[Sfor]

Indeed. DOS running with Windows GUI is not the same as DOS running without GUI.

I would rather say, the GUI DOS is just a DOS emulation.

The DOS itself can not work with multiple CPU cores. But, it could be possible to run two DOS instances on two different CPU cores. Still there is a problem dividing the system resources (ports, graphics cards and other). The DOS can not do it by itself.

A multitasking multi CPU enviroment, requires some OS layer between application and hardware resources. It is necesary to divide hardware between multiple threads. The windows 9x systems do provide such a layer. One of the kernel fuctions is thread managing through dividing the CPU time between threads. All it is necesary to replace is part of the kernel with a multi CPU one. This makes a multi CPU driver idea completely wrong, as the kernel core is responsible for the multi CPU support. Drivers are working on a higher layer then the CPU support. So, I see no possible solution with a multi CPU driver.

As for replacing the Windows 9x kernel with NT kernel. I think it will not work. As, the NT kernel is a completely different design. The NT kernel does have some security features on the lowest layer, this makes it completely different than 9x kernel. NT based drivers are working on a higher layer than the 9x ones, as far as I know. This makes the system more stable, as a driver can not stop the system kernel.

Also, nobody mentioned a non paralel multitasking feature such as a system kernel running on one CPU with all remaining application threads running on the other CPU. This could increase the system stability mainly. The speed can go up a bit, but not very much. Also, one CPU can do all the hardware access, while the other would do the other things. With such a solution, it would not be necesary to divide system hardware between two CPU cores. The kernel core can be much simpler, that way.

[submix8c]

Well said! You seem to be very knowledgable (a compliment).

GUI=Graphical User Interface (Interface being the keyword; symantecs again)

And don't anyone suggest a new/modified kernel. There's much more to the design than just that single item. Ask the KernelEX Project guys. Even in the Mainframe world there are multiple modules that need generated to support a single new configuration. The kernel is only the basis (ref. IBM's DOS/VSE/AF and OS/390, the latest I'm familiar with).

Still, all-in-all, seemingly not worth the effort. There doesn't appear to be any takers to code from scratch the necessary support. How many months/years of dedicated coding/testing for such a project to satisfy a "want" and not a "need" (without recompense)? That's why MS quit bothering (IBM DOS is history, IBM OS is not; same rationale).

On a side note, the SETI project allows for Computer Clustering via specialized software, but I hesitate to suggest this as relevant to Multi-CPU support, since it's task-specific (as are the other Cluster projects). Check out MS' latest "server inventions" to what's involved.

@ohmss006, how about a little "light reading" on the subjects/objections, 'k?

'Nuff said. As MCP said in Tron (the movie) "End of line..."

This post has been edited by submix8c: 19 July 2007 - 12:36 PM

[Fredledingue]

That's why I suggested an application specialy dsigned to ddtect unused second core and use it.

[Sfor]

As far as I remember, every memory block does have the access rights. Each thread does have the rights to access certain memory blocks. If a thread tries to access wrong memory block a protection fault error condition is raised by the CPU. The system core has the access rights to all memory blocks, while higher level processes do have some limits.

As the system kernel keeps the best access rights, the application running on the second core would not be able to work in the same manner, I think. The memory block list is maintained by the system core. The second CPU has to have the access to the memory block list, as a CPU can not work in protected mode without it.

The protected mode virtual memory block system used in Windows makes the second CPU used by another application idea looking impossible. As, the second CPU would not have the access to the memory block list.

[oscardog]

submix8c, on Jul 19 2007, 06:30 PM, said:

eidenk, on Jul 19 2007, 11:13 AM, said:

ohmss006, on Jul 19 2007, 08:37 AM, said:

Quote

I think this is irrelevant as Windows does not run on top of DOS.

infact, Windows 98 is more or less the GUI that runs on a DOS-based layer, so assuming now correctly i assume, Windows 98 does run on top of DOS.

No, I don't think Win9x does run on top of a DOS layer.

http://win32assembly...r/vxd-tut1.html

DOS is apparently used in the boot sequence of Win 95 and 98 (but not ME which is why it boots faster) but that's it. Once Windows is running there is no DOS layer running underneath it.

Unless I am wrong but in this case you'll have to show where is this DOS layer that runs underneath the 9x GUI.

