Tripredacus

I don't think the Discord is official. You can use whatever you want. The site has been running good lately, so hopefully no one will ever need to use it.

Websites can be weird at times, idk why. I got banned on Venmo just by registering an account. And I got banned on Whatsapp by opening the app on my phone after I took the sim card out.

An alternate/emergency forum is in the process of being completed and I will post more once it is ready.

I only go there while logged in but they could really use a better search overall. Besides shorts, it often gives results for things other people have watched which are never relevant to what I typed in. I wish all sites that had a search even Google or Ebay had an advanced option you could set where it would just give you results based on what you typed and not all that fluff they want to make money with.

Somewhat annoying is that some Youtube videos now start with an unskippable ad despite using ad blockers. I didn't know if there was a thread in regards to this in a non GD location. While I find the ad intrusion to be extremely annoying, it is still funny to me that the ad I see on Youtube videos is for a paid ad blocker.

Considering that the Asus PDF is a second revision and from April of 2006, it would be before PCI-E 2.0 would actually be available Gemini says that there is no Intel Specification named as such, but is likely refering to an Intel implementation guide and chipset-specific capabilities document in relation to the PCI-SIG specification. So they may be referring to a document like this, which is very difficult to read if you ask me: https://www.intel.com/content/dam/doc/datasheet/i-o-controller-hub-7-datasheet.pdf But there are some bits such as "Ports 1-4 can be statically configured as 4x1, or 1x4". Here you can build off this if you want: https://g.co/gemini/share/07cb56274eb7 I know working with these LLMs sometimes you can get them going in the wrong direction and they can be less reliable. Maybe I got a good start on that instance. But I have to preface again that these public LLMs are trained on public information and not on in-depth technical subjects like chipsets or hard drives. And because they are trained on public posts, they have also read forum posts or comments from people who may have not quite used the correct answer or an incorrect answer. You've been on MSFN for a long time, so you've seen posts that were inaccurate or wrong. And LLM doesn't know what is true or not and doesn't have the ability to test if anything is true or confirm things. That's why it is never actually helpful when you want it to help in regards to advanced hard disk related things such as trying to get more than 2.2 TB on an MBR disk, LLM think this is actually impossible due to it having read official support statements and the like. So try not to get caught up in the trap that LLMs know everything or are the solution to all of our problems. They are just a powerful tool and until AGI happens, we should treat them as such.

This looks to be the same PDF that I downloaded (pdf page 13 and 42) http://oldschooldaw.com/asus/2006/e3037_p5m2_series.pdf It does not say exactly what I wrote, it does not mention mechanical vs electrical but that is what it is actually refering to in a broader sense. It may be that those particular terms were not common at the time this was written. I am not even sure that in a general sense that electrical is even the right term, since I doubt it is actually voltage based. Rather the limitations boards can have on their PCI-E slots are typically caused by bandwidth limitations. Nonetheless, any time you see any mention of a 16x running at a lower spec when a video card is inserted typically means that slot was not actually designed to have a video card put into it. The manual doesn't seem to mention putting a video card into the PCI slot, I don't know if it would work with the onboard since they would both be on the same bus. This kind of reminds me of that one configuration I had to get working where I had a requirement that a computer be able to use the onboard sound in addition to 3 PCI sound cards and for them all to work independently. We tried everything but the only OS where it would work correctly ended up being Windows XP Pro 64 bit, which wasn't ideal because XP was going EOS very shortly afterwards (at least in comparison to a device lifecycle) and that configuration would not work on Vista of any version.

Modern games tend to ship with more capabilities than any one person needs. This is the result of the dev environment and I personally consider it to be laze. Take a simple example where a game will have all of the textures included that the game supports using, despite the fact that any person who plays the game will only use one of them. This is why you can find a lot of people complaining about games being so big, or 4K textures being included even for people who don't/can't use them. But on the code side, since these games are created using suites, they also include support for a lot of extra stuff. Let's say this particular thing is only relating to the multiplayer aspect, if you only played in single player the missing function calls would likely never be used. So the result of removing a dependency is that it is likely something isn't going to work, but that doesn't mean the thing that won't work is a function that a player will ever use or need. Eventually it can lead to a problem with future patches or with mods.

