Dietmar

@Mov AX, 0xDEAD I make a brandnew install of original XP SP3 with the new acpi.sys on the MSI z690-A Pro DDR4 board. When I look in Device Manager at processors, I see the first 10 cores 0,1,2,3,4,5,6,7,8,9 with the correct name 12600. But the last 2 cores A, B have the name "Intel Processor" Dietmar

@Mov AX, 0xDEAD @Damnation I test my idea with the 12900k processor definition on an MSI z690A Pro DDR4 board with 12600 cpu. On first start I see processor 0,1,2,3,4,5,6,7,8,9 with correct name 12600 in Device Manager. The last 10,11 are still listed as 12900k cpu. I delete in Registry in the entry Enum in Services for intelppm those 2 wrong entries. And voila, after next reboot all processors 0,1,2,3,4,5,6,7,8,9,A,B are shown correct as 12600 cpu. One thing I noticed in Enum for intelppm in Services: The 2 before as 12900k named cpus are not in sequence with the others: They appear in the middle: I mean 0,1,A(wrong name 12900k),B(wrong name 12900k),2,3,4,5,6,7,8,9. And this strange sequence stays in Enum of intelppm, even they are now named correct as 0,1,A,B,2,3,4,5,6,7,8,9 Dietmar

@Mov AX, 0xDEAD @Damnation May the reason is just, that now the processor is defined as normal device and the driver for it has to be given by hand, when it was not there during first installation. For Intel boards it is intelppm.sys for AMD I dont know Dietmar

@Mov AX, 0xDEAD I add all new changes to the Source Files for acpi.sys and compile it. Yessssa works also on the Optiplex 780 Dietmar https://ufile.io/qwlndr5c

@Mov AX, 0xDEAD I think, that this is the solution. Just now I dont remember all steps for compiling and @George King has also already added the new instructions for the new processor definition, me not. So we have to wait for you (via mail acpi.sys), for him or for me (until tomorrow) Dietmar

