DOS ain't dead

Jack updated his UDMA drivers! Main change seems to be a merged UDMA/UDVD drivers into one driver called UIDE. I haven't tested it yet but i looks impresive. Good work Jack!
Download site is as usual: http://johnson.tmfc.net/dos/

Sadly, there is a multi-controller ERROR in the 10-11-2007 UIDE driver.
I have sent a corrected 10-13-2007 UIDE driver to Johnson and Lucho for
testing. Until it is posted, be-advised that the 10-11-2007 UIDE may
not work properly on systems with more than two IDE controllers.

MASM v5.1 is needed to assemble the source without errors.

---
MS-DOS forever!

The corrected 10-13-2007 UIDE has now been posted by Lucho as part of his
COMBOOTF "boot" diskette files, which are available at:

<http://linux.tu-varna.acad.bg/~lig/dos>

For 17 years, MASM V5.1 has been my assembler. Those who prefer another
assembler are free to convert UIDE's source file for it.

> MASM v5.1 is needed to assemble the source without errors.

It's offered for sale at E-bay for a few bucks, together with printed documentation

If one searches for it, one can find some places to freely download it, too.

It was also a part of the free OS/2 DDK, but recently IBM have discontinued it.

The source can easily be converted to the newer (and now free) MASM versions.

> The source can easily be converted for the newer (and now free) MASM versions.

...if needed

They're now available at their principal location - the Johnson Lam's site: http://johnson.tmfc.net/dos/

> The source can easily be converted to the newer (and now free) MASM
> versions.

Latest MASM versions are not free! Microsoft prohibits MASM's use for developing open-source software.

---
Forum admin

> The source can easily be converted to the newer (and now free) MASM
> versions.

it's "easy" for those being very familiar with MASM and its quirks, but "beginners" may have some difficulties. One Example:

    btr [bx+IOF-@],bl

According to Intel docs the second operand must be 16/32 bit. So using BL is somewhat confusing. MASM v6 won't accept BL.

No big issue, of course.

---
MS-DOS forever!

>     btr [bx+IOF-@],bl
>
> According to Intel docs the second operand must be 16/32 bit. So using BL
> is somewhat confusing. MASM v6 won't accept BL.

In fact, MASM 5.10 silently translates in the object file the above instruction to

    btr word ptr 0x4a[bx],bx

> "Beginners" may have some difficulties [with MASM V5.1]. One Example:
>
>     btr [bx+IOF-@],bl
>
> According to Intel docs the second operand must be 16/32 bit. So
> using BL is somewhat confusing. MASM v6 won't accept BL.

"Beginners" will have difficulties in ANY assembly-language, which is
why it is taught at colleges so little, and "C" is taught so much.

Upon changing UIDE's 2 commands of the above form to use "BX", rather
than "BL", UIDE assembles to an identical .SYS file. Thus MASM V5.1
"forgives" this harmless misreading of the 80386 manual.

> Latest MASM versions are not free! Microsoft prohibits MASM's use for
> developing open-source software.

As far as I know, it's Steve Hutchinson (http://www.masm32.com/) who prohibits this, not Microsoft. Where you've read this and for which version?

Like Lucho, I have not seen any "prohibitions" by Gates & Co. against
using the MASM assemblers for open-source software.

If you believe such prohibitions do exist, you are welcome to "Do the
right thing" and DELETE all links from your website to drivers I have
written beginning with QHIMEM/QCACHE/QCDROM, as those drivers were all
assembled using V5.1 MASM.

> If you believe such prohibitions do exist, you are welcome to "Do the
> right thing" and DELETE all links from your website to drivers I have
> written beginning with QHIMEM/QCACHEQCDROM, as those drivers were all
> assembled using V5.1 MASM.

But in this case he'll have to delete also his links for Japheth's HX: it's also developed for MASM! (See http://www.japheth.de/HX.html#hxsrc)

Fortunately, I'm pretty sure that Robert is wrong to claim that Microsoft prohibits open source development with MASM

> But in this case he'll have to delete also his links for Japheth's HX:
> it's also developed for MASM! (See
> http://www.japheth.de/HX.html#hxsrc)
>
> Fortunately, I'm pretty sure that Robert is wrong to claim that
> Microsoft prohibits open source development with MASM

I believe the "not for non-Windows OSes" clause only pertains to the latest free MASMs (8.0? 9.0?), not any of the previous ones that you may have bought in years past. (Of course, I personally like FASMD, but your mileage may vary.)