Tidbits from the manual for reference - a video card in the PCI-E x16 mechanical slot will run at 1x electrical - board has an onboard VGA enable/disable jumper VGA_EN1 As I understand, the 1x electrical probably won't make a difference for 2D compositing but 3D could have performance issues. BIOS apparently has primary boot graphic adapter options either Auto or Force x1. Since this is a server class system, we can likely presume that this slot was not intended to be used for a video card. Even the manual talks about using storage controllers and while it does mention the electrical difference, I would bet that it was intended that the board with use either the onboard or a discrete adapter but not necessarily both. Even with PEG present, it still makes the onboard available. The fact it has the jumper would indicate to me that the intended configurations would be to have VGA_EN1 enabled if using onboard, or disabled if using a video card. There is no mention of using multi-display for this board. What you describe (if I read it correctly) is that Windows 7 is only using the first display adapter that the BIOS makes available to the system. It is not necessarily disabling the other, which is why it appears in Device Manager. I have a couple of ideas but being able to find out if they are the right answer or not is difficult. This is another one of those types of questions that AI can't figure out because they aren't trained on the technical information. But basically this is what I can put together. First, WDDM 1.0 (from Vista, I don't know if just RTM, SP1 or what) was not designed to handle multiple display adapters and typically would use one and disable another. It was always third party implementations that had to be used to get multi-display to work, especially once hybrid graphics in notebooks started appearing. WDDM 1.1 in Windows 7 was supposed to correct this, and largely has. However there is a possibility that using a Vista driver on Windows 7 may have something hard-coded in it that is replicating this, or if your Win7 is modded in any sort of way to make this behaviour possible, although finding out what would do that might be impossible. The other thought that I had was that perhaps something is being set in an ACPI table by the system that 2000 and XP doesn't read, 10 may or may not read (but obviously the DDM in 10 is way more robust) and Win 7 does that causes this issue. Again, I don't know how to verify this, but just thinking of what Win7 could read from the system to make it behave this way. But I still think that the intended way to use this board was to only have 1 display adapter, and that Win7 is working correctly and that the 2000, XP and 10 experiences are actually the outliers.

What motherboard is it?

No issue with activations today. I can't speak about what was actually installed onto that refurbished computer. Quite possibly it was the Direct version of the OS. Any new install using System Builder or Retail will still activate but as with any version of Windows, you may have to use slui 4 (telephone activation) if you are re-using a key. I have no experience with the downloaded version of Windows 10 so I can't speak about that.

The forum has open registration, anyone can join. Just make sure they know about the forum rules, as you know sometimes these types of projects can be tricky to fully talk about here because of that.

I found this driver for this model. https://pcsupport.lenovo.com/us/en/downloads/ds001073-integrated-camera-driver-for-windows-7-32-bit-64-bit-vista-32-bit-64-bit-xp-32-bit-64-bit-thinkpad

Computing has plateaued over the past decade, so there is not really any point to buying new for a general purpose system. A few weeks ago I finally decided that I will need to transition my daily driver home computer because there are indications that some things are going to be more work than it is worth the work, such as being able to access bank websites. I ended up buying a refurbished Lenovo Thinkpad P50 with an i7 CPU, 32 GB RAM, 512 GB M2 SATA and it also has the Quadro/Intel hybrid graphics. It cost just over $300 on Newegg. How I came to this particular choice was I looked at what models were available and then searched those models to find reviews. I was particularly interested to know about display panels and thermals. The P70 seemed an even better option but then I went onto Ebay to see what part availability was like. There were way more parts for P50 than P70 and being able to find parts will just get harder in the future, so I went with the P50. @Karla Sleutel There is no issue to post links to products on commercial websites.

That script is running FSZ.exe which is part of dsfok. It is on archive: https://web.archive.org/web/20070222124242fw_/http://members.ozemail.com.au/~nulifetv/freezip/freeware/

Helpful hint in regards to notebook drivers, from back then or even to today. Nothing is really proprietary when it comes to the stuff inside of notebooks. So usually if I can't find a driver for a particular model from the manufacturer's website, I will find out what a relevant cross-over model from another manufacturer is and look on their website. A real world example is that sometimes I have to build an image for a Clevo notebook before the drivers get put on Clevo's site or the ODM's site, but MSI and Asus sell notebooks that contain the same hardware so I will look on their sites for drivers. For the drivers you are missing, if you can post the hardware IDs of the devices in device manager (Just the VEN_/DEV_/SUBSYS lines) then others may be able to know where to find the stuff you need.