Re-read the referenced link. VMM (I believe) is interfacing with IO.SYS (the "DOS layer"), ergo DOS!

All Win 1.x/2.x/3.x and Win9x (including WinME) run "on top of DOS" That's why you can Upgrade from Past (DOS) through "current" WinME (with the appropriate Windows Upgrade sequences). Note that certain DOS modules are either replaced or eliminated as part of the Upgrade process (compatability). Also note that there is a "patch" for WinMe that allows you to "boot into pure DOS Command Mode" available on the internet. WinME from MS simply eliminated the access to it to phase Consumers into WinXP (an offshoot of NT, no DOS involved). Kind of like the difference between IBM Mainframe DOS-based (now defunct) and OS-based (still going strong).

Trust me, I've installed them all (including the IBM stuff)! It is definitely DOS-based (symantecs aside)!

And forget modifying 9x to use dual-core! It's definitely not worth the effort (ref. IBM stuff; look it up...). The systems specs are too radically different...

Windows 9x uses a new system file, IO.SYS, which replaces the MS-DOS system files (IO.SYS and MSDOS.SYS). This real-mode operating system file contains the information needed to start the computer.
The actual underlying os in win9x is the base system and the real core is the virtual machine manager hence it allowing multiple threads in multiple memory address spaces

[submix8c]

oscardog, on Jul 20 2007, 10:06 AM, said:

Windows 9x uses a new system file, IO.SYS, which replaces the MS-DOS system files (IO.SYS and MSDOS.SYS). This real-mode operating system file contains the information needed to start the computer.
The actual underlying os in win9x is the base system and the real core is the virtual machine manager hence it allowing multiple threads in multiple memory address spaces

I beg your pardon? Press F8 on Start-up, select "Safe Mode Command Prompt", key in WIN.COM and watch Windows start up...

Win3.x, same deal. Is Win3.x then the base system/real core? The only reason IO.SYS changed (ref. my statements/posts above in "configuring a new IBM OS") was to recognize the "new design". VMM only interfaces with DOS (IO.SYS/COMMAND.COM). Also changed to directly utilize 32-bit addressing (Win3.x cheated).

Ever use Linux/Unix (any Mainframe OS's)? When the Computer (not PC) "boots", it is called IML (Initial Microcode Load), which is, in essence, the equivalent of the "hard wired" code within the CPU of a PC. Next, is IPL (Initial Program Load), which is the "base system" (DOS/OS/Unix/Linux, etc. Command Prompt). Next, start the Partitions/Address Spaces (VMM). Now you have the option of either manually or automatically starting other software on top (GUI's like CICS, the equivalent of W-I-N-D-O-W-S) within a given Space, which in turn allows other software (eg GUI editors like MS-Word and Inventory Systems like SQL+Apps) to run on top of it. Where do you think Billy got the ideas from? Remember IBM (all source code supplied)? Fact!

It's all in the design of the operating system. But there's still an underlying "DOS/OS" that needs to be changed to get the rest to work (Windows components changes). IO.SYS (in DOS/Winx.x) is the primary Hardware Interface in conjunction with Command.com, hence the underlying OS (ref. previous paragraph). Windows (utilizing VMM interface) is written to simply route its commands to the underlying DOS (single-threaded). It does indeed "ride on top of" DOS!

Winx.x is not capable of what is being suggested unless the basis (IO.SYS) is modified as well and that is the reason NT-oriented Windows is so radically different (NTLDR, not IO.SYS; ref. IBM DOS vs OS). Are you suggesting eliminating/rewiting IO.SYS and changing the MBR (which is also different)? How many modules in VMM? One, two, a gazillion? How many to change? One, two, a gazillion?

Again, go ask the KernelEX Project people whether a SINGLE module can be changed/written to access multi-cpu's. It's in the overall design. I refuse to disassemble the key components of Windows just to prove the point of "forget it, not worth it". "Migrate" NT/XP code to 9x if anyone thinks they can; it would be faster!

Systems Mainframe 2themax! Done it all and recognize the direct relationships to Windows x.x/NT... (p.s. Mainframes also have SQL)

Sheesh! Symantics (mis-spelled before; was that what confused you all?)

