Nomen

Firefox 2.0.0.20 decided to crash on me today, and threw up this error: Firefox caused an invalid page fault in module XPCOM_CORE.DLL I didn't even know there was such a DLL. I searched my system and found 3 different versions of it. Strange info in the file-properties: Version "Personal" 2/19/2008 Version 1.8.1.20: 2008121709 12/17/2008 (probably the one I was using) Version 1.8.1.22pre: 2009041504 4/15/2009 Copyright: License: MPL 1.1/GPL 2.0/LGPL 2.1. Company name is either Netscape or Mozilla. No file description. Just a fluke, or is there a better (more stable) version somewhere?

This add-on also provides the desired functionality for new tabs: NewTabURL 2.2.3 August 13, 2011 https://addons.mozilla.org/en-US/firefox/addon/newtaburl/ Select the default URL when new tabs are opened. The options are: "A blank page", "The home page", "The current page" or a user defined URL. When there is an URL in the clipboard, it is also possible to open automatically that URL in the new tab. It does function correctly for FF 2 running on win-98.

This is really useful. I downloaded and installed the zip file (mentioned in the older thread, modified the .cb file as described, and restarted my system in safe mode. My screen came up as 1024 x 768, 256 colors. The screen-resolution slider bar was fixed at 1024 x 768 (I couldn't change it) and 256 was the only color choice. This is with an Nvidia 6200 AGP video card.

Over the past few days, I've been setting up a new install of win-98se on a clean drive on a 3 ghz P4 Soyo i845 motherboard. (And I've installed 98 SE SP 3.18 just prior to installing KernelEx). Something I noticed is that I wasn't able to install any version-11 version of flash, at least not as the first version the system had ever seen. I eventually got some version of Flash-10 working, and then was able to get the most recent Flash-11 working. So for someone working with a new install of win-98, you might have to install flash-10 first before trying flash-11. Oh - yea, I might as well mention that even though Firefox 8 does install and run, I don't see the point of using it when the bookmarks and recent-history don't function.

I'm not sure if I should have asked this as a new subject under > Windows 9x Member Projects or (like I'm doing now) adding it as a reply to the > 98 SE SP 3.18 thread, but... Here's a dumb question. What is the suggested (or required) starting point when installing 98SE SP 3.18? I mean, after installing 98se on a virgin computer, and installing all drivers (motherboard, video, etc), what else should I go out and install *before* I run 98SE SP 3.18? If I intend to have IE, should I go and get IE6sp1? What about Directx 9c? And KernelEx? Are any of those included in SP 3.18, and if not, do they get installed before or after SP 3.18? Also - does SP 3.18 include all of the updates / patches that I would-have-gotten from WindowsUpdate server (if it were still running) ?

You know, I realize full well that I'm going to have to open ports (and I know how to do that), and it might be a bear to set up. All I'm asking is if it's been done, or has anyone done it, does it work, etc. And if it does work - then what software (out of a small handful of possibilities) am I looking at. What about net meeting? This came with win-98, didn't it?

These answers aren't helping me. I want to control / manipulate / access the desktop of a XP-sp3 system FROM a win-98se system. Which system is the "server" in that situation? And will TightVNC do what I want? Will anything do what I want?

I know that remote desktop sharing is common between NT-based systems - can I do the same between a win98 system controlling the desktop of a remote win-XP system? If so, what do I install?

There apparently does not exist a sound font that simulates the SB-16 (FM synthesis, aka "OPL3") sound-cards that existed prior to wave-table synthesis used in AC-97 sound cards or integrated sound devices. If such a sound font did exist, then it could be used by software like Fluidsynth (or vlc) to play old midi files such that they would sound *exactly* they way they did 10 - 15 years ago (when they were created).

I was using import patcher to see if I could get Adobe Audition 1.5 to run. I ran Audition.exe through Importpatcher.37 and then ran the modified file (auditio#.exe) and got the error "Error loading language module". Looking at the import patcher logs, it seems that I'm missing apphelp.dll. I obtained that file and placed it in c:\windows, c:\windows\system, and c:\windows\system32, as well as the directory where I have import patcher (IP). I ran IP again, and this time it created apphel#.dll which I'm not sure what to do with. Auditio#.exe still gives me the same error. Looking at auditio#.ini seems to indicate something is wrong with ntdll.dll (lots of entries ending in "=") and some entries for Kernel32.dll [KERNEL32.dll] BaseDumpAppcompatCache= BaseFlushAppcompatCache= BaseCheckAppcompatCache= BaseUpdateAppcompatCache= So am I supposed to replace apphelp.dll with apphel#.dll? What do I do with ipstub.dll? Is there any hope of getting audition.exe (or auditio#.exe) to run? (yes I have KernelEx installed).

