user57

a alignment is made for a "piece of area" if you have a "align area" for 1 MB (aka 1´048´576) then 2048 sectors of 512 bits would fit in there (or 256 sectors of 4k (4096)) if you dont have a "align area" that rounds up correctly it might lead into a sector, cluster, track, cylinder, whatever in between a other the 512 bit and 4096 bit sector question been around for a while now, the hardware/firmware can also handle this - in this case this in then made in the harddrive not the operation system anymore if the firmware handles that right it can emulate logical and physical sectors - a drive is made at least of 3 maybe more parts (cluster, cylinder, sector) - its a combination not a single piece like "i want this specific sector" this is very common for other electronic parts to translate for example a "logical address" to a "physical address", or segmentation (thus 32 bit can write past 4 GB) whereehere the electric signals in the ram apear is then part of that -hardware piece- or in case of a harddrive in hand´s of that firmware this also make the -32 wires question- come out again - a USB only has 4 wires - and a usb cable can write to a "terabyte + harddrive" - you dont need always 32 or 64 wires to do this i wonder why noone has made the related change in the windows xp operating system yet - basicly it sounds simple to ask the harddrive what sectors it uses (sometimes called the AF (advanced format)) but its just 4 k sectors the term AF sounds a bit to high for opinion after that the right controlment for the 4 k sector has to be written - if it reads out a different sector size it handles for that size https://learn.microsoft.com/en-us/troubleshoot/windows-server/backup-and-storage/support-policy-4k-sector-hard-drives the x-box solution maybe ? if not it was like everything is a paragon driver problem

actually i have often delays opening a youtube video with my win7 machine it might actually has something to do with this but im not sure useally it are around 10 seconds that problem apeared rather recently

to me to that sounds very weird, but when fixing a pc problem i often had someone putting it that way this would be a case for a "being on the place" for a better meaning the idea to use a cable is not a bad idea, but if that connection problems persists in that case it´s likely to be that device i would disable wlan in that router, just in case someone of your neighbors somehow can connect to your router

ok next besides assembly would be c++, windows drivers, physical harddrive knowledge and how its programmed to (not engine/api like createfile and writefile or script like "copyfile c:\a.txt c:\b.txt) to write this to a windows xp driver you also have to have knowledge about pe headers sections, relocs ect.

right NTSTATUS NTAPI CsrNewThread(VOID) { return NtRegisterThreadTerminatePort(CsrApiPort); } while CsrApiPort is defined as HANDLE (HANDLE CsrApiPort;) in this case that CsrApiPort variable is empty

it looks like someone has tryed to debug a driver without having any knowledge debugging is not something you can do without any knowledge its not easy stuff either for example its a lot of math, its not like a script language a bare minimum to start with would be to know assembly https://www.ic.unicamp.br/~pannain/mc404/aulas/pdfs/Art Of Intel x86 Assembly.pdf but that´s the bare minimum its not stuff you can just do, like lifting a few weights around, or trying some drivers or some different versions, maybe putting a few different hardware types into a pc you litterally trying to make a own CPU - and not just putting it into a computer no you making it this time a-z

there CsrNewThread is just a redirection to NtRegisterThreadTerminatePort if thats all you just have to use NtRegisterThreadTerminatePort, xp has that function

Csr functions useally came from csrsrv.dll i think CsrNewThread is just a different name for CsrCreateThread those are often very similiar to CreateThread or NtCreateThread it would be about looking its parameters and purpose for exact detail, but xp has many of these functions https://www.geoffchappell.com/studies/windows/win32/csrsrv/api/process/thread.htm

