: https://web.archive.org/web/20071220062213/http://mateusz.free.fr/mscab/ What is queer is that the C:\ and D:\ (i.e. first primary and first volume inside extended) are as well problematic. It is possible that *somehow* the "later" partitioning affects also the recognition of the first two, but it is "strange", particularly it is strange that the first primary (the C:\ drive) is considered Unformatted or damaged, it is likely that you fell into one of the little "traps" we haven't mentioned: http://support.microsoft.com/kb/151414/en-us Are you sure-sure that the partitioning and formatting was done correctly? Check the Partition ID's. The FAT volumes C:\ and D:\ should have a partition ID of 06 (and NOT of 0E) And possibly the Extended Partition should have - at least temporarily - a partition ID of 05 (and NOT of 0F). It is also possible that you will need to make a smaller Extended partition (in order to get the 05 ID) and later "expand" it to the final size. Can you run Partinfo (from the working 98) and post it's output? jaclaz

Well, to be fair, the SATA connector has been designed for 50 connections/reconnections http://www.wdc.com/en/products/resources/drivecompatibility/ In real life the connectors should last much longer than what the specs allow, still it is not a good idea to insert/remove frequently the SATA connector. The "right angle" ones (WITH clips) are said to be more reliable, but cannot really say, but whenever possible I would use the "latching" ones, straight or right angle. It is now a rather common issue, or at least it happened to me more than once (when you move, or send a PC through a courier, etc.) that on destination the PC doesn't boot up and on inspection the issue is the SATA connector on the cable, usually on the hard disk end, has come loose, and i have seen more than a few "mail-ordered OEM" PC's arriving with a bit of "hot glue" or "silicon" over the connector on the hard disk to keep it steady or even with an elastic band to prevent them from getting loose. As always having the contacts (on the hard disk and motherboard/card ) cleaned possibly using a specific product helps as I believe that such issues are due to the actual "springy effect" in the (cable side) connector losing some strength. jaclaz

I am starting maybe to understand, but I have not yet fully clear the situation/constraints. You are not exporting a key, you are redirecting the output of REG.EXE to a file, and then you want to edit this output "raw"? Why not export using a .reg file and the re-import/merge the .reg file? Or, while still using for the search/replace the output of REG.EXE, why not re-importing by creating a .reg from the "raw" data? jaclaz

Let's say that your lifetime experience led you to have non-conventional opinions and approaches to configuring and running windows systems and to their data storage organization, as said no doubts about you having good reasons for your choices. Sure . Though RAID 0 by itself means "little", most of the tricks are in the choice of the "right" high reliability (and high performance) hardware.. Here you are definitely wrong , most people cannot even dream about the throughput of a bunch of SSD's in RAID 0. jaclaz

I am not sure to understand (actually I am sure I do not understand) the "nature" of the the data 80Kb seems like a lot to be needed parsing/changing nor in what consists the actual find/replace. Can you post an actual example (using "dummy" data if needed)? A tool often used by me is gsar: http://home.online.no/~tjaberg/ but cannot say, without an actual example, if it might do in your case. Or there is the need to only do it with "built-in" tools? Also, which specific OS? jaclaz

Sure, I understand your point of view about having the largest possible free chunk and have it in one single place, I mentioned them only in the case (for some strange reasons) you wanted a single, fixed, drive letter. But still you have to (actually you do not "have to", but you well may) realize how your approach (for which you surely you have your very good reasons) is very "uncommon" in the sense that usually people that care about their systems tend to have the OS volume (and possibly even "device") separated from Data. Of course everyone has his/her own ways, and likes and dislikes, but personally I would not even think of having my data scattered over multiple devices in a RAID 0 (which is what you are planning to do if I get it right ). I guess that ReFS and "Storage Spaces" is exactly what you need/want, but still it requires data being separated from boot/system. jaclaz

