LoneCrusader

Yes, you can install them on a copy of Windows that is already set up, but you may need to manually remove several things from the Device Manager first.. or it may just auto-update them when it detects a newer-dated INF file. Never tried using that method, so YMMV.

DDR4 working fine on my Intel X99 system that I discussed here.

The actual speed limit that causes the error varies between AMD and Intel CPU's. There are two different errors each with different speeds. The first problem (Windows Protection Error in IOS.VXD) occurs at around 350MHz on AMD and somewhere around 1000MHz on Intel (I used to know the Intel limit more specifically, but it's been a long time.) The second problem (Windows Protection Error in NDIS.VXD) occurs at 2.1GHz on Intel and I'm not sure of the limit on AMD; rloew may know this. Both of these are fixed for all versions of Windows 95 by my FIX95CPU v3 patch. More info here if you want. Microsoft did provide a HotFix solution for 98 First Edition. They later revised that KB article with a LIE that no fix was available. The HotFix can be found online. I think I gave MDGx a copy, so it should be there. Microsoft also provided a solution for 95 OSR2 but they failed to make this known (intentionally, or just oversight?). NDIS.VXD 4.00.1113 from Dial-Up Networking 1.4 for Windows 95 contains updated code that fixes the issue. All of this is covered at length in the thread I linked above.

Yes, that's it.

9x compatible PCI-E sound card. On my system the 95 .VXD driver worked better than the 98 WDM .SYS driver. 9x compatible PCI-E network card. (sold) BEWARE - you MUST get Version 1.x of this card. Version 2.x IS NOT 9x compatible. On eBay many sellers use stock photos of v1.x card. Do not believe this, if you buy one make them show you a picture of the actual item. Drivers here. How to tell them apart? Version 1 pic. Version 1 pic2. Version 2 pic. Version 2 pic2. I am using the equivalent of NUSB 3.5, however my 98SE system is slipstreamed to include this out-of-the-box. I have seen problems in systems that have only USB2 controllers and no USB1 controllers. This is caused by USB1 drivers not getting copied to the system because there is no USB1 controller to install. USB2 drivers depend on USBD.SYS, a USB1 driver that the USB2 INF does not copy. Try copying USBD.SYS to your \WINDOWS\SYSTEM32\DRIVERS\ folder or manually adding USBD.SYS to the CopyFiles section of USB2.INF before installing and see if it helps. I have most of these boards because for years I tried to stockpile a bunch of 9x-compatible hardware. This is why I have so many 8xx systems. Later on some friends of mine who work for a large organization gave me a bunch of old hardware they were replacing. This is how I ended up with many 9xx systems. Beyond those it has become a hobby to try running these older systems on newer hardware, and when I can afford it I get a newer board and see what happens. I don't have much AMD stuff except very old Super Socket 7. I have a Socket 939 nForce 3 AMD system and one slightly newer than that, but that's all.

I checked my INF's against xRayer's for this device. xRayer's INF is missing the "INTEL_PCI" install section that the device ID is directed to, so I can only assume it does nothing if no error is reported. My INF's point this device to a blank section like the original 9x compatible Intel versions, so my INF "does nothing" for this device as well. Both sets of INF should name the device but that's all. (It's really all chipset INF's are good for anyway.) Neither of them should have any effect on the nVidia driver. The only way I could see any different condition arising from using xRayer's INFs or my INF's would be if the entry for this chipset's LPC controller on xRayer's INFs does not load ISAPNP.VXD as mine (and the original Intel 9x compatible ones) does. Keep me posted about what you find.

You can discuss 9x issues here to your heart's content (or till you're blue in the face), take your pick. But you must learn to separate different issues and keep discussions on topic. Different specific forums and threads exist for a reason. Use them as such. Do not post about 9x things in the XP forum. Do not discuss subjects from one thread in a different thread. Failure to do this results in a nightmare for anyone looking for specific information or trying to follow a discussion. In the event there is an issue with "overlap" across multiple operating systems, you need to keep the discussion in the forum for the OS where the issue originated and where it will receive the most relevant attention.