@ohmss006 - Give it up; ain't gonna happen. Be happy with a jet-engine Win9x PC using 1 core. Genuises many on MSFN, but unless there are any takers for such a vast project (ain't heard from any volunteers yet) forget it... Pipe dream...

[marxo]

submix8c, on Jul 20 2007, 05:38 PM, said:

"Migrate" NT/XP code to 9x if anyone thinks they can; it would be faster!

AFAIK for this "operation" u need a source code of a kind... Probably a source code of nt-based Windows

[eidenk]

submix8c said:

All Win 1.x/2.x/3.x and Win9x (including WinME) run "on top of DOS" That's why you can Upgrade from Past (DOS) through "current" WinME (with the appropriate Windows Upgrade sequences).

You can upgrade from anything to anything if a proper installer/setup is written. Please avoid that kind of nonsense arguments.

submix8c said:

Win3.x, same deal. Is Win3.x then the base system/real core? The only reason IO.SYS changed (ref. my statements/posts above in "configuring a new IBM OS") was to recognize the "new design". VMM only interfaces with DOS (IO.SYS/COMMAND.COM). Also changed to directly utilize 32-bit addressing (Win3.x cheated).

It's all in the design of the operating system. But there's still an underlying "DOS/OS" that needs to be changed to get the rest to work (Windows components changes). IO.SYS (in DOS/Winx.x) is the primary Hardware Interface in conjunction with Command.com, hence the underlying OS (ref. previous paragraph). Windows (utilizing VMM interface) is written to simply route its commands to the underlying DOS (single-threaded). It does indeed "ride on top of" DOS!

IO.SYS is the OS loader, after that it is not there/in use anymore AFAIK. As for command.com, it is not needed to either boot or for the operation of the OS (not even for running DOS programs). At least not on WinME.

I have deleted all instances of command.com (as well as the C:\Windows\Command folder) from an install of ME in a virtual machine and I can boot it and run it normally, I can even run DOS programs except of course for command.com because it is not there anymore.

So you'll have to find something else because it obviously does not work as you say.

The DOS programs seem to be running thanks to DOSMGR.VXD, the MS-DOS Emulation Manager and V86MMGR.VXD, the MS-DOS Memory Manager.

Here is a bit of information about them from the Windows 95 Device Driver Kit Virtual Machine Manager (VMM) Reference Manual :

Quote

Chapter 32. The MS-DOS Manager

The virtual MS-DOS manager virtualizes the elements of the MS-DOS operating system, such as device drivers and internal flags. This device also manages instance data for MS-DOS.

Chapter 35. The V86-Mode Memory Manager

The V86MMGR manages memory for V86-mode applications. It supports the Expanded Memory Specification (EMS) and the Extended Memory Specification (XMS), is responsible for allocating the base memory for new virtual machines, and translates calls made from protected-mode applications to V86-mode API functions.

http://rapidshare.co...kernel.doc.html

Initially downloaded from the MS FTP site but I could not remember where in this labyrinth

So ?

[submix8c]

IPL (on Mainframe DOS) = IO.SYS
IPL (on Mainframe OS) = NTLDR

Yep, the Boot Sector "loads" and "executes" it. Now what does IO.SYS do? AFAIK it defines basic hardware, grabs and processes some parameters, runs Config.sys/Autoexec.BAT, then tosses you to the command PROMPT awaiting input (didn't even imply you needed command.com).

So go on and boot, delete it from both the HDD and Memory. What now?

I repeat - THAT and every single bit of code inside is DOS and nothing else! Without it, no Programs can access one whit of the hardware. Therefore, Windows is DOS-based. And in NT-type Windows, the NTLDR makes them OS-based. Without these, NOTHING runs, get the picture? I must assume NTLDR (not having looked inside it yet) does very similar processing. These two items prepare the remainder of Programs to do their job (eg the HAL which is a pain in the a@@ to change without reinstall because of other dependencies).

The thread was started suggesting that some poor soul sucker-up to making Win9x use more than one CPU/Core. It's the interaction between modules that makes this "project" unfeasible. If YOU had the Source Code, you would see what I mean...

