CharlotteTheHarlot

@Ponch, actually I said lost, not overwritten. They cannot be overwritten since nothing is being written into B. His plan as he said is "move all the files, along with their folder structure, from B to A, as long as they don't cause any overwriting.". His folder A becomes the fatter one, aka the master, with B becoming skinnier, aka the remainders. Consequently, any files in B which somehow became updated or better than A prior to this merge will be lost during the transaction. Not sure what you are asking here. But first note there is no folder C to start with (that was my temp thingie). Also note that I had a (and click YES to all) after that first copy (since he mentioned in Post#5 that the second folder structure is similar so there will likely be a dialog about overwrite, hence the YES to all.). Maybe you are looking for less steps? Looking at it again I see it can be done this way as well ... Copy folder B to new folder C Copy A into C (and click YES to all) Delete Folder A Rename Folder C to A But it is certainly academic at this point since he has since made clear this is ongoing and thus he really needs to obtain a file sync program (one that allows source delete or file moves). P.S. it looks like that excellent For Loop you posted should be in a CodeBox instead of Code as it causes that one post to exceed the width of the page. We'll see in a moment how this post goes in (whether it inherits the runoff). EDIT: yup, just Post#12 exceeded the width. Mods ... I will make a post in the forum issues area about this. I took a look at the HTML and CSS and think I see the problem.

horus, let me ask you, is this folder merge is a on-time task or an ongoing procedure? If this is something ongoing that you will need to do often and consistently then you will want a programmable solution, and I would suggest a lightweight file synchronization utility (keyword is sync). If this is some one-off job it can be easily done as I stated above but with an added step to address what you stated in this reply ... First, let me point out a possible flaw in that algorithm. If you make any edits to B they will get lost with this strategy. In other words, you better be sure that there are no files in B that are newer/better than identically named files in A. Secondly, please note that this is a safe method. This suggestion does not involve a move when a copy plus delete can be used instead, this is due to the small but still existing element of risk of data loss in the middle of the operation should the power fail (yes, I know that NTFS is journaled, but I always play it safe!) If you want to use a Move or Xcopy with a switch, then this suggestion is not for you. Here we go: Copy folder A to new folder C Copy B into C (and click YES to all) Copy A into C (and click YES to all) Delete Folder A Rename Folder C to A We are now half done per your requirements, now we just need to skinny down folder B by deleting files that are identical to the now fatter A. The next step requires a shortcut icon to WinDiff, I keep one in QuickLaunch. Grab both Folders A and B and drop them on WinDiff. In Windiff, click options and make sure that Show Identical Files is checked. The other three options: Left-Only and Right-Only and Show Different Files must be unchecked. Ok? Now the list shows the filenames with folder paths pre-pended of every file that needs to be deleted from B. This list needs to be pasted into an editor and then fixed up for use as a batch file. If you don't know how: Single-Click the first file, hold down the Shift and Ctrl keys, press the End key, the entire list should be selected (release keys). Ctrl-C copy to clipboard, paste this into your editor. Two fix-ups are necessary (beginning and end of each line). Beginning of each line ... Search-Replace the string directly preceding the first FilePath, usually 3 chars: <space><tab><dot> and substitute it with DEL and your exact FilePath to the Folder B like so: del "C:\B End of each line ... Search-Replace each LineFeed and substitute "LineFeed (for example in UltraEdit each ^p becomes "^p Example here to make it crystal clear ... let's say this is the raw list from WinDiff ... [color="#FF0000"] [color="#000000"].\[/color]1\1\$$$5.reg [color="#000000"].\[/color]1\1\(((.reg [color="#000000"].\[/color]1\1\---.txt [color="#000000"].\[/color]1\1\1.b64 [color="#000000"].\[/color]1\1\1.xap [color="#000000"].\[/color]1\1\100_2013-15.jpg [color="#000000"].\[/color]1\1\100_2013.jpg [color="#000000"].\[/color]1\1\100_2014.jpg [color="#000000"].\[/color]1\1\2.b64 [color="#000000"].\[/color]1\1\2.xap [color="#000000"].\[/color]1\1\3.b64 [color="#000000"].\[/color]1\1\3.xap [color="#000000"].\[/color]1\1\4.b64 [color="#000000"].\[/color]1\1\4.xap [color="#000000"].\[/color]1\1\85888067.jpg [color="#000000"].\[/color]1\1\@.htm [color="#000000"].\[/color]1\1\^$#.txt [color="#000000"].\[/color]1\1\_k_2.txt [color="#000000"].\[/color]1\1\_k_4.txt [color="#000000"].\[/color]1\1\_rar.bat [color="#000000"].\[/color]1\1\ddd.htm [color="#000000"].\[/color]1\1\debugtrace.log [color="#000000"].\[/color]1\1\deltree.pif [color="#000000"].\[/color]1\1\dvd_win7_1_1.xap [color="#000000"].\[/color]1\1\dvd_win7_1_2.xap [color="#000000"].\[/color]1\1\dvd_win7_1_3.xap [color="#000000"].\[/color]1\1\dvd_win7_1_4.xap [color="#000000"].\[/color]1\1\edit1.html [color="#000000"].\[/color]1\1\file01 [color="#000000"].\[/color]1\1\filelist.dat [color="#000000"].\[/color]1\1\firmtools_test_panorama.jpg [color="#000000"].\[/color]1\1\get.htm [color="#000000"].\[/color]1\1\listst.txt [color="#000000"].\[/color]1\1\macro.txt [color="#000000"].\[/color]1\1\macrooo.txt [color="#000000"].\[/color]1\2\page_47_d.jpg [color="#000000"].\[/color]1\2\page_47_d_steg.jpg [color="#000000"].\[/color]1\2\page_52_h.jpg [color="#000000"].\[/color]1\2\page_53_a.jpg [color="#000000"].\[/color]1\2\page_53_j-ic.jpg [color="#000000"].\[/color]1\2\page_53_j.jpg [color="#000000"].\[/color]1\2\page_60_d1.jpg [color="#000000"].\[/color]1\2\panotools_result0.0.psd [color="#000000"].\[/color]1\2\ppp.htm [color="#000000"].\[/color]1\2\prefs.js [color="#000000"].\[/color]1\2\report.html [color="#000000"].\[/color]1\2\rrrttt.txt [color="#000000"].\[/color]1\2\rrtt.txt [color="#000000"].\[/color]1\2\sss.txt [color="#000000"].\[/color]1\2\temm.txt [color="#000000"].\[/color]1\2\temp#@.txt [/color] Your edits should result in this ... [color="#FF0000"][color="#000000"]del "C:\B[/color]\1\1\$$$5.reg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\(((.reg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\---.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\1.b64[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\1.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\100_2013-15.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\100_2013.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\100_2014.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\2.b64[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\2.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\3.b64[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\3.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\4.b64[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\4.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\85888067.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\@.htm[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\^$#.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\_k_2.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\_k_4.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\_rar.bat[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\ddd.htm[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\debugtrace.log[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\deltree.pif[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\dvd_win7_1_1.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\dvd_win7_1_2.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\dvd_win7_1_3.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\dvd_win7_1_4.xap[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\edit1.html[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\file01[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\filelist.dat[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\firmtools_test_panorama.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\get.htm[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\listst.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\macro.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\1\macrooo.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_47_d.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_47_d_steg.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_52_h.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_53_a.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_53_j-ic.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_53_j.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\page_60_d1.jpg[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\panotools_result0.0.psd[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\ppp.htm[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\prefs.js[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\report.html[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\rrrttt.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\rrtt.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\sss.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\temm.txt[color="#000000"]"[/color] [color="#000000"]del "C:\B[/color]\1\2\temp#@.txt[color="#000000"]"[/color] [/color] The batch file is done (unless you have some file attributes to worry about! If so, respond back if you need to learn how to clear ReadOnly etc). Save it and run it. But remember, if you have any nagging questions, ask them first! EDIT: needed Shift+Ctrl then End to select everything in a WinDiff window! Yes, believe it or not there is no Select All! But there is a File > Save File List ... EDIT2: arrgh just noticed I misread what you said: and a skinny B, containing only the files that would cause overwriting. That means non-identical are left in place! All fixed now. Sorry for the mistake.