do i understand this right ? the youtube downloader for xp (aka yt-dlp) only could get the h.264 format in windows xp ? but the chrome-player (thats the thing what plays the video from youtube ect.) seems very capable to do the "h.265 codec" one of the h.265 decendent´s is called VP9 - in xp the browser (like supermium) it also could not handle VP9 ? when i tryed the "KI" the KI did the same thing to nicolaasjan what the KI did to me, it pretended like to know the problem but everybody could say something like the KI say like "it is because the fluctuation in the internal functions" -> but basicly that says nothing, rather to write something like this describe "he/KI dont know either" the KI did the same to me i asked it up what REG_MULTI_SZ_APPENDED is and why these can have 2 constand´s then it firstly say "that is right" - but the correct answers would be that it is not - the inf installer has a own set of constand´s that´s why it has 2 constands - it has nothing to do with a c++ compiler or the registry constand´s nor the registry functions so the KI told me some bullcr.. and i was more confused then before (but we get the point the KI´s answer said it like it knows that, but actually the KI dont) VP9 would be doable it is open source, openssl if it has a age-problem would also be doable - there would be the possiblity to write the cryptografic instructions if they are in SSE - then you still could translate them to normal instruction, that also would be possible in sence of vista-10 function - we would have some on our list - one example would be the SRW-locks, those dont exits in XP but there implementations that works (they are function tested) it would depent on how many functions python would be asking from what i understand it was going like "hmm python 9.11 + didnt compile up anymore" now we done !? so do we know the problem more precisely ? openssl ? vista functions (in this case what ones) ? vp9 ? the thing with the frames should be mentioned, a video is not a game in a game i could not even play with 60 frames even with 80 frames i could reconize that framerate with 100 frames it was ok but a video this is however not the case often 24-30 are enough - the motion blur rather depents on the shutter-speed a shutter speed is not a framerate a shutter speed is how many pictures are taken in a certain speed like 1/60 thats 0,016 seconds to take 1 frame you dont need 60 frames to make this happen you can take 24-30 frames of 1/60 frames either (in 1 second intervals = 24 frames with 1/60 taken pictures) if you dont want the motion blur at all you take at least 1/500 for barly motion blur, 1/1000 for close to none , 1/2000 for no motion blur at all for fight scenes that look well (then you just take 24 frames that was taken in 1/2000 speed) 1 thing to tell here is that often for 24 frames also 1/24 shots are used 1/24 can make the picture to wash (not only motion blur) - higher speeds here dont have that problem so much or not at all if very fast - but they do not depend having 60 frames - in this case you can also have lets say 24 pictures/frames taken with 1/1000 speed what you can say about a faster shutter speed is : it tends to sharpen the picture (if the iso dont has to be increased) sharpen the picture also means the video looks sharper or better - a video is just a picture repeated 24 times (sometimes 50 or 60) but in a video that is often not as reconizeable as with a game - that would remind a 3d-object discussion and a 2d video discussion - while 2d would do just fine for a video more frames are not bad ... but the benefits are mostly (at least for video) are not that high anymore - and not to compare with the frame discussion from a game however for shutter speeds of that level you need a fast lens the so called t-stop or f-stop, an alternativ would be to be in a good light (like sunlight) (in older movie´s you often see that they used outdoor lights to have a better iso, that was when FILM was still around - i saw them and even in a shadow scene the video looked very bright - that was because a outdoor light was used) if either you have a slow lens or a bad light you have to increase the iso - and increasing the iso means less pixels and less dynamic range! (dynamic range means that the texture in bright and dark parts have more detail - a problem special small sensors have - thats why they use tricks - one trick is to take multiple pictures at different shutter speeds - and overlap them) so far i have not seen what is really missing, so more information would be very welcome

its not a security measure, its security against the user or "non-microsoft" (if i remember right they had always a grude against a bootkit (even when it was totally legal)) so they decide what you do there - that would describe it right i would not call it security for that reason - its a missleading term that trys to confuse and lead it to a different discussion point its a old talk like about the monopol position microsoft is taken action for (not the first time) there was something like that even in very past times like with netscape - where microsoft where sued for taking monopol positions after that list they doing that then is very long - its litterally about billions of money where microsoft where sued for this surely this is about child-abuse or terrorism ? i dont think so - its just a company interest - calling out that is the same missleading term - if you would accept it like that you actually ruined yourself and fall for the trick