That's the thing - I'm asking which VM gives good audio support to win-98 running as a guest. As your links indicate, other people have asked, and haven't found workable solutions when it comes to midi support (as far as I could find).

Looking for a solution where win-98 can run in a virtual machine (under XP on i7 hardware) that is emulating AC-97 audio - with a functional midi interface. Seems that most emulators only provide SB-16 sound device - which does not have a midi player. Looking for known solutions or hard information, not "did you try this" or "did you try that".

Still doesn't explain how FF 2 was coded back in 2008 to know that when loading the flash plugin, that it should search for all NPSWF32_version_number.DLL files and choose the most current version - when Adobe wouldn't even start creating/distributing NPSWF32_version_number.DLL files until 4 years later in 2012. I guess whoever was coding FF2 was just so clever to anticipate that method of flash file naming 4 years in advance, and build it into FF2 eh?

From here: http://kb.mozillazine.org/Flash They describe this change when the file-name for the flash plugin began to include the version number. This started with flash version 11.2 (March 2012). How it is that Firefox 2.0.0.20 (released Dec 2008) has some "awareness" that when it loads the flash player, that it must search the specified directory according to a pre-set pattern, interrogate each file that matches the pattern, and load the newest version of NPSWF32_version_number.DLL ? Or is there some other "middle-ware" piece of software between FF and NPSWF32*.DLL that is doing the searching / loading?

They call registry keys which contain the file locations ( see list above ). Consider them pointers to files on disk. How can the registry keys exist for a brand new file that I drop into C:\WINDOWS\SYSTEM\MACROMED\FLASH\ ??? I didn't "install" the new version of flash. All I did was drop NPSWF32_11_6_602_168.DLL into the macromed\flash directory. So how can Mozilla FF/Netscape know that it exists? It can't be present anywhere in the registry, because I didn't install that version (didn't run the flash installer) and I didn't register it manually using regsvr32. And (I ask again) why do these flash DLL's exist as both NPSWF32.DLL *AND* NPSWF32_version_number.DLL ??? Why does Adobe create / distribute them using dual-names like that? Who else (what other company) has ever distributed / installed new versions of DLL files by including the version_number in the name of the file? You keep talking about the ocx files - I don't care about the ocx files because I don't care about IE6 or how flash works (or doesn't) with IE6.

Don't paste the same Flash file with different names in the same folder and don't use different versions at the same time! How does Mozilla FF/Netscape know to look for both types of files? How do they know to look for NPSWF32.DLL *AND* NPSWF32_version_number.DLL ??? Why would they look for both types (both names) for the flash DLL file? What tells them to do that? Why does the flash DLL exist with both types of names? What purpose does that serve?

Back during the time-frame July 2000 through Dec 2005, we had about a dozen win-98 machines (and about 6 other machines running NT4 and 2K) in a small organization sharing an ISDN connection to the internet. We had a net-block of 64 static IP addresses, and each of our PC's was assigned on of those addresses. No firewall running on any of the PC's (at least not the win-98 PC's) and no NAT router. All of the PC's ran some version of Norton Antivirus, and it caught most viral email attachments as incoming mail was spooled on our NT4 mail server. Over that time frame, none of the win-98 machines got infected by anything. Our 2 NT4 servers were discovered to be hosting someone's private FTP site (or at least they were trying, but our slow connection was a problem I suppose). Our Win-2k machines were periodically hit with network worms and other stuff. I would argue that those years (2000 - 2005) were the prime years for win-9x to be targets, and we had a completely open network topology that would have facilitated it, but in the end our win-98 machines sailed through those years cleanly. In the years since then, our PC's shared a DSL connection behind a nat-router, and although I maintained NAV 2002 on about 6 machines until about 2008, they continued to be devoid of malware, trojans, virii, etc. So for the past 5 or so years I've abandoned any AV protection on these win-98 machines. On the other hand, I'm quite aggressive at adding entries to my hosts file, which is based on the MVPS hosts file. I examine my router's out-going logs periodically to see what domains or hosts are being accessed, and any that don't look right are added to my hosts file. There is a lot of web-metrics, click-tracking, ad-serving and god-knows-what servers out there that are hooked into the web-surfing experience that have no place being there, and I'll be damned if I'm going to expose my PC's to that crap. Especially when it's those servers that are likely to be hacked and serve up malware. But it all comes down to this: Win-9x, either by design or dumb luck, is simply not vulnerable to even a fraction of the exploit vectors that have existed for the NT-based line of Windows. The website Secunia.org lists all the known security issues for many many hardware and software products. As of July 2006 (when win-98 went EOL) Secunia was listing 33 advisories for Windows 98. That's for the entire life-span of the product. They were listing well over 200 advisories for Win 2k/XP at the same time. And as of Dec 2012, they were listing 408 advisories for Win-XP pro (44 of which were un-patched). So that should put things into perspective.