@Mov AX, 0xDEAD @George King With the new acpi.sys I get this strange BSOD on the Optiplex 780, which I have never seen before. @Mov AX, 0xDEAD Any idea, what this can be? Can you write me more instructions for acpi.sys, how to give more output of this Bsod Dietmar EDIT: In the original 5512 acpi.sys there is no "div" in _ExprOp2 proc near and so no Bsod at this place. Microsoft (R) Windows Debugger Version 6.3.9600.17200 X86 Copyright (c) Microsoft Corporation. All rights reserved. Using NET for debugging Opened WinSock 2.0 Waiting to reconnect... Connected to target 192.168.2.103 on port 50000 on local IP 192.168.2.101. Connected to Windows XP 2600 x86 compatible target at (Sat Dec 3 11:50:12.921 2022 (UTC + 1:00)), ptr64 FALSE Kernel Debugger connection established. ************* Symbol Path validation summary ************** Response Time (ms) Location OK C:\Symbols ************* Symbol Path validation summary ************** Response Time (ms) Location OK C:\symbolssss Symbol search path is: C:\symbolssss Executable search path is: C:\Symbols Windows XP Kernel Version 2600 MP (1 procs) Checked x86 compatible Built by: 2600.xpsp.080413-2133 Machine Name: Kernel base = 0x80a02000 PsLoadedModuleList = 0x80b019e8 System Uptime: not available ************* Symbol Path validation summary ************** Response Time (ms) Location OK E:\binaries.x86fre\Symbols ************* Symbol Path validation summary ************** Response Time (ms) Location OK C:\Symbols ************* Symbol Path validation summary ************** Response Time (ms) Location OK C:\symbolssss OK C:\symbols OK C:\symbolss OK C:\symbolsss OK E:\binaries.x86fre\Symbols Deferred https://msdl.microsoft.com/download/symbols Deferred srv* Break instruction exception - code 80000003 (first chance) nt!DbgBreakPoint: 80ac37e0 cc int 3 kd> g MM: Loader/HAL memory block indicates large pages cannot be used for 80100000->8012777F MTRR feature disabled. KiInitializeMTRR: OS support for MTRRs disabled KiInitializeMTRR: OS support for MTRRs disabled *** Assertion failed: IopInitHalResources == NULL *** Source File: d:\xpsp\base\ntos\io\pnpmgr\pnpinit.c, line 1455 Break repeatedly, break Once, Ignore, terminate Process, or terminate Thread (boipt)? i i *** Fatal System Error: 0x0000007f (0x00000000,0x00000000,0x00000000,0x00000000) Break instruction exception - code 80000003 (first chance) A fatal system error has occurred. Debugger entered on first try; Bugcheck callbacks have not been invoked. A fatal system error has occurred. Connected to Windows XP 2600 x86 compatible target at (Sat Dec 3 11:50:22.078 2022 (UTC + 1:00)), ptr64 FALSE Loading Kernel Symbols ..................................... Loading User Symbols ******************************************************************************* * * * Bugcheck Analysis * * * ******************************************************************************* Use !analyze -v to get detailed debugging information. BugCheck 7F, {0, 0, 0, 0} Probably caused by : ACPI.sys ( ACPI!ExprOp2+102 ) Followup: MachineOwner --------- nt!RtlpBreakWithStatusInstruction: 80ac37ec cc int 3 1: kd> !analyze -v ******************************************************************************* * * * Bugcheck Analysis * * * ******************************************************************************* UNEXPECTED_KERNEL_MODE_TRAP (7f) This means a trap occurred in kernel mode, and it's a trap of a kind that the kernel isn't allowed to have/catch (bound trap) or that is always instant death (double fault). The first number in the bugcheck params is the number of the trap (8 = double fault, etc) Consult an Intel x86 family manual to learn more about what these traps are. Here is a *portion* of those codes: If kv shows a taskGate use .tss on the part before the colon, then kv. Else if kv shows a trapframe use .trap on that value Else .trap on the appropriate frame will show where the trap was taken (on x86, this will be the ebp that goes with the procedure KiTrap) Endif kb will then show the corrected stack. Arguments: Arg1: 00000000, EXCEPTION_DIVIDED_BY_ZERO Arg2: 00000000 Arg3: 00000000 Arg4: 00000000 Debugging Details: ------------------ BUGCHECK_STR: 0x7f_0 TRAP_FRAME: b9d2fca4 -- (.trap 0xffffffffb9d2fca4) ErrCode = 00000000 eax=00000001 ebx=80000000 ecx=8acb0158 edx=00000000 esi=8acb1d7c edi=8acb0000 eip=b96e872c esp=b9d2fd18 ebp=b9d2fd1c iopl=0 nv up ei pl zr na pe nc cs=0008 ss=0010 ds=0023 es=0023 fs=0030 gs=0000 efl=00010246 ACPI!ExprOp2+0x102: b96e872c f7711c div eax,dword ptr [ecx+1Ch] ds:0023:8acb0174=00000000 Resetting default scope DEFAULT_BUCKET_ID: DRIVER_FAULT PROCESS_NAME: System ANALYSIS_VERSION: 6.3.9600.17237 (debuggers(dbg).140716-0327) x86fre LAST_CONTROL_TRANSFER: from 80a30d7b to 80ac37ec STACK_TEXT: b9d2f7f4 80a30d7b 00000003 b9d2fb50 00000000 nt!RtlpBreakWithStatusInstruction b9d2f840 80a319e6 00000003 b96e872c 8acb1d7c nt!KiBugCheckDebugBreak+0x19 b9d2fc20 80a31f52 0000007f 00000000 00000000 nt!KeBugCheck2+0x574 b9d2fc40 80b817b9 0000007f b96e872c 8acb1d7c nt!KeBugCheck+0x14 b9d2fc98 80adfbfa b9d2fca4 b9d2fd1c b96e872c nt!Ki386CheckDivideByZeroTrap+0x41 b9d2fc98 b96e872c b9d2fca4 b9d2fd1c b96e872c nt!KiTrap00+0xaa b9d2fd1c b96eb187 8acb0000 8acb0180 00008000 ACPI!ExprOp2+0x102 [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\type2op.c @ 756] b9d2fd38 b96e590a 8acb0000 8acb1d7c 00000000 ACPI!ParseTerm+0xf7 [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\parser.c @ 588] b9d2fd60 b96e70a3 8acb0000 8ac5d390 b96f0218 ACPI!RunContext+0x5a [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\ctxt.c @ 588] b9d2fd74 b96e7123 8acb0000 00000000 b96f0218 ACPI!InsertReadyQueue+0xa5 [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\sched.c @ 278] b9d2fd8c b96e1d8a 8ac5d390 00000000 8acbe478 ACPI!RestartCtxtPassive+0x28 [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\sched.c @ 384] b9d2fdac 80bd81ac 00000000 00000000 00000000 ACPI!ACPIWorker+0x8c [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\nt\worker.c @ 325] b9d2fddc 80ae4212 b96e1cfe 00000000 00000000 nt!PspSystemThreadStartup+0x34 00000000 00000000 00000000 00000000 00000000 nt!KiThreadStartup+0x16 STACK_COMMAND: kb FOLLOWUP_IP: ACPI!ExprOp2+102 [e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\type2op.c @ 756] b96e872c f7711c div eax,dword ptr [ecx+1Ch] FAULTING_SOURCE_LINE: e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\type2op.c FAULTING_SOURCE_FILE: e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\type2op.c FAULTING_SOURCE_LINE_NUMBER: 756 FAULTING_SOURCE_CODE: No source found for 'e:\software\windowssourcecode\microsoft.leaked.source.code.archive_2020-10-04\nt5src\source\xpsp1\nt\base\busdrv\acpi\driver\amlinew\type2op.c' SYMBOL_STACK_INDEX: 6 SYMBOL_NAME: ACPI!ExprOp2+102 FOLLOWUP_NAME: MachineOwner MODULE_NAME: ACPI IMAGE_NAME: ACPI.sys DEBUG_FLR_IMAGE_TIMESTAMP: 63853811 IMAGE_VERSION: 5.1.2600.7777 FAILURE_BUCKET_ID: 0x7f_0_ACPI!ExprOp2+102 BUCKET_ID: 0x7f_0_ACPI!ExprOp2+102 ANALYSIS_SOURCE: KM FAILURE_ID_HASH_STRING: km:0x7f_0_acpi!exprop2+102 FAILURE_ID_HASH: {af20677d-c060-3ffb-ceae-0764beb57b50} Followup: MachineOwner ---------