well it raise questions why they are not able to make it xp compatible, they are deeply into this stuff is it a problem of skill ? like somebody just used someones code that was wrote for win10, or a must to a certain engine version (that would also be a python version) it also could have come that python is a scriptish language like java - if you do write a code there it might just depents on the java engine personally i have never seen this to happen with a normal c++ style but going a little of-topic - this is bad for vista 7 and even 10 users - this version around play might just hit them at some point either - that might not be now but that would be a thing - they are trying with the vs2022 version already (and trying to get rid even 10) if it comes over a next newer python version then you run into the same problem - again it raise the question if you have insight into the real code - not just it´s low level/script functions

from my understanding (is not absolute - or maybe someone just correct me up) it has 5 main things related to the harddrive cylinder cluster track sector and head those get translated by the firmware of the harddrive to a LBA address the LBA registers seems 48 bit in total what makes sence (since its 3 times 16 bit = 48 = 48bit LBA) (LBAlo, LBAmid, LBAhigh) those are a combination, the real thing not neccesary always needs 32 wires, it actually could use less wires to transfer the data either the speed rather would be limited to what these wires can progress by frequency a USB stick for example dont have so many wires (even tho running with a 32 bit os or maybe 64 bit os) https://techdiagrammer.com/wp-content/images/usb_a_wire_diagram_hihdn.jpg but we getting the point - we not neccesary need always 2, 32 or 64 bit wires this formula often came up: LBA = (C × HPC + H) × SPT + (S − 1) C, H and S are the cylinder number, the head number, and the sector number LBA is the logical block address HPC is the maximum number of heads per cylinder SPT is the maximum number of sectors per track lets say the sector is 512 bytes then you can multiply the indicator by 512 with 4096(4k) by 4096 that would also not be limited if you have a track containing lets say 1000 sectors the next track then would point into a next 1000 sector track if then the sector 0 would have 512 bytes then the indicator 0 would point into track 0 and sector 0 the same things would be the case if its a head a cluster or a cylinder this cylinder sector head would actually be translated by the firmware to a pysical position - that would be like its job it also can do this software side - it actually dont need to have these things - as long they are transtated correctly by the firmware of the harddrive for a drive it seems that you have to set these things before you output your data stream (outsb (byte/8 bit), outsw (word/16 bit), outsd(dword/32bit)) if you see a code like "move me 4 mb of ram" - no its not doing that that way for memory there are useally 32 bit moves (this is a small piece of data) - this then is repeated as soon 4 mb are reached for the harddrive it seems that you actually dont have just an "offset" you have to set where first - then you write the output again either 8 bit 16 bit or 32 bit wise the reason 8 16 32 bit maybe 64 bit or even 128 bit (sse) are not slower is because the cpu/hardware can detect this progress and translate this to a faster progress (aka faster speed) what xp (or windows useally give us) at usermode is the overlappend structure - that one is made of 64 bits it probaly gets translated into the right progress that the harddrive is needing ... here the logic would play a role - maybe they limited it to a dword (32 bit) * 512 (common sector size) = ~ 2,19 TB that with the 4k sectors are not a bad idea because then the data just increases 8 times 2,2 * 8 = ~ 17,52 TB it would be nice to have insight into that paragon driver, there was also a 3 tb+ unlocker from asrock, the information might be out there somewhere

i think it originates from this kernel extender : it allow some drivers from 7-10 to use on xp, but i do not think it can fix your sound card driver problem dont forget that win7 is a successor to vista, win7 is just a extended vista, win7 has more functions while holding same compatibility after win8 the next windows versions gone a different direction for xp its a different story it is successor to win2000, and the previous nt versions also server2003 originates from xp

so just using the right paragon driver solved it up ? thats good news that the /pae question related to that problem would be finally closed

i dont think that will work in this case you litterally asking to write a driver for that specific device, without having knowledge of that device (like that company that created it has) if its not a open source driver sounds a close to mission impossible to me porting an existing driver is not easy either you are likely to create all the missing functions that win8 driver lacks in windows xp - what actually is a lot of work also you could ask their driver devlopers if they are doing that - but normally they ignore a such request

