Please, write any suggestion and report about 2.0 in this topic. I read all your posts carefully.

My to-do list for 2.0 FINAL:

- Fixing of bugs.
- Patching process (for new desktop icons).
- Testing.

Is it possible with new 98 SE SP to slipstream all updates into original Windows 98 SE source CD?

Only hot-fixes (Main Update) are slipstreamable. It is different from Windows XP's slipstream method.

Firstly, you must copy your Windows 98 SE Source CD's WIN98 directory into hard drive (like D:\WIN98). After that, you must only extract SP2.CAB into D:\WIN98. Windows Setup will use these newer fixed files during installation.

Note: You can see setup file's (SESP20B2.EXE) contents with Winzip, Winrar or PowerArchiver (Right click - Open with Winzip). SP2.CAB is in the setup file.

Hello !
Gape, how about to keep original Windows 98 SE icons. ?
The colour scheme from Windows 2000 seems to be fine to me, but I really don`t like the icons.
And...how about to replace the original boot logo with Windows 98 Second Edition boot logo.

A release date ? [I can hardly wait ... ]

shuKaru,

As I said, Windows 2000 desktop icons and Windows 2000 color scheme features will be optional. 2.0 BETA 2 contains a new boot logo as optional feature.

All,

I have just got 16 new hot-fixes, so I think perhaps there will be a Release Candidate (RC) before FINAL. Sorry about this but we must be sure about these new hot-fixes' compatibility.

Gape!

I know that vxd fix is myth.

Read this please!
http://www.infinisource.com/techfiles/vmm32.html


And do you have a 320798 hotfix ?
Microsoft Office Documents Open Slowly After You Visit Korean Web Pages

That is only Korean-localized versions of Windows 98 and Windows 98 Second Edition.
I'm using Korean-localized versions of Windows 98 Second Edition.

If you have a 320798 hotfix, send me email please.
My email address is myinbox@empal.com.

Thanks.

HSI,

I read the link, thank you. According to this link, "missing vxd files" is NOT a bug. They are already in the VMM32.VXD. I totally agree.

From the link: "M$ says that extracting system files that are rolled into VMM32.VXD will not increase system performance..." I don't agree with M$, sorry. Extracting some VXD files into Windows\System\VMM32 directory improves performance. I made some tests on my system with WinTune and PCPlayer 3D Benchmark, and these benchmark applications say my performance increased approximately %5 after VXD extracting. I know some other people reports similar results.

VXD-Fix contains 7 files. Service Pack contains updated versions of 4 of them. So, I already have to install these 4 files into Windows\System\VMM32.

I don't have 320798, sorry. Perhaps, you can request it from M$.

what about adding that task manager for win98 se and replace the windows 98 se one with this one instead?
here is the link:
http://www.niresoft.com/tm.asp
also would it possible to add the choice of adding or removing the windows 2000 color scheme and icons to the add/remove in the control panel or add it to the windows setup like what 98lite does so that if you make a mistake you can reinstall it without running the service pack?

Well I think there isn't too much point adding too many 3rd party applications, we can download them anyway ... they merely increase download size

My suggestions -

1. Conservative Swap file enabled only on systems with 128MB or more RAM ...

More later

ujjwal, with the version 2.0, conservative swap file is in the "Performance Tweaks" option.

soldier1st, we can discuss this task manager for 2.1.

ok great Gape

sounds good to me gape

The edit to the System.ini file of ConservativeSwapFileUsage=1 is meant for systems with 256MB RAM or more only. ;)

Actually I got a good performance increase on a PIII 450 with 128MB RAM with conservative swap usage, but I run on a celeron 266 + 64MB RAM at home

Great to see that 2.0 organises updates in such a fashion, good work

the AlwaysUnloadDll Registry Trick works wonders for windows 98 se,not very good under 2000 or xp though

Supported operating systems: Windows 95, OSR 2.x, 98, 98 SE, ME. Applies also to Windows NT 4.0, 2000 and XP, but only for 32-bit DLLs.

Open a notepad and enter the following code:


File --> Save As... --> DLLTweak.reg

Double click and say Yes to apply. It's that simple.

but what i meant was i've tried it under 2000 n xp it actualy slows it down,under win 9x or me it works wonders

Hey all,

I have an opinion for version 2 of this service pack

There is a Windows Installer 3.0 Redistributable

Can you put it in version 2 FINAL of the Service Pack ?

Windows Installer 3.0 does not support Windows 9x/ME.

I have another question ...

You know that in Windows XP, when you go to System Properties you have there written information about processor type, processor's clock frequency and the amount of RAM installed on the computer ...