Try this horus ... We'll call the finished product C ... Copy folder A to new folder C Copy B into C (and click YES to all) Copy A into C (and click YES to all) Done! B is integrated with A preserved. Delete A and B for cleanup. Its kind of like an XOR. This is an often needed procedure in an infinite number of areas, and most dedicated tools make it too complicated. Besides, IMHO this is the fastest way.

I hear that! Well if you need it, you need it, but pulling it out and making sure the legacy serial and parallel is turned off in the BIOS usually guarantees that IRQ 3 and 4 is free, ... and in your case, one less device on 11 .... UPDATE: ok i am looking at the Device Manager (DM) listing and see both IRQ3 and IRQ4 are wasted by serial ports. I would suggest they get deleted in DM and then you immediately locate that BIOS setting and disable them so they do not return (note for example : I have USB channels on IRQ 3 and 4 plus separate IRQ Holders) .... UPDATE: looking at the manual dencorso provided, they hid this setting under I/O Device Configuration and a submenu where Serial Port A and B are found. Actually this *is* how the BIOS Routing Table changes. You may have noticed when an upgrade is available for a motherboard BIOS it usually adds some features and fixes some bugs. Among the actual changes made when you perform an upgrade are corrections and additions to the Routing Table. The table is located in the programmable BIOS (and in many cases we ourselves can edit the image with BIOS hacking tools and later Flash this modded image back to the Eprom). Anyway, was there a BIOS flashing done before you added the USB reader? I tend to agree. FWIW, I find that having any USB plugged in during bootup to add to the overall bootup time on all versions of Windows. Some Win9x systems also can suffer from periodic hangs with USB devices in during power on/reset/off. And even though it is sometimes impossible I still try to keep USB ports empty to avoid problems. Contrary to what far too many people think, a USB printer does NOT have to be plugged in and powered on before the computer boots! USB actually is Plug and Play (not to be confused with Wintel's PCI Plug and Pray ;-) None reserved in here, yes, but those serial ports are effectively doing exactly this ;-) Yours must have spooled to the Default Printer! Sorry about those Print To File instructions, I just assumed everyone had the Generic / Text Only printer driver installed. Do you have it in your Control Panel > Printers? If you do then this is how you get that Device Manager listing into an ASCII text file: Device Manager > Computer > Print > All Devices > Print To File > Setup > Specific Printer > Generic / Text Only on FILE:. If you don't have it, you do Start > Settings > Printers > Add Printer > and install the Generic/Text Only. I tend to hook printers to XP and newer machines so I happened to have that Generic / Text Only as the default! Sorry again! Yup, that would be the problem. F8 Command Line DOS is actually not true DOS (it has HIMEM, IFSHLP, DOSHIGH, and other stuff I can't remember now). Most low level tools like HwInfo, RAM Testers, Intel Utilities and such require pure bare true DOS since they supply their own memory management in their code. What you need do is FULL format a floppy and check the Copy System Files box if you are using the Win9x GUI (essentially the same as format a: /s from a command line). Then place the HwInfo files on there and boot to the floppy. These are the same instructions you follow for many true DOS utilities. (sigh) I had some hope for that one. I believe for this to work you will have to go entirely Medieval on this box: nuke the Windows\INF bins, kill the serial ports, and temporarily disable USB, ethernet, sound. Re-enable one-by-one with USB first. Cross fingers, reboot often and look for a brand new IRQ arrangement in DM. P.S. Very nice motherboard BTW. And, I see a BIOS setting you have called: USB Device 9, Function 7, with its not so helpful description: This setting controls USB functionality. Options are Disabled and Enabled.. This is a new one on me, least I can't remember seeing it before. Anyone else know what it is?