i do not know python but if its like a normal compiler it searches for a main() or winmain(), the next error talks something about __main__ looking these 2 lines might be of interests it seems to be some kind of call "PyInstaller.__main__" a import useally is a function that a modules exports (such as a dll (or dllmain()) or in case of executable winmain) combined it says in "PyInstaller" + "__main__" was not found or pyinstaller was not reconized as "package" File "C:\Users\Admin\Documents\GitHub\xl-converter\misc\pyinstaller\PyInstaller.py", line 15, in <module> from PyInstaller.__main__ import run "ModuleNotFoundError: No module named 'PyInstaller.__main__'; 'PyInstaller' is not a package [end of output]" one way i could think of this is happening is that winver in the peheader is set to 6 or higher (6 are vista and 10) in this case a executable caller (such as createprocess) cant open the module - also from consideration the answer from compilers can sometimes not be very precise, it also could be a small type error/syntax error near that area aka PyInstaller at spot __main__ however i do not know the syntax in python

the answer from the developer is not bad, but he also say he dont know it so exactly either but i would have to go with close to nothing to just write something to this connections are made via winsock - here TCP/IP procolls and TLS (aka the named openssl) can be involved if its a winHTTP request the HTTP request also sends what OS you are on - some actually then refuse the connection if they see this that would be changeable the idea going to a changed python yea - i also said that - even tho i had much less useful detail´s a other small idea is to look if the same code work for a other OS, if so its likely the internals (the functions that come next) somehow mess up thats also what the python devloper said with other words the openssl problem might be possible to determinate by just trying a different OS - if it has the same code then its not openssl a problem i can see here that this rely´s on script functions (low level) the next tier functions are deeper (like the named winsock) if its somewhere at this spot you need to debug these functions - and see why they are breaking up this is the other problem i talked about - if you would have a normal c++ style you could see the most internal functions (instead of only "connection failed" or something like "urllib3.connection.HTTPSConnection object at 0x032F9910>: Failed to establish a new connection")

wasnt that dietmar´s terretory ? it is nice to have a kernel extender that provide these missing functions or maybe a missing piece that might be a solution there is maybe a different approach to solve this problem - a driver from that company today´s OS such as win10 provide standart functions to solve this for a device - then something like this project just works without having a own driver the problem we faced here is that win10 has newer codes in the relevant drivers - its like a standardized code to control certain devices the problem with that is that this is not very OS-independant the most drivers/or and devices useally define 3 basics: writes control read these are somewhat somehow written into that standardized code a driver from the company could also provide the same thing - they neither have to bind a such driver to windows 10 functions it would be doable - then they might would have a few more customers if lets say a XP see their product he could see "xp driver present" or something like that and buy it up like that it is like limited to win10 maybe you could ask in their forums about this

im not someone who programs in python but they say its intensional to drop support for python 3.9 the commits say that they could support it still (so also that devloper says there) ... but they dont want to and that error says it has to do with "urllib3" of python - what is part of their removement in newer python versions they say everything below "urllib3" is dropped urllib3 would be the thing to look for, like what functions of other then XP they used here - and if you can make this version compatible with xp

that´s a important thing to talk about again often newer app´s depending on side things, like runtimes(c-runtime, sdt), they require cmake, they require a new c++ style (not only the ones that allow new types of writings, rather something like "currentpath" - this is going into a direction of a script rather then a language (in java a scriptish language that is for example the case)), they requie a new version of compiler like vs2019 or even vs2022, they require a engine like "cuda, like dwrite, like LAV", MinGW,maybe also a second programming language like masm thats only to name a few ! it rather are even more those rather also are already bond to win10 or win11 like directx11 is for example so this is a problem you have today if only one of these things break up (in this case python) it might be a direct end - if more break up you have a even bigger problem the reason behind this are scriptish languages (what c++ in newer styles is also going into that direction) - and its going very into that direction a elder style is very independed of it´s operating system, like something "currentpath" would have a own code rather then somewhere in the code "std:CurrentPath" -> that might looks small but its not - that code is getting translated to a "real next code" what then up uses windows functions ... - its not faster it´s like an illusion that looks small and the knowledge is getting lost how these functions actually work - so a new programmer will always come up with the std:CurrentPath (or whatever std´s, c-runtimes, engines) there now are ... in a bigger program there are then many of these things - you cant just unbound all of them - a older compiler would not compile them, and a newer compile (translate) them to win11 functions ... so it might be better to have a solution in a normal c++ style - then you are fully independed of all these things - they cant just release a new phyton version that you have to fix - then you have the control about your code ffmpeg in the past was relativ independed today it´s going a different direction it begin to use engines rather then having own routines that can do the things today ffmpeg is rather like a command line - and internal it already started to use many engines and these use win10 functions, drivers, dependencies - up to some point i found ffmpeg a very nice tool, today im a bit disapointed of it a few years ago i mentioned that problem - now we face it again