Now if this is going to be a flame-fest instead, have at it; we're not ALL geniuses. And the thread-starter is probably laughing his bootie off at us "geniuses" (or should I say fools?)... Bye Bye

edit - just did the same in VPC on 98SE "Safe/Command Prompt" then rebooted "Normal" - "The Following File Is Missing Or Corrupt - Command.com - Type the name...". NO WINDOWS! Same message when "Safe/Command Prompt". So how DID you test that on the WinME, hmmmm??? Did you even reboot?

This post has been edited by submix8c: 20 July 2007 - 06:34 PM

[Sfor]

It is possible to say Win9x systems are DOS based and they are not DOS based, in the same time. Everything depends on what "based" word means to the person speaking in the particular moment.

There are no "real" DOS in the NT based systems, except for the emulation.

It is true the DOS is used to run the Win9x kernel. The DOS is still present in the memory while the 9x OS runs. The windows has a feature to shut down the kernel and run a DOS application, without a reboot. Also, all DOS drivers loaded before Win9x kernel starts are available when it is working. Windows can use DOS supported disk access if needed, as well. So, DOS is a part of the Windows 9x design.

Windows 3.xx was more DOS based for sure. All disk and network acees was done trough DOS layer drivers (except for the 32bit disk acces feature, I think).

In relation to the system boot and design, Windows 9x and 3.xx are DOS based.
In relation to the system kernel operations only the Windows 3.xx is DOS based, I believe.
Windows NT is not DOS based, for sure.

What matters in the multi CPU systems is the hardware and memory access management. And, the windows kernel is the one holding both jobs, in the 9x systems.

[submix8c]

@Sfor - Thank you! (see? I knew you were ejumukatid! you even know the definition of "symantics"...)

@all-who-argue-DOS
...read here for more info on MS-DOS (or is MDGx also wrong? ) -
http://www.mdgx.com/msdos.htm

@ohms006 - no takers, too much hassle; guess you're on your own

edit -

Looks like IO.SYS looks in SYSTEM.DAT to get the Configuration (which is where ALL used Hardware/Drivers is defined to set CurrentConfiguration) before starting WIN.COM (which it does). From there I guess you go for the Kernel, VMM stuff (all those VXD's), etc etc (an assumption of course, not having dug into that). Still proves that there is more to this than just one single module mod.

And AFAIK NT-type HAL has something to do with multi-cpu support, or tell me I'm wrong on that too? Seems that you would have to support multi-CPU in Win9x in the same fashion. More than just the kernel module or am I wrong again?.

IBM (used to for extra $; don't know about now) provides Source Code for every last scrap of the DOS/OS and all standard (exception COTS-type) software they provide. MS does not. Makes the project a bit harder unless you're willing to break the EULA, don't you think?

This post has been edited by submix8c: 21 July 2007 - 10:04 AM

[oscardog]

submix8c, on Jul 21 2007, 04:23 PM, said:

@Sfor - Thank you! (see? I knew you were ejumukatid! you even know the definition of "symantics"...)

@all-who-argue-DOS
...read here for more info on MS-DOS (or is MDGx also wrong? ) -
http://www.mdgx.com/msdos.htm

@ohms006 - no takers, too much hassle; guess you're on your own

edit -

Looks like IO.SYS looks in SYSTEM.DAT to get the Configuration (which is where ALL used Hardware/Drivers is defined to set CurrentConfiguration) before starting WIN.COM (which it does). From there I guess you go for the Kernel, VMM stuff (all those VXD's), etc etc (an assumption of course, not having dug into that). Still proves that there is more to this than just one single module mod.

And AFAIK NT-type HAL has something to do with multi-cpu support, or tell me I'm wrong on that too? Seems that you would have to support multi-CPU in Win9x in the same fashion. More than just the kernel module or am I wrong again?.

IBM (used to for extra $; don't know about now) provides Source Code for every last scrap of the DOS/OS and all standard (exception COTS-type) software they provide. MS does not. Makes the project a bit harder unless you're willing to break the EULA, don't you think?