Wowza. Looking at that IRQ11 makes me wonder how Win9x is even working ... System IRQ 5, INT# C ... Intel Corporation 24C7h 82801DB/DBL USB UHCI Controller #3 (ICH4/ICH4-L B0 step) System IRQ 10, INT# B ... Intel Corporation 24C4h 82801DB/DBL USB UHCI Controller #2 (ICH4/ICH4-L B0 step) System IRQ 11, INT# A ... Adaptec Inc 8178h AHA-2940U/UW/2940D Ultra/Ultra Wide/Dual SCSI Host Adapter System IRQ 11, INT# A ... ATI Technologies Inc 5961h Radeon 9200 Series (RV280) System IRQ 11, INT# A ... Conexant Systems 2F00h HSF 56k HSFi Churchill Data/Fax Modem System IRQ 11, INT# A ... Intel Corporation 100Fh 82545EM Gigabit Ethernet Controller (Copper) System IRQ 11, INT# A ... Intel Corporation 24C2h 82801DB/DBL USB UHCI Controller #1 (ICH4/ICH4-L B0 step) System IRQ 11, INT# A ... Philips Semiconductors 7146h SAA7146 Multimedia Bridge Scaler System IRQ 11, INT# B ... Intel Corporation 24C3h 82801DB/DBL SMBus Controller (ICH4/ICH4-L B0 step) System IRQ 11, INT# B ... Intel Corporation 24C5h 82801DB/DBL AC'97 Audio Controller (ICH4/ICH4-L B0 step) System IRQ 11, INT# D ... Intel Corporation 24CDh 82801DB/DBL USB 2.0 EHCI Controller (ICH4/ICH4-L B0 step) System IRQ 15, INT# A ... Intel Corporation 24CBh 82801DB/DBL (ICH4/ICH4-L) UltraATA/100 EIDE Controller Jeez, unless he has changed something in this version it should work. Just for the heck of it, try adding the himem.sys file to the floppy and this line to config.sys: device=himem.sys /testmem:off I am not sure why you would get that error. On my HwInfo floppies (I just checked) I only have lastdrive=z in the config. Anyway, there is a nice thing in HwInfo where he shows the Supported IRQ(s) for the devices it finds. I also like comparing its IRQ summary against the Win9x summary (see below). Looks good from PCI.EXE (The ROM PCI IRQ routing table appears to be OK.), though I am not exactly sure what PCI/PCI32 can tell except for the fact that the table can be successfully read or not. But as stated previously, as long as you have not flashed or physically replaced the chip itself, the routing table is a constant and not part of the problem. But a few other things now come to mind: Any chance that you have some resources reserved in that Device Manager > Computer > Properties > Reserve Resources Tab? Does your BIOS allow disabling USB completely? If so, you might try it and reboot and see if the sound works. Could you please explain exactly the cards installed (I thought 3, but it looks like there is a modem too?). I looked throughout this thread but do not see it detailed. Are you actually using both Ethernet and modem? Can you post the the Device Manager > Computer > Print > All Devices > Print To File. This is the Win9x view of the environment and essentially tells us what Win9x did with the BIOS IRQ Routing Table it called up. Does the motherboard have onboard video (yeah I do see the ATI card)? The reason I ask is that sometimes it helps to first install the chipset INF without any video card and get only the mobo native devices working, then install the video card and its drivers. This results in a different IRQ arrangement, most importantly the onboard USB gets first pick of the IRQ litter (typically four different ones).

Doh!, my bad. You have to cause a re-detection somehow, this is usually done by deleting the whole key or just branches in HKEY_LOCAL_MACHINE\Enum or removing devices in Device Manager or pulling or swapping PCI cards or even by disabling a couple of items in the BIOS (like USB, serial, parallel, ethernet, sound etc), anything that will trigger a shuffling of interrupts. Let the record show that my real suggestion is to install Windows on another HDD to find out if this mobo is still 100% functional, because if it is not, we are just spinning wheels (not that wheel spinning isn't fun mind you ;-) Well, I always restore to an exact known state, so as mentioned in that last post I would restore those two BINs and the Registry as well (four file matched set). This we do not know yet. There was a time when IBM machines had BIOS based or 5.25" floppy disk diags which made this simple. Unless you can scour the Supermicro website and forums and locate some DOS based utility that queries their hardware, well, your are only left with deductive reasoning. yeah, I saw that, although after re-reading this thread I am confused a little, I need to ask: Are you still dual-booting Win2K and the audio is absolutely positively definitely ok in Win2K? This seems to answer that question as yes, and that means there is NO hardware problem! We got our answer. Yeah but AC97 or SB card is all the same to Plug and Pray. Disabling the onboard ethernet is like removing a NIC, disabling the onboard sound is like pulling out a SB card, etc. Does this mobo also have onboard video which may have somehow become re-enabled thereby re-arranging IRQs? I should have asked earlier but have you already killed un-necessary devices in the BIOS (serial, parallel, etc)? From the BIOS is the first choice in that infamous IRQ Steering dialog box. This table (located in the BIOS Eprom) actually can change if you flash your BIOS with an updated (or modded) image. If you have not done any flashing whatsoever, and the sound did in fact work at some time, the problem is most definitely not in here. Doublecheck that Control Panel > System > Device Manager > System Devices > PCI Bus > (Properties) > IRQ Steering and make sure only the 3rd box (Protected Mode) is unchecked. Dave, it seems clear to me that this is just a configuration error in Win9x. If you handed me this mobo, I would ... - Pull all the cards (3 right? AGP included if there is onboard video) - Clear the CMOS - Disable practically everything in the BIOS (USB, ethernet, etc) - Install Win9x on a spare drive - Install Chipset INF - Install Audio drivers - Ensure Audio is working (save REG/BINs) - Re-Enable one-by-one - Ensure Audio is working (save REG/BINs) If all is ok I would transplant the Hardware tree to the old system. It is difficult. But this is the nature of Plug and Pray (don't get me started on this subject!). You see, you need to jigger the IRQ arrangement which is a real PITA. Sometimes you can manually change them in Device Manager under certain Resources tabs, and then you change another and another, etc, but it sucks in a big way. Near as I can tell the sound problem began when the USB reader was added. This is a clear case of Interrupt conflict and Windows not really sharing devices as Plug and Pray promised. In addition to what dencorso said (Craig Hart's software) there is also Martin Mal¡k (REALiX) HwInfo, get the DOS Version. Both of these (Craig Hart and Martin Mal¡k) query the hardware directly. They are ran from true bare DOS on floppies.

