Drugwash

I got a similar setup, but as always, didn't make notes on the exact steps taken, so it's up to you to figure out a few things. After setting up the 3G modem with drivers and all, set the respective connection as default Internet connection (Control Panel > Internet Options). Then enable Internet Connection Sharing (ICS) which will basicly do all the hard work for you: IP of the main machine will be changed to 192.168.0.1 and the connection will be "forwarded" to all other machines connected to the local network. At some point it will ask you if you want other machines to be able to dial in the connection when needed (phrasing may be slightly different but the basic idea is the same); choose 'Yes' so that any of your machines would be able to dial in the connection without you having to manually connect (this way, the main machine could work keyboardless and monitorless if you're short of peripherals). The nice thing is you don't need a router since the other machines could connect directly through a switch (if you have one). But if they're unable to connect through wired connection, then you'll have to set up your router so that the IP of the main machine will be its gateway to the Internet. It's been a couple years since I last used my router and kinda forgot its internals, so you'll have to poke around. Just make sure to set the internal network class in the router different from 192.168.0.x as it would interfere with the IP of the main machine.

At this point, I don't think any of us needs any more proof that Win9x users are going to be swept off of the www soon. While logged onto Gmail, I switched to Standard view for a quick glimpse - a lot of useless gimmicks that make loading slower (I'm on GPRS which is some kind of dial-up with limited traffic and speed), but the ugly trick was no way to go back to Basic view! One could only do that after logging out and logging back in, when there's a short "window" of time when such choice is visible at the bottom of the screen. When that choice dissapears (and it will, at some point), then we're left out with a crawling, bloated (and possibly, soon non-functional on Win9x) Gmail - not that they'd care or anything. My IE6 (with or without SlimBrowser as front-end) is crashing on more and more sites, including here at MSFN. Sure, it may just be my machine, fed up with all kinds of libraries ranging from WinME to Server2003; the crash is in MSHTML.DLL, which is an XP version, but may be related to something else. K-meleon, the browser I chose for a slightly higher speed than Firefox 3.6.x, is killing the video driver every so often, which leaves me no choice but reboot. Same behavior as with Firefox, on top of the lack of page refreshing, which produces artifacts (especially on pictures) when scrolling up/down. Dunno of any other browser able to run on Win9x - with or without KernelEx - that would not exhibit such behavior. And messing with the system now would be just asking for trouble, considering it's my development environment.

It's built into Revolutions Pack AFAIK and I already have RP installed; always had. Firefox might have a severe GDI leak that's too much for GDI Heap Extender to handle. No worries, I don't like it anyway; never did.

Thanks, I'll check the Firefox site for an update, but for now I'm fairly at peace with K-meleon. My system is a cocktail of 98/ME/2000/XP/2003 files, which has endured countless crashes during testing with alpha-staged and/or badly coded apps (including my own ), but it's still standing and going strong. The folders Temp, Temporary Internet Files, Cookies, History are automatically deleted upon each start-up (granted, this can happen every 2-3 weeks, at times, unless I screw up ); as for the registry - I'd rather clean it up manually, when needed, but - except for recent browser testing - I haven't installed foreign apps in a long time and my own scripts/apps usually don't use the registry, so it should be in a fairly OK state. Defragmentation is however a delicate matter, since I have a 160GB HDD but the BIOS is not LBA48-aware so it only gives me 137GB of available space, which required LLXX's BigHDD patch, therefore the ScanFrag package must be used. All in all, I'd rather leave it as is than risk losing all the unsaved work and especially the system settings that proved almost perfect for the last 5 years. Apologies for the off-topic.