Yes. I have a representative board from almost every "family/level/generation" (whatever you wish to call it) of Intel chipset. On the "newer" ones I have chosen however to only spend money on Xxx chipset (X58/X79/X99) boards because I don't want the integrated graphics that come with Zxx boards. I have at least one board using each of these Intel chipsets: 845, 865, 875, 915, 925, 945, 965, 975, P35, P45, X38, X48, X58, X79, and X99. There's a Z87 or Z97 (not sure which) somewhere but I think it had a power problem. I have already used each of these systems at some point or another in test setups just like the one pictured. No audio or LAN yet. I have several of these newer systems but I have not tried to use them beyond proof-of-concepts yet. However I do already have solutions for this. I found a Chinese PCI-E sound card using a 9x compatible chipset on eBay and ordered a couple. They do work. I also have PCI-E Network cards that have 9x drivers. This X99 board has two PS/2 connectors so I'm using PS/2 Keyboard and Mouse. My X79 board has one PS/2 connector so in this case I use a USB Keyboard and PS/2 Mouse. Onboard USB2 ports are working. I just haven't tried the USB3 ones yet. I used the older ones out of force of habit, lol. The RAM patch would have no affect on device detection, this is absurd. Without the RAM patch the system probably would not boot, and even if one had the desire to play around with other "free" RAM tweaks and solutions (which have never worked for me), and if they did miraculously work long enough to boot to the desktop, the system would not be stable. Bottom line, I don't waste my time trying to run 9X on a modern system without the right tools to do it. And rloew's RAM and SATA patches are mandatory for that. Once again the RAM patch cannot affect device detection. Not having the SATA patch will not prevent you from running SETUP or reaching the desktop. Of course these systems could "work" without the SATA patch. If you're willing to run in DOS compatibility mode, go right ahead. The specific Device ID's come from my INF's. They can be found here. You will need the later versions that came along around post #31 or so for an X99 system. I have a 512MB version of the 7950GT that I have used successfully before with rloew's NVidia patch, I just didn't have it in this system. rloew wrote the Shutdown Fix for the 82.69 driver but you must apply it yourself. It's here in a thread somewhere. I haven't tried running 2K or XP on the X99 system yet. I do have XP x86 + PAE and XP x64 on my X79 system though (once again this system is not daily-use yet).

What are you talking about? I'll have you know I have no trouble "detecting" my motherboard. It's right where I left it on my testing bench. It is not mobile and does not have cloaking ability. Stop hijacking threads and cross-posting multiple subjects across different threads. This thread is about MrMateczko's Windows 98SE machine. It has nothing whatsoever to do with VBE9X, DOSBOX, or any of my motherboards.

I have a Gigabyte GA-X99-UD4P motherboard + GeForce 7950GT 256MB with 95 OSR2 and 98SE up and running for testing. Using rloew's RAM and SATA patches, the NVidia 82.69 driver (patched for shutdown), and my updated Intel chipset INF's. I did notice that on this board the SATA controllers will only show up if they are in use. I had to connect a drive to another controller to make another one show up in the Device Manager. Apparently the BIOS disables devices that are not in use without consulting the user.

I'm not insulting your intelligence. I'm just pointing out the fact that moving a pre-installed system is asking for trouble, plain and simple. It's not that there is anything wrong with doing so, but moving a pre-installed system is something to be done when necessary and when you have already tested the target system with a clean install to verify compatibility. You can't just throw it on a new machine and expect everything to work properly. Sometimes that happens, but you can't assume that it will always (or even usually) work. In all honesty I have probably tested rloew's patches on more systems than anyone else, maybe even rloew himself. I have quite a stockpile of various hardware, mostly Pentium 4 era and later that I have either stockpiled specifically for 9x compatibility, been given by some friends I have in a large organization when they were replacing, or purchased specifically to test 9x compatibility. Not too long ago I started out on a "quest" to test Windows 95 and Windows 98 on each subsequent set of Intel chipsets from the time when they were supported (845-865-875) up to the present (X99, I don't have X299 yet). I have tested 9x boards using Intel 9xx chipsets, P35, X38, X48, X58, X79, and X99, all without any strange issues except those cause by Intel-branded motherboards that have a garbage BIOS incompatible with 9x. (Do I need to prove it with screenshots? lol) I only have 2 older AMD nForce3/4 systems but the patches work there as well. I have the full range of nVidia 7xxx cards; 7200GS, 7600GS, 7800GS, 7800GTX, 7950GT, and possibly others. I actually did a lot of testing for the nVidia problems to help rloew as well. It's possible that your hardware combination can cause issues. Any motherboard with an Intel BIOS newer than 8xx chipset is garbage for 9x. It's possible other things may cause issues as well, but you need to be able to pin it down to a very specific point by removing ALL other variables from the situation. Once you do this, if there is indeed a bug of some kind, I'm sure rloew will address it.

