dencorso

Two Gig won't hurt any either. Point is: you can have 1995 MiB of windows usable RAM with ME, just out of the box (i.e.: without any patching)... why settle for any less?

Upgrade. You won't regret it!

Sorry, I guess I should know it already, but please do tell me: which cpu does your machine currently have?

Yes, there is a max RAM Limitation, but it's not the same value for ME as for the rest of 9x. No, there isn't, sorry! Quouting myself from other posts: So, to go beyond that, you *need* RLoew's one-and-only RAM Limitation Patch. [...] You must add both the MaxPhysPage and MaxFileCache to both SYSTEM.INI and SYSTEM.CB (Usher's Method) to have the ability to boot OK, both normally and in Safe Mode, because 2048 MiB is actually too much, see loblo's (I) machine configuration here.

As the owner of 3 Athlon XP (Barton) machines, which I do like a lot, BTW, I think it's wise to avoid the multiple legacy problem: Since you're already intending to give it Windows ME, I'd recommend you avoid any processors not having SSE2 (and this means no single-core AMD is an option). I'd suggest you go to the highest-frequency single-core Pentium 4 available (say, either an HT 661 or 672, or even a probably much cheaper and easier to find 571) and 2 GiB of RAM onboard, if your aim is to get a real very high end 9x computer.

Like the unforgettable NEC-V20?

You mean this, right? This issue mentioned by Tihiy is precisely the one due to USBHUB20.SYS v. 5.0.2195.6891 and VIA chipsets. While one can recover from the BSOD, the only sensible thing to do at that point is to restart or shutdown the machine.

But, at the end of the day, it all boils dow to modifying .INFs, regardless of whether it'll then be update.exe or some other installer module that will end up interpreting them. And, yes, it'll break checksums and cause the need to dump security .CATs, all the same, too, at least in some cases. It's OK, bphlpt, It's OK! You did capture my meaning all right! And no, I don't recall any more-specific posts describing how to do it, specifically for drivers, across the various sites this knid of info is usually spread (I mean at least here, at RyanVM's, at wincert, and at reboot.pro), although it's the underlying technique behind things like Maximus-Decim NUSB and some unnoficial nVidia drivers, as well as update packs and the like. And receive my best wishes for your prompt recovery, too! Ear infections are horribly painful experiences,,, may you get well fast!

The 5441 driver seems to be the last one that supports XP. None of the later drivers have even a leaked or unofficial version for XP, AFAIK. Am I wrong? I think you are wrong. And 5441 is previous official driver, but the link closed now.(Cause of it is old) Plz see #5 link lastest comment. You can download XP lastest driver (7th Mar 2014) from link. 5445 and 5449 do not install correctly for my i7 3777k (on XP SP3) But by substituting igcodeckrng700.bin and igvpkrng700.bin by the files of the same name from 5441, it then installs OK. I guess substituting igcodeckrng700.bin and igvpkrng700.bin, as well as igcodeckrng750.bin and igvpkrng750.bin, by the files of the same name from 5441 should fix all the problems reported. Please test this and let me know how did that work for you all.

YMMV. IMO it's preferable to have a board for which I do have all the necessary drivers (including, if needed, UNIATA and/or unofficial drivers), than a newer board lacking them. Then again, modifying existing x64 2003 .INFs for XP x64 should be a trivial exercise in most cases. This is just my 2¢, of course.

What did you expect? It's interesting, nonetheless. But hardly useful, or else it would have remained undisclosed, of course.