"excuse me"
Microsoft marketing adopted Windows 95 as the product name for Chicago when it was released on August 24, 1995. Microsoft had a double gain from its release: first it made it impossible for consumers to run Windows 95 on a cheaper, non-Microsoft DOS; secondly, although traces of DOS were never completely removed from the system, and a version of DOS would be loaded briefly as a part of the booting process, Windows 95 applications ran solely in 386 Enhanced Mode, with a flat 32-bit address space and virtual memory. These features make it possible for Win32 applications to address up to 2 gigabytes of virtual RAM (with another 2GB reserved for the operating system), and in theory prevents them from inadvertently corrupting the memory space of other Win32 applications. In this respect the functionality of Windows 95 moved closer to Windows NT
You try and run a 32 bit app from a pure ms-dos prompt,i.e ping,traceroute not via windows but f5 on boot up,just because ms-dos is included for backwards compatibility does not mean that windows is a gui for 16 bit dos.

[eidenk]

Now what does IO.SYS do? AFAIK it defines basic hardware, grabs and processes some parameters, runs Config.sys/Autoexec.BAT, then tosses you to the command PROMPT awaiting input (didn't even imply you needed command.com).

So go on and boot, delete it from both the HDD and Memory. What now?

Well you said that IO.SYS (in DOS/Winx.x) is the primary Hardware Interface in conjunction with Command.com. If you say that you imply it is needed. Besides IO.sys appears to be a 32bit file. DOS is exclusively 16bit AFAIK. So where is DOS in IO.SYS ?

Quote

The 32-bit IO.SYS is really a combination of DOS's IO.SYS, MSDOS.SYS, CONFIG.SYS, and AUTOEXEC.BAT files. The new IO.SYS contains all the instructions the operating system needs to initially interact with your hardware. IO.SYS also loads some default CONFIG.SYS and AUTOEXEC.BAT commands, and it controls much of the remaining boot sequence.

A combination of several files. IO.SYS is the Windows 95 equivalent of DOS's IO.SYS, MSDOS.SYS, CONFIG.SYS, and AUTOEXEC.BAT files.

http://www.microsoft.com/technet/archive/w...4.mspx?mfr=true

I repeat - THAT and every single bit of code inside is DOS and nothing else!

Inside what ? In 32bit IO.SYS ?

Without it, no Programs can access one whit of the hardware. Therefore, Windows is DOS-based.

Programs access the hardware through VXDs under the supervision of VMM. See below.

And in NT-type Windows, the NTLDR makes them OS-based. Without these, NOTHING runs, get the picture? I must assume NTLDR (not having looked inside it yet) does very similar processing. These two items prepare the remainder of Programs to do their job (eg the HAL which is a pain in the a@@ to change without reinstall because of other dependencies).

NTLDR means NT (OS) LoaDeR. It loads the OS. Period. You are confusing things. IO.SYS on 9x anf NTDLR on NT INITIALLY interact with the hardware. They are not responsable for hardware access after the OS is fully loaded AFAIK. Can you point out to some docs or specs proving your point ? Especially for 9x.

edit - just did the same in VPC on 98SE "Safe/Command Prompt" then rebooted "Normal" - "The Following File Is Missing Or Corrupt - Command.com - Type the name...". NO WINDOWS! Same message when "Safe/Command Prompt". So how DID you test that on the WinME, hmmmm??? Did you even reboot?

Yes mister I did reboot. And I even searched "My Computer" to make sure it was not loading a copy of it from an obscure location.

BTW, in safe mode Windows MAY BE running on top of DOS but have you noticed how poor it is with regards to supporting hardwre notably ?

Quote

Paging Through MS-DOS

When run on hardware for which 32-bit drivers are not available, Windows 95 may be forced to use MS-DOS and/or the BIOS for access to the paging device. When paging through MS-DOS, the VMM changes its behavior in significant ways, and new rules apply to existing VxDs. Windows 95 will also page through MS-DOS if the system is running in safe mode.

(From the Windows 95 Device Driver Kit Virtual Machine Manager (VMM) Reference Manual again)

The normal behaviour of VMM in normal mode is as follows :

Quote

The virtual machine manager (VMM) is the 32-bit protected-mode operating system at the core of Windows 95. Its primary responsibility is to create, run, monitor, and terminate virtual machines. The VMM provides services that manage memory, processes, interrupts, and protection faults. The VMM works with virtual devices, 32-bit protected-mode DLLs, to allow the virtual devices to intercept interrupts and faults to control the access that an application has to hardware devices and installed software.