Gmail started replacing the log-in page and will soon remove the Basic HTML page as well. My main mail account is also Gmail and I've been through this. Firefox has far bigger issues: I have 3.6.20 (beta, probably) and at some point - not sure what the trigger is - the GDI system gets screwed up, artifacts appear and stay on the screen and the only way out is a reboot. Same thing happens with K-meleon 1.7.0, which uses the same Gecko engine as Firefox, just a little bit behind with the version. Apparently after a longer usage time, however. Symptoms usually appear after navigating through Wordpress blogs while logged in (I have a blog there - don't bother unless you know Romanian); there's a stupid toolbar attached to each WP page visited while logged in. Until last week I've been using SlimBrowser on top of IE6-SP1; since quite a while ago, SlimBrowser started crashing MSHTML.DLL when trying to open Wordpress blog pages, but also here at MSFN, reason why I eventually installed alternative browsers. BTW, anyone knows other browsers that'd be up-to-date with latest web/HTML changes and might run on 9x, with or without KernelEx? I also have Yahoo accounts but so far I denied the switch to new version; won't be long until they force it upon everybody though. Call me paranoid, but I don't think it's a coincidence IE6 crashing hard on updated (presumed HTML5) pages. This already cuts out 9x's chances to access the web (what could Win95 guys use?) and next step will probably be IPv6, a subject always buried here. As for Spybot, I only had it installed once, long time ago and it blocked a few sited I used to visit; promptly uninstalled it and never again ever thought of reinstalling it. SpywareBlaster on the other hand still supports 9x (at least it installs and runs with KernelEx active) and offers a certain degree of protection. Useless, I'd say - a couple days ago I made the silly mistake of connecting a foreign, knowingly infected machine to my local network and it instantly popped Explorer on my XP machine (which holds the Internet connection); my main machine - the 98SE I'm typing on now - was whistling, unaware of XP's tragedy. But this topic is already worn out, no need to add to it. (darn typos...)

Looks like I forgot to post back: after taking all required steps "by the book" and reinstalling the new 3.5 package, everything is fine and works correctly. Thank you maximus-decim for another fine release!

@ Tihiy: Apologies if I've raised this issue before; can't remember and search doesn't help. One of my applications, designed to load and display VersionInfo and icons stored in exe, dll, ocx and other similar files, crashes the current thread when enumerating the dlls in the System folder. Particularly, it crashes on uberskin.dll and comctlv8.dll. Fortunately, it's a separate thread so the application continues enumerating the rest of the files, but the crashes are somewhat disturbing, more so for a user unfamiliar with the application, who might think of it as being buggy. Of all the libraries in the System folder, only those two exhibit that behavior, therefore I consider this as being the fault of RP. The application is using LoadLibrary() with the flag LOAD_LIBRARY_AS_DATAFILE. Same thing happens in Total Commander while launching the Lister on the files with the FileInfo plug-in installed, which essentially does the same thing (amongst others). The application is called VersionInfo Lister and can be found on my AHK homepage (see sig below). Both executable and source code are available for download. If that would be too hard or impossible to fix, at least I'd like to know it so I could add a warning/disclaimer for the user, in my application. Thank you.

That'd be my intention but it'll take quite some time before I figure out all the tricks. The big problem is, I don't have access to any of those pieces of hardware you're talking about., so actual testing would have to be performed by people who do. All I have at hand is a floppy drive removed from an Atari before throwing it away; no idea about it since I never managed to properly connect it to a PC. The boot disk seems to be OK and I just made an image of it; it's in German though. Don't hold your breath on the direct boot manipulation with BootMaker.

I'm getting old and productivity is not that high now so development is slow. However, here's a glimpse of what will be: Obviously, I will try my best to create a good editor. All formats in the drop-down list will automatically yield the proper values and the File system dropdown will add the corresponding boot code and - when possible and allowed by the license - also the required files or otherwise prompt for manual addition by the user. At some point I thought of adding a disassembler, but that's still far from becoming a certainty. I also plan on reading/editing available HDD/FDD/USB boot sectors but that's also far from implementation. For now it can edit and save .ima images of either floppy or hard disks. Currently, editing can be done at bit and byte level (hex), since I haven't yet found a reliable way for ASCII editing. String editing is also available but it doesn't update the editor window, that's why I haven't released it yet, since it could create big problems. If anyone has other suggestions, please speak up; I'd be glad to comply, if and when possible.

I think a line should be drawn, and fast, because it's been too much talk about sticking "this" or "that" format into one standard or another. Is anyone here member of a Standards Association of some sort?! Personally I'm getting more and more confused by all this information which deviates the course of the initial topic: "bootable CD floppy emulation". Since it should be about booting, why don't we focus on BIOS capabilities instead? I've read about some old Dell BIOS versions that wouldn't clear a carry flag, thus malfunctioning in certain situations and there may be other issues that need dealt with. If the purpose of the current topic is still making the largest possible floppy emulation image, then first find a common denominator for BIOS capabilities. I've had machines that wouldn't boot a ISOLINUX CD but would gladly accept Windows 9x or XP install CD. I've had machines that wouldn't boot any Linux flavor. How exactly does BIOS makes the difference between Floppy and HDD (image): is it the media descriptor (IBM's 11111RED byte)? Does it have any internal limitation on the number of C/H/S recognized for a certain media type (aside from or along with LBA/LBA48)? More questons I have not yet found a definite answers for: - why, exactly, choose floppy emulation over HDD emulation? - why, exactly, make a single huge image instead of the classical 1.44/2.88 image plus creating a RAM drive, when the superfloppy wouldn't be writable anyway? - what file system would be best for such boot image, considering all capabilities as well as limitations of different file systems? There's probably more questions waiting for an answer, but that's all that came to my mind, for now. As for BootMaker, it'll still take a few days before I can release a safe version, after the changes I've made. That is, if it's still needed, because I haven't yet heard any constructive criticism about it, other than the labeling issue.

