[pinecloud]

The following Chipsets have Windows 9x Support:

Intel Chipsets:

910
915
925 (mobile)
945
946
955
975
G965
P965
Q963
Q965

FISR (Intel 865PE chipset)
MSI K9MM-V
[/quote]


Intel 900 Series with socket 775 MOB no longer supports Windows 98/ME/SE
Intel infinst_autol.exe or intelata_enu.exe did work winth Win 9X until 2006

When you try to install newer 9XX chipset.inf file today “infinst_autol.exe or intelata_enu.exe”, they refuse with “Operating System not supported” or even “rundll32 export failure” messages. If you carefully extract the file content to many of *.cat file level and examine them carefully. I see the $chicago$ headers and ‘donothing’ instructions within them. I am not even a coder but just a photographer and graphic artist, I can still see that Intel was influenced by somebody in the background. Perhaps one of software giants stopped Intel’s driver development team from releasing drivers suspected detrimental to the sales of new software campaign. The time and motivations seem to co-incide with discontinuation of innovative and altruistic Winternals projects from Texas. The Winternals was well capable of providing add-on component pack for Windows NT and Windows 9X at $29 (meaning such pack can appear on eBay for $4.99 range p&p inclusive) per seat to perform the today’s role of Windows XP platform (application compatibility wise) aside from ADS (alternative data stream) which I do not care for in any means. Games of software giants play are not as devastating as oil price schemes detrimental to the future economy of our country but it’s really something. Question of morale of the country strived on philosophy of freedom.

Pinecloud

Economic interest was not build into hardware enumeration hierarchy until 486 or even early P5. Pci/ven code started to appear in Windows registry after P5 era. Windows block device mapping has no logical relation to physical device mapping e.g., c0t0d0p0s:b (hba ID, target ID, disk ID, partition ID, slice ID) which makes more sense over Windows device vendor signature based mapping perhaps arising from its economic interests. Playing with pci/ven codes cut and paste using established models already running on Windows 9X and XP as reference guide at “Brutal Force Locksmith Method” eventually works because it has worked on setting up PCI-E video card. for me. Doing it on disk enumeration has much more threat of losing data. That is why I have not tried as hard.

[pinecloud]

Quote

The following Chipsets have Windows 9x Support:

Intel Chipsets:

910
915
925 (mobile)
945
946
955
975
G965
P965
Q963
Q965

Intel 900 Series with socket 775 MOB no longer supports Windows 98/ME/SE
Intel infinst_autol.exe or intelata_enu.exe did work winth Win 9X until 2006

When you try to install newer 9XX chipset.inf file today “infinst_autol.exe or intelata_enu.exe”, they refuse with “Operating System not supported” or even “rundll32 export failure” messages. If you carefully extract the file content to many of *.cat file level and examine them carefully. I see the $chicago$ headers and ‘donothing’ instructions within them. I am not even a coder but just a photographer and graphic artist, I can still see that Intel was influenced by somebody in the background. Perhaps one of software giants stopped Intel’s driver development team from releasing drivers suspected detrimental to the sales of new software campaign. The time and motivations seem to co-incide with discontinuation of innovative and altruistic Winternals projects from Texas. The Winternals was well capable of providing add-on component pack for Windows NT and Windows 9X at $29 (meaning such pack can appear on eBay for $4.99 range p&p inclusive) per seat to perform the today’s role of Windows XP platform (application compatibility wise) aside from ADS (alternative data stream) which I do not care for in any means. Games of software giants play are not as devastating as oil price schemes detrimental to the future economy of our country but it’s really something. Question of morale of the country strived on philosophy of freedom.

Pinecloud

Economic interest was not build into hardware enumeration hierarchy until 486 or even early P5. Pci/ven code started to appear in Windows registry after P5 era. Windows block device mapping has no logical relation to physical device mapping e.g., c0t0d0p0s:b (hba ID, target ID, disk ID, partition ID, slice ID) which makes more sense over Windows device vendor signature based mapping perhaps arising from its economic interests. Playing with pci/ven codes cut and paste using established models already running on Windows 9X and XP as reference guide at “Brutal Force Locksmith Method” eventually works because it has worked on setting up PCI-E video card. for me. Doing it on disk enumeration has much more threat of losing data. That is why I have not tried as hard.