Both the VMM and virtual devices run in a single, 32-bit, flat model address space at privilege level 0 (also called ring 0). The system creates two global descriptor table (GDT) selectors, one for code and the other for data, and uses these selectors in the CS, DS, SS, and ES segment registers. Both selectors have a base address of zero and a limit of 4 gigabytes (GBs), so all the segment registers point to the same address space. The VMM and virtual devices never change these registers.

The VMM provides multiple-threaded, preemptive multitasking. It runs multiple applications simultaneously by sharing CPU (central processing unit) time between the threads in which the applications and virtual machines run. The VMM is also non-reentrant. This means that virtual devices must synchronize access to the VMM services. The VMM provides services, such as semaphores and events, to help virtual devices prevent reentering the VMM.

The virtual machine manager provides a variety of services, messages, macros, and structures to help virtual devices manage their corresponding hardware devices and services. The following chapters introduce the services and provide details about how to call the individual services, messages, macros, and structures.

(Still from the Windows 95 Device Driver Kit Virtual Machine Manager (VMM) Reference Manual)

It seems that overall you are very eager to prove multiple core support for Win9x cannot be done. Are you on a mission or something ?

And forget modifying 9x to use dual-core! It's definitely not worth the effort (ref. IBM stuff; look it up...). The systems specs are too radically different...

Still, all-in-all, seemingly not worth the effort.

@ohmss006 - Give it up; ain't gonna happen. Be happy with a jet-engine Win9x PC using 1 core. Genuises many on MSFN, but unless there are any takers for such a vast project (ain't heard from any volunteers yet) forget it... Pipe dream...

@ohms006 - no takers, too much hassle; guess you're on your own

[submix8c]

oscardog, on Jul 21 2007, 07:44 PM, said:

<snip>Microsoft marketing adopted Windows 95<snip>although traces of DOS were never completely removed<snip>386 Enhanced Mode<snip>up to 2 gigabytes of virtual RAM<snip>closer to Windows NT<snip>included for backwards compatibility does not mean that windows is a gui for 16 bit dos.

And your point is? Even remotely in relation to the topic? Ever heard of Software Enhancements (Win1.0 thru WinME)? Ever hear of Redesign? Notice the keyword "virtual" in "virtual RAM"? What does VMM stand for? What is the definition of "VIRTUAL"? And the last statement quoted seems to be an opinion, not necessarily fact (although it might be).
From eidenk -

Quote

The virtual machine manager (VMM) is the 32-bit protected-mode operating system at the core of Windows 95.

Seems that says it all...

Do you have a degree in Systems Analysis, Design, and Implementation (I have)? Do you have access to the source (I did on IBM Mainframes)? Do you specifically know the methodology behind a Page File (I do)? How many languages (not scripting) have you programmed in (I learned Assembler on IBM, went from there)? Sheesh, we all (maybe) have credentials to fall back on; is this a case of "who has more credentials"? Or who is "smarter" than who?

Or are you responding to the link provided in an attempt to refutiate it? My basic premise stands, even if I am incorrect on a number of points (thx to all for corrections/clarifications/information/links).

Win9x is not easily transformed for the support requested. Just too dang many modules and variables. Is that single sentence too much to understand? Do you agree with that sentence or not?

Stick to the subject! Give suggestions for or arguments against. Quit being so danged defensive just because someone else is also relatively (if not completely) informed. I was not slamming, merely attempting to correct as others are so willing to. Fine by me... I attempt to search on it and either confirm or give references. It makes me that much more the wiser. Are you then above all of that?

Finally, why has no other informed and talented member suggested such a modification (perhaps they have)? Multiple CPU boards have been around a loooong time (relatively speaking).

Someone step up to the plate and specifically propose a "fix" to Win9x for multi-CPU support or specifically state that you are against such a proposal. Simple information (accurate or not) without the rationale as to why it is being provided (for or against) is a waste of keystrokes.

Again - and your point was? For myself, this seems to be degrading into a bar-room brawl... started with "I like that picture on the wall; But I don't" and turned into "Tastes great; No, less filling". DUH! So many toes to step on and so little time. Wisdom dictates leaving the bar and leaving the drunks to their little bruhaha (got the analogy? http://www.m-w.com/dictionary/analogy )...