But in Windows 98 you don't have the processor's clock frequency displayed there. So my question is: is there any way to make that happen' in Windows 98 too, through this service pack ?

10x Gape for the previous answer!

MasterT
U can use Start\Run\dxdiag.exe to get that info in any WinOS

2 Gape 'bout infex.exe
how about to make short description frame (or like a frame) for each selected item in add/remove list? Of course all "descriptions" have been described in infex.ini in special section...

egorz,

Good idea, but I think it's not necessary for 2.0. I can do it for 2.1. I'm very close to finish 2.0, so I'm working on only missing features and testing now.

Workaround: Write the option name in a long format. For example, we can write "Adaptec ASPI Drivers Update (Version 4.60.1021)" instead of "ASPI Update".

any expected eta?maybe 1 day or 2?

Only patching process and uninstallation are missing now. Perhaps, if Tarun finds a better way for recreating of VMM32.VXD, I'll add it , too.

I hope I'll release RC (Release Candidate) in this weekend. If RC works good, I hope FINAL will released in 1 December.

Is 2.0 final going to fix the problem where deleting files in EXPLORER.EXE gradually causes system resources to be leaked away (viewable in Resource Meter)? This leads to crashes and extreme sluggishness (e.g. after deleting a file, 1 minute wait for Explorer to wake up).

Is there any fix for this problem?

I think I'm not the only one because after searching it sounds like others have this problem and it comes from IE 6.

This problem comes from IE 6. Because if you use IE 5.5 or older version, you don't have this problem.

Problem files are BROWSEUI.DLL and BROWSELC.DLL. The only fix I know is that you must replace these two files with IE 5.5 SP2 versions.

For more information, please look at this link.

A fix for this problem is not in the SP 2.0, because adding a IE 6.0 specific fix is difficult. I hope I can add it in the 2.1.

i got this really annoying problem too!!

perhaps you can add some fixes/features optional... selectable or deselectable during installation?

It also happens with Windows ME.

I have also encountered the problem when deleting a large number of files
in Explorer and it hangs when running WinME. Or it stalls for a few minutes
and then recovers & I can continue on normally. Workaround is to delete the
files in good old-fashioned File Manager tool. Go to the Windows folder and
double-click on the winfile.exe file to run File Manager. Even though File
Manager doesn't handle long filenames, it deletes files faster than Windows
Explorer and doesn't lock up Windows. You can even change the 'System'
attribute to a file which cannot be possible under Explorer.

Windows ME with IE 6.0?

Thanks for that link Gape, I'll try it out (the fix for slow deletion of files).

At one point you said that IE6 install was suggested before installing your SP.

Therefore my installation procedure will be to install IE6 SP1, your SP, and then this new fix

That is correct. As you of course know I'm on Windows ME and running IE 6.0 with SP1 and all the other junk filled fixes.

I may just revert to IE 5.5 now that I use Firefox. Annd, it fails to revert. Heh. "Uninstall failed." Though I'll recommend BugOff from Merijn's site.



Merijn's Download Page

i don't use Internet Explorer 6
i use Opera so i'm protected from this

Incorrect. Even if you do not use IE you are still vulnerable. What it says in the program is incorrect, because if you go to a favorite through a folder in Explorer, it uses IE. IE is rooted all through of Windows.

i removed ie from windows 98 se and use opera n don't use any ie based apps that need it
now you see i am not vulnerable cuzz i have no ie

hey thanx for the service pack .... i am greatful to u people

i find a small problem after i install is i have to ping some site to connect to browse untill then i cant

hey all i have all the updates of win 98 se .... almost 125mb and windows me 220mb if any ones wants me to upload them i will be happy to do so .... i downloaded these file from microsoft .... for installin it later .... they are in folders just like from thge site ... i downloaded all the updates on 01 - 04 - 2004 .... dd - mm - yyyy

like i read ..... before please add tweak UI ..... in the new version ...

and if u could add this feature i would be very happy ....

in windows me and higher like win2k win xp if we opena file with different applications ... the next time we right clike it will ask us open with .....

but in win 98 se every time we want to open the file other than the default programe we have to hold down shift then only open with comes on the right clike see if u clould add this in the new update

once again thanx ....... for every thing

greetz
craberos
craberos@yahoo.com

hey since you dled the updates for win me you could upload em to gape n that would help him get a win me sp going faster but this one should include ie 6 with all updates or give a choice to install it or not

Thanks to George [Axcel216] for pointing me to this forum!