i would suspect that paragon driver is a filter driver filter drivers are something between the OS and the HARDWARE it can perform the same actions as the OS can do with the hardware there are those IRP requests where filter drivers often would be if the controlment is taken from that spot the paragon driver has the control so putting things right, the paragon driver actually could handle the harddrive instead of the OS (in this case xp) then interact with the OS and show it as if the harddrive can control more then 2 TB of discspace this might have a problem but, it needs that paragon driver in that case to figure out what the inner XP code is doing is more of work, that require a lot of debugging - i think we might have candidates to do that

someone should correct me if im wrong but PAE is something for memory managment, is actually involved ? for a disc it dont have a 32 bit offset so directly, rather it makes use of 3 register of 16 bit size LBAlo - 16-bit LBAmid - 16-bit LBAhi - 16-bit = LBA48 https://wiki.osdev.org/ATA_PIO_Mode#Registers (the 8 bit are longly outdated - they are now always 16 bit) after these are set (LBAlo,LBAmid,LBAhi) it just use a I/O code - it dont use a 32 bit offset - it rather tells the disc where to write its data over these 3 filled registers its a bit like with memory paging, the old norm where 4k pages what are normed with a 32 bit PTE entry (and maybe a PDE) -> then PSE (page size extension (similiar to PAE)) apeared and made these 4 k pages to 4 MB pages thus for 4k you need less bits in 32 bit norm (20 bits to represent the highest 4 GB page) (4 k * 20 bit = 4 GB) for a 4 MB page it would be only 10 bit´s (4 mb * 1024 = 4 GB) to represent the highest 4 GB page, the rest of 4 MB pages would point to the upper pages above 4 GB i heared for PAE its only the half, they said its because PAE has the NX bit that means -1 bit = 2 MB pages, this is the result they often describe (also they say the 4k pages are ignored, aka PSE is always on) https://en.wikipedia.org/wiki/Physical_Address_Extension#/media/File:X86_Paging_PAE_2M.svg i do not know it exactly for that PAE mode, but even if it would be like that i could programm it like that the thing with the sector-size for a HDD/SSD/drive is very similiar if you have a 4 k sector instead of a 512 sector you have 8 times as much data it might not be the PAE extension itself, it might be the licence check like starter, home, pro - but that is only a guess just trying to make things better, if someone have knowledge let me know i also can improve myself

im not certain, it would requie to read out all the stuff that has been done related to this patches the /3GB is known to have some problems too, i dont know the problems with it j7n maybe has some knowledge here ? if its a problem with the /3GB switch there would be also the /USERVA switch to set a different value (between 2048 and 3072 (3 gb)) https://learn.microsoft.com/en-us/windows/win32/memory/4-gigabyte-tuning the one from dibya only use a small patch - maybe that would be worth a shot daniel k.´s patch use more of changes, maybe static offsets - that would be a reason why they dont work together also the method daniel k. described could trigger a crash not certain what "the russian patch" do to solve this problem is at the moment to much things around, i might can tell the norm´s, talks, idea´s a debug analize for example would take some time but if it comes to new work im busy at the moment sorry i also dont have the function test for dibya´s patch (aka if it passes the 4 gb limit) there should be someone to try these patches with the method j7n did to do so is simple just disabling the pagefile, then starting up app´s/executables that eat up more then 4 GB of ram if that is possible that patch is likely to work