@eidenk - Try reading my first post in this topic, then those following IN ORDER and then THINK of why I continued to respond (my bad for paying attention to y'all). I'm assuming I'm arguing against those who apparently want to do this little project. I wish you luck. Let us all know when it's completed.
BTW - your own words, dude -
http://www.msfn.org/board/index.php?act=fi...&pid=671343
@oscardog - and here's yours -
http://www.msfn.org/board/index.php?act=fi...&pid=673442

http://www.m-w.com/c...onary/semantics
(yep, misspelled it again; anyway there's the link)
I've clearly (if not accurately) stated my case against; where do all of the rest of you stand? 1 means yes, 0 means no (simple enough?)...

Guess I'll have to start a thread in Site & Forum Issues questioning the validity of "bashing"... NOT! Probably everyone who THINKS I am bashing them would probably go there to bash some more. Have you no shame?

Summary -
1 You have my answer to this proposed project - IMHO, NoGo!
2 I have wasted enough time and server space here.
3 On to more fruitful endeavors (how about you all?)...

This post has been edited by submix8c: 22 July 2007 - 04:51 PM

[oscardog]

I suggest you study this subject a little bit more, then perhaps you may have answers rather than to create further inaccurate questions. My latter answer dealt directly with windows in this particular forum in responding to your questions I would have thought you might have ascertained that(for windows take it that it means the relevant os being discussed hence win9x). The original poster asked Microsoft Windows 98 to recognize Dual-Core processors (project?), A place where we can gather all together to make this happen. The differences between the 3 different win2k uni and multiprocessor files, perhaps the examination of the previous use of a modified ntoskrnl.exe etc when dealing with ntfs, and above all the start of vxd,wdm programming might be a start.
I suggested clustering mainly because it could suffice until this forum gets up to speed on stubbing and/or filtering vxds,wdms.

[ohmss006]

hello everyone, its me, now, im sorry if i just came in after so long, but i didnt know when or where to come in, after this thorough discussion between you lot.

dont u start EVER thining that im laughing my head off because of this, but this is knowleging myself and many other Windows 98 people out there! including my friends as well, as htey are very intrigued to know if something like this was possible, i also wanting to try and convince them to join here, also to prove i am not joking around.

i now know for a fact, that Windows 98 is based on DOS, and running on that, and that NT itself doesnt have DOs, but more like an emulation, like XP, but its processes are very much closer to 9x than XP itself.

maybe i should have made this thread in the Unofficial service pack forum, trying to find a way to move this place a well.

from what u lot are saying, it is still possible to make such a thing, but yes, effort wil need to be made into it, and i see not alot are willing to take that challenge, unless someone got the source code to Windows 98, then people can start messing around with it.

unless, another theory, is that if there was a DOS that can do multitasking or multiprocesing, then Windows 98 on top of that, is another possible way to make the OS see dual core.

u lot may thingk im a dumb a**, but no, just no, by making htis thread isnt ridiculous, we are acknowledging as to wats right and not, and this has given me more of an idea on what to do, this isnt the end here. i am very aware there are still loads of people who hasmt came by this and probably knows what to do. i did mention donations as well didnt i?

thats all im going to say right now, need to start finishing bulding my new system.

[submix8c]

oscardog, on Jul 23 2007, 09:06 PM, said:

I suggest you study this subject a little bit more, then perhaps you may have answers rather than to create further inaccurate questions. My latter answer dealt directly with windows in this particular forum in responding to your questions I would have thought you might have ascertained that(for windows take it that it means the relevant os being discussed hence win9x).

I don't recollect requesting any answers based upon creating inaccurate questions. I had already cited both outside reference and cross-reference (e.g. your own posts, ref. VMM), i.e. questioning your own answers. I may have created inaccurate answers (the opposite) and my questions have been perfectly accurate and have been totally ignored and unanswered (ref. previous post). Non-Win9x had been brought into the subject to discuss "which does and does not support multi-cpu"; this is perfectly valid and supports the suggestion that the obstacles to be overcome are numerous. I, sir, suggest that you should be the one to study the subject a bit more, unless you are again ignoring my previous posts on the relationship between IBM and Mr. Gates (source code and my former personal access to IBM's)...

oscardog, on Jul 23 2007, 09:06 PM, said:

The original poster asked Microsoft Windows 98 to recognize Dual-Core processors (project?), A place where we can gather all together to make this happen. The differences between the 3 different win2k uni and multiprocessor files, perhaps the examination of the previous use of a modified ntoskrnl.exe etc when dealing with ntfs, and above all the start of vxd,wdm programming might be a start.