---
Know your limits.h

I agree with Rugxulo's post, re: MASM "prohibitions" applying only to its
new releases. I would especially like to read such prohibitions against
my 17-year-old V5.1 MASM, released at a time when the idea of open-source
was largely UNHEARD of.

I also note that the "links" to QHIMEM/QCACHE/QCDROM, and later V5.1 MASM
work by me, are still present on this website ...

> > Latest MASM versions are not free! Microsoft prohibits MASM's use for
> > developing open-source software.
>
> As far as I know, it's Steve Hutchinson
> (http://www.masm32.com/) who prohibits this, not Microsoft.
> Where you've read this and for which version?

Sorry for the confusion! I mixed up those things. From the MASM 8.0 EULA: You may install and use one copy of the software on your device to design, develop and test your programs for non-commercial purposes. You may not use the software to develop programs you either intend to distribute for a fee or use to maintain your own business or IT systems.

---
Forum admin

> Fortunately, I'm pretty sure that Robert is wrong to claim that
> Microsoft prohibits open source development with MASM

Yes. This confusion is due to Mr. Hutchesson, who repeatedly claimed such nonsense in his forum and in newsgroups (ALA).

---
MS-DOS forever!

> > Latest MASM versions are not free! Microsoft prohibits MASM's use for
> > developing open-source software.
>
> As far as I know, it's Steve Hutchinson
> (http://www.masm32.com/) who prohibits this, not Microsoft.
> Where you've read this and for which version?

IIRC this issue was discussed in other forum in the past and result has been documented in Dr-DOS Wiki by Japheth and me:

MA$M32: Mr. Hutch has an odd deal with M$, claims a "true freeware status" but at same time prohibits non-M$ and OS development - avoid it absolutely !

MASM ("ordinary" paid versions): no problem. Japheth reported to have a paid version - HX is absolutely legal OTOH allowing to compile HX with something else still would be cool

---
This is a LOGITECH mouse driver, but some software expect here
the following string:*** This is Copyright 1983 Microsoft ***

> IIRC this issue was discussed in other forum in the past and result has
> been documented in Dr-DOS Wiki by Japheth and me:

What's the URL?

> MA$M32: Mr. Hutch has an odd deal with M$, claims a "true freeware status"
> but at same time prohibits non-M$ and OS development
> - avoid it absolutely !

And for some really nasty things read the alt.lang.asm newsgroup.

---
Forum admin

> > Latest MASM versions are not free! Microsoft prohibits MASM's use for
> > developing open-source software.
>
> As far as I know, it's Steve Hutchinson
> (http://www.masm32.com/) who prohibits this, not Microsoft.
> Where you've read this and for which version?

"Any person, party or entity that undertakes any form of legal action whatsoever against the project, its authors, the owners and/or copyright holders of any software contained in the project is licenced to completely and unconditionally remove the software from their computer(s) and nothing else."

I will swear as a matter of law and English grammar that this is garbage, especially "licenced to completely and unconditionally remove the software from their computer(s) and nothing else." This implies that 1) a license could be required to remove software from a computer, and 2) that software might be removed from something other than a computer, but is not allowed to be, or maybe, not 2), but 3) only software and nothing else can be removed from a computer (which means, don't remove that old dead hard drive, etc.). Total nonsense. (I would guesss that they meant that anyone who brings legal action against them is not permitted to use MASM, and must remove it from computers - but they didn't say that, and what is not said is not binding.)

As a practical matter, in present company, this is a killer: "You cannot use the MASM32 Project to write software for Non-Microsoft Operating Systems." But how could they prove a violation? If a prog works under MS-DOS, but also under FreeDOS, etc., then their only remedies would be to teach programmers how to 1) write to undocumented and unique features of MS-DOS, and 2) break other OSes. However, the truth is, they will not do 1), and if they tried 2), they would face legal action from owners and users of the other DOSes.

This is not on solid legal ground: "The MASM32 project cannot be used to create open source software or any other project under any form of licence that requires the user of the MASM32 project to surrender the rights they are afforded under the MASM32 licence. In particular the MASM32 licence completely excludes projects licenced under the GNU organisation's published GPL licence and/or variants."