hmmm in the past we had often a discussion about "32 bit is the 4 gb limit" or = 32 wires 32 bits can form 2^32 bits therefore 4´294´967´296 what represents 4 GB the problem with that is that even in 8 or 16 bit´s there are so called selectors those do actually still exits in 32 bit too also very important is that these 4 GB of possible "addresses" are not converted into that rather the representing PTE´s and PDE´s represent to a 4 KB page (in old normal 32 bit style) so going with that if you have 4 GB/ 4 k pages you end up in only using 20 bits ! (20 bits * 4k = 4 GB) a idea (very similiar to the discussion about 4 k sectors and 4 MB sectors with the harddrive(hdd/ssd)) would be just to increase the 4 K pages to 4 MB pages that would be 1024 times as much RAM the discussion about this are elder thats why that bit (PSE (page size extension)) was 1 candidate to pass that 32 bit limit the other candidate was the PAE bit (physical address extension) here is 1 catch PSE can extend the 4 GB by increasing the page size (even with a 32 bit PDE´s and PTE´s) in PAE mode PDE´s and PDE´s are 64 bit (yes even in 32 bit mode), that´s one of the reasons why PAE can address more then 4 GB of ram (to be more precise paging are mechanism´s there also some more things they play a role the CR3 register (also called the page directory base register) or the GDT, it would be to long to explain all but lets stay like this for now) so let us take the next important step in the old 32 bit system without PSE,PAE or "concrete use of segments", an app/executable had to use up the 4 GB ram because there was a user-mode (ring3) and a kernel-mode (ring0) there had to be a part in kernel mode and a other part in user-mode offsets are not a problem (as you can map a address like 0xFFFFFFFF to page 0 if you wanted) deviding the kernel mode and user mode is important, if that would not be the case every crash on a normal app/execuble would be a entire OS crash (BSOD or wrose) so the solution was simple usermode 0x00000000 - 0x7FFFFFFF - kernelmode 0x80000000 - 0xFFFFFFFF thats how the eldest 32 bit norms work it is not to bad to have kernel mode ram because many modules actually need space in memory (its not like an app is just made out of the app itself the app consits of many modules that also use ram internal) - also the video frame buffer often has a kernel mode area that takes memory for example so here we finally have our split 0x00000000 - 0x7FFFFFFF (2 gb) + 0x80000000 - 0xFFFFFFFF (2 gb) = 4 GB the app/executable now has the problem that it cant directly use the kernel mode offsets (it certainly use them but rather passiv (its some memory you need less in your app)) for this microsoft then made the /3 GB option in this one usermode has 3 GB and kernelmode 1 Gb (0x00000000-0xBFFFFFFF) + (0xC0000000-0xFFFFFFFF) = 4 GB but now the important part those offsets are not physical addresses ! actually you can map every of that "offset´s" to different places in memory (including the ones above 4 GB RAM) so thats why multiple applications/executables can use up the entire 64 GB of RAM (in j7n example he had 20 GB of 64 GB possible) thats more then 4 GB of ram and the proofes XP 32 bit can do it that way now we also have to mention the ramdisc solution, if you have like 20 GB of data and map then into the physical ram then the ramdisc use a chuck-wise solution it then maps for example 1 GB of that 20 GB data into an offset (that goes like fast) (when it wants the next part it just maps the next part - but it is chuckwise) now we have to mention the pagefile, in the pagefile windows stores memory data on the harddrive (HDD/SSD) when windows wants it use exactly that trick (it loads the data from the harddrive into that virtual offset) but this time there is an important differenz - the data is not from a RAM memory - it has to load that from the harddrive (and that is slow) for a SSD its faster - but still not as fast as it would be in memory so i guess some have reconized the increase in speed with SSD`s ? now you have a little more precision why this is the case so now to the part why a "normal app/executable" actually dont pass that 4 GB ram ram limit i already explained why it is rather like 2-3 GB ram + some kernel mode memory that is rather used like passiv but now the reason why it actually dont go beyond the 4 GB ram in "just one application/executable" here you have to know that all compilers form the past-now dont use segment selectors to pass that 4 GB ram i never seen a compiler that would be able to do so. probaly windows xp (even up to win10) dont have code for this either. but here is the thing normally in a code flow of a debugger you see a offset (or better called vitual offset this time) for the code flow it use a register called EIP (instruction pointer) - the E stands for 32 bits - for example only IP for 16 bit and REIP for 64 bit (R means 64bit) but here is the next catch ... the most common debuggers dont show the selector but for the code flow the CPU use a combination of a segment selector + virtual offset (the one you see in a debugger for example) this is called the CS:EIP combination - where CS is the segment selector (called the CODE SEGMENT(short CS) selector) EIP E for 32 bit and IP for instruction pointer now might be a good time to call out a wiki page about that: https://wiki.osdev.org/Segmentation that CS selector is 16 bits wide (in 32 bits) and 16 bits means 65536 possible selctors ... so the combination 16:32 or CS:EIP would be 65536 * 4 GB - that would be theoretical 262144 GB that it could select (+ you have 6 of segment selectors all in range 0-65536) but here is the catch windows (either xp - 10) dont use this to pass the 4 GB limit the only things windows actually did with these registers is that they point to small pieces of memory (like the EPROCESS structure) that is possible - but it leaves out the use to pass the 4 GB of ram limit by just using a few kilobytes with that method ... here is an example where microsoft accesses a small piece of ram using a selector : https://devblogs.microsoft.com/oldnewthing/20220919-00/?p=107195 (the problem with the way this is used, it not used that to pass the 4 GB ram limit, rather is use that as small memory storage) so the next step: back to the app/executable as mentioned there is no compiler that actually generate the useage of segmentation ... therefore you would need a specific app/executable that can do so - and that means chrome still would not pass the 4 GB ram limit (because it cant use segmention) (and then you also would need the windows operating system code also to control this kind of memory managment) also to mention is the FPU unit even the elder ones are 64 bit (or bigger) in modern cpu´s they are 512 bits (AVX) so actually you can make 64 bit moves in 32 bits or calculations, compessions like h.265 do so also some routines are written in the FPU, for example cpu intense sort mechanism´s - you might heared MMX, SSE ect. so in short 32 bit has 64 bit commands that it can use over the FPU unit (what is actually done sometimes) it can pass the 4 GB of ram over multiple app´s/executables it cant pass more then 4 GB of ram with just one app/executable - unless it use a chuckwise method (like the ramdisc) - the segment registers are actually not used to pass the 4 GB ram limit(that also goes for win10 32 bit) so i tryed my best to explain this as simple as i could

