98 Guy

Update: Up until now, I've been preparing (partitioning) this drive by connecting it to a SATA port on another machine and running the Samsung version of On Track's disk manager software (via boot-CD). Most recently, I've kept the drive in the destination machine (Levono R61i laptop) and ran the disk-manager software on it. One of the first questions that the software asks is what OS you intend to use. It lists the options as: Win-XP Win-2K Win-Me Win-98se My previous attempts were using the Win-ME choice, but this time I chose XP. It then wanted me to check (or not-check) a box if I was installing SP1 or higher (which I was, and I did). I then went on and re-partitioned the drive as a 64gb primary (4-kb cluster size) and 96 gb secondary(8kb cluster size). I then booted the XP-pro CD and began XP installation. > Two questions: > > 1. Aren't you offered the option of re-formatting the drive with the same file system? No. During XP installation, once you select the destination partition or volume, you have (basically) only 2 choices: Reformat the volume as NTFS, or keep the existing file system. No choice to re-format as FAT-32 (if such a choice ever existed, I don't see it probably because of XP's intentional limitation of 32gb for FAT-32 volumes). > 2. Is the partition you are installing XP on marked as active? Yes - in all previous attempts it was active. OnTrack's DM software does set the first volume as active. So at this point, I've told DM to create a 64 gb Fat-32 partition, and that I intend to install XP on the target drive (instead of ME). The XP installation proceeds with about 10 minutes of solid file copying from the CD, and then upon first boot it actually does work (no missing NTLDR message). So what's different here is: a) The hard drive was connected to and OnTrack DM was run on the destination machine instead of a surrogate. b) DM was told that XP was the target OS instead of ME. I'm thinking that (b) was the item responsible for the success, but I wouldn't have thought that a FAT-32 volume needed any special preparation or setup to allow an NT-based OS like XP to boot from it. PS: After the machine booted the first time during XP install, and it began to ask questions as part of the rest of the install, I shut the system down and copied the XP-cd to a subdirectory on the hard drive. I then resumed XP installation and told it where it could find the install files. Problem is, I was CONSTANTLY being asked where the files were, and I was CONSTANTLY browsing to the same location. Why doesn't XP remember the alternate source location during install?

No, I used On-track's Disc Manager, which has been customized for use by various hard-drive makers and branded as their own software for use only on their own hard drives. In my case, I obtained Samsung's version here: http://www.samsung.com/global/business/hdd...iskManager.html Specifically, the bootable CD ISO image here: http://www.samsung.com/global/business/hdd...CDROM_Drive.zip During use, you specify a) the OS you're planning to use (win-95, win-98se, ME, 2K/XP). For some reason they separate Win-98se and Win-me as separate options. b) the type of file system (NTFS, or FAT16/32 as determined by your choice in A) c) then you specify the size of each partition, and again you can choose FAT-16 or 32, and then the cluster size When the operation is completed, the partitions and volumes have been created, and the volumes have been formatted, but they are completely blank (no bootable sector or system files present). I have found that very few other hard-drive preparation tools (PowerQuest, etc) allows for custom cluster sizing when creating FAT32 volumes. Seagate and Western Digital also have their own versions of this DiskManager software. Just FYI - using the above method I've been able to install win-98se on the drive in question (as installed in the laptop in question) as well as in other situations where I've installed win-98se on a 500 gb SATA hard drive prepared as a single fat-32 partition with 4kb cluster size. I'm under the impression that Win-2K and XP are both capable of dealing with large FAT-32 volumes (larger than 32 or 64 gb) but they've been handicapped in that by design their own native drive preparation tools are prevented to create FAT-32 volumes larger than 32 gb. So, now you know how I prepared the hard drive, do you have any ideas why XP is throwing up this error message upon the first restart after OS installation?

Please ignore for a moment why I'm trying to install XP on a FAT-32 partition. What I have is a 160 gb drive that I've used manufacturer's software to partition as a 64 gb primary FAT-32 partition (using 4 kb cluster size) and the remainder of the drive (96 gb) also as FAT-32 (8kb cluster size). The drive is SATA, the computer is a Lenovo R61i laptop, and the drive was set to compatability mode in the bios (so no SATA drivers are needed during XP install). I've booted the computer with an XP CD (XP-pro, OEM "System-Builder" version, version 2002 with integrated SP2). During XP installation, XP sees the drive, sees the 2 partitions, asks which one I want to install XP on, asks if I want to keep the existing file system (yes), and then performs the first phase of installation (a lot of file copying). It then performs a re-start, and that's when I get the message "missing NTLDR". Some fixes for this include making sure that NTLDR and ntdetect.com are present in the root directory (they are). Microsoft seems to think that the only way this can happen (installing XP on a FAT-32 drive) is when you're attempting to install XP over a pre-existing Win-9x install (which I'm not) so I can't see how a lot of their solutions would or could apply here. About 2 years ago I did something similar - I installed XP on a large (250 or 320 gb SATA drive) that was formatted with a couple of FAT-32 partitions (again with 4 kb cluster size) and the installation went smoothly in that case. Any idea what's going on here?