This turns "rights" into requirements. It is true that some other licenses do not permit charging money, but, and this is important, it is not required by law that anyone charge money for anything - not even in the most capitalist of countries. It is as legal to give away software for free as it is to give away any other property, including actual money.

If a contract (what a software license really is) that requires giving something away for free is entered into voluntarily, then that contract is legally defensible. In a US court, all a clever lawyer has to say is "What do you mean, my client can't give software away? What's next - can't invite friends to lunch without charging them for the food because it's cooked on a MASM stove?" Case over, friends eat for free.

Garbage.

> In fact, MASM 5.10 silently translates in the object file the above
> instruction to
>
>     btr word ptr 0x4a[bx],bx

Yes, MASM "silently " generates the opcode which is expected ... not that surprisingly, since an opcode which uses a 8bit register as second operand simply doesn't exist.

On another occasion, however, MASM v5.1 "silently" generates code which most likely isn't exactly what was expected:

125E  33 DB                    I_Term: xor     bx,bx                   ;Zero BX-reg. for below.
 1260  80 7F 0A 20                      cmp     [bx+DvrNam-@].lb,' '    ;Is driver "name" valid?
 1264  75 0A                            jne s   I_SetN                  ;Yes, eliminate any spaces.
 1266  C7 47 0A 4455                    mov     [bx+DvrNam-@],"DVDU"    ;Set our default "name".
 126B  C7 47 0E 2031                    mov     [bx+DvrNam+4-@],"   1"
 1270  BE 0F79                  I_SetN: mov     si,(TTLName+8-@)        ;Set driver "name" in title.

Here the opcodes at 1266 and 126B will move a WORD, but I bet it is expected that a DWORD is moved instead.

I detected this issue because I tried to assemble UIDE.ASM with Open Watcom's WASM. Although WASM is far away from MASM v6 compatibility, it should - almost - work with v5.1 sources. With UIDE source, WASM has 2 problems:

1. A label followed by a DB pseudo-code in the very same line (EndIO):

        pop     ax              ;Reload returned I-O error code.
        stc                     ;Set carry again to denote error.
EndIO:  db      0EAh            ;Return to driver "exit" logic.
        dw      (Exit-@)
@RtnSeg dw      0               ;(Driver segment address, Init set).
        db      0               ;(Unused alignment "filler").

is perfectly legal with MASM v5.1, but v6.x complains and WASM generates a wrong opcde. Since it doesn't cost much, one might consider to separate such lines:

        pop     ax              ;Reload returned I-O error code.
        stc                     ;Set carry again to denote error.
EndIO:
        db      0EAh            ;Return to driver "exit" logic.
        dw      (Exit-@)
@RtnSeg dw      0               ;(Driver segment address, Init set).
        db      0               ;(Unused alignment "filler").

2. WASM doesnt like the brackets around "short":

s equ     <short>           ;Default conditional jumps to "short".

This is a WASM bug. Adding a comment for WASM users might be useful.


After these "trivial" adjustments, WASM/WLINK generated a perfectly valid UIDE.SYS.

---
MS-DOS forever!

> Adding a comment for WASM users might be useful.

Another "non-issue", by one who has ADMITTED he enjoys being the
cause of only TROUBLE, as anyone can read here:

<http://www.bttr-software.de/forum/forum_entry.php?id=914>

I am not the "babysitter" for NASM, TASM, WASM, or "WhateverSM".
For 17 years, V5.1 MASM has been my assembler. People who wish
some other assembler must take RESPONSIBILITY for this decision.
If that assembler is "not compatible" with previously acceptable
assembly-language standards, there is an easy and fast solution:
Use V5.1 MASM instead.

> 1266 C7 47 0A 4455 mov [bx+DvrNam-@],"DVDU" ;Set our default "name".
> 126B C7 47 0E 2031 mov [bx+DvrNam+4-@]," 1"
>
> Here the opcodes at 1266 and 126B will move a WORD, but I bet it is
> expected that a DWORD is moved instead.

Yes - fortunately, DvrNam already contains the same static string "UDVD1 "

> After these "trivial" adjustments, WASM/WLINK generated a perfectly valid
> UIDE.SYS.

