pappyN4v2

Earlier I mentioned that modern Intel was a lot faster in a specific software compared to modern AMD. After investigating, turns out that the issue is not with XP/XP64 on modern hardware, but with the company compiler the program was created with. Back in the XP era, a compiler from a certain lawsuit happy company compiled in such a way so that CPUID was checked, and if the cpu was not from that company, lets say AMD, then it used basic instruction set only instead of SIMD (MMX/SSE/SSE2/etc...) https://www.felixcloutier.com/x86/cpuid snippet here is just after cpuid was run and loaded ebx,edx,ecx into memory, you can see the vendor identification string being compared to the memory locations cmp [ebp+...], 75 6E .. .. h if it matches then flag is set to "1" mov eax, 1 if cpu was does not pass vendor identification string, then it sets flag to zero xor eax,eax code continues on to test if flag was set to 1, and if it was continues on to the rest of the code where the SIMD occurs test eax,eax if not then goes to the end of the subroutine to the same place as if it was a "good" vendor identification string but one that reported no SIMD capabilities. cmp [ebp+..], 0 jz loc_5431EC To fix problem, you can search in IDA for the 3x 4 character grouping starting with 756E.... and replacing the first conditional jump JNZ "75 xx" with an JMP "EB yy." so that it skips the checks and goes straight to setting the flag "mov eax,1". Fortunately the certain company compiler was not really popular for games, but more for computational/scientific software. By making the change the AMD processor got a 3x speed boost thanks to now being allowed to use SIMD instructions.

Tried to replicate this problem but couldnt. Running XP64 with latest 2019 updates, GT710 with two monitors, DVI 1280x1024, HDMI 1920x1200 either 368.81 or 355.98 worked no BSOD. Downgraded videoprt.sys from latest 5.2.3790.4032 to 3959 from SP2 ISO, still worked. Only nvidia change from original is that I removed LargePageMinimum from INF as it causes Samsung and Int** NVME drivers to not work (doesnt have problem with win7 NVME drivers). Looked over 368.81 patch by Tal Aloni, and logic in 355.98 seems like it should work the same. nv4_mini.sys x64 368.81 jnz short loc_35611 movsxd rcx, dword ptr [rsp+68h+var_40] ;swap mov eax, dword ptr [rsp+68h+var_40] jnz short loc_35611 ;swap, address-4 mov [rbx+rax*8+2F4h], ebp mov [rbx+rcx*8+2F4h], ebp ;rax to rcx mov [rbx+rax*8+2F8h], r15d mov [rbx+rcx*8+2F8h], r15d ;rax to rcx mov dword ptr [rdi+rbx+0B8h], 6 mov dword ptr [rdi+rbx+0B8h], 6 jmp loc_357C3 jmp loc_357C3 ;address-1 cmp r11d, 2 cmp r11d, 2 ;loc_35611 jnz loc_35727 jnz loc_35727 ;address-1 mov ecx, dword ptr [rsp+68h+var_40] nop nop nop ;nop nv4_mini.sys x64 355.98 jnz short loc_37CD2 movsxd rcx, dword ptr [rsp+88h+var_60] mov eax, dword ptr [rsp+88h+var_60] jnz short loc_37CD2 mov [rbx+rax*8+2F4h], ebp mov [rbx+rcx*8+2F4h], ebp mov [rbx+rax*8+2F8h], r15d mov [rbx+rcx*8+2F8h], r15d mov dword ptr [rdi+rbx+0B8h], 6 mov dword ptr [rdi+rbx+0B8h], 6 jmp loc_37E89 jmp loc_37E89 cmp r11d, 2 cmp r11d, 2 jnz loc_37DE8 jnz loc_37DE8 mov ecx, dword ptr [rsp+88h+var_60] nop nop nop Not quite exactly the same code in assembley, as jmp loc_357C3 in 368.81 is a little different location (E9 B1 01 00 00) compared to jmp loc_37E89 in 355.98 (E9 B6 01 00 00). nv4_mini.sys x64 368.81 @255ec 75 23 8B 44 24 28 89 AC C3 F4 02 00 00 44 89 BC C3 F8 02 00 00 C7 84 1F B8 00 00 00 06 00 00 00 E9 B2 01 00 00 41 83 FB 02 0F 85 0C 01 00 00 8B 4C 24 28 -> 48 63 4C 24 28 75 1F 89 AC CB F4 02 00 00 44 89 BC CB F8 02 00 00 C7 84 1F B8 00 00 00 06 00 00 00 E9 B1 01 00 00 41 83 FB 02 0F 85 0B 01 00 00 90 90 90 plus PEchecksum.exe nv4_mini.sys x64 355.98 @27cad 75 23 8B 44 24 28 89 AC C3 F4 02 00 00 44 89 BC C3 F8 02 00 00 C7 84 1F B8 00 00 00 06 00 00 00 E9 B7 01 00 00 41 83 FB 02 0F 85 0C 01 00 00 8B 4C 24 28 -> 48 63 4C 24 28 75 1F 89 AC CB F4 02 00 00 44 89 BC CB F8 02 00 00 C7 84 1F B8 00 00 00 06 00 00 00 E9 B6 01 00 00 41 83 FB 02 0F 85 0B 01 00 00 90 90 90 plus PEchecksum.exe Since I didnt have dual monitor BSOD problem 355.98 patch is not tested other than to say that it continues to work fine as unpatched version.