Gape [Alper I presume]:

I am still working on my auditing of 1.6.2, but I have been side-tracked by having to fix FAR TOO MANY XP-based machines that come to me to be de-malwared, etc. Hopefully, I will have all of my points in order fairly soon. Here are a few highlights:

Relative to the list of the first seventy [not counting the last three added to make 73] hotfixes, I would dispute the "dominant" hotfix number in a few instances simply because of the quirky way MS defines them. I don't have access to the specifics here [as I am typing on an "alien" machine at the moment], but there is at least one occurance of something like the following:

KBxxxxxx refers you to download hotfix yyyyyy. However, there also is KByyyyyy which also references hotfix yyyyyy. In my view, that means that hotfix xxxxxx is unavailable and replaced by yyyyyy. As such, the dominant article is the yyyyyy one, not the xxxxxx one, etc.

In some cases, there are "co-dominants" where there are multiple KB articles that all point to the need for the same file revisions. If applicable, perhaps in this case they are either equally dominant or perhaps only one of them has an obtainable hotfix? Whatever the outcome, I think there needs to be an alternate way of categorizing these cases, etc.

In at least one instance, you are taking a file from a known package [such as dsclient] and as such, there doesn't appear to be a KB article to attribute the SP upgrade of the file to. In point of fact, there IS a KB article that precisely fits the description. Thus, this allows the number to be upgraded to 74 [Whether or not the hotfix is actually available is irrelevant to any rollup package, but it's nice to have the individual hotfixes as well.]

As I understand the overall action of the SP, files that aren't currently installed are not immediately upgraded. As needed, they are installed from the SP1.CAB file to prevent taking from the original .CAB collection, etc. A prime example would be a machine where there is currently no USB hardware, but later a USB add-in card could get installed.

My point is that the installation of the relevant, for example, USB-oriented upgrades from MS as hotfixes ALWAYS install these files, while the SP merely only ULTIMATELY might install them as needed.

The subtle point is that if you include QFECHECK information for the hotfix, then installing only the "necessary" files leads to false-positive errors in the hotfix info for that update, which in turn causes the confusion and waste of time reconciling out which red-flags are merely for not-yet-installed hardware specific updates, as opposed to genuine errors regarding corrupted files, etc.

A related point is: Can the QFECHECK information be installed along with the hotfix-related files? This would work exactly like a fairly recent update rollup that MS did for XP months before SP2 was released, etc.

Other topics:

I can pass along some horror stories related to MS's installers of certain hotfixes, mostly covered by the SP [other than the QFECHECK issues I raised above]. For example, Q249973 cannot be installed unless you haven't upgraded to IE60 or higher due to some internal logic error; it just bails and fails to upgrade any file or provide the QFECHECK program or relevant registry updates.

The root cause seems to be the mislogic that because one of the files it provides is already updated to a higher-still revision by the IE6 install, it just exits, even though there are two other files that still need to be upgraded as well as the other QFECHECK and registry stuff, etc. [Note that the SP correctly installs the higher rev files!]

Again, I apologize, but not being at "home base" I cannot get details I don't have committed to memory, but there is a serious problem with an available hotfix that disrupts the SP and even other MS-provided hotfixes. Goes something like this:

hotfix xxxxxx installs 4.10.2223 of a file; hotfix yyyyyy installs 4.10.2224 of same file; hotfix zzzzzz installs 4.10.2226 of same file. However, if xxxxxx is installed BEFORE yyyyyy or zzzzzz or the SP, then the file stays at 4.10.2223! If the file is manually deleted, QFECHECK info for each hotfix correctly state what rev of the file they require. Yet, if xxxxxx gets installed, then each QFECHECK section gladly accepts 4.10.2223 as the revision they "like".

If xxxxxx is never installed, all else works as intended. Apparently, something within the installer for xxxxxx makes the installation of the file at 4.10.2223 too "permanent".

Should this update already be installed, the only way to fix it is to manually delete 4.10.2223 of the file, then apply any other update [yyyyyy or zzzzzz or the SP]

I suppose an appropriate mod to the SP would be to check if xxxxxx is installed and pre-delete the file? Or perhaps delete some "overly strong" reg info?

Yet another topic [so I can get off of this machine which is not mine!]:

I have periodically heard references to this "heresy" topic about either VMM32.VXD is always a faster-loading panacea as compared to providing the relevant "component" files, presumably loaded into either \windows\system directory or perhaps elsewhere? [iosubsys? vmm32? both? all three?] and people reporting either no change or dramatic loading time changes or performance changes or all of the above or none of the above, etc.