Very good! So WASM isn't just an "unusable toy", but a perfectly usable tool, albeit poorly documented...

Michal Necasek will be glad to hear that!

> Very good! So WASM isn't just an "unusable toy", but a
> perfectly usable tool, albeit poorly documented...

No, my points mentioned in the link are still valid, the first one now even more after this "label + db in one line"-error has been detected. If you have to scrutinize every byte the assembler has generated I wouldn't call such a tool "perfectly usable".

---
MS-DOS forever!

> No, my points mentioned in the link are still valid, the first one now
> even more after this "label + db in one line"-error has been detected. If
> you have to scrutinize every byte the assembler has generated I wouldn't
> call such a tool "perfectly usable".

Yes, still buggy, but "imperfectly usable"

There is possibly a misconception in the second part:

> This turns "rights" into requirements. It is true that some other licenses
> do not permit charging money, but, and this is important, it is not
> required by law that anyone charge money for anything - not even in the
> most capitalist of countries. It is as legal to give away software for
> free as it is to give away any other property, including actual money.

the MASM32 license talks about "Open Source", not about "Freeware". So you can charge people for using your software or not, that doesn't matter, but you must not reveal the source.

Sounds slightly less absurd, but still "strange".

---
MS-DOS forever!

> The MASM32 project cannot be used to create open source software or any
> other project under any form of licence that requires the user of the MASM32
> project to surrender the rights they are afforded under the MASM32 licence.

I've always wondered what are these "rights afforded under the MASM32 licence" that he means, which would be "surrendered" by the open source licence?!

I wouldn't dare to ask Hutchesson about that

> > The MASM32 project cannot be used to create open source software or any
> > other project under any form of licence that requires the user of the
> MASM32
> > project to surrender the rights they are afforded under the MASM32
> licence.
>
> I've always wondered what are these "rights afforded under the MASM32
> licence" that he means, which would be "surrendered" by the open source
> licence?!
>
> I wouldn't dare to ask Hutchesson about that

lucho, the little I do know about you is that both you and I definitely don't understand some of the cryptic licenses in use today (and this is one of the more oddball cases, too). This MASM32 one is very very strange, stupid, self-defeating, paranoid, and just incomprehensible. I know MASM is a semi-decent assembler, BUT I prefer others:

* FASM -- very popular but no .OBJ support, no linker needed, almost no cmdline use (by design), very fast, runs on 9 OSes, x86-64 since 1.64, various official and unofficial IDEs, auto-optimizes encodings, FOSS
* NASM -- very popular, supports .OBJ (among a billion others), finally has x86-64, can optimize encodings, LGPL
* YASM -- very popular NASM fork, no .OBJ support but supports Intel or AT&T/GAS syntax (plus x86-64 since a while), auto-optimizes encodings, mostly BSD
* Octasm -- small, fast, supports DOS or OctaOS, can run snippets from cmdline, partial ELF reloc read support, up to SSE3, only outputs raw binary (and/or optional simple debug info text .SYM), auto-optimizes encodings, FOSS
* LZASM -- supports only TASM Ideal mode, up to SSE4, outputs .OBJ only, optimizes LEA (like MASM), closed src

All of these run on pure DOS w/ DPMI support (so no worries about memory limits).

---
Know your limits.h

> For 17 years, V5.1 MASM has been my assembler. People who wish
> some other assembler must take RESPONSIBILITY for this decision.
> If that assembler is "not compatible" with previously acceptable
> assembly-language standards, there is an easy and fast solution:
> Use V5.1 MASM instead.

If there was no fear of using MASM V5.10, FreeDOS could have had a much better kernel, written entirely in Assembler, with built-in LFN support, which just needs fixing some bugs and further development to potentially become the best - RxDOS by Mike Podanoffsky (who has also written the book "Dissecting DOS").

Alas, what FreeDOS got instead was DOS-C by Pat Villani (who has also written a book for his kernel), written in "C" initially for the Motorola MC68000 CPU, later ported to x86 without knowledge of the undocumented internal structures of DOS, with its famous "f-nodes" instead of the System File Tables (SFT).

All this because of the free-software / open-source purism of some open source adopters!

Is it a so big crime to search for MASM510.ZIP and get this old but still rather useful rusty junk?