My experience is limited only having used Bard and Grok. Neither are good for technical questions about hard disks. That being said, both of those are good at general things like being able to read hex dumps or assembly. So just as an example, here are some info from Bard to your questions above but as I warned, you cannot take the answers as gospel. And there is a lot of fluff which I'll leave out. If it provides relevant links, I will post those. I prefaced the conversation about being relating to 2000/XP era. 2. Why do some MBRs have a copy of sector 0 in sector 1? So it doesn't actually answer the question really. All of these public LLMs are designed to add a bunch of explanatory content to answers as well. Sometimes the question needs to be re-worded as well. 3. Why do some FAT tables start at logical sector 32, others at 33, 34, or 37? Seems to be general information, no specifics. 4. Some partitioning programs don't create spaces between partitions. Doesn't this prevent proper alignment? In this response it is doing a comparison to AFd/SSD/4Kn, so I left out the extra text it generated about the above items that was referencing modern implementations. Links provided: https://knowledge.broadcom.com/external/article/152308/alphabetical-list-of-ghost-command-line.html https://thestarman.pcministry.com/asm/mbr/NTFSBR.htm look familiar? 5. My friend is having trouble interpreting the binary code in an MBR. He can only recognize the boot signature 55AA in bytes 510 and 511. Is there any reference material that can help him understand the boot code? Understanding the MBR Structure: A good idea about searching in Github, of course, jaclaz could just come in here and answer all of these questions. 5.1. Why do some programs show the VBR OEM name as MSDOS5.0 while others show MSWIN4.1? The OEM name identifies the formatting tool's origin. MSDOS5.0 from MS DOS 5.0 and MSWIN4.1 from Windows 95 (OSR2) based tools. the LLM seems to indicate that MSWIN4.1 name means the file system supports VFAT/long filenames. I can't ask it about question 6 because I don't know what "the same program" refers to. Sector 12 (at least from my reading on the Starman site you linked earlier) has the NTLDR in it. 6.1: Why is backup sector 6 empty in logical partitions? It doesn't actually know the answer to this. LLMs typically won't have a response of "I don't know" and they end up speculating (or hallucinating) and I think the answers it gives to this question is one of those two results. It does provide one link that might be useful: https://sysplay.github.io/books/LinuxDrivers/book/Content/Part14.html 6.2 What do sectors 0, 1, and 2 contain in logical partitions? The further along questions during this instance seemed to get more helpful as time went on. If you have a google account you can talk to this LLM here: https://gemini.google.com/ While people seem split on using an AI/LLM for anything for various reasons, a few things are clear. First, they are just a tool and second this is (currently) where the future of the internet or computing is going. So I personally feel that people should get experience in dealing with these things, to understand how to read their responses, understand how to word things for interaction, etc. I have posted some of my own observations in this thread about that. And worst case scenario, maybe the future will be like that meme where the robot overlord spares the human from being killed because they used to talk online years ago.

To get this out of the way, the MBR is a standard but all of the different tools do something a little different. Things are not really as documented as we'd like or perhaps they are but are stuck in the dead internet somewhere. As for specific tools and how they manipulate disks, some may be more generic or some may be based on an intended or observed use. For example a tool that was tested on a *nix vs a Windows 98 vs a Windows XP disk may write the bits a little differently. And unfortunately, a lot of what we know to be true is not found through any sort of documentation but through actual testing. My input on the linked thread was not so much in regards to the disk but rather the BCD aspect of things. At first I would consider to ask an LLM some of these questions but so far none I have experience with are any good at figuring out disk related questions of this sort because it is mostly theory and public LLMs are trained on support articles. I had considered running my own LLM and only train it on hard disk related info, but I don't have the time for such an endeavor.

My thoughts on the matter end up being tainted by Embedded SKUs. I absolutely hated Windows 10 but I haven't really used retail Windows 10. Windows 11 seems to be an even worse disaster but my experience with that in any form is quite lacking. There were some changes to deployment that don't make any sense and I'm pretty sure that those issues are actually bugs. Unfortunately, I don't have a TAM anymore to interface with to resolve those issues. Even if I did have that access still, I feel like nothing would get resolved, especially since MS still hasn't fixed that bug in Office that I reported to them over 10 years ago. I do manage one system with Windows 11. The only thing that I like about it is the theming options. That system runs as a digital signage system. If whenever I have to personally move to Windows 11, I'll probably use an embedded version there also.