Note that it was removed from the {cf2524c0 START key which I read as Windows attempted to start some device (maybe Audio) on this shared IRQ and it failed. This could mean that the current PnP arrangement is ineffective and may require re-arranging cards, BIOS resets and a boat-load of patience ... or ... there is an actual hardware failure. Two ideas that might help if it is not a hardware failure. Please be sure the BIOS has been flushed and settings are then set correctly: (1) The In-Place Nuclear Option: back up the twin files Drvdata.bin and Drvidx.bin located in Windows\INF (also save the registry files at the same time and consider them a four file matched set). Then you delete those BINs in true DOS (F8 Menu: Command Line, this is much fast than Safe Mode), reboot and enjoy the show while Win9x performs drastic Plug and Pray brain surgery. After several more reboots some chance exists that the audio might be working. (2) Re-arrange the PCI cards and then do the Nuclear Option. (ACTUALLY: you boot to DOS (F8 Menu), then delete the BINs, then shut the power off, then re-arrange the cards, then power up. Make sense?) Save the files beforehand as mentioned above! In either case: if it works and the audio is fixed, grabbing the new four file matched set as a working prototype to be WinDiff'd against the saved set will lead to the answer as to what had changed (NOTE that if the cards have changed slots, there will be new registry key names and a comparison requires some skill as well). Both of the above is still a rough approximation to (3) a clean install on a different HDD which later can be slaved to the system, allowing you to cherry pick fresh working objects (registry entries, INF files, INI files, etc) and shoe-horn them into your current system. To be clear, I do not recommend the above (deleting the INF database) for casual users because it is a truly destructive approach: you may blue screen halt. You would then have to know how to go into true DOS and manually restore those four files in order to recover. This is why working on a spare drive (while the system drive is safely sitting on the shelf) is so much more preferable. Or, you could even clone the current system to a spare HDD and experiment away on the clone without fear (including drastic measures like deleting the INF database, re-installing the Chipset INF package, swapping card slots). Really, if all three of these drastic measures (especially the clean install) results in a system that still has audio problems your culprit is likely a hardware problem. These do happen (perhaps static discharge while changing RAM DIMMs) although rarely after a proper burn-in period. I highly doubt that SFC will lead anywhere promising. But the only thing you've got to lose is a little time, so go for it! BTW, for anyone wondering why not just re-install Windows and start over, you must understand the need for this preservation method: you have tons of stuff installed with all their attendent registry keys and files spread all over the disk, countless updated Windows core files and runtime libraries. Not to mention private data, passwords, users, customizations, etc. Blasting away a working system has consequences, especially when you ponder reverting the majority of the environment to 4-23-99 or older. IMHO, with enough effort any problem can be rectified this way, except for hardware failures of course. However, for this to work you must have a clean working set of registry/INF data for a reference. Hence the repeated suggestion to install to a spare drive. If the problem cannot be rectified on a nice new (and smaller and faster) registry, it will never be rectified under the current one. Yikes! this should never happen! Golden Rule: never experiment on your one-and-only system drive! Backup, Backup, Backup (or Clone, Clone, Clone). Then you can try anything with nothing to lose.