I saw some comments on amazon from other people also having issues with XP and IDE on patriot SSDs and led me to this post among others. I'm curious what model/chipset yours was running. I bought a Patriot Burst Elite 120GB and an ADATA SU655 120GB for an older retro PC and both worked for me either as fresh install or as secondary drive. I used XP with the latest 2019 updates. Tested on a FM2 motherboard with SATA port, 939 motherboard with ATA133, and socket7 motherboard with ATA33. ADATA was a MAS0902 controller based on the info from JMicron tool from: http://vlo.name:3000/ssdtool/ Patriot was some type of Phison based on a similar firmware version from different a model: https://smarthdd.com/database/Patriot-P210-128GB/HDFED3.2/ Patriot Burst Elite 120GB (Phison ??): PRO Can use HPA (host protected area) PRO Detailed attributes CON works with SATA-IDE, ATA133 and ATA33 with adapter, but massive slowdown! ADATA SU655 120GB (MAS0902): CON Can not set HPA! CON Minimal attributes, temp reading also fluctuated to below room temp! PRO works with SATA-IDE, ATA133 and ATA33 with adapter, normal slowdown

Just incase someone else might be interested in patching tcpip for the latest versions. Run PEchecksum.exe after patching below. XP tcpip.sys 5.1.2600.7510 @4fbc6 0A 00 00 -> FF FF FF XP tcpip6.sys 5.1.2600.7510 @33EB3 0A 00 00 -> FF FF FF XP64 tcpip.sys 5.2.3790.6527 @c00cc 0A 00 00 -> FF FF FF XP64 tcpip6.sys 5.2.3790.6527 @5e9a0 0A 00 00 -> FF FF FF where 0x0A = 10 is the default, and 0xFFFFFF = 16777215 is the maximum

I use this https://github.com/temerkhanov/SetTimerService on XP guests in VMware that has distorted audio on modern CPUs. Noticed it when i upgraded to Ryzen (Zen2). Host XP was fine, but got audio distortion when running XP as guest VM. I for sure was not using the "frequency divider fix" at the time. I can't remember if "HAL_acpitimer_fix" was being used. My XP ISO that i use for VMs has SetTimerService set up to run by default now, so I never did do a proper test. If someone wants to try it: XP host + VM Workstation 10 + XP guest on Ryzen using pure XP ISO should get the audio distortion issue. Then try the hal with timer+frequency fix and see if it stops distortion by itself. If it doesn't, then SetTimerService is still needed for XP VMs.

I had the same "unsupported browser" issue for CAPTCHAs as @raddy for NewMoon. Noticed it a couple of days ago, i think Oct 1 or 2 is when it stopped working. Both on an older build and on the latest build of NM. Works fine on basilisk. Changing the useragent also fixed the issue for me. I ended up using one of the other overrides strings that was already in NewMoon.

@DietmarDo you remember what is difference between v8 and acpiSkull3new.sys from 2024.11.11? On Intel Asrock B760M-HDV/M.2/D4, both 2003x64 and 2003x86 ACPI 7777.8 - 2023.04.17 BSOD A5 (03,...,C0000034,...) 7777.8 - 2023.04.17 - A5x11+A5x03 BSOD 0A (00,02,00,...) acpiSkull3new.sys - 2024.11.11 works I would like to make same change to 2003x86 v8 to test something, but error is not C0140010 so must be something else that was patched between v8+A5x11+A5x3 and this one? New file shows very different in hex compare, I guess maybe compiled from new source

Bought an Intel for the first time in over a decade. Patched in the TSC frequency fix that should be used after the PmTimer is patched With CPUz started and WinTimerTester running a test I see the known clock drift problem. Ryzen does not have the issue so I wasnt able to test it until now. I'm assuming still no fix with windows patch or BIOS? For now, I use classic XP era Dimension4 time to keep sync. Like with XP, add shr to shift right, value is 0A hex or 10 decimal, so 2^10=1024 https://msfn.org/board/topic/183489-windows-xp2003-x32-on-modern-hardware-faq/page/4/#findComment-1235566 x64 hal.dll 5.2.3790.6912 https://ufile.io/j28adslm PmTimer @38881 73 -> EB @40186 73 -> EB TSC frequency @F890 74 0C 65 48 8B 04 25 C0 00 00 00 48 89 01 0F 31 48 C1 E2 20 48 0B C2 C3 CC CC CC CC CC CC CC CC -> 74 10 65 48 8B 04 25 C0 00 00 00 48 C1 E8 0A 48 89 01 0F 31 48 C1 E2 20 48 0B C2 48 C1 E8 0A C3 in KeQueryPerformanceCounter ========================================= 74 0C jz short loc_XXX 65 48 8B 04 25 C0 00 00 00 mov rax, gs:0C0h 48 89 01 mov [rcx], rax 0F 31 rdtsc ;loc_XXX 48 C1 E2 20 shl rdx, 20h 48 0B C2 or rax, rdx C3 retn change to ========================================= 74 10 jz short loc_XXX 65 48 8B 04 25 C0 00 00 00 mov rax, gs:0C0h 48 C1 E8 0A shr rax, 0Ah <--------- 48 89 01 mov [rcx], rax 0F 31 rdtsc ;loc_XXX 48 C1 E2 20 shl rdx, 20h 48 0B C2 or rax, rdx 48 C1 E8 0A shr rax, 0Ah <--------- C3 retn

Hmm, guess I missed the boat. Last time I logged in was before the summer with just my username. Password is fine, but I dont remember the email address I used to register @Tripredacus Can I get back into pappyN4, or shall i just be pappyN4v2 from now on? for bonafides, my IP address should be the same as before... or PM on wincert or mydigitallife with same username