Jump to content

All Activity

This stream auto-updates

  1. Past hour
  2. Too bad this poll isn't multiple choices, because I really like more than 1 Windows OS.
  3. It was a joke. Clearly, the title is misspelled, not to mention the overall misleading, clickbait style name of the topic and improper, poor English grammar. No articles to support that "theory", nothing.
  4. I'm glad the 6.14.13.6881 worked better - I reiterate - the only thing changed in all inf-files was the version - to make driver compatible with fireferret foolish blocklist.xml so we can have browser graphics acceleration. If your GT710 isn't stable, it might be overheating peradventure? I mean it's not necessarily software fault. It's a nvKepler card, so it should work as any other, I had GTX 680, 770 and 780Ti with this driver and all worked well, no instabilities. Any yeah GT730 is better than 710, theoretically 2x, it has 2x the amount of shaders (384) in 2 texture processing clusters.
  5. Today
  6. Did you try this website? A huge drivers collection. https://driverscollection.com/
  7. I had XP running on a GT 1030 but only in super-VGA mode. I have a GT 730 but it has disappeared under mounds of equipment...can't find it. Now, when I try to start XP on the same Asus B360, I get a black screen for a long time and sometimes it BSODs after 5 or 10 minutes. I was too slow to get the error code but it is display related. I am trying to repeat the BSOD but all I have is a black screen. Oddly, I get the XP welcome screen but it does not reach logon. Any ideas? Also, I can no longer get W7 running on the B360 machine because I re-installed the USB driver for my new B760 and erased the A36D identifier for the older machine. I am looking for a way to re-install the old hardware ID but I can't even find INFfile for it.
  8. Then use the v344.75 driver, I don't think OpenGL version 4.5 is required anywhere. And I'm sure you don't use OpenCL and CUDA, so the fact that they're not there won't be a problem. PhysX should probably just be reinstalled, here are links to it, you need to install two versions, 9.13 Legacy is required for games released before 2010 https://www.nvidia.com/en-us/drivers/physx/physx-9-13-0604-legacy-driver/ and version 9.19 is required for games after 2010 https://www.nvidia.com/en-us/drivers/physx/9_19_0218/physx-9-19-0218-driver/ It also happens that the new GPU-Z does not show PhysX correctly, so I check with the old version, 0.8.5 https://www.upload.ee/files/16448328/GPU-Z_v0.8.5.exe.html
  9. The list in that dialog box is empty for me. I have seen something there in the past, but not now. When I looked through the entries using SQLite Manager, I saw my username in a few of them. In Opera 12, I don't see anything from Wikipedia in persistent storage. Other wikis have 500-700 kB. Perhaps it considers the browser too old to use. Its global quota looks to be 10 MB for all sites together.
  10. Is this where the "Offline Web Content and User Data" shown when you go to Tools / Preferences / Advanced is stored? I think logins and passwords are stored in a different table, but I'm not sure.
  11. yeah it looks like so, I reported to upstream (i.e. AF) to see if there is something missed.
  12. Yessaaa, I understand to 100%, what this function is doing. During crazy thinking today about the few lines in Assembler of the function ExInterlockedFlushSList, I start to wonder, why this is there: jnz short loc_40B0BE (a loop) In my eyes, no loop is needed, when I understand the opcode correct. Now I understand, what happens: cmpxchg8b is not really atomic, even not on a single cpu. So, it can happen, that the content in memory does NOT match witch the content in EDX EAX, because another process killes during this operation the 64 bit content in memory or just disturbs operation. Crazy. Oh my, what kind of errors may be there in other OS too. Cutler solves this problem with his loop. When the content in memory is disturbed, or the operation cmpxchg8b is broken, there was a next try and a next try and a next try.. I testet this with disabling the loop with 90 90. Oh my, Bsod. So it happens nearly always, that this operation is disturbed. And I can overcome this to be disturbed, when I change the opcode from cmpxch8b ==> lock cmpxch8b in the original ntoskrnl.exe from XP SP3. Now, without the loop, XP boots. This means, only with the lock nobody is allowed to disturb the operation of cmpxch8b, and so for any simulator, real atomic is needed. Waaaoh, this is the first time, that a modd in this function works for me Dietmar
  13. I have been using GT 730 with v344.75 for a while with same PC and it is doing very stable unlike GT 710. Games I tested and 3dmark2001se are not so power hungry softwares i guess. GT 730 with latest drivers such as v368.81 also caused freeze issue but has been very stable with v344.75. I dont know what else test I can do with PC and GPU health. I can say (except OpenGL showing 4.4 instead of 4.5 and not showing some other features on GPU-Z comparing native supported drivers) seems modified v344.75 is stable with GT 710. But need to run it for some days for more experience.