Does anyone know if the driver packages available from driverpacks.net have been re-worked to remove any win-9x drivers? Or do they / will they contain any win-9x drivers if they were part of the original source?

If you do a google search, such as: http://www.google.ca/search?hl=en&q=ic...earch&meta= it appears that there are some ICH8 components with win-98 drivers. Is there a single intel chipset utility file for ICH8 that might (or does) contain some win-98 drivers? Is there a modded driver file somewhere? For example, on this page: http://www.xpvistaworld.com/intel_r_ich8_f..._2_2841_643.htm They show this: Intel® ICH8 Family PCI Express Root Port 2 - 2841 driver ----------------------------- Developer: Intel Version: 8.13.1 Filename: Intel_infinst_autol.exe Size: 10.03 Mb OS: Windows XP,Windows ME,Windows 98 SE,Windows 2000 SP2,Windows NT SP2,Windows Vista,Windows 95,Windows 2000 SP4,Windows NT SP1,Windows 2000 SP3,Windows NT,Windows 98,Windows NT SP4 License: shareware http://downloadmirror.intel.com/13326/a08/infinst_autol.exe ------------------------------- But that link doesn't work...

> win.com /B /D:M The /B argument is either wrong, or undocumented. The above command (when typed in as written) results in a safe-mode startup, but an error saying something like "can't find" or "can't run" the file "/B". But in any case, it does seem to create a bootlog.txt file. And BTW, is it normal to see these entries: [0015EDFE] Starting Unknown (HTREE\ROOT\0) [0015EDFE] Started Unknown (HTREE\ROOT\0) [0015EDFE] Enumerating Unknown (HTREE\ROOT\0) [0015EDFE] Enumerated Unknown (HTREE\ROOT\0) I'm also seeing lots of these font load errors: LoadStart = C:\WIN98\fonts\MARLETT.TTF LoadFail = C:\WIN98\fonts\MARLETT.TTF Failure code is 0016 The following is the last few lines of my safe-mode bootlog.txt: LoadSuccess = USER.EXE LoadStart = MSGSRV32.EXE LoadSuccess = MSGSRV32.EXE Init = Final USER InitDone = Final USER Init = Installable Drivers InitDone = Installable Drivers Init = TSRQuery InitDone = TSRQuery [0015F051] Starting Unknown (HTREE\RESERVED\0) [0015F051] Started Unknown (HTREE\RESERVED\0) [0015F052] Enumerating Unknown (HTREE\RESERVED\0) [0015F052] Enumerated Unknown (HTREE\RESERVED\0) Are these unknown entries caused by the /b startup option?

I've got a win-98se system that boots ok most times but sometimes immediately after all the desktop icons are rendered I'll get a blue-screen "Fatal Exception OE" message. When I hit a key, the desktop comes back, the icons are flashed/rendered once or twice, and then the blue screen comes back. This exact sequence happens in safe mode as well. It basically gets stuck in a loop where it shows the Fatal Exception message, then you hit a key to see the desktop come back, then at the very moment the desktop looks ready for use the blue-screen fatal exception 0E comes back. When started with the logging turned on, the very last line in bootlog.txt is something like "enumerated microsoft kernel audio mixer". Scandisk with full surface scan turns up no errors. Scanreg also says nothing is wrong with the registry, and restoring several older versions of the registry does not make this problem go away. I'll power it down for a day or overnight, then turn it on and it will usually boot and work fine. Any ideas? PS: Is there any way to turn logging on (create bootlog.txt) for a safe-mode startup?

> A convenient way would be to borrow rules from spamassassin, > but they don't appear to do this I have many rules for filtering spam based on the X-mailer line that I use with my e-mail client (Netscape Communicator 4.79). I don't know how powerful or flexible spam assassin is, but I really don't need to use a third-party pre-filter. > presumably it is so easy for a spammer to find a real version number to use? Yes, many spams have real or actual legit versions of OE, but as I've posted, I'm seeing new OE versions every week, and I can't find any indication they are real. > The long winded way would be to download all the historic OE > patches for all windows versions (and all languages?) and check > them all... I do web and usenet searches for the OE versions I've posted, but (a) I get only a handful of hits, and (B) all the usenet hits are in NANAS. In the past day I've got 3 more: 6.00.2800.2963 6.00.2720.1409 6.00.3790.2963 Does anyone know what 6.00.2720.xxxx is possibly for?

