http://sweetlow.orgfree.com/download/esdi_506.zip

Patches for ESDI_506.PDR from Windows 98SE [98SE] /
RLoew's Terabyte Plus Pack 2.1 [TBP2.1].

Fixes:
1. [TBP2.1 - BigHDD] Handles drives with LBA48 support using algorithm
from Windows NT: LBA48 commands are only used for drives that
really need them, i.e. >= 128GiB, for drives < 128GiB LBA28 is used.
Needed as work-around the rare situation of badly functioning hardware
that reports support for LBA48 commands, but in reality does not
process them.
2. [TBP2.1 - SATA] Interrupt handler:
- Based on code from RLoew's TBP3.0.
- For dedicated interrupts (Legacy ATA controllers) the code is completely
identical in behavior to the original one (no unnecessary access to ports).
- DMA error flag is preserved, only the Interrupt flag is reset. 
3. [TBP2.1 - SATA] Data synchronization with BIOS:
- LBA usage is enabled for drives with LBA support 
without any additional checks, drives without LBA use the old
(full) algorithm
- rollback to PIO if DMA fails
4. [98SE] Driver retrieves shared interrupt flag not from the CONFIGMG
shared interrupt flag, but from the level/front interrupt flag.
The CONFIGMG incorrectly changes the shared interrupt flag
for multi-function devices (which are the vast majority of standard
PCI ATA controllers).
5. [98SE] Fixed minor bug of non-saving register with the value of error
of command execution for drives with removable media and media status
notification enabled.
6. [98SE] Changed drive characteristic flags for the default record.

Additions:
1. [98SE] Added IOR_IDE_PASS_THROUGH request processing for ATAPI devices.
Also this request is correctly processed for ATA devices
with sector size other than 512 bytes. Allows you to execute
drive identification commands and SMART commands for such drives.
It is recommended to use together with updated SMARTVSD.VXD
for full functionality.

Fixes 2 - 6 and Addition 1 are non-specific to TBP2.1.

Corollary projects updated accordingly:

https://github.com/LordOfMice/Tools/blob/master/atanames.zip - Right names of ATA drives plus version of device firmware in Device Manager for Windows 9x/ME

http://sweetlow.orgfree.com/download/smartvsd.zip Patches for SMARTVSD.VXD - works for any combination of ATA channels and devices on these channels and more

 

http://sweetlow.orgfree.com/download/esdi_506.zip

Added 2025/03/01:

1. [TBP2.1 - SATA] Data synchronization with BIOS:
- Returned physical geometry extraction from IDENTIFY for drives with LBA.
2. [98SE] Drive parameters setting error in drive reset handler
no longer causes a general read/write error for drives with LBA.
3. [98SE] Uninitialized CMOS drive type definition variable.
For all but the first two drives on Legacy Primary Channel, it could have been
anything, usually what was on Legacy Primary Channel (i.e. type 47),
and if there was no Legacy Primary Channel, it was 0 or really something
random. Now anything other than real CMOS data will give you a 95 disk type.
This reflects the FFFFh value that was already used in the code for the
unset parameter in one place, and now it is also used for the default value.
4. [98SE] Minor change: IDENTIFY read error no longer resets other flags
and sets IDENTIFY error flag, but only sets the IDENTIFY error flag
(OR instead of MOV).
 

Well, knowing now internals of ESDI_506.PDR I solved this puzzle completely from the second attempt
(the first attempt was disabling IOR_IDE_PASS_THROUGH handler to prevent BSoDs):

IDENTIFY (including ATAPI devices) and SMART commands work in AHCI.PDR
Plus fixed few other bugs seen in process of main development, I personally recommend to pay attention to N5.

http://sweetlow.orgfree.com/download/ahci.zip

readme:

Patches of RLoew's AHCI.PDR 3.0
1. Fixed broken handler of information about internal structures
(system hangs/reboots when trying to read them).
2. Fixed accessing an uninitialized pointer variable (wild pointer)
in the power-down handler (SYS_POWER_DOWN) leading to BSoD / hang-on /
reboot.
Thanks to https://github.com/PluMGMK
3. Fixed bug in the handler for ATA commands without data transfer,
such as standby drive or working with removable media.
These commands were previously just not executed.
4. Made new working handler of IOR_IDE_PASS_THROUGH request,
IDENTIFY (including ATAPI drives) and SMART commands are supported.
5. Fixed random CPU port write access in some cases on executing
Input/Output Request.
6. Fixed bug in enable Media Status Notification function.
7. Memory copy speedup in Input/Output Request handler.
 

https://github.com/oerg866/win98-quickinstall/issues/22#issuecomment-4488077576
RLoew's Terabyte Plus 3.0 ESDI_506.PDR data corruption bug. Fix:
http://sweetlow.orgfree.com/download/esdi_506_tbp3.zip

http://sweetlow.orgfree.com/download/esdi_506.zip

Fixes:
1. [TBP2.1 - BigHDD] Fixed - data underrun in the initialization procedure of
PM Only drives. This resulted in slowdown on booting (sometimes significantly)
and caused other problems.
2. [98SE] ATA devices are no longer considered to be in a stopped state
(SpinDown) after initialization.
3. [98SE] The CMOS type of a PM Only drive on non-primary ATA channels
when added after the boot stage is now consistent with that during boot.
4. [98SE] Fixed the setting of the I/O request call method in the drive
initialization procedure. This did not affect functionality in the vast
majority of cases, but has been improved nonetheless.