@Damnation Geekbench 2.4 bit32 is nice. Or you can just use the tip from @Mov AX, 0xDEAD and take 7zip bit32 time compress for a large folder, full with different files Dietmar

@Damnation Do you make some Benchmark tests with the old acpi.sys and the new one on XP SP3? How is power consumption now on your board Dietmar

@Mov AX, 0xDEAD I found out, what happens. The old acpi.sys and also after my Bios Mod, XP SP3 uses only the 8 power cpu cores with 16 threads. But the new acpi.sys for XP SP3 makes use of the 8+8 cores with 24 threads! And the Win10 version, bit 64 version 1803 also uses only the 8 power cores Dietmar

@Mov AX, 0xDEAD I cant install Win7 SP1 , win8.1 , win10 all bit 32 versions on the Asrock Extreme z690 board. Just now I remembered my tries from before on that board, also without success. So I tried an already installed win7 SP1, win8.1 32 bit. Both crash and on repair mode no input is possible at all Dietmar PS: I check other benchmarks under XP SP3 also. Now with this new acpi.sys, they are all about 30% faster than with old acpi.sys or modded Bios. EDIT: So I use win10 bit64 on the Asrock z690 Extreme board with 12900k cpu. Results: Cinebench 11.5 bit32 XP SP3 43 points win10 29 points win10 22H2 39 points with 24 cores with 7 zip compress of whole system32 folder (1.3 Gbyte) on Optane drive XP SP3 46 sec Win10 68 sec bit32 7zip, 69 sec bit64 7zip 72sec Windows 10 22H2 with 24 cores(!) geekbench2.4 XP SP3 37428 Win10 31900 Interesting is also the power consumption: Under win10 bit64 it also reaches 250 Watt with prime95 and with zero load 29 Watt. This is about double as with my modded Bios under XP, thanks to @Damnation. 2000000! with Mathematica 5.0 on this nice XP SP3 on Optane 905P 27 sec. 2000000! with Mathematica 5.0 on i920 cpu, XP SP3 needs 62 sec on my daily compi, also nice.