I've been looking at some spam I've been getting recently and noticing specific OE version numbers in the X-Mailer line. I don't see those same versions in my own e-mail. Here are some examples: 6.00.3790.4682 6.00.3790.2962 6.00.3790.1106 6.00.3790.181 6.00.2900.2963 6.00.2900.2969 6.00.2800.2962 6.00.2720.4682 6.00.2600.4682 6.00.2600.1409 Is there a conveinent way I can determine if any of those are actual OE versions? Possibly for which OS, and when they would have been the current OE version?

This is a system running NT4 server. It runs in a basement machine room, and can run for weeks without anyone checking on it. I noticed last night that it was performing periodic heavy disk access - about 1 minute of continuous disk access every 5 minutes. During the 1 minute of disk access, I see this process in the task manager: Image Name: Error-Handler PID: 282, 290, (I guess this is not important) CPU: 90% + memory usage: 2 to 6 Mb (seems to fluctuate) When disk access stops, that process goes away. When I look at the various logs in the event viewer, I see no entries related to an error condition. How can I track down what error or condition is causing this process?

I've got a win-98 system (one of several that I operate, own or otherwise maintain) and this particular system should just be wiped and have win-98 reinstalled from scratch... Anyways, on this system I will experience a situation at least once a day where if I have outlook open, and if I open a browser (ie6) it will immediately cause outlook to crash, with the message something like "outlook caused something or other in module riched20.dll". I've turned up a very very few web hits in various obscure (and old) forum posting that describe pretty much the same issue, but there's never any resolution or fix posted. I've replaced riched20.dll with another copy (same version - 5.30.23.1200) but that doesn't seem to help. I'm also not sure where this file needs to be located. Is it c:\windows or c:\windows\system ? Any ideas?

What's the deal with XP-SP3 ??? Is a non-beta version available yet? If not - when? Will it be rolled / stamped into the OEM/VAR System Builder CD before Jan / 2009?

I'm trying to track down a problem with a graphics app that uses OpenGL. I'm concentrating on 2 different win-98se systems. One one system, the app runs. On the other system, the app doesn't run properly (and upon closing, I get an invalid page fault in module <unknown> message). I've compared the files in the process list between the two systems, and they differ on two files: System 1: user32.dll 4.10.2227 msvcrt.dll 6.10.9844.0 System 2: user32.dll 4.10.2231 msvcrt.dll 6.10.9359.0 The app runs on System 2, and doesn't run on System 1. I'm going to swap those files around and see if that fixes the problem, but I've got lots of different versions of those 2 files and was wondering what are the last, or maybe "best" versions of those files?

Regarding USB-attached mass storage devices such as memory sticks, external hard drives, external device with their own internal storage, etc: Does an Operating System (Windows 9X, 2K/XP, etc) interact with these devices such that it sees them *exactly* as, say, an internal hard drive, to the extent that the host computer can format the storage device with a file system of the user's choice (FAT-12, FAT-16, FAT-32, NTFS, etc) or does the host computer interact with these devices at a higher "layer" where it does not know, or does not see, or does not have control over the file system being used by the device?

The following is a list of AGP and PCI-e cards showing current price and availability as quoted to me by a parts vendor. Given a motherboard (such as the VIA-based Asrock) where there is a working PCIe bridge being shown by device manager, and also given the right hacked or modded driver, then: Which of these are known to work (or to NOT work) on 98se Asus EN7300GT pci-e Fanless -- $69 - Readily Available XFX 7600GT pci-e -- $119 - Limited quantities eVGA 7600GT pci-e -- $119 - Limited quantities Asus EAX800XT pci-e -- $579 or $749 - Different versions - Limited XFX 6800GT AGP -- $289 - Very Limited Albatron 6800GT AGP -- $359 Very Limited BFG 7800GS AGP -- $279 Readily Available Sparkle nVIDIA 6200 Ultra 128MB AGP - $49 I would think the Sparkle 6200 would work no problem, but it would be my last resort because it's a low-end card. I'd rather try the Asus EN7300GT or eVGA7600GT but the messing around I did about a year ago with basically any PCI-e boards on the Asrock motherboard didn't work. The ATI board above is crazy-priced so it's not an option. What sub-$200 ATI boards (PCIe or AGP) should I look at?