Which Thinkpad? What driver is installed?

It is not always just registry keys to look at in these PMLs. First thing to remember is that the process you attach to may not actually be the process that is doing all of the work. A child process may be spawned by w2ksp4-2_en.exe that is doing more things that are not captured here. Also, this seems to be extremely filtered. Two things for next step: 1. your PML shows that this file was written to, see what is in it for the same timestamp: C:\WINNT\system32\config\software.LOG. You could also check the system for *.log around the time you executed the EXE (you do this check right after running it) to see if any other files get updated with anything useful. 2. Check for a child/spawned process doing anything. I do not know the completeness of this PML but if there is nothing more than this then it doesn't appear that this process launched anything. What you may or may not be able to do is the next step to use Process Explorer to pause the w2ksp4-2_en.exe process. You can do this before the error message appears, but you can also do the following (without pausing/halting the process) once the message box is on the screen. In ProcExp (at least XP era versions) you can look at the properties of a process and it will show you all of the files and regkeys that it has open, as well as view the Strings tab for both image (not usually useful) and Memory option which may show paths not visible in lower pane.

It seems that you need to get a matching SP4 from the OEM who originally sold the installation of Datacenter and install that version. Maybe it wasn't always different bits, perhaps there were just instructions you needed to follow before running the update. I found information here, there is a table in the 2.1 section: https://web.archive.org/web/20041130034246/http://www.microsoft.com/windows2000/downloads/servicepacks/SP4/ReadMeSP.htm I have not tried to install Datacenter myself because I tend to try using old software in their optimal setup... which means running on a cluster of which i don't have the hardware for., This document seems to indicate that IBM's implementation of Datacenter is locked down in some way. It mentions not being able to install software that has kernel access: http://ps-2.kev009.com/basil.holloway/ALL PDF/sg245845.pdf What you could try, potentially, is to attach procmon to the processes that are involved with running that update file to see if you can see what it might be looking at when it decides that the system isn't qualified and stops the installation to show the message. It might be that there is some old version of Procmon that works in Windows 2000, but maybe not. Here is a thread on superuser talking about other tools that you could try. https://serverfault.com/questions/129044/where-to-get-working-sysinternals-tools-for-windows-2000

It would be helpful if you could post the exact errors you get. And also get any error messages from log files.

Even on an old computer, I doubt you'd see a measurable speed difference in copy speed with what you are using. Gemini says that /v was added in MS-DOS 3.2 but doesn't give a reference link. But let's say that is the case, the reason for needing to add /v was likely to handle copying issues on floppy disks. You would be more likely to see a speed difference on a 286 and copying to/from actual floppy disks. In regards to /V/Y arrangement, this could be an issue with MS-DOS 7.1. While outside of your scope, you could see how other (actual) versions of MS-DOS handle this. COMMAND.COM should parse the entire command, read the options and then execute the command. If it does it in a FIFO method, it could be that (for some reason) it actually is totally ignoring the /Y or would (somehow) try to set that option after verification, and it shows you the overwrite prompt because of that. That is just a guess, but I wonder if DOS 5 or 6.22 has this issue. You could also see if it behaves the same way without setting the COPYCMD variable, just run COPY /V/Y manually to see if it has the same result. If it doesn't then obviously it has something to do with the environment variable.

It seems /V is silent and you'd only know that it is being used if verification fails. Of course there is certainly a way to test that, but I only have theory on that process and not any actual instructions. How I'd imagine that would have to be done is you would need to attach/use a debugger or other process manipulation tool, halt the copy process at some point and you'd need to manually change something. Not knowing exactly how the copy command works, I don't know if you could simply edit the source file or if you'd have to change something in memory, before resuming the process. Another thing that you'd probably need to figure out is how the command is actually verifying. Does it do a CRC check or a bit compare? You'd also want to know at which points in the process that copy gets the info from the source and destination file occurs, so that you would have a better idea of when to interject to cause a failure. Your ability to cause a verification to reliably fail may be very difficult to do under DOS.