True; lots of modules (as I had stated). However, here is where it started (you also commented) -
http://www.msfn.org/...showtopic=98123
Notice that the topic appears to be a dead issue. Here is where it migrated to -
http://www.msfn.org/...showtopic=99663
Also apparently a dead issue. And it finally wound up here... hilarious!
@eidenk - I see you never posted in those topics...

oscardog, on Jul 23 2007, 09:06 PM, said:

I suggested clustering mainly because it could suffice until this forum gets up to speed on stubbing and/or filtering vxds,wdms.

Here is a link to cluster definition -
http://en.wikipedia....omputer_cluster
Yes, I had several years ago read up on the subject... and the last Mainframe I worked on was OS/390ESA on an Amdahl. I had assumed you were being facetious (need another definition?) when you suggested "firing up" multiple diskless Win9x', since it doesn't appear to be relevant (not a viable solution). Directly (opposite of indirectly) answer posed questions and the original poster will then have a solution or final answer to his/her overall request to insist on loading Windows 98 SE (specifically, BTW) under the hood of his Core-2-Duo Mobo. If you re-read the original topics you will see that this is the case...

"Stubbing" is a cake-walk (in Assembler). True, I don't have specifics on, eg., VXD or WDM constructs, but I'm sure others on this forum understand the concept as well as I do. "ZAPping" (not do be confused with MSZAP) is a method by which code to be "modified" is Branched-Around/Branched-Back-To or Branch-And-Linked/Returned-From tacked-on-code, potentially NO-OPing original binaries. The modified/new code is then executed (relocatable addressing aside). Granted, a simplified explanation (used it many a time), but common since day-one of BAL. That's Basic Assembler Language; the very first god-awful-to-manipulate since computations involved adder-registers for + and - ops and register-shifting for * and / ops. Look it up; I used it when I was trained to nuke cities (back in '69, gulp)! See, I not so toopid. But, sadly, EULA, EULA, EULA!. And DON'T YOU DARE argue ZAPping with me... I will eat you alive!!! With Specifics!

@ohmss006 - I'm quite sure your intentions are good. Recommend only accurate constructive criticism, suggestion, or actual code. I can't help unless I get super-smart super-fast (not gonna happen soon). I have given what I deem a "slightly educated" assessment of the subject and its feasibility. You need some real gurus, none to be found (IMHO) as of yet. Suggest looking back occasionally to see if that happens. I anxiously await the final solution as I too will someday have a multi-cpu MOBO/Server (already have NT-type opsys; not naming it...). So far, only vague relationships and possibilities. As far as "laughing", I would have (at the "I'm smarter than you" syndrome)...
BTW -

Quote

i now know for a fact, that Windows 98 is based on DOS

Careful... get ready for another flame!

Come on, yay-sayers, let's get to debugging/disassembling and recompiling/patching (EULA?) or generating "replacement" modules instead; with examples! Or else point to someone who can (if not you); some support tech you would be if you didn't (I was and did)! Done tooting horns? Getting noisy again... then Chef Ramsey (of Hell's Kitchen) said "STFU and get busy, donkeys" (another analogy). That having been said...

[eidenk]

@submix8c

Delusion, is that what you suffer from ?

Shall I answer you point by point or shall I not ?

[submix8c]

eidenk, on Jul 24 2007, 03:17 PM, said:

Shall I answer you point by point or shall I not ?

Yes. All of them and all questions (you know, those with ? at the end?). Offline, in a zip, with backing, not just your "word" (want a copy of my Tech history? I want yours !). Enough here...

[Sfor]

ohmss006, on Jul 24 2007, 03:24 PM, said:

unless, another theory, is that if there was a DOS that can do multitasking or multiprocesing, then Windows 98 on top of that, is another possible way to make the OS see dual core.

Well, it should not solve the problem. The Windows 9x kernel takes direct controll over the hardware. So, the DOS layer can not do multi CPU support.

If the DOS would support multi CPU cores, the support would stop in the moment the Windows 9x kernel is started.