Please save them now! ... ok ... some time travel ... You should know that you can use this awesome utility called RegDat (website: http://freenet-homepage.de/h.ulbrich/) Win9x version is regdat.zip. It allows you to open the binary files SYSTEM.DAT and USER.DAT directly within the program and then export an ASCII registry file complete or any selected section. Very Very nice. Then you can try using these previous registry values in .REG patches among other things. But here is another idea which I am always suggesting. Grab a spare HDD and replace your system drive with it, and then re-install Win9x (nothing else), and you can grab that new registry as a parallel working prototype to manually fix any holes in your current one. I suggest that before you install Win9x again, that you first flush the BIOS (maybe pull the battery and/or short the reset jumper too) and reset defaults and then make sure that all settings are correct (ACPI, USB, etc). With that new registry in hand and a lot of spare time, anything should be able to be fixed by carefully porting the known good values. I can tell you exactly how to re-enable them (though the suggestion above to re-install on a spare drive is better). But, in a nutshell here it is (export the registry and use an editor!): First, find out exactly how many IRQ Holders you have defined by counting the subkeys under: HKEY_LOCAL_MACHINE\Enum\Acpi\*pnp0c0f Note: that red color indicates that it is probably within *pnp0c0f but I am not sure if that is a constant. Anyway, you are looking for and counting all the subkeys that contain the string value pair: "DeviceDesc"="ACPI IRQ Holder for PCI IRQ Steering". For example I have eight of them: [HKEY_LOCAL_MACHINE\Enum\Acpi\*pnp0c0f\00000001] ...thru... [HKEY_LOCAL_MACHINE\Enum\Acpi\*pnp0c0f\00000008] Now, these devices must appear under two other keys, ENUM and START ... [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{9b4e7760-3196-11cf-97ea-00aa0034319d}] ;;; "Problem"="Enumerating a Device" "Acpi\\*pnp0c0f\\00000001"=hex:00 "Acpi\\*pnp0c0f\\00000002"=hex:00 "Acpi\\*pnp0c0f\\00000003"=hex:00 "Acpi\\*pnp0c0f\\00000004"=hex:00 "Acpi\\*pnp0c0f\\00000005"=hex:00 "Acpi\\*pnp0c0f\\00000006"=hex:00 "Acpi\\*pnp0c0f\\00000007"=hex:00 "Acpi\\*pnp0c0f\\00000008"=hex:00 [color="#FF0000"];;; NOTE: lots of other devices are in this list![/color] [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{cf2524c0-29ae-11cf-97ea-00aa0034319d}] ;;; "Problem"="Starting a Device" "Acpi\\*pnp0c0f\\00000001"=hex:00 "Acpi\\*pnp0c0f\\00000002"=hex:00 "Acpi\\*pnp0c0f\\00000003"=hex:00 "Acpi\\*pnp0c0f\\00000004"=hex:00 "Acpi\\*pnp0c0f\\00000005"=hex:00 "Acpi\\*pnp0c0f\\00000006"=hex:00 "Acpi\\*pnp0c0f\\00000007"=hex:00 "Acpi\\*pnp0c0f\\00000008"=hex:00 [color="#FF0000"];;; NOTE: lots of other devices are in this list![/color] If there was a problem the hex:00 will either show up as hex:01 or be deleted completely. If they are deleted completely, they show up as disabled in Device Manager. For example if #4 and #7 were disabled from above, those two keys would look like so ... [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{9b4e7760-3196-11cf-97ea-00aa0034319d}] "Acpi\\*pnp0c0f\\00000001"=hex:00 "Acpi\\*pnp0c0f\\00000002"=hex:00 "Acpi\\*pnp0c0f\\00000003"=hex:00 "Acpi\\*pnp0c0f\\00000005"=hex:00[b] <-- see?[/b] "Acpi\\*pnp0c0f\\00000006"=hex:00 "Acpi\\*pnp0c0f\\00000008"=hex:00[b] <-- see?[/b] [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{cf2524c0-29ae-11cf-97ea-00aa0034319d}] "Acpi\\*pnp0c0f\\00000001"=hex:00 "Acpi\\*pnp0c0f\\00000002"=hex:00 "Acpi\\*pnp0c0f\\00000003"=hex:00 "Acpi\\*pnp0c0f\\00000005"=hex:00[b] <-- see?[/b] "Acpi\\*pnp0c0f\\00000006"=hex:00 "Acpi\\*pnp0c0f\\00000008"=hex:00[b] <-- see?[/b] Keep in mind that the contents of those two keys are not necessarily in sorted order (although in REGEDIT they will look sorted). So in your editor you have to scroll all the way through all the devices under those two keys to see everthing (or just use your editor's search/find command ;-) Anyway, this REG patch would restore the missing ones (#4 and #7 from my example only! you would need to make a custom one for yours!) ... REGEDIT4 [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{9b4e7760-3196-11cf-97ea-00aa0034319d}] "Acpi\\*pnp0c0f\\00000004"=hex:00 "Acpi\\*pnp0c0f\\00000007"=hex:00 [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Asd\Prob\{cf2524c0-29ae-11cf-97ea-00aa0034319d}] "Acpi\\*pnp0c0f\\00000004"=hex:00 "Acpi\\*pnp0c0f\\00000007"=hex:00 Note that when playing in this part of the registry it is good practice to reboot several times before you go back in and compare registry settings because Windows can change things very quickly from INF files. This is very good advice for highly CPU bound busses. I do the exact same thing on Win9x and on slow processors as a rule. (though Vista/7 with Dual/Quad Core finally makes it a moot point).

Well, since this is Win9x, live experimentation is nice and easy because you can copy SYSTEM.DAT and USER.DAT somewhere and replace their later possibly FUBAR versions from DOS for a perfect 100% registry rollback. The only other thing that needs to be done is get a complete filelist so that you can manually compare it to a later one for rolling them back as well. Now you can play in device manager and with USB driver packages all you want with the guaranteed ability to restore the system exactly to a previous state. This is not easy on Windows later than WinME. But there is a sure way to rule out the driver as having disabled the port. This is what I do when a computer hits my shop with an apparent defective port, be it Parallel, Serial, PS/2 or USB, that cannot be repaired by normal methods. Note that this method is to avoid blasting away a working system drive! ... (1) Grab a spare HDD like 10, 20 40 GB whatever. Drive contents are expendable. Remove the working drive, install the spare. (2) Double check the BIOS so that everything is enabled (reset to default or even shorting the reset jumpers or removing the battery for a few minutes to clear the CMOS might be in order here). (3) Boot with a Windows setup disc, I usually grab WinXP SP2 (USB issues were pretty much ironed out at this point), if that's impossible SP1, or SP0, or Win2K or maybe WinME might be OK as a last resort. Now do a full and clean install. Finally test the ports! If they cannot be seen here they are likely blown (or a pair are blown). At this point I just grab a $9 USB (or whatever) card and just fix it. (4) If the ports are working ok here, I grab a copy of the registry and use that in my detective work later to repair the possible registry discrepancies. In your case, you should probably now grab the Win98se disc and install that next on the spare drive, try to get the ports working and if successful grab the registry and files (slave it later if necessary) and repair what you now know is Port Death by Windows. When you are satisfied that you have the answer just reverse (1) and voila, no harm has befallen the working system disk. When I have a customer's computer, blasting their system is just not an option naturally. BTW, perhaps a Linux live CD can check this as well (I have not tried this since I have plenty of drives and Windows discs). And there are some really low level tools like HwInfo that boot from a floppy that might now have USB testing. FWIW, I believe the OS port drivers, particularly USB can disable the power and data lines to effectively kill the port. So the whole point of this method is to see if the motherboard parts are recognized and utilized by a completely independent version of Windows that is completely untainted by any user tweaks, BIOS or otherwise. EDIT: doh! ... defective <> detective