@Mov AX, 0xDEAD @George King @Damnation Waaoooh, this acpi.sys counts the cpu's correct!!! I have never seen an 12900k running faster. Power consumption is now 25 Watt with zero loading. And with Benchmark it reaches 250 Watt. But this is the by far best benchmark result for Geekbench 2 until now for XP Dietmar PS: With my modded Bios, this cpu is not so fast, 36% less but also less Watt. @Mov AX, 0xDEAD Any idea, why this happens?

@Mov AX, 0xDEAD @George King @Damnation I make some more tests with the new acpi.sys. First I notice, that the DSDT in Bios 9.04 and 10.04 for the Asrock z690 Extreme board are exact the same. Then I modd this 10.04 Bios the way I did before, for to get the old processor definition. After reboot, I get with the new acpi.sys Bsod 0xA5 (0x11, 0x08, xxx, yyy), know acpi crash because of error in _AMLILoadDDB() . Ok, quick dirty hack in the new acpi.sys with IDAPro with 90 90 to go over this Bsod. Now, compi starts with new acpi.sys, showing the correct 24 cores for the 12900k cpu (8+8). But now comes the most crazy: The speed of the cpu now again only 100%, as before. Power consumption nearly zero with 0% load and maximum <200 Watt via Prime95. This means: Even not all is correct with the new acpi.sys, the processor definition there in gives a jump in cpu power + 36%(!) under XP. I checked this with different benchmarks Dietmar This new acpi.sys https://ufile.io/d82p7icl

@Mov AX, 0xDEAD @George King @Damnation The new acpi.sys with new processor definition works on my Asrock z690 Extreme board with 12900k cpu. But it shows not correct the numbers of cores and cpu's in Device Manager. Core Temp shows 16 cores and 32 threads, CPU-Z shows correct 8+8 cores. With my modified Bios, XP gives Bsod with this acpi.sys. So I flash the original brandnew Bios 10.4 from Asrock Webside for this board Dietmar PS: It looks, as if processors are counted double: This cpu has 24 threads, Device Manager shows 48. Power consumption is now 25 Watt with 0% load of each core. Those 25 Watt are much higher than it was with my modded Bios. EDIT: But the most interesting is: ALL benchmarks show an increase from 100% => 136% (!!!). So i think, those different benchmarks are real. In this case, power consumption reaches 250 Watt. My cooler Dark Rock Pro4 can handle this just at its limit, temperature on all cores go up to 91 degrees Celsius, without any throtteling. The cpu Benchmarks show an unbelievable +36%(!) increase.

@ragnargd Motherboard needs to have in Bios CSM enable possibility. Today I found a board from MSI pro b660m-p ddr4 which has CSM in its Bios, but you cant enable it. So, until now no possibility to install XP SP3, win7 32bit, win8 32bit there. I think, all the Microsoft OSs with bit32 need CSM. No graphik driver for >geforce 980 under XP. No lan driver for Intel >i218 and for Realtek only Lan with 1Gb. Sound works for Realtek until 1200 with driver R2.74. This limits you can overcome with PCIe Slot cards, only CSM has to be there. There is a program called Quibble for to overcome this. For me, it does not work. For Sata, until now always Kai Schtroms Sata driver works. The from @Mov AX, 0xDEAD modded Microsoft driver for USB3 from Win8 works on about 80% from boards. This nice USB driver works also on boards with win7 or win8. Nvme 1.3 from Kai or the modded Microsoft 8 nvme driver from @daniel_k work everywhere. The acpi.sys for XP SP3 drivers work on about 99% of all boards, on more boards than win7 or win8 and even win10(!) support. Until now I have no motherboard, where XP not works Dietmar