Update is recommended due to point 1.

On 4/19/2026 at 3:15 AM, SweetLow said:

Well, knowing now internals of ESDI_506.PDR I solved this puzzle completely from the second attempt
(the first attempt was disabling IOR_IDE_PASS_THROUGH handler to prevent BSoDs):

IDENTIFY (including ATAPI devices) and SMART commands work in AHCI.PDR
Plus fixed few other bugs seen in process of main development, I personally recommend to pay attention to N5.

http://sweetlow.orgfree.com/download/ahci.zip

readme:

Patches of RLoew's AHCI.PDR 3.0
1. Fixed broken handler of information about internal structures
(system hangs/reboots when trying to read them).
2. Fixed accessing an uninitialized pointer variable (wild pointer)
in the power-down handler (SYS_POWER_DOWN) leading to BSoD / hang-on /
reboot.
Thanks to https://github.com/PluMGMK
3. Fixed bug in the handler for ATA commands without data transfer,
such as standby drive or working with removable media.
These commands were previously just not executed.
4. Made new working handler of IOR_IDE_PASS_THROUGH request,
IDENTIFY (including ATAPI drives) and SMART commands are supported.
5. Fixed random CPU port write access in some cases on executing
Input/Output Request.
6. Fixed bug in enable Media Status Notification function.
7. Memory copy speedup in Input/Output Request handler.
 

SweetLow, can any of your updated versions work on Windows 95 Osr 2.x? I was thinking i can use the updated ACHI.PDR, the original package of Rloew mentions Windows 95. If i'm not mistaken, ESDI_506.PDR is only relevant when keeping devices in IDE mode when using Sata devices ( requiring also PATCHSATA ), though may require the 1116 version of esdi_506.pdr on WIN95 ( i have not tested ), but the specific version of ESDI_506.PDR becomes irrelevant when switching to AHCI mode and using the ACHI drivers. 

Can users replace all updated files in your package into the original package of Rloew's files and use in the same way, or are there certain dependencies, like needing certain esdi_506.pdr versions, other updates, etc?

>can any of your updated versions work on Windows 95 Osr 2.x?
EHCI_506.PDR - probably:
https://msfn.org/board/topic/187782-silicon-image-si680other-unsupported-pci-ide-controller-under-windows-95-osr-20/page/2/#findComment-1286966
I don't use any of windows 95 now. So check and tell us.

>I was thinking i can use the updated ACHI.PDR
Yes.

> requiring also PATCHSATA
No. This ESDI_506.PDR is prepatched standalone version (including enhanced version of PATCHSATA). You can use new .INF file too for "SATA" adapters:
http://sweetlow.orgfree.com/download/WIN98SE_ATAPCIBM.ZIP
but there is not strict dependency between binaries and .INFs. And "WIN98SE" in the name of file is just the source of original file, it should work on any Windows 9x.

>but the specific version of ESDI_506.PDR becomes irrelevant when switching to AHCI mode and using the ACHI drivers
Yes. ESDI_506.PDR and AHCI.PDR are completely independent (as the hardware they handle are independent).

>Can users replace all updated files in your package into the original package of Rloew's files and use in the same way
Yes. They can be used this way too. But not with RLoew's packages only. Anywhere you meet ESDI_506.PDR (of according OS, of course; Millenium ESDI_506.PDR has some additional features for example and is not recommended to be replaced) or AHCI.PDR just replace them with these patched versions.

Thanks, this helps tremendously, I see that your updated inf from "WIN98SE_ATAPCIBM.ZIP" matches what the original sata.inf from TBPLUS package provides, but also adds some extra device descriptions, so i am using that version renamed to SATA.INF in the TBPLUS package. 

Since you say the updated ESDI_506.PDR is a prepatched standalone / enhanced version of PATCHSATA, does that mean PATCHSATA is no longer required, since your updated esdi_506.pdr already does what patchsata does but better? Also is the situation similar to windows 95? Lastly, if keeping PATCHSATA, can i also use the older sata.inf with the updated SATA.INF ( renamed one ). 

13 hours ago, SweetLow said:

Millenium ESDI_506.PDR has some additional features

which "additional features"?

35 minutes ago, MERCURY127 said:

which "additional features"?

Hibernate support AFAIK.

1 hour ago, cov3rt said:

does that mean PATCHSATA is no longer required

Yes for _PATCH_ itself. But you can use SATA.INF, of course, if you wish. I said about binaries and .INFs already above.

1 hour ago, cov3rt said:

can i also use the older sata.inf with the updated SATA.INF ( renamed one )

I don't understand what did you ask. You can not use two .INFs simultaneously. You should select one of them for particular adapter.

@SweetLow Sorry if it's an ignorant question, but is it possible to install some sort of patch in Win98 SE so that hibernation gets enabled? I ask this because of the reference to Millenium ESDI_506.PDR above

12 hours ago, sk9392 said:

is it possible to install some sort of patch in Win98 SE so that hibernation gets enabled?

No. S4OS is supported by Millenium only (and in OEM versions only AFAIK). 98 supports S4BIOS, but enabling this is not the OS task.

Edited July 2Jul 2 by SweetLow
clarification