AND From Adapter (GND) to >>> GND (on Seagate) That would be READ-ME-FIRST point #7 http://www.msfn.org/board/topic/143880-seagate-barracuda-720011-read-me-first/ and FGA's #4 and #7: http://www.msfn.org/board/topic/147532-fga-for-the-seagate-720011-drives/ Sure. The generic idea is that the more links you put into a chain the more probable it is that one of the links is weaker of the other (or if you prefer that something goes wrong). It is of course "better" if you plug that RS232-to-TTL converter directly in the port of your PC, as you remove a link from the chain, in any case you DO NOT want a null-modem cable (which is crosslinked): http://en.wikipedia.org/wiki/Null_modem you want - in case - a plain RS-232 extension cable. The general idea is the following, you are using a serial console (or Terminal program) in the OS, so a direct RS-232 to TTL is preferred because the data goes like this: Serial Port TTL converter SHORT piece of cable Device (your Seagate disk) (and back) The above is "better" (or "shorter" if you prefer) than: Serial Port Serial cable TTL converter SHORT piece of cable Device (your Seagate disk)(and back) And possibly even better than: Virtual Serial port (or if you prefer Serial port simulated through a software driver) USB port USB to TTL adapter SHORT piece of cable Device (your Segate disk)(and back) And - to repeat myself - GROUNDING together all devices involved is ALWAYS a good idea. jaclaz

If you (somehow) put together 8 devices each 512 Gb you get approximately 4 Tb total, so: The single 4 Tb cannot, as it needs to be GPT (unless you use an additional disk - or other device - to boot or use one of the available methods that you deem too complex or "unsupported") Here is from the mouth of the wolf: http://support.microsoft.com/kb/2581408 AFAIK exactly the same limits of Windows 7 apply to Windows 8. This is perfectly possible, and BOTH disks (or stripes of SSD's) can be MBR. The 1+3 earlier proposed would work if the first is MBR and the second GPT. Also, as hinted before, if you are after having a "single drive letter" you can have "C:\" volume on a 2 Tb disk capable of containing more than 2 Tb , as long as it is OK for you to have it NTFS and the other 2 Tb mounted - say - to a mountpoint *like* C:\Data\. I hope you have clear this latter possibility of mounting a non-boot volumes to a directory on a NTFS filesytem. jaclaz

I thought I was clear enough initially: So, if you still insist in having a monolithic boot and system volume/partition, and you find 800 Gb "tight" (though the "scratch space" as well as the pagefile can be - JFYI - somewhere else), you can make a 1 Gb disk (be it a single device or a striped volume over two devices) MBR partitioned, and assemble the other devices the way you want and make them - say - a huge 3Tb GPT partitioned striped set. jaclaz

It's not about being good at electronics . If you are not good at electronics you should NOT use the CA-42 or DKU-5 cable EXACTLY for the reasons expressed in point #10 of the READ-ME-FIRST, but if you already have such a cable and you already cannibalized it on one side, you can as well cannibalize it on the other one and map the cable colours as in the provided link in FGA #6: http://buffalo.nas-central.org/wiki/Use_a_Nokia_Serial_Cable_on_an_ARM9_Linkstation Even if you don't have a multimeter, you can see the different colours soldered to which pin on the telephone side, a multimeter is something that you can buy anywhere for anything between 3 and 15 US$, but a very poor man continuity tester to check which wire goes to which pin (still on the phone side connector) is not really difficult to make out of scrap parts you surely have: http://www.seekic.com/circuit_diagram/Measuring_and_Test_Circuit/CONTINUITY_TESTER_FOR_LOW_RESISTANCE_CIRCUITS.html In any case, you posted the picture of the wrong side of the board, post one of the other side (where the cables are actually soldered), maybe there are the actual "names" of the connection. jaclaz

Sure , that is a "perfectly normal" setup in a multi-boot environment, but if you remove the need for the XP and the other Operating Systems, you remain with 64 Gb for Windows 7, let's double it for 8 and double again the result to get the space for all the programs on Earth that you don't have+ huge slack space and you still get around 500 Gb. I doubt that anyone could find that a "tight" space. I mean, once someone has entered the idea/approach that a single huge system and data volume is not possible (besides not being particularly handy in case of needing a CHKDSK or the like) it would make sense to have ONLY the OS + programs in the smallest (while still comfortably) sized volume and have more space for the data on another volume. Until everything is on the same volume as NoelC pointed out the available space is one "chunk", while when there are two volumes it becomes "two chunks", it is less inconvenient if the "chunks" are 0+3.5 Tb instead of (say) 0.5+3 Tb. I am still missing how exactly NoelC plans to access anything beyond the LBA corresponding to 2.2 Tb on a MBR scheme, however. Maybe Windows 8 (and those SSD's and his RAID card(s)) can somehow work in "native" 4K sectors , but if this is the case then there is no need to make two volumes, and I am not at all familiar with a RAID 0 stripe across 8 devices (actually any stripe with more than two devices), so I may well be waaaay off target. jaclaz