I have a concrete example of something that makes be a bit "nervous":

I am testing out the SP on a slightly souped-up Compaq Presario 7212 [now has a mighty 64 MB and a P133; originally was 16 MB and a P75]. [The main reasons to test on are: a) It's available and now not being used as a doorstop, It has virtually none of the "optional" hardware thus allowing me to document which files the SP doesn't initially install.]

Specifically, this machine uses the OPTI "Viper" chipset instead of the more common Intel versions. This is not a problem per se, as 98SE has full support built-in.

I initially installed a quite vanilla 98se system on this box, and all built-in stuff was totally recognized [all except my Realtek 8139 NIC for which I added the latest driver and no problems there].

However, I noticed that DMA is not set on the hard disk controller, so I enabled it, got the standard warning, etc. Unfortunately, with DMA set, the machine became quite unstable as can be the case, etc. In safe mode, I was able to get things reset, etc. and basically ignored DMA mode for some time.

Later, I was reinstalling the same basic system, but had run across the whole VMM32 .VXD debacle, and noticed that the drivers for the hard disk controller referenced a few files included inside of VMM32.VXD, so I decided to follow someone's directions and placed the de-CAB'ed originals into a couple of relevant directories.

After a reboot, I do notice that it DOES take a bit longer for the system to come up, which is what most of us would expect, since VMM32.VXD is being thwarted, etc.

However, I was able to set DMA on the controller and IT TOTALLY WORKS NOW!

Thus, it would appear that the compaction process that creates VMM32.VXD could be buggy; the fix in that case is to not use it!

Have I blown any holes into anyone's "theories" about this?

cjl (will write more when I get back to my own machine!)

Indeed it does sound like your VMM32.VXD had problems when it was built during install. You may want to try and reconstruct it.

You're still wrong soldier1st. Even if you use non-IE apps that does NOT mean
you're out of the woods with security problems. Even non-Microsoft software
can have potential security flaws. But at least they're harder to exploit than with IE.

Tarun, try using IEradicator to remove IE6. Get it here:
http://www.litepc.com/ieradicator.html
I use this tool to uninstall IE when IE's uninstaller doesn't work.

The IE uninstaller in WinME won't work if WinME's system file protection feature
is disabled. I found that out several months ago when I prevented 'Statemgr' from
running by disabling it from the MSConfig tool and restarting WinME, I couldn't
remove IE. When I re-enabled Statemgr and run the IE uninstall tool, it worked.

Forget about the Q249973 update, CLASYS. It's CRAP! Go to Axcel216's addons page:
http://www.mdgx.com/add.htm
And click on the links to download newer riched20.dll, riched32.dll & usp10.dll files.
They're better than the ones found in the Q249973 patches.

This is a testbed system created thus:

• Format drive C:
• Install 98SE on C: drive selecting all options allowed except WebTV
• After last reboot, resolve Realtek 8139 NIC using latest (616) driver on D: drive
• Don't bother enabling DMA on Opti Hard Disk Controller because it will make system lockup predictably [Tested on a previous otherwise identical test installation on this machine]
• Add files indicated as being imbedded in VMM32.VXD into IOSUBSYS and SYSTEM directories. System Device Manager acknowledges them as now independent of VMM32.VXD
• Reboot; takes longer to come up; this was expected!
• Enable DMA in hard disk and CD-ROM
• After reboot, major speed improvement since DMA enabling makes it run faster as expected; no system lockup whatsoever. System behaves exactly like similar systems based on Intel chipsets. DMA can be enabled and disabled at will with predictable performance changes. Yet, if you do NOT remove the VMM32.VXD dependancy, you cannot ever enable DMA!

Clearly, there is no bug per se in the 98se built-in support code for the Opti chipset, or else it couldn't ever work. However, something gets "lost" in the process of creating VMM32.VXD by Windows itself when it self-creates that file, etc.

So, what would you have me reconstruct?

What are the files Gape is adding to \...\VMM32 to override VMM32.VXD? As I understand it, some are specifically because of hotfixes beyond 98SE release, but some are because he also noticed something "works differently".

[Note: "works differently" is official MicroSpeak for any change in how something works, no matter how big or small, important or not important. It is part of MS UML [User Manipulation Language] SP2 as used with Windows XP Service Pack 2 to deflect and minimize the impact SP2 has on many people's systems after installing SP2. UML cannot be uninstalled and all support contracts are automatically entered into and cannot be cancelled. ]

cjl

This has been discussed in this thread and many others. Remember, the Search feature is your friend.

Article on how to rebuild VMM32.VXD

A utility for those who do not wish to use VMM32.VXD titled WinPatcher. Basically scans the system and VMM folders and allows extraction of the files from the CAB files.