On my computer, the only place where I have the file(s) NPSWF*.dll is: C:\WINDOWS\SYSTEM\MACROMED\FLASH\NPSWF32.dll C:\Program Files\Mozilla Firefox\plugins\NPSWF32.dll There are no such files anywhere else. Opera says (under plugins) Shockwave Flash 11.6.602.168, located at C:\WINDOWS\SYSTEM\MACROMED\FLASH\NPSWF32.dll I realize that NPSWF32.DLL is also available (or also exists) as NPSWF32_11_6_602_168.DLL (why that is - I don't know). If I add NPSWF32_11_6_602_168.DLL to both of the above locations where NPSWF32.DLL already exists, nothing changes. Opera still places a > (play) symbol where flash content exists on web pages, forcing me to press the > symbol to "play" (render) the content. And nothing changes for FF2 - I still get that error message when I bring up the same web-pages with flash content. I seem to recall that I tried this same thing (replacing the flash DLL file) the last time I upgraded the flash version - and it didn't work, and I ended up running the installer, and that's how I got the current version (11.3.300.265) to work. Also - I don't know if you're supposed to see "Flash" show up on the FF2 add-on screen (Tools -> Add-ons) but I don't see Flash showing up there (I don't recall if I ever did). But a copy of the DLL does exist in the Mozilla Firefox\plugins\ directory. So if FF uses the flash plugin from that directory, then why would FF be messed up if I change the flash DLL file in the SYSTEM\MACROMED directory? Edit: So I removed NPSWF32_11_6_602_168.DLL from: C:\WINDOWS\SYSTEM\MACROMED\FLASH\NPSWF32.dll C:\Program Files\Mozilla Firefox\plugins\NPSWF32.dll And I restored NPSWF32.DLL version 11.3.300.265 to both those locations, and FF2 and Netscape 9 are happy and Flash works for them. I then added the file NPSWF32_11_6_602_168.DLL to C:\WINDOWS\SYSTEM\MACROMED\FLASH\NPSWF32.dll and FF2/Netscape 9 stop rendering flash content. They don't give any errors - they just don't render flash. Can anyone explain that?

I tried running several version of FF 3.x.y (don't recall what x.y was) a few years ago, and couldn't tolerate it because it was drawing thin white lines across any bitmapped images that were on the web-page I was viewing. These lines would appear when I scrolled the page such that the images were partially out of view and they were scrolled up or down into view. The white lines would appear at the cut-point where that portion of the image was out-of-view. I very likely might have had KernelEx version 3.6 installed at that time - don't know if that's a factor or not. Have also changed a few system DLL files since then...

So if I substitute either of these versions: 11,5,502,149 or 11,6,602,168 of the DLL in place of what I had, and if I try to bring up a web page with flash content using either FF2 or Netscape 9, I get this error message (pops up in it's own window): --------------- Title Bar: Illegal Operation in Plug-in The plug-in performed an illegal operation. You are strongly advised to restart Firefox. ( ) Don't show this message again during this session [OK] --------------- On Opera 12.02, flash appears to work - but instead of automatically rendering flash content on any given web-page, I get a big > symbol that I have to press to render the content. If I hover over the > I get this text on the bottom line of Opera: "Please click to download and activate plugin". For example, when I view this page: http://www.investing.com/currencies/aud-usd-advanced-chart Where the chart should be, I get a big > symbol that I have to click. Funny thing - I thought those graphs were done in java - but I guess they use flash. With the previous version of flash, the graphs just appear right away. On this page: http://helpx.adobe.com/flash-player/kb/find-version-flash-player.html In section (1), when I press the > symbol, nothing happens (I don't see the ball rolling back and forth). In Section (2), when I press the > symbol, it tells me the flash version (11.6.602.168) and my OS (Windows 2000 32-bit) and the browser (Opera). Do I have to do something else in these browsers to properly register or install these new(er) versions of flash? There seems to be more needed than just replacing the DLL file. Oh, this PC has a socket-478 Intel P4, 2.5 ghz.