> If there was no fear of using MASM V5.10, FreeDOS could have had a much
> better kernel, written entirely in Assembler, with built-in LFN support,
> which just needs fixing some bugs and further development to potentially
> become the best - RxDOS by Mike Podanoffsky (who has also written the book
> "Dissecting DOS").
>
> Alas, what FreeDOS got instead was DOS-C by Pat Villani (who has also
> written a book for his kernel), written in "C" initially for the Motorola
> MC68000 CPU, later ported to x86 without knowledge of the undocumented
> internal structures of DOS, with its famous "f-nodes" instead of the
> System File Tables (SFT).

FreeDOS does what it is supposed to do, running DOS software sufficiently fast and stable. You're fighting phantoms.

> Is it a so big crime to search for MASM510.ZIP and get this old but still
> rather useful rusty junk?

No, but it is disrespect for other people's intellectual property, or, in other, slightly old-fashioned words, DISHONORABLE.

---
MS-DOS forever!

> FreeDOS does what it is supposed to do, running DOS software sufficiently fast and stable. You're fighting phantoms.

Or, like Don Quixotte, windmills?!

Still, RxDOS could have been a much better and faster kernel of FreeDOS. Maybe in the future EDR-DOS could be such a kernel, if there is no fear of using it there.

> > Is it a so big crime to search for MASM510.ZIP and get this old but still rather useful rusty junk?
>
> No, but it is disrespect for other people's intellectual property, or, in
> other, slightly old-fashioned words, DISHONORABLE.

Sorry, but I can't imagine how this can be so in this case.

> Still, RxDOS could have been a much better and faster kernel of FreeDOS.
> Maybe in the future EDR-DOS could be such a kernel, if there is no fear of
> using it there.

Well, it's shareware, but the true "problem" is:

99% of DOS users run DOS in a VM under Windows/Linux and dont care about speed
0,9% run DOS - sometimes - natively, and are happy with its speed

The rest of 0,1% sits in front of their DOS machines, running speed benchmarks and counting the free available memory. These 0,1% are you, Jack and Udo.

---
MS-DOS forever!

I absolutely agree with Lucho. "C" for systems work is the PITS!!

One of my associates once ran 8 partitions, including both an IP and Novell
"stack"(as he called them), on an 8-Megabyte 80386. Not a "Ball of Fire",
but it got the job done. How did he do this?? Using DOS and "Desqview".
And "When was that", you ask?? 1988, almost TWENTY YEARS AGO!!

But, it made-no-money for "Chairman Gates", who even THEN had 1500 "Clowns"
working on what came to be known as Windows/NT, now called "Vista", and the
reason for everyone's "Mad Rush" to get quad-processors and 1-GB of memory!
What language does Windows use?? "C", God HELP us all!!

"C", quad-processors, and 1-GB of memory, What A JOKE!! They never really
figured-out how to use ONE processor and maybe 32-MB of memory efficiently,
and NOW we must all "dash out" to be SOLD DOWN-THE-RIVER by Intel and Gates
AGAIN??? "Get REAL, People"!!!

If there are still "doubters" about the power of assembly-language, perhaps
the next UIDE, for which "I have a plan", may make "believers" of them. A
"hint" about it: Same features/capabilities, 3.5K max. HMA, and 1728 bytes
of upper-memory, that is ALL! Johnson and Lucho already have its "model".

And I do NOT consider it at-ALL "dishonorable" to continue using V5.1 MASM,
which has been in my possession for 17 years!!!

> 99% of DOS users run DOS in a VM under Windows/Linux and dont care about
> speed 0,9% run DOS - sometimes - natively, and are happy with its speed
>
> The rest of 0,1% sits in front of their DOS machines, running speed
> benchmarks and counting the free available memory. These 0,1% are you,
> Jack and Udo.
>
>

Good joke, but how did you figure out the percentages?! Do you have statistics?!
If they're really correct (which I doubt!), it'd mean that the real DOS is dead!

I am PROUD to be among your "0.1% group", along with Lucho and Udo, who
still value something YOU must never have HEARD of: EFFICIENCY!!!

> Good joke, but how did you figure out the percentages?! Do you have
> statistics?!

my estimates (DosEmu counts as VM, NTVDM and DosBox don't count as DOS).

> If they're really correct (which I doubt!), it'd mean that the real DOS is
> dead!