MS-DOS and Word for Win early versions sources made available

Reverse translation of the quoted pseudo Italian (into pseudo English ) Intended meaning of the above-quoted pseudo-Italian: "Come on! Go ahead and install precisely this [one] here" Although my aging mind did better cranking my rusty knowledge of Italian than Google Translator does on its own: Then again, even a doorknob translates better than Google Translator ... Well, thinking about it, it sure is a better idea than asking people on the 9x/ME forum to produce one for NT4, ain't it? Then again, my suggestion stands, as there is a SNAP for NT4, although it's not that one I linked him to. The FAQ I ponted to also covers that, and the SNAP for NT is even easier to find. However, your comment made me read again the title of this thread, and only now did I notice the lone "2k" appended to it: Given the forum the thread was created in, I bet most everybody'll tend to read that title as I did, and never even notice that lone "2k" coda appended to it. ... and re-reading the whole thread seems to confirm that impression, too.

Yeah... I guess so!

Computer History Museum Makes Historic MS-DOS and Word for Windows Source Code Available to the Public!

Ma... dai! Cerca d'installare addirittura proprio questo SNAP qui. Here's the SNAP FAQ.

Hence, because amoebae can be thought-of as being roughly bidimensional (at least part of its lifetime), the ocean giant has an area of ca. 314.16 mm2 and Proteus has ca. 0.20 mm2 (assuming both are perfectly circular), so that the giant is 1570.80 times greater. Which is just about half-again your estimate, Flasche, but mathematically sounder, IMO, with all due respect. Of course, I used the area of a circle, but I might conceivably have used the volume of a sphere, although I don't think an amoeba is a drop-like (or ball-like) creature, during most of its lifetime. If so, the giant would have 4188.79 mm3, while Proteus would have 0.07 mm3, so that the giant would be 64,442.93 times greater than Proteus! Now, while this latter model (viz. drop-like) makes little sense by itself, IMO, but illustrates quite well that any estimate depends on the selected model... and since amoebae are not infinitely thin, may contribute to show that this fact tends to widen the difference betwwen the compared organisms. So I'd hazard a guesstimate that the giant may be up to 2000 times greater than Proteus, in real life.

Of course, a DOS boot floppy (or floppy image) may be needed, capable of getting a SATA CD/DVD going with a DOS driver: GCDROM.SYS + SHSUCDX.COM are the usual best choice, but there is also AHCI.SYS (findable inside sp39596.exe), as an alternative to GCDROM.SYS, and, of course, MSCDEX.EXE may be used, instead of SHSUCDX.COM. In any case, GCDROM.SYS should be the driver to be used, because Win 9x/ME cannot run SATA in AHCI mode, but I've added here the info about AHCI.SYS (distributed by HP but created by Intel, AFAIK), just in case someone needs to boot DOS from a machine having a SATA CD/DVD/BD in AHCI mode (which is common today). Alternate, direct download link for sp39596.exe, which contents can be extracted with the Universal Extractor. Some more info and an enlightening brief description of AHCI over at DOS_ain't_Dead. SataDosBoot_MWP.pdf

It depends... Have you ever watched The Blob?

Read the forum threads... it has been done and is all around you.

I think you are wrong. And 5441 is previous official driver, but the link closed now.(Cause of it is old) Plz see #5 link lastest comment. You can download XP lastest driver (7th Mar 2014) from link. You're right. I stand corrected.

The 5441 driver seems to be the last one that supports XP. None of the later drivers have even a leaked or unofficial version for XP, AFAIK. Am I wrong?

There is the unofficial Graphics Drivers 6.14.10.5441 (aka 14.56.0.5441), which support XP, and Haswell and IvyBridge, floating around the net. Attached is an inf for it, with the erroneous (but cosmetic) "Intel® HD Graphics 3000" corrected to "Intel® HD Graphics 4000", as it should be. igxp32.inf for Intel Graphics 6.14.10.5441.7z

Sure. It seems to be a bug on Auslogics Disk Defrag. Or it may be "by design"... The best way to find out is to ask their support, by following the link in the quoted text below:

In fact, MS will be actively inviting it, if they offer patches to Vista+ and say "look, this vulnerability also exixts on XP, but we won't patch it". Because, then, they'll be directing people to obscure vulnerabilities no one ever imagined existed.