  14. Hi, On my gaming PC when I try to start a VM on VMWare Pro 16 I get a BSOD. I get: EXCEPTION_ON_ INVALID STACK BSOD. I've enabled everything needed for virtualization in BIOS. This is my build list (I got a different GPU, but it’s still an RTX 4070) https://newegg.io/8f3ca36. I have an ASrock Z690 PG Riptide motherboard (here's the manual: https://download.asrock.com/Manual/Z690 PG Riptide.pdf). How do I fix this?
  15. for resolution i used VBEMP driver
  16. Thanks, I don't know SQL wizardry. Last time I touched it was almost ten years ago. Looks like the biggest consumers are various Wiki sites: Wikipedia, Wiktionary, Fandom, Wikinews. Each language domain has an entry between 1 and 3 megs. Wikipedia has become noticeably slower over the years with the same basic Monobook style and visual editing turned off. On talk pages they discuss how pages for some popular subjects need to be shortened, and on an old PC it is clear why.
  17. Maybe you have a problem with the PC hardware. I have GTX 750 Ti with the latest iCafe drivers and it works. I wonder if it would be stable under NT 6 on the same harware. Some games do not work under XP because they run out of 32-bit memory. Then the system spectacularly crashes. Direct3D9 seems to take double the memory, in video RAM and in system RAM (in the Paged Pool address space, which can be listed by PoolMon). If I alt-tab out of a game, it copies some of the video memory into system memory. Video memory use fall to near zero. OpenGL doesn't do this. Maybe it helps when alt-tabbing back in. On XP SP1, the paged pool was around 100 MB. On XP SP3 and the latest Server it is about 300 MB, and the problem may not be experienced. But possibly can because these new cards have a lot of memory on board. I can compare memory use using Euro Truck Simulator 1.35, which is a very demanding game, and supports both Direct3D and OpenGL. It will crash in Direct3D after driving through new maps for 15-20 minutes. A crash also occurs with GTA SAMP using the default cache memory setting, which is optimized for modern PCs. The Sims 2 FreeTime actually causes a BSOD when launched repeatedly. I couldn't understand the reason. It does not use much memory. Perhaps the driver has overlooked something when reloading a cache, and they've stopped testing it in depth for an obsolete Windows version. With this game there was some blame passed between developers of Maxis and nVidia, but supposedly it was resolved, but not for me.
  18. SELECT originkey, SUM(LENGTH(value)) AS total_size FROM webappsstore2 GROUP BY originkey ORDER BY total_size DESC;
  19. I think, that a make a mistake in my description of the work of CMPXCHG8B. It works really on ALL bits of the 64 bit in memory. The lower 32 bit are stored in EAX, the higher 32 bit in EDX. Now comes the check: ONLY, when all of the 64 bit in memory match with the EDX EAX combination, the combination of ECX EBX is written back into those 64 bit in memory. In ECX are the higher 32bit for the exchange and in EBX the lower 32bit for exchange. ONLY when this exchange has happened, the Zero flag ZF is set Dietmar PS: This is the base for the Simulator.
  20. It is obviously a result of one or a few sites storing oversized "cookies", possibly to sneak some advertising in. SQLite Manager lists 19801 entries inside the table webappstore2. After an exercise of reading backwards, on the first page I recognize entries for hdtracks (possibly a culprit), softonic, pristineorganics, reddit, msfn, wikipedia, google, deezer (suspect), imgur (suspect), rateyourmusic, gtaforums, facebook. Looks like it may hold useful settings and logins that I need to keep, or I would need to log in to every place again, and likely don't remember the passwords. Any way to sort this table "by size" to narrow down the bloat?