Please, those of you who had the "C:\Program" error on setup, would you be so kind and provide a little bit of information to help in fixing this error? If so, open the registry editor (Start > Run > type REGEDIT > Press OK or Enter), then navigate to the following path: HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Setup\VarLDID Then please post the exact values for the subkeys 28700, 28701, 28702. If the path and/or respective subkeys do not exist on your machine, please post all the same so we acknowledge it. The language of your Windows OS might be useful too, in case it's not English. Thank you. @MDGx: I'm not at all familiar with VB scripts, but when trying to run the script that finds the "Program Files" environment variable in registry (created by O2.BAT), I get the error stating "This program cannot be run in DOS mode". Since the users have not reported this error, I suppose I'm just doing something wrong; however, running the script manually under Windows does pop up a MessageBox stating SET PF=C:\PROGRA~1, which is the DOS (short 8+3) name of "Program Files". Since the error reported by users is "cannot find C:\Program", I believe they might have the Windows (long) name in that registry location and the space in the Windows name confuses the script, because it is not "escaped" in the O2.BAT script. All it needs is to "escape" the Program Files name so it would yield SET PF="C:\Program Files" in case that's what the 28700 registry subkey holds. The change I suggest for the corresponding line in O2.BAT is the following: @ECHO WScript.ECHO "SET PF=""" %AN% sh.RegRead("%LM%\%VL%") %AN% """">>%TEMP%\$PF$.VBS (note the added quotes after SET PF= and the %AN% """" before >> ; I would've made them red but the colors panel won't show up, for some reason) My values for the aforementioned keys are: - 28700 > C:\PROGRA~1 - 28701 > C:\Program Files - 28702 > C:\PROGRA~1 [EDIT] I just noticed the same technique of finding the registry value of Program Files is used in O3.BAT for Movie Maker and in O6.BAT for Pinball (again) and GameZone bunch of files (I presume the MSN games mentioned prevoously in the topic). Therefore the fix I posted above should be applied to these two batch launchers as well. The only thing I'm not sure of is whether the DOS interpretor will correctly parse the quoted PF variable when it contains the correct short form of C:\PROGRA~1.

Sorry for the delay - just woke up minutes ago. @dencorso: Yes, that's the package.I had downloaded it on July 22nd. Let's clear up the hardware: - motherboard Soyo SY-6VBA 133 - chipset VIA VT82C693A (Apollo Pro 133) - on-board USB 1.1 UHCI controller VIA VT82C596B - PCI adapter with USB 2.0 EHCI controller VIA VT6202 Both USB controllers have been working correctly with NUSB33 (and previous versions, back at the time). @LoneCrusader: Yes, I did install the USB 2.0 Stack package on top of NUSB33, primarily because I was in a hurry and secondly - and most important, IMO - because such upgrade should be compeltely transparent to the user, that is it should require absolutely no prior uninstall/tweaking of other system services or whatnot, since not all potential users are knowledgeable. As such, I have not removed any of the USB devices that appear in Device Manager, especially because I have other USB related problems (one USB stick doesn't get assigned a drive letter and the brand new bluetooth USB adapter got VID & PID zeroed out) that I probably will elaborate on some other occasion and I was afraid that particular stick wouldn't be recognized anymore (which happened previously). The files I added manually right after installing USB20DRV were: - openhci.sys 4.90.3000.1 - uhcd.sys 4.90.3000.1 - usbccgp.sys 5.1.2600.5512 (from XP-SP3) - usbd.sys 4.90.3000.1 - usbhub.sys 4.90.3002.1 - usbser.sys 5.1.2600.5512 (from XP-SP3) I believe the issue with Composite devices is deeper than it may seem, since the class doesn't appear to be registered on a normal 98SE system and probably other changes and system files may be required as well. But for the time being I'm not gonna dig into this, as I got other important work to do (dencorso knows).