dang! I just looked at a bunch of mine and yep, Manufacturer and Version are always listed, but OEM and Retail is sometimes missing. The system builder edition tags seem to always say OEM (this is probably not you) but the retail store model tags mostly do not have it. So, I agree with you, assume OEM. MS has a good page on this here: http://www.microsoft.com/howtotell/content...g=en&pg=coa And that pretty much proves it is OEM. This is what I was trying to avoid doing. Haven't exactly had much luck with HP support in the past.I hear you, but, talk to them anyway. They really should offer the disc for like a $10 meda and shipping charge. They might even have a CDROM disc image you can download and burn. And yes, I was referring to contacting HP. Definitely don't bother MS over this, they washed their hands clean from any involvement with all things OEM. Microsoft gets the phone call only after you have the CD and activation fails. You should realize that you are much better off with this solution even though the correct CDROM for this model is going to be SP2 at best or possibly even older, because, you can then process this CD with nlite or similar techniques and create a fresh SP3 custom install disc. And if your motherboard is still the same (actually the BIOS), and you mind your OEMBIOS files, you will have a fully legal no activation scenario. This is as good as it gets on Windows XP legally. It is very likely that a so-called recovery disc in this case means it either contains a raw HDD image (with the OS and other apps all set up on a set of discs) to be plastered onto your system ... or ... it has software to access a hidden partition which may yet still exist on this computer and again plasters an image from there to your system. I guess a 3rd possibility is that it has no operating system functionality at all and it merely screws things up by re-installing all the bloatware. IMHO, I can envision no good coming from the use of one of these discs. Any of these recovery scenarios can be accurately described as icky. This is common practice nowadays for laptops and even desktops (accessed through some BIOS menu selection), usually after the internet and viruses have caused the system to go FUBAR. But real men (and women ) have custom CDROMS and/or recent disk images for doing it the best way.

Not enough info here. Exactly what did you use?To cut to the chase, you always need to use a disc that matches your retailer (HP, Dell, eMachines), the version of XP (Home, Pro, MCE), and the type of install (OEM or Retail). Your sticker will tell you everything you need to know. What you will most likely need to locate is an HP OEM branded re-install CD-ROM for WinXP Pro. This is what you need to find (try contacting them directly before going to eBay). Using this disc will likely result in a simple re-install without activation assuming it was a standard OEM Retailer SLP arrangement (the installer will sniff your motherboard and see if it is qualified). Worse case scenario is a free phone call to MS with you describing your legitimate sticker and the circumstances. Somewhat related discussion in this thread, read post #6.

Exactly what are the specs of your machine with Opera 9.27 working (including JRE version)? Be aware that un-installing Opera may result in the loss of saved settings, passwords, bookmarks etc. Opera 10 should work on any machine that runs 9.xx provided the JRE and stuff remain the same. Having said that, 9.2x was definitely faster than 10. Can you estimate how long the delay was for 9.27 from the time you clicked the icon until Opera appeared? There is a delay while the JRE initializes. I have noticed a slightly longer delay from running Opera 10 versus 9 under the same exact JRE. Note that I find it helpful to start Opera with no home page and no open tabs. (1) It is possible that the shortcut you mentioned may not have pointed directly at the Opera.exe file, instead it may have been an MSI generated pseudo-shortcut that used the Application Data folder. (2) Also, it is possible that the registry settings for Opera may have similar MSI generated bogus data. If you cannot get it to run after re-installing, we should look at both of these (1) and (2) as likely suspects. Also, please post here the exact path on your HDD to the Opera.exe file after you install Opera 10. This will make it easy to whip up a quick registry patch should you need it.

Get yourself the so-called 'Classic' Installer. You used to have to click around to find it but lately it is found right there on the main Opera download page. You probably downloaded and used the normal setup which is typically an MSI installer wrapped in an Installshield package, which is indicated by your error message. Its a miracle the thing works at all on Win9x these days. The Classic package is a basic Wise wrapper that only leaves a single registry key uninstall entry which is a very tiny footprint. Win95 has severe registry size limits so my advice is to use as few MSI installer packages as possible (their registry footprint can be gigantic!). This is FYI only since your error message appears to be locale related and not registry size related. The really nice thing about the Classic setup is that you can just grab the files and drop them into folders yourself (if you know what you are doing naturally ). This is very easily done with WinRar as it opens the file directly. This can also be accomplished with WUN (Wise Unpacker) and 7-zip (7z.exe). But before doing it manually, try running the Classic installer of course.

Nice work as usual. I just updated Post#28 with that page you found. It has working downloads directly from the web.archive.org archives of the old dos.li5.org site. Anyone who still needs it can grab SrcTools and some other stuff as of this writing.

All I can say is brilliant minds ... Nice job there. It never amazes me how folks spread out on the four corners of the globe wind up tinkering away accomplishing identical achievements without some how-to guide!