  21. @user57 Give me the assembler code of the emulator and I test. After 1 day of work on this few lines, I think, now I understand how this function works. But to emulate it, is another hard job Dietmar
  22. you making the same mistake that command sets flags, and react if the compare was correct or not there 2 problems i can certainly tell in the first step cmpxchg can have 2 results (if equal it makes the mov if not it makes the mov to a register) (and it should not do that because it has to compare 64 bits) if you have 32 bits with the compare it reacts already to the 32 bits (the other 64 bit are ignored) then the following happens : the flags are lost and the reaction - for equal 32 bit already reacted or not then you do the code again but here sits the same problems now the flags get changed a second time (and it should not) the compare depending if equal reacts to the next 32 bit (while igoring the first 32 bit) if that compare was equal it sets the values and if not it sets no values (but you need the 64 bit) the flag registes (ZF) is that readed as if the first 32 bits are not there with other words the results are gambled up the solution looks not that hard to me you need 2 compares to see if the wanted to compare 64 bits are equal before you set the 64 bits reactions if those 2 compare where equal you set the values, in the other case you need an extra reaction to set the other case the reaction stores them into EDX and EAX (the flag should still be activ, unless you start to use a command that affect flags) cmp edx and eax (destination operand) if equal store ECX EBX to destination operand (The destination operand is an 8-byte memory location) // CMPXCHG8B - 32 bit emulator cmp dword ptr [ebp],eax // eax suppose to be the low part jne skip_and_load_edx_eax cmp dword ptr [ebp+4], edx // edx suppose to be the high part jne skip_and_load_edx_eax // 64 bits where equal, change with ECX and EBX mov dword ptr [ebp], ebx // suppose to have the low part mov dword ptr [ebp+4], ecx // suppose to have the high part jmp end_of_CMPXCHG8B // they where not equal do as the command is described and load those to EDX and EAX skip_and_load_edx_eax: mov eax, dword ptr [ebp] // suppose to be the low part mov edx, dword ptr [ebp+4] // suppose to be the high part end_of_CMPXCHG8B: // CMPXCHG8B - 32 bit emulator end this code should replace the code after : loc_40B0BE: // emulator code here loc_40B0CD: <--- at this place mark " end_of_CMPXCHG8B:" // notice i could not test that command yet if the order is right (like upper and higher parts) it might said something about the upper and lower part but as i remember right you never can be exactly certain about this (in memory if you have 11223344 - the 44 are the bits that control the high values (very old architecture stores that differently too - but that we dont have even in a 486) if that dont work i certainly can fix this , i need a test to make certain the command reaction the command description however says EDX and ECX contain the high part https://www.felixcloutier.com/x86/cmpxchg8b:cmpxchg16b if different: // CMPXCHG8B - 32 bit emulator cmp dword ptr [ebp],edx // if different edx suppose to be the low part jne skip_and_load_edx_eax cmp dword ptr [ebp+4], eax // if different eax suppose to be the high part jne skip_and_load_edx_eax // 64 bits where equal, change with ECX and EBX mov dword ptr [ebp], ecx // if different ecx has the low part mov dword ptr [ebp+4], ebx // if different ebx has the high part jmp end_of_CMPXCHG8B // they where not equal do as the command is described and load those to EDX and EAX skip_and_load_edx_eax: mov edx, dword ptr [ebp] // suppose to be the low part mov eax, dword ptr [ebp+4] // suppose to be the high part // your 55667788 example say so end_of_CMPXCHG8B: // CMPXCHG8B - 32 bit emulator end
  23. Hi mackid1993, Thank you for your donation of 5.00 USD. We look forward to improving the forums with your donation. Thanks MSFN
  24. Yesterday
  25. So I found a nasty Dark Magic edge case when taking ownership of a file. This doesn't happen when Dark Magic is disabled. I've tested this in Directory Opus. When the error is dismissed Directory Opus is left in a completely frozen state and needs to be killed. This also happens with Explorer which also causes a freeze which requires an Explorer restart.
  26. Looks like mica acrylic to me, do you have Dark Magic on? There have been some changes recently with that. Edit:
  27. Install the extension SQLite Manager 0.8.3.1 from the Classic Add-ons Archive! This extension can manage sqlite files such as your webappsstore.sqlite inside New Moon. BTW, my file is only about 12 MB. With SQLite Manager 0.8.3.1, you can view all entries, add new ones, edit or delete old ones. My webappsstore.sqlite files in different profiles are of different sizes from 96 KB up to 12 MB.
  1. Load more activity
×
×
  • Create New...