Searching through the forum as you suggested, I realize there has been some furor over this subject. Since this is the time to "get it right" for the benefit of releasing the new SP, I just want it stated succinctly:

1) There are factions for and against the notion of even having VMM32.VXD and whether or not it either helps some things or hurts others. Within this discussion are claims of performance improvements/lack thereof and in particular case a report of a vanilla system build that probably creates the file initially corrupted, thus the "loose" files are better than the corrupted one, or at least this is the theory behind what I have observed.

2) There is the possibility that it gets created wrong when the system was first installed or somehow gets corrupted later.

3) There is an article from the people who insist that MS is correct and that there are no benefits whatsoever from displacing VMM32.VXD, for the purpose of rebuilding VMM32.VXD when it is discovered to be corrupted for whatever reason, etc.

4) There is posted a utility to carry out the method of eliminating the need for VMM32.VXD entirely.

5) There is posted a link to a VMM32.VXD known to be stable, but was created for Win ME; does this particular file apply to 98SE?

6) The SP will include several VXD files in part because it is necessary as the files are updated by the SP itself, and partly because Gape believes that there is a performance boost if certain others are additionally added in "loose" form despite not being updated. Are there any other files that would be considered as a portion of VMM32.VXD beyond this set?

7) Most importantly, what is the method the SP will use to best implement a stable result after applying it?

Part of what I read here is that unless you can prove that VMM32.VXD isn't corrupted, it's always safer to run with all of the "loose" files, albeit it boots slower. Conversely, rebuilding VMM32.VXD to a known non-corrupt state should be as good as having the loose files with the added benefit of shortening boot-up times. However, some report that VMM32.VXD eats some resources as compared to some form of perhaps selective override at least some "loose" files, etc.

Do I have it right or what? Someone please show me the errors of my ways, if any.

cjl

1.) I run Windows ME and have found my system to be much more stable and faster when using the VMM32.VXD

2.) That has indeed been confirmed as a bug on Windows 9x, it all depends on the hardware.

3.) I have checked with many PC techs and asked on several forums about the Infinisource website; it is correct and reliable.

4.) If you are referring to WinPatcher, that would indeed be a creation of mine to help automate the process if anyone chooses not to use VMM32.VXD (which I do not recommend unless you are experiencing problems)

5.) Another file that I constructed with the help of several PC techs, however I do not have a Windows 98 machine around and thus am unable to create one for 9x versions. The ME version provided will not work on Windows 9x machines as it asks for MS-DOS 8.0.

6 and 7 will be left for Gape since he's constructing the SP.

VMM32 is a good file to have in your system and running if it is constructed properly and without bugs (As some Windows OS' have issues with the creation of this file).

You can also check this link where I asked about this instance. DjLizard is a certified PC technician and he definitely knows his stuff.

tarun was talking about internet explorer not non microsoft software,stay on topic please

Ah, then that would be why. I disabled and removed System Restore. ;)

VMM32.VXD is a compressed file of a collection of VXD files. Its contents depend on system configuration.

Because it's a compressed file, any corruption affects most of VXD files in it. For example: Assume that you have a ZIP file which has 10 files in it. If this ZIP file corrupts, you cannot extract ANY of these 10 files with using normal ways.

On my tests, an uncompressed VMM32.VXD file (still ONE file like a TAR package) has a small performance improvement. But, the booting process is slower a bit, too.

In the 2.0, I only copies updated VXD files onto VMM32 directory. Perhaps, a reconstructed VMM32.VXD will be better, but it is not an easy process. Perhaps it will be on the future versions.

but will this vmm32 ask for msdos 8?cuzz i hope not

I'm still slightly confused:

1) How did you test an uncompressed version if it's a compressed file? What process makes it an uncompressed variant? Or is this just the "natural" result of adding in the updated component .VXD files necessitated by the SP?

2) What performance improves? Is this possibly merely because there is redundant resources being used, i.e., for both the VMM32.VXD monolith being in memory as well as the authoritative drivers also being in memory [i.e., the ones you added because they are updated per the SP, etc.] "wasting" memory space?

3) Can I assume that if we carry out the infinisource-based method on the updated collection of files [after the SP] we can get both the performance and the resources back?

Part of why I ask is that my test example indicates at least the possibility that Tarun suggests, i.e., the installation failed to make a non-corrupted VMM32.VXD. Thus, to fix the problem, I have no choice but to rebuild the VMM32.VXD file anyway. Thus, it would be best to defer this to the point where all the driver considerations are stable to avoid having to do this multiple times, correct?

cjl