is someone making that test spoken of ? like j7n did with server 2003 x32 bit´s (its a XP based OS) someone has to disable the pagefile (because theoretical windows can also pass the 4 GB using a pagefile, a pagefile is just data on a harddrive/HDD/SSD) then someone has to use up more then 4 GB of ram, it would be a good indicator, but perfect would be to look of the pages are really in the upper 4 GB ram i meant to turn off the pagefile quite often - but then it didnt do that anyways the pagefile was still not offline maybe daniel k. should say something about this he wrote in a other post that in his opinion both ntoskrnl and hal has to be changed to make xp able to do the 64 GB (dibya said the oposite for example) - who knows who is right daniel k.writes that he modified hal and ntoskrnl (that suggest that that ntoskrnl´s and hal.dll are patched) also he mention symbol files - that also suggest that a ntoskrnl version and a hal version was used to do so you cant just select the patched ntoskrnl.exe and hal.dll by a symbol file searcher you have to figure out what version that ntoskrnl and hal are... then you have to force the disassembler to use the file from the non patched version the filecompare tool will find the changes if you have the right ntoskrnl and hal therefore also daniel k. wrote that he used the free space method for doing the patch this can be risky because some codes only access some free space sometimes, or on certain conditions (this often makes random crashes) a better solution is to add a section or increase a section - you have to know what you do here but - you quickly can kill the files like that if you dont know what you are doing are there more then 3 versions to do so ? its gambling around like "dibya´s" "the russian one" and the "daniel k. one"