Not yet. This needs further testing. Today I made a second attempt at installing the USB 2.0 stack together with all other system files recommended in post #628. Upon reboot(s), the USB Root Hub entries for the on-board USB 1.1 failed to load the drivers (NTKERN.VXD could not load the drivers for this device). Tried reinstalling drivers, removing and rebooting, etc. to no avail. Upon uninstalling USB 2.0 stack and restoring system files from backup, the machine locked up. Tried a couple reboots - same result. Safe mode did work but wouldn't fix anything, so after a few failed boots I had to restore the registry from the most recent valid backup, at which point the machine got back to a working state. This 98SE installation is 5 years old and lived through countless alpha tests, crashes and other mishappenings but only once or twice was I forced to actually restore the registry to get it back afloat, so I'd say there might be something fishy with this package, since it breaks the 1.1 controller and then corrupts the registry upon uninstallation. Or am I Mr. BadLuck?

Just couldn't help it... First alpha of BootMaker is so incomplete it makes me ashamed for ever releasing it. So I put a little bit of elbow grease in it and got to a more acceptable version. Please note it's still in the viewer phase and the File system selector is not yet functional. The Save buttons have been temporarily disabled to avoid any misleading. The Code viewer will eventually become an editor (hopefully); for now it doesn't respond to user input, save for the right side where code can be copied from. Oh and the acronym was wrong in v0.0.0.1: it's BMF, not BFM. Sorry! Now it's fixed. Enjoy! (and now I'm going to sleep; it's past 6 AM) BMF 0.0.0.2.exe

Thank you for the warm welcome, Dencorso! I missed you all too. Since I got drowned in a sea of documentation following the links kindly posted throughout this topic, I'll be spending a fair amount of time studying the available information before coming up with a usable version of the BootMaker tool. I also intend to make it translatable so it could be used widely even by non-English speaking people, being able to display labels as they see fit (to avoid issues such as hidden vs before). So if you don't see me posting here, don't think I'm AWOL. Will keep reading new posts.

If you checked out UltraISO you'd notice that is the "friendly name" by which a file system is selected. Currently it doesn't respond to image file loading because I was too tired this morning at 6 so I had to wrap and go to sleep (yeah, bad habit, I know...). But when working, it should just be the friendly correspondent of OEM string. Sectors are not 'small' or 'large', but the count is 16bit (when sector count doesn't exceed 0xFFFF) or 32bit when it does. So I'd rather modify to 16bit/32bit than use a misleading denomination (which actually made me wonder at first glance what this small/large is all about). Same goes for 'Sectors before' - I kept wondering what that meant until I found documentation that called it as it should: hidden. Anyway, the source is bundled in the exe so it can be easily modified and recompiled if it's so critical. Beer is still not out of my system... for the simple reason I'm still pouring it in.

I wonder where bit 8 and 9 reside in a one-byte media descriptor... But I'll try to digest that once beer gets out of my system. While I enjoy having fixed my broken roof, have a look at this alpha-stage tool try the newer version below. It can only read and display (partly) floppy image info, for now. It will be able to create/edit one later on when I would have figured out all the pieces of the puzzle.[attachment removed - new version below]

Oh, but I did. For the alpha version it's been a good start., but after reading a few web pages linked to in this very topic, I got more and more lost in the fog. There's no fixed standards, speculations are being raised and finding a reliable formula for all situations is not an easy task. Moreover, information comes from unknown sources so the level of trust is not too high. For now, my app can read and display most of the attributes of a floppy image in ima format. I've tested with images created by UltraISO, which is a handy tool I've already had on my system. BTW, how come it has options to create 120MB and 240MB floppy images, while this topic is focused on 36MB images that aren't even listed in UltraISO's options? Oh and don't worry - all my applications are open-source, therefore free of any charge and only bound by the GPL license - if anyone wondered. I'm poor, but still refuse to pray to the god of money. Anyway, this is off-topic.

Please, could any of you be so kind and take some time to send me a PM or e-mail with a very detailed description of the goal of this endeavour as well as any usable information that could help me understand the boot process and especially the why of using a huge floppy image instead of the other classical way(s)? Admittedly, I've never delved too far in the boot process so there are many blank spots that need filled with information. I have however started to build a simple application to create/edit floppy/HDD images, which should ideally avoid the use of that not-so-classy Excel sheet, so any help would be appreciated. In the mean time I'm gathering as much information as possible, but I need time to process since I'm a (very) slow thinker. Thank you in advance!