I have Flash version 11.3.300.268 working on this computer. I obtained NPSWF32.dll version 11.6.602.168 and simply replaced the older file with the newer one where it sits in the \SYSTEM\MACROMED\FLASH directory. This file-substitution did not work. I got some sort of error message in FF2 when trying to bring up the Adobe flash-version/test page. I thought that replacing the DLL file was all that was needed. No?

The " KernelEx Apps Compatibility List" thread lists version 10.0.2 as being compatible with Kex. Explain? When it comes to removing IE from a win-98 system, I was under the impression that there are some (many?) dll files that are associated with IE that are needed (or are useful) in terms of rendering web content. Especially scripting (like jscript). For example, the 5.7 scripting host update for win-98 requires IE 5.5 or 6. So my basic question is, if you go to the trouble of setting up a win-98 system and strip IE from it, do you lose something in terms of web functionality? One last question: If someone does not like or want to use Opera, and if they have a Win-98 system without IE, and if they have KernelEx, then what is the most recent, most advanced, most functional (working history, bookmarks, etc), most HTML5-compatible browser available to them?

What is the latest (or last) version of Firefox that can run more-or-less trouble-free on a win-98se system with KernelEx? Also, for any win-98 system that has had IE removed from it - does that have any impact on web-browsing (using other browsers) in terms of causing problems with proper rendering of web pages, handing scripts (jscript, etc) ?

After messing with the msi file with Orca, I managed to get Adobe Audition 1.5 to install without any errors. The first time I ran it, it asked what drive I wanted to use as the temp drive, but I didn't get any further with it after that point. When I try to launch it, I get a spash-screen for a fraction of a second. That's it - no error messages. Is there some way I can figure out what it doesn't like that even KernelEx can't help with? Was there some way to use KernelEx in a diagnostic mode? (for a program written in 2004, I'm surprised that it doesn't run under win-9x with Kex) PS: When I saved my edited msi file with orca, it seems to have stripped out the data1.cab file. The original file was 30 mb in size, and the edited file was 10 mb. What am I doing wrong such that the cab file gets lost in the saved file?

Just to clarify, I did use ultradefrag yesterday (I know I didn't mention it in my last post). I used it more today - I went back and forth between these methods: - native XP defrag - ultradefrag - jkdefrag - using Contig on individual files The root problem is not the fact that copying files from NTFS source to FAT32 destination can be "messed up" if the source file is fragged as it sits on the NTFS drive (it wasn't, btw) or that copying using the gui (copy - and - paste) can invoke a multi-threaded process that will results in a fragmented destination (it doesn't, because command-shell copy method gave exactly the same results). The root problem is that there seems to be something about a "well used" Win-98 volume that, over time, results in immovable files (seems to be very small files) that prevents large contiguous blocks of clusters from being created during defrag sessions. Of the methods I've tried so far, I was not impressed by ultradefrag (just based on how "hard" it worked at re-arranging the little blocks on the screen during it's defrag runs). Sometimes it didn't even do anything after I made changes the volume. Jkdefrag, on the other hand, always worked very hard at re-arranging the little blocks (reminded me of Norton Speed Disk). I remember that when Norton was done with a defrag job, the graphical layout of the drive was completely nice and tidy looking. No stray blocks or streaks of files strewn about the layout. Everything was compacted down and contiguous. I haven't tried norton yet. Something else I've been doing is going crazy with re-setting file and directory attributes - removing System, Hidden and Read-Only when ever I see them. Also deactivated XP's desire to store restore points in System Volume Information folder on this slaved FAT32 drive. I've been focusing on the "C" volume on this drive. It's 32 gb FAT32, with about 20-25% free space, about 7,000 folders and maybe 100k files. I haven't used win-98's native defrag because I don't like to defrag a drive that's in-use by the OS, and at that location I don't have a second win-98 system handy that I can commandeer for several hours to act as a master to defrag this drive as a slave.