Confused..Must the field lengths be maintained or not exceeded? ie if the DeviceDesc filed is Generic USB Hub can I not replace it with Dell USB Hub? I see no problem there, go for it. But since you asked, don't exceed them (maintain the original lengths). Note, if you export the registry before you edit anything, you will be 100% safe because anything can be undone later with a simple registry punch-in via importing a small registry script. The field lengths are the size of the data values (number of characters) found in the "DeviceDesc" entries. Note once again: Original vs Edited ... |----------------------| <--- original field 24 chars long [color="#2E8B57"]"DeviceDesc"="Generic IDE Disk Type47"[/color] |----------------------| <----- edited field 24 chars long [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (PM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (PS)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (SM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (SS)"[/color] Here is an example with multiple spaces ... [color="#2E8B57"]"DeviceDesc"="Generic NEC Floppy Disk"[/color] You see, the CodeBox above uses a fixed width font. Out here in the normal forum space is a variable font which displays fat characters wider than skinny ones. Compounding that problem is the fact that web browsers typically collapse multiple white-spaces (unless told not to). Now, take a look at how that exact same string looks out here with a different font: "DeviceDesc"="Generic NEC Floppy Disk" When you look at these things in GUI tools like REGEDIT, Device Manager and UsbView, we are viewing with the standard Shell font, and this can lead to inadvertent mistakes. This is why it is imperative to use an editor with a fixed width font for viewing/editing REG scripts (not notepad, etc). But like I said previously, probably the worse case scenario when sticking too many characters into a fixed length field is truncated information in some GUI tool. I say probably because Microsoft apparently stinks at using error checking within their code judging by the enormous quantity of buffer exploits patched over the years (these theoretically can be forced by overloading data input fields in certain situations). Years ago I was editing these entries and noticed that some of my information wasn't displayed in Device Manager. I simply went back and carefully matched the exact size of the Original fields and all was well. I was able to quickly correct that by referencing a saved exported registry so save yours first. Anyway, I have no explanation as to why there is such a wide range of field sizes! There seems to be no discernible pattern. I'll end on this note: here are some more actual entries demonstrating vastly different field lengths and white-space padding: [color="#2E8B57"]"DeviceDesc"="Iomega ZIP 100 " "DeviceDesc"="Microsoft USB Wheel Mouse Optical (IntelliPoint)" "DeviceDesc"="Advanced Configuration and Power Interface (ACPI) BIOS" "DeviceDesc"="Standard 101/102-Key or Microsoft Natural Keyboard" "DeviceDesc"="Intel® 82865G/PE/P/GV/82848P Processor to I/O Controller - 2570" "DeviceDesc"="Intel® 82801EB USB Universal Host Controller - 24D2" "DeviceDesc"="Intel® 82801EB LPC Interface Controller - 24D0" "DeviceDesc"="Ic35l020 Aver07-0 "[/color]

Cool! And thanks for posting back. Yup. I've been doing this myself on Win9x systems for a long time with very few problems. To be safe, just don't exceed the length of the original entry, some of these "DeviceDesc" appear to be fixed length fields (within their respective keys). Here are some I have played around with in the past. Original vs Edited ... [color="#2E8B57"]"DeviceDesc"="Generic IDE Disk Type47"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (PM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (PS)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (SM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 47 (SS)"[/color] [color="#2E8B57"]"DeviceDesc"="Generic IDE Disk Type80"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 80 (PM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 80 (PS)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 80 (SM)"[/color] [color="#FF0000"]"DeviceDesc"="Generic IDE Type 80 (SS)"[/color] [color="#2E8B57"]"DeviceDesc"="Plextor CD-R PX-W2410A"[/color] [color="#FF0000"]"DeviceDesc"="Plextor CDR PX-W2410A PM"[/color] [color="#FF0000"]"DeviceDesc"="Plextor CDR PX-W2410A PS"[/color] [color="#FF0000"]"DeviceDesc"="Plextor CDR PX-W2410A SM"[/color] [color="#FF0000"]"DeviceDesc"="Plextor CDR PX-W2410A SS"[/color] These denote the various positions (PM=Primary Master; etc...). And they each need to be under their proper respective branch naturally in order to work. Notice the field lengths were maintained after editing. Many entries use multiple white spaces that are easy to miss when messing around with GUI tools like Device Manager and REGEDIT that don't use monospace fonts. The length for yours will probably be different, so, just keep that exact length always. Save a registry export, and always view/edit REG files using a fixed width font. Most likely the GUI in Device Manager will just truncate any extended descriptions (or not! buffer overflow?). And of course, sometimes the stupid Windows detection gets re-initiated for no apparent reason and the defaults from some Windows INF file overwrites them!.

I take it you mean trying to configure an INF file and set up registry settings is going to be very difficult. Maybe I should just rebuilt this pc with a USB2.0 enabled board. Just to expand on my previous post a bit, from some haphazard research on USB activity: The usual hardware identification strings (&Ven &Dev &SubSys &Rev &Bus &Dev &Func ...) which get concatenated into registry key names are consistent from computer to computer. They certainly should be consistent since they originate from a common source: the INF file from the driver package or Windows INF folder, and the physical bus it is on, position, etc. Likewise for the values found underneath these keys ("DeviceDesc" "MFG" ... ). What I was referring to up in Post #8 was the apparently exclusive to USB "SymbolicName" entries that are pointers to within that massive HKLM registry structure under System\CurrentControlSet\Control\DeviceClasses. For example: a basic hardware item like a native PS/2 mouse that later gets used on a USB port (via adapter) acquires new registry entries under Enum\Usb where that "SymbolicName" value will appear and it points into that key. As a PS/2 device the entries across multiple machines are consistent, but as USB there *can* be variation underneath ...\DeviceClasses usually from the sequential numbering found there (\##.#0000000000000013# ... etc). So what I was getting at is that manually porting USB devices can be interesting to say the least. Well I'd be 100% certain first that the INF is for the exact same Hub that Dell stuck into the monitor. However the next post was from dencorso which seems to suggest the really good news for you that no INF file might be needed at all. Try that first! But still, I'd snapshot the registry beforehand so you can manually rollback if necessary (save the System.dat also which makes for a real easy rollback ;-).