Back after a two year leave from the Internet. Looking at my desktop, the 32bit icons show a black alpha channel. Can't recall anyone having reported this for the reason I suspect (in my case): long time ago I changed shell icon size from 32px to 24px and most likely the GDI function(s) that resize the 32bit icons don't play nice with the alpha channel. Now, I don't care much about this issue since I'm mostly working in maximized state with whatever application needed so the desktop is way behind all the time, but just for the sake of it: can it be worked around by using oher GDI functions (or GDIplus if not too cumbersome)? System specs (if needed): P III 667MHz, 512 MB SDRAM, video nVidia GeForce 4 Ti 4200 AGP8x (running in 2x mode due to chipset limitation), video driver 45.32 (latest available from manufacturer MSI, since recent ones disable AGP texture acceleration in DirectX). OS: 98SE heavily upgraded, RP 9.7.2, KernelEx 4.5.1 installed minutes ago (former version 4.5 RC5). Glad to be back here among friends, but sad to be forced to use XP for Internet due to lack of drivers for the GPRS modem. *sigh*

For what it's worth, I've been working for a little while with the Docx2Rtf team back in 2009, submitting a few GUI bugs and a secondary translation for the Romanian language and in the process I've tested a few versions that were being released at the time. As such, the last version I've installed back then was 3.8, which - as I've just tested and can be seen below - does work under 98SE, with KernelEx either disabled or enabled. It does have a lot of dependencies though, and it's possible one may have to update a few Windows system files before getting it to work. However, I do not have a personal webspace to upload it to (if anyone cares to try it) and those free hosting sites are not a valid choice for me, so if anyone cares to host it, just PM a mail address and I'll send it over (hopefully the DVD I've saved it on still works).

Indeed, Tihiy's site appears to be down. If it doesn't come back stable, soon, could anyone please PM me an alternative download link (or an attachment, if the board allows the size) for v9.6.5 or whatever may be the latest at the time? It might take me more than a week to get back here and I won't have much time on my hands. Thank you. @ Tihiy: As mentioned, I haven't had a chance to test v9.6.5, but with 9.6.3 the horizontal scrollbar went far right when clicked, in an "Open" dialog. Vertical scrollbar seems to behave correctly, Metapad included.

Confirm issue #4 on CharlesF's list (scroll thumb frozen until mouse button is released). To me it happens in Metapad (so far) and that is unfortunate because it's my only editor and I use it for developing my scripts. Another minor issue is related to themes: the slider (trackbar) bitmaps - which are 32bit - appear incorrectly on a 24bit display (Intel 82801 integrated graphics). Reducing color depth of the related bitmaps to 24bit (magic pink transparency) fixes the issue. It strikes me as odd having both 24bit and 32bit bitmaps in a theme - in regard to consistency - but it's not that big of a deal. Sorry for being no help lately - still no Internet access at the new location and my only machine at the old location either crashes or connects very slowly to the board, especially since the theme was changed. However, Tihiy, I'm still interested in RP and anything else 9x-related. Keep up the good work!

OK, assuming that one has a previous install of Windows with RP 6 lying about, perhaps backed up via backup9x.bat, then one grabs the command.com and io.sys files from it, putting both in c:\ and also the former in c:\windows\, and one grabs logo.sys and wll.com, putting them in c:\windows\. Then it works like a treat. Edit: well, not entirely, as there is a problem with scandisk when restarting after a crash. Personally I used TweakUI 1.33 (the Boot tab) to disable DOS Scandisk's autorun. If required, I always run the Windows version of it. This would take care of the WLL issue but there would still be minor similar issues after (un)installations or system updates. There's also a topic I opened some time ago where some of us tried to fix/improve Tihiy's original WLL but we never got to a final, stable version and unfortunately development stalled; links most likely expired too but I still have last versions somewhere around if needed - might take some time to post new links due to my very limited access to the web lately.

Apologies if I'm repeating myself: for (basic) image editing I use FastStone Image Viewer and Chun Sejin's Imagine (as both standalone and Total Commander plug-in). However I still miss the ability to crop and paste from different image files using positive/negative zoom levels; having scanned some documents I tried to copy/paste certain parts to create a printable image that fits to paper size, but failed even under XP with its MSPaint. If anybody knows of a free utility that can do that, please do tell.