[RetroOS]

SeeAScot, on Nov 29 2008, 11:16 PM, said:

The HP Laserjet 6P works with everything from DOS to XP. Been running a refurb one since 1998 or 1999
The HP Laserjet 1012 works with 98SE and XP. Replaced my sisters 1012 with a older refurbished 6P as the 6P is more durable and will handle card stock.

The HP Photosmart 2610 All-In-One works with 98SE and XP
The HP Photosmart 375 (370 series) 4" x 6" printer works with 98SE and XP
The HP Photosmart 475 (470 series) 5" x 7" printer works with 98SE and XP
The Photosmart 2610, 375 & 475 all use the same series of ink cartridges.
...

That may be the case, but if you don't already have the drivers for 9x, then you won't get them from HP anymore...
HP have pulled all 9x drivers from their website some time ago, although they left the Windows 3.x drivers!!! - Go figure???
So much for post sales support from HP!
The problem is that HP are in bed with M$, and when M$ say jump, HalfPie jumps!

Here's HP's official statement:
ALERT! Printer Driver Support for Windows 98/ME
Effective December 31, 2006 HP support for Windows 98/ME will end.
- No further support of the HP traditional print drivers for Windows 98/ME will be provided.
- No further development or enhancements of the printer drivers for Windows 98/ME will be completed.
- No support for defect resolution of the printer drivers for Windows 98/ME will be provided.
Effective June 30, 2007 all Windows 98/ME printer drivers will be removed from hp.com:
- HP’s licensing agreement with third-party providers granting use of their components will expire.
- Basic support for your HP product will be available through the use of prior printer or ‘legacy’ printer drivers.
Hewlett-Packard will provide Windows 98/ME drivers for a very limited number of HP LaserJet low-end printers introduced Spring and Fall 2006.
HP's currently available drivers for Windows 2000 and later are the recommended drivers for all supported HP printers from January 1, 2007 forward.
Background
Microsoft discontinued Windows 98/ME shipments in June 2002/2003 respectively.
HP’s Windows 98/ME driver discontinuance is directly related to the fact that there is no longer Microsoft resources allocated to issue management.

Blah, blah, blah, yes Microsoft, as you wish Microsoft, of course I'll jump over that really large cliff Microsoft...

Epson on the other hand still produce 9x drivers for their consumer printer range, and you will always be able to download existing drivers!
I've always had Epson printers since the earliest dot matrix days.
Go Epson!

[Analada]

There's always the MS win98SE CD to fall back on - it contains a lot of drivers for HP printers, including Laserjet.

The newer Laserjets which *aren't* on the win98SE CD, *may* be driven quite successfully by an older win98SE Laserjet printer driver since the family is software compatible to a large degree. There's an interesting discussion about this topic here, which includes some more background as to why HP pulled win98Se support.

http://forums13.itrc.hp.com/service/forums...hreadId=1173447

[rloew]

Quote

dencorso Posted Nov 13 2008, 12:39 AM
Let me add my 2 cents...
IMHO, it's too risky to use:
1) FAT-12 above 32 MiB (127 MiB is possible with patches), it's better to use FAT-16 instead.
2) FAT-16 above 2 GiB (8 GiB is possible with patches), it's better to use FAT-32 instead.
Otherwise, sooner or latter, some program will trash the partition, at the worst possible time (Murphy's Law).

Generally that would be true but there are exceptions. Sometimes there is no choice as to which FAT type you can use. I used a 4GiB FAT-16 formatted CF card in my camera which does not recognize FAT-32. Crashing is unlikely as the biggest problem seems to be Programs treating the Partition as having zero size.
Windows NT supported 4GiB FAT-16 Partitions. Only >4GiB FAT-16 Partitions require Patches.

RetroOS, on Nov 17 2008, 04:59 AM, said:

rloew, on Nov 13 2008, 09:53 AM, said:

...
AOL terminated their web hosting service so I had to move my website...
...
The standard FDISK can be used in Percentage mode up to 512GiB. I haven't tested the updated version since it is buggy...

rloew, sorry to hear about AOL retiring hometown - looks like AOL have made a lot of people unhappy...
I've been using the updated FDISK for years - care to expand on what's buggy in it?

The upgraded FDISK was not adequately debugged. If you use it to create a small Partition in the first 8GiB, it may produce a corrupt entry if an odd number of cylinders are assigned to the partition.

This post has been edited by rloew: 17 March 2009 - 12:40 AM

[RetroOS]

rloew, on Mar 17 2009, 07:35 PM, said:

...
The upgraded FDISK was not adequately debugged. If you use it to create a small Partition in the first 8GiB, it may produce a corrupt entry if an odd number of cylinders are assigned to the partition.