Definitely not fun! No way to kill the USB 1.1 and legacy serial ports in the BIOS? Do you need LPT? Maybe kill the NIC or modem too (lots of systems have both, but few need both)? This should free up one or possibly two IRQs. Just looked at it. Typical Dell file, just puts the monitor into the registry. Apparently their plan is for the computer to PnP the hub, unless they have some supplemental file somewhere. It is coded for Chicago, so it will be fine on Win98. Its worth a shot to see if Win98+NUSB will recognize the hub. Be sure to do the monitor without the hub first (reboots, and make sure it sticks in display properties), then only move on to the USB wire after it is ok. No sense confusing things If it fails on Win98 my next option would be to try a WinME system (hey, they are good for something! ). They are better USB equipped out of the box naturally. There is much less registry and file activity on Win9x systems which is a great benefit when capturing file and registry settings. BEFORE connecting the monitor you want to first export the registry. After rebooting and doublechecking that the hub is working take the second snapshot. NOTE: porting settings is the hardest method for sure! And porting USB settings is really tough because of the nature of the ConfigMg detection. To put it plainly, most hardware are using identical identification strings and registry from one computer to another, USB stuff almost never is ***. So, when undergoing the massive effort of manually porting a USB device, it had better be worth it! Now if we look at it this way: if we are turning a USB 1.1 mobo port into four downstream USB 1.1 ports on a monitor, hmmm. ***: see later post for a little more!

Most likely an executable program was ran, this program was a 'setup' (aka 'installer'). Setup programs, not surprisingly, set up software by copying files to your HDD and modifying files and the registry. They also usually leave a folder with an executable file that performs the reverse, an 'uninstall'. That is what your folder on the desktop likely is (although it is typically placed in the Start Menu). Visual Basic Set Up Tool Kit was the description of the EXE file found in Microsoft's setup packager circa MSVB/MSVC/C++ 4/5/6 which probably got deposited onto hundreds of millions of computers that ran a Setup/Installer packaged by those compilers. The name of your folder 'sssheet' means nothing to me, and very little to Google. There is some mention of a spreadsheet, but that looks like a zip file at first glance. It doesn't ring any virus/spyware bells. Maybe someone else knows. You need to research your system carefully. Note the time/date of the Folder you found, and then search (Start > Find > Files or Folders) the entire computer for all files with time/dates that are close. Click the middle tab, use the same day, then after the search is done: click the date column in the list to sort the files and look for clues.

Agreed. mikejf, if you don't get the sticker don't buy it. Else you will become one of many who are getting scammed this way. If you have the actual 3D hologram-ish sticker there is practically nothing to prevent activation after a simple phone call and honest explanation. I say almost because if it happens to be a WinXP OEM SLP arrangement that in effect ties this copy of WinXP only to one exact model of motherboard, it will not be legal and most likely will never work anyway. Well, not unless your motherboard matched his original. What you are looking to buy is the so-called 'System Builder' OEM edition that is transferable to another motherboard and comes on a hologram CD-ROM (with the sticker and Key of course). Better yet is the obscenely overpriced full retail WinXP Pro.

Well, since you are planning on springing for a new display I'd suggest tacking on an extra $9 onto the order and get a USB 2.0 PCI card (if you have a slot free naturally). This is because legacy USB 1.1 is too painful to comprehend (IMHO)! Flashdrives, Cameras and External HDD's on USB 1.1? You will be much happier after changing this! Next, I would concentrate on getting this USB 2.0 working first, without ever plugging in the USB wire from the new monitor. Once it is working OK then I would try to get the monitor's USB hub working, but this is definitely secondary (at least to me). I suspect all that is really needed is an INF file detailing the ID's for the USB hub, because all the system files are probably present already from the NUSB package. The pertinent data may even be included in the new monitor's INF (as you probably are aware Dell monitors typically come with 3 files, for example: E173fp.cat , E173fp.icm , E173fp.inf). The INF for a new monitor with a USB hub *may* have the data inside. Or, maybe not. If you have no luck tracking down a Win98 compatible INF file for the new monitor's USB hub, I can think of one way to kind of 'capture' them. I just might hook the new monitor to a WinXP system (or maybe even WinME) insert the driver CD, and make careful snapshots of the registry (and file copy activity) before and after. At the end of this, a .INF or .REG can likely be created. Of course, it's also possible that only a few edits to the INF file from the driver disc might be needed in order to allow Win98 to use it. Finally, failing all of the above, there are always cheap external hub/card-readers. This may be the smartest option since popping it into the USB port only when needed simplifies the whole multiple drive letter or hanging on shutdown that sometimes happens on Win9x with permanent mounted readers/hubs.

@Multibooter ... FYI, Svend Olaf Mikkelsen also developed a method for live booting from floppy images a few years back. I won't bother explaining it since he has it pretty thoroughly documented on this page The Boot Floppy. It is 9x based (well, FATxx actually). I meant to mention this a while ago, sorry I got busy elsewhere!

You can report freely your results regarding the patching of System Internals utilities. What you cannot do (Rule 1.b) is to actually redistribute the patched binaries (nor the unpatched binaries, but, for them, you can provide links to the original software provider/vendor site, if you so want). I thank you sir!

UPDATE TO POST #1 The top post has been re-written to describe the current ZIP package suite (downloaded on 2009-08-26). Win9x fans will notice that one more utility no longer operates. -CTH-