I see.. I didn't take into account the Quad support, so I suppose the ConRoe board has some degree of advantage there. However Windows 9x cannot use these other cores anyway, so a single-core P4 with a higher raw clock speed should still be better for 9x than a Quad. Also, if the ConRoe board doesn't support DDR2 then I'm not certain how much having a 1066MHz FSB CPU is any better than 800MHz FSB. Bottom line, if you're looking for 9x performance then any advantage would be negligible, and you could use a 9xx chipset board that also supports DDR2 and have a better result. If you're thinking of other OS such as XP, then there are far, far, more performance-capable systems for it than dropping all the way back to Quad + DDR1. Also, based on reports I've seen I'm not certain ASRock has the same level of quality as Gigabyte or DFI but I've never used ASRock so I don't know.

SoftICE 1.99? I don't think I've ever heard of that... The versions I see mentioned most are 4.05 and 4.2.7., apparently both of which were produced pre-2KSP4 and pre-XPSP2 and require fixes to work under those systems.

Eh.. what's so great about it? DFI Lanparty 875P-T is the king of the 8xx chipset + LGA775 boards.

You seem to have a very strange set of problems with your various experiments. Being a frequent user and tester of most of rloew's programs and patches I can say I have never yet seen any issues like the problems you are having. I believe most of your problem comes from trying to move around previously installed copies of the OS. I know it is annoying to have to re-setup updates and programs for each new machine, but this is questionable at best to start with. Any time you do this you should at LEAST go into the Device Manager in Safe Mode and remove any and all devices BEFORE even attempting to boot the system on different hardware. Failure to do this is asking for trouble. Better yet, you should do that step and then manually go into the ENUM key in the registry and manually delete anything that is left under \PCI, \MF, \USB, \USBSTOR, and possibly others (do not remove \ROOT) after the Device Manager cleanup. You should try clean installs using only the base system + rloew's patches and see if you have any issues under that configuration before you try moving a pre-installed system that exponentially multiplies the variables for failure. If it works in the clean setup and doesn't in your moved setup, then the problem lies somewhere in your pre-installed environment.

What version? On what hardware? I'm debugging a driver, or rather tracing what a couple of specific 2K-compatible drivers do when running in their normal environment under 2K and comparing it to how the driver operates under 98SE. (we are trying to backport two 2K-compatible drivers to 98SE) So I need SoftICE to be working under 2K SP4 on the same machine where I already have it working under 98SE to compare.

Hello my Win2K friends! Has anyone here been able to successfully run SoftICE under Windows 2000 SP4? If so, what version of SoftICE did you use? did you have any updates for that version installed? any special registry settings or other settings you had to modify? any updates beyond SP4 installed on your Win2K? what hardware did you run it on? what specific video card and what video driver version did you use? I've combed the Internet and tried just about everything that I've been able to dig up but SoftICE keeps crashing out when the machine boots and causes the machine to restart in an infinite loop. Thanks in advance!

I'm a sponsor. And as you can see it's disabled anyway.

Really now? Me too.

Using PaleMoon 24.7.2 under XP, 1280x960 resolution I see some overlap issues with formatting. This goes away if I use Firefox 52.2.0esr. Hopefully a simple problem.

Since answers don't seem to be forthcoming about what, if anything, can be done to at least alleviate the immediate crisis long enough to archive everything I suggest we all get busy with HTTrack and try to archive as much as possible. I don't know if the entire site can be somehow indexed by the Wayback Machine but we'd better not depend on it to properly handle multi-page forum indices and threads. EDIT: Pick a specific forum (copy the link into the specific forum) to try to archive with HTTrack if you attempt it. If you try to do the entire forum at once it may never finish. I backed up a site a few years ago and HTTrack literally ran for a month straight.

I never knew that. I wish this had been made more widely known... maybe we could have come up with a salvage plan at least.

I remember that thread... and I understand completely how hard it is to find time for online things and for real life as well. Kudos to xper for keeping it up as long as he has. Has the possibility of someone else taking over the forum ever been considered?

Anyone have any idea what the problem is? If it's financial I'm sure several of us could step up and help. I'll help as much as I can... This is a disaster equivalent to losing the Great Library of Alexandria - information exists here that exists nowhere else.

Good point about the BIOS. I recently obtained a Gigabyte X99 motherboard and it was using the first BIOS version and was limited to 1873MB for 32-bit. Flashing to the current version allows 2910MB. I actually just got through emailing their tech support requesting better optimization of the BIOS MMIO for 32-bit. Probably won't get anywhere but it doesn't hurt to ask.