That's not good!
Any chance of a community patch?
We could call it rloew's FDISK Patch.

[rloew]

RetroOS, on Mar 20 2009, 04:42 AM, said:

rloew, on Mar 17 2009, 07:35 PM, said:

...
The upgraded FDISK was not adequately debugged. If you use it to create a small Partition in the first 8GiB, it may produce a corrupt entry if an odd number of cylinders are assigned to the partition.

That's not good!
Any chance of a community patch?
We could call it rloew's FDISK Patch.

Due to more fundamental limitations, particularly the inability to define a Partition without erasing it, I created my own Partitioning tool called RFDISK.
It provides many more features than FDISK including the setup of a Dual-Boot or Multi-Boot system.
It is also not limited to 512GiB. It can easily handle 2TiB. An experimental version should support 512TiB.

[RetroOS]

Microtek's scanners currently support Windows 9x, and they are also updating the drivers.
I'm using ScanWizard 5 v6.61, released February 2009 in Windows 98SE.
So, if anyone needs a scanner that works in Windows 98/SE/Me, then Microtek is the answer.

[dexter3d]

Sony Vaio VGN-A115M laptop works fine on 98SE, except wireless lan.

- Intel Pentium M 715 Dothan (2MB) 1.5 GhZ - - Intel Speedstep utility v3.1 - Allows multiplier/voltage switching, keeps CPU cool
- Intel 855PM Chipset - Official driver from Intel Version: 6.3.0.1007
- Mobility Radeon 9200 (both ATI Catalyst 6.2 and corresponding Omega drivers)
- Intel 100Pro/VE 100mbps network connection - Intel website
- RealtekAC97 Sound
- ESI Waveterminal U24 USB soundcard


No drivers for:

- Intel 2200BG network connection (wireless lan)

[Philco]

RetroOS, on Dec 13 2008, 03:57 AM, said:

...

Epson on the other hand still produce 9x drivers for their consumer printer range, and you will always be able to download existing drivers!
I've always had Epson printers since the earliest dot matrix days.
Go Epson!

EPSON with support win98:

Epson D92, Epson D120 and Epson R285

[JustinStacey.x]

I have a Medion MIM 2080 cheapie laptop, with mostly VIA components, and I run Windows 98 stably with drivers for sound, LAN, modem, VGA, and everything. One or two ?s in Device Manager but as far as I can see everything does work. And to make it even faster, I ripped out all the webby 'enhancements'.

When I can get the exact hardware spec I will edit here. I bought it in 2006.

[pointertovoid]

Nearly everything about disk capacity limit is wrong.

I regularly use W95b on disks exceeding 32GB and exceeding 137GB. No particular measure for it.

W95-98-Me-2k-Xp-2k3 (but not Fat32) is limited to about 2TB disk capacity for coding sector numbers on 32 bits, a limit of both Windows and of Mbr-type partitions. And Fat32 is the same from W95b to WMe.

W95-98-Me is not limited to Lba28 and doesn't even notice the 137GB barrier. It completely relies on the Bios, and operating these Win over 137GB works without any effort if the Bios is able of it.

Added drivers - like Intel's one - do not modify Windows' behaviour at 137GB, but shortcut a limited Bios to enable Win accessing bigger disks. Just like the "disk installers" provided by disks manufacturers do.

About processors, I've just installed a W95b on a 1400MHz PIIIs 80GB without any patch. This confirms once again that the 350MHz refer to Amd processors.

[rloew]

pointertovoid, on May 24 2009, 04:50 PM, said:

Nearly everything about disk capacity limit is wrong.

I regularly use W95b on disks exceeding 32GB and exceeding 137GB. No particular measure for it.

W95-98-Me-2k-Xp-2k3 (but not Fat32) is limited to about 2TB disk capacity for coding sector numbers on 32 bits, a limit of both Windows and of Mbr-type partitions. And Fat32 is the same from W95b to WMe.

W95-98-Me is not limited to Lba28 and doesn't even notice the 137GB barrier. It completely relies on the Bios, and operating these Win over 137GB works without any effort if the Bios is able of it.

Added drivers - like Intel's one - do not modify Windows' behaviour at 137GB, but shortcut a limited Bios to enable Win accessing bigger disks. Just like the "disk installers" provided by disks manufacturers do.