@we3fan Yepp, I tested a lot of different xp.iso CDs, all work Dietmar

I found this https://github.com/jeremitu/PatchPae2 and https://ufile.io/su0lnnvh Has somebody tested it, if it is stable Dietmar

Hi, I just test win7 SP1, win8.1 on the Gigabyte z690 UD DDR4 board, both bit32 versions. In win7 SP1 I have to change the acpi.sys against my modded version, because it shows Acpi A5 BSOD during install. The for XP modded USB3 driver works there. But compi is slow. A check shows, that only 790 MB ram are free, the same for win8.1. Any idea, for to get more ram Dietmar

@Computer511212 You can take your original XP CD, name it to the *.iso from Ramsey in the folder and integrate all the needed driver. This is a fantastic possibility of the Ramsey XP Patch Integrator Dietmar PS: Later you update it with all the Posready Updates using this: https://legacyupdate.net/

@Damnation I also cant find the DSDT table in this Bios from ASRock QC6000M . May be, that you can extract it on running compi via RWEverything or acpidump -b Dietmar PS: Do you find a speed compare of the cpu? Interesting board.

@Damnation In win98 you can disable any acpi with the setup /P i switch. Interesting board, for all its devices there are XP SP3 driver, sound, lan .. And Bios is from 2020. I just check, if there are Acpi2.0 orders in their DSDT Dietmar

And here is the same tested Bios Mod for XP for to show processors in Device Manager and bring power consumption down as max as possible and no more "1000" unknown devices there, for the MSI z690-A Pro board. Procedure is the same: Copy only MSI.ROM (BIG letters) to an fresh with Fat32 formatted USB stick and put it into the MSI board, very left down, and press the Flash botton, left direct next Dietmar Biosflash is always a risk. So no garantie. For me it works. https://ufile.io/qr41nwzc You can easy go back to original Bios from webside, copy for this the original Bios to the USB stick and rename the original Bios there to MSI.ROM .

Hi, look for Ramsey XP Dietmar

You can easy go back to your original Bios: Just copy the Original Bios from Webside on the with Fat32 formatted USB stick, rename it and repeat the procedure. The name of the Bios file on this USB stick depends on manufacturer: It is creative.rom for Asrock GIGABYTE.bin for Gigabyte (letters big size and small for bin) MSI.ROM for MSI . I just check this with reading out its new build DSDT, works Dietmar

I found the reason for the Dxdiag.exe crash on z690 boards under XP. It is the LED device under Hid conforme devices in Device Manager. Just disable this (not the USB Hid), and crash is gone Dietmar

I did the same Bios Mod (last version F20b from Gigabyte Webside) for the Gigabyte Z690 UD DDR4 board, for to enable the old processor definition thanks to @Damnation , which shows normal Processor in Device Manager and brings power consumption under XP down as much as possible. All the "Unknown Devices" are gone. You have only to copy the Biosfile GIGABYTE.bin to an fresh with Fat32 formatted USB stick. Put this stick in USB port left side, most up with name Bios written on this port. Compi is shut down, but power is connected. Press Qflash_Plus botton (on board). The orange LED on the board begins to blink, and on your USB stick also. Whole process laasts about 5 min. Dont shut power off as long as the LED is blinking!!! At the end, all LEDs are out. Restart compi, voila you see Bios version F20b, which also works for 13900k etc. Dietmar EDIT: Now no crash because of USB on this board Gigabyte Z690 UD DDR4 when running dxdiag.exe . Before, this Bios even do not recognice a CD-Rom device. Now it does. Gigabyte repairs a lot in their faulty Bios. Now this is also a nice z690 board for XP. The Asrock z690 Extreme board still crashes with dxdiag.exe . Biosflash is always a risk. So no garantie. For me it works. BiosMod for Gigabyte Z690 UD DDR4 https://ufile.io/zpe0ndhy