The 32GB limit is a flaw in the original version of ESDI_506.PDR in Windows 95. It will access more than 32GB, but if for any reason a retry occurs, it switches to CHS mode and will corrupt the Hard Drive if the access was above 32GB.

All of the Windows versions mentioned are limited to 137GB as they only support 28-Bit LBA. Unless running in compatability mode, they do not use the BIOS.
The BIOS must also support 48-Bit LBA, for proper operation, since it is used during Boot before the Hard Disk Driver is loaded.
The 2TB limit is a separate limit due to 32-Bit Sector Addressing. I have written software for Windows 9X to go beyond 2TB.

You may have allocated Partitions extending above the 137GB limit but the problem won't occur until you actually try to put data into sectors above the limit. You will then encounter errors or corruption of files in lower sectors. Newly created files may seem OK but they will have overwriiten other files. SCANDISK will not detect the damage if it is confined to the data sectors of files.

Drivers such as the Intel Application Accelerator or my High Capacity Disk Patch do add support to Windows above 137GB. The Disk Manufacturers Installers supplement the BIOS as does my BOOTMAN. I have had problems getting the Disk Manufacturers Installers to work properly so I wrote my own.
These are two entirely separate fixes.

[dencorso]

The 137GB limit is real all right. Much has been discussed about it. Please refer to the index topic about it that I've compiled, to which there is a link below, in the signature part of this message:

[LoneCrusader]

(Repeat of a post made in another topic, but I felt it was relevant here as well as this is a compatibility thread.)

pointertovoid, on May 24 2009, 04:50 PM, said:

About processors, I've just installed a W95b on a 1400MHz PIIIs 80GB without any patch. This confirms once again that the 350MHz refer to Amd processors.

I'm not saying you're wrong, but I recently installed 95C on a P4 2.0GHz machine and had to use the amdk6upd patch. Probably even the Intel processors at some point above the processor you used hit a limit where the patch is needed.

[rloew]

I have developed a Patch that should allow the use of 512MB Graphics Cards with Windows 9X.
I do not have a Card to test.
If anyone has a 512MB Graphics Card and a Windows 9X driver for it, and is interested in doing some testing, contact me at rloew@hotmail.com

[eidenk]

galahs, on Nov 7 2007, 01:27 AM, said:

Supported Processors:
Opteron
Athlon 64 (FX-57 2.8 GHz is the fastest)

The Opteron 154 is also clocked at 2.8 Ghz and the rare Opteron 156 seems to be the fastest AMD single core at 3 Ghz. Just like the FX-57 they are socket 939 CPUs but I am unsure whether they would work on my motherboard (see my sig) as they are not in the list of supported processors. I did not buy a rather cheap Opteron 154 on eBay last week because of that uncertainty.

This post has been edited by eidenk: 30 May 2009 - 05:46 AM

[LoneCrusader]

rloew, on May 29 2009, 06:43 PM, said:

I have developed a Patch that should allow the use of 512MB Graphics Cards with Windows 9X.
I do not have a Card to test.
If anyone has a 512MB Graphics Card and a Windows 9X driver for it, and is interested in doing some testing, contact me at rloew@hotmail.com

I would be interested, but I also don't have a 512MB card. I might look into buying one if I can find one that I like. I prefer ATI cards, does anyone know of an ATI 512MB AGP card that has 98 drivers? I am already using the Catalyst 6.2 driver pack with my 9800 XT, and if I go into my adapter driver settings and view all hardware, the newest card listed is the x800 XT Platinum Edition which only came in 256MB versions.

[rloew]

I have purchased and tested a GeForcd 6200 512MB AGP Graphics Card.
The Installation CD was limited to XP and Vista, but NVIDIA offers their Forceware 81.98 which works fine.

The NVIDIA Driver doesn't appear to allocate the entire AGP Aperture so it doesn't appear to be limited to 256MB. A 1024MB may work also.

The AGP allocation issue occurs in the ATI RADEON 9600. Since it is only 128MB, there is no problem, but a 512MB Card, with a similar driver, would cause problems.

One person, who answered my request for Beta Testers, has a GeForce 7600GT 512MB AGP Card. He had no problem related to the AGP size but had a Shutdown problem that appears to be unrelated. He has used the NVIDIA 82.69 Driver without Problems.

I am still looking for Testers who have access to 512MB or larger Graphics Cards with Windows 9X compatable drivers to test my Beta AGP Patch.

This post has been edited by rloew: 20 June 2009 - 05:58 PM