Ok, if the numbers are not 99/0.9/0.1 but 96/3.6/0.4, would this mean that DOS is alive?

---
MS-DOS forever!

I couldn't agree more, Jack!

A quote from http://www.abc.net.au/sydney/stories/s1958814.htm:

Bloatware

It seems bigger more complex computer programs are outstripping the mammoth gains made in computer hardware technology.

A comparison was done between a 1986 Computer and a modern day PC.

The old computer, with just 4 Megabytes of ram and a tiny 8 MegaHertz processor outperformed the modern dual-core beast on nine out of seventeen tests.

The trial looked at how each computer handled similar tasks like word processing or making spreadsheets. It turns out modern programs waste a lot of processing power on unnecessary routines.

> Ok, if the numbers are not 99/0.9/0.1 but 96/3.6/0.4, would this mean that DOS is alive?

Of course! 4 times more alive

> I couldn't agree more, Jack!

But this "C vs. ASM" debate is such old hat that it is bloatware itself.

People believe what they want to believe, and usually the perception is deformed as well to be able to ignore the facts which don't support one's belief. This is a common disease of human beings, everyone suffers from it, but sometimes it may help a little if you become at least aware of it.

---
MS-DOS forever!

You, who have shown absolutely NO appreciation for EFFICIENCY, who only
like to "Pick FIGHTS" with others, and who now want not-another debate,
can achieve your ends very simply: "Keepen Zie Mouth SHUT", and do not
begin any OTHER "threads" on this forum!!!

> 99% of DOS users run DOS in a VM under Windows/Linux and dont care about
> speed
> 0,9% run DOS - sometimes - natively, and are happy with its speed

Please quote your source of information or it could be interpreted that you are a misinformer and a liar. OTOH those "0,9%" as you say, are happy as long as they don't know something better. When they will know, they will discover new forms of "happines" and will forget the "happines" of today.

Yes, I know. Your post was just a joke, an example of your "intelligent" sarcasm. But you are too repetitive in your attitudes and stances. You are incapable to bring something new, so it's time for you to be disposed without any semnificative loss. Be sure that no one will cry after you.

> Please quote your source of information or it could be interpreted that
> you are a misinformer and a liar.

Please feel free to think what you like!

> Yes, I know. Your post was just a joke, an example of your "intelligent"
> sarcasm.

Where did you detect "sarcasm"?

> You are incapable to bring something new, ...

This might be true. Old dogs ...

> Be sure that no one will cry after you.

Good to know, but how did you manage to get a mandate to speak for all members of this board?

---
MS-DOS forever!

> Where did you detect "sarcasm"?

I didn't detect, but this was your intention.

> This might be true. Old dogs ...

To be incapable to bring something new is not a question related to age...

> Good to know, but how did you manage to get a mandate to speak for all
> members of this board?

I assumed that the members of this board are thoughtful. I must be ashamed because of my presupposition?

> I didn't detect, but this was your intention.

Ok. You somehow felt that my intention was sarcasm. <irony>I understand</irony>.

> > Good to know, but how did you manage to get a mandate to speak for all
> > members of this board?
>
> I assumed that the members of this board are thoughtful. I must be ashamed
> because of my presupposition?

Ok, you spoke for others without mandate. That's impolite in any case and yes, you should consider to be ashamed. It doesn't matter to what extend you regard your opinion as a matter of course.

---
MS-DOS forever!

> But this "C vs. ASM" debate is such old hat that it is bloatware itself.

It's not just C vs. Assembler. The applications in 1986 were also written mostly in C. But they were still far from being a bloatware. It started a bit later...

> People believe what they want to believe, and usually the perception is
> deformed as well to be able to ignore the facts which don't support one's
> belief. This is a common disease of human beings, everyone suffers from
> it, but sometimes it may help a little if you become at least aware of it.

Don't worry, we're aware, but it's equally true that obvious things are often most difficult to see (especially given the brainwashing of the media and ads!), and here's hardly anything more obvious than bloatware in the computer business. So let's not turn blind eye to it.

> Ok, you spoke for others without mandate. That's impolite in any case and
> yes, you should consider to be ashamed. It doesn't matter to what extend
> you regard your opinion as a matter of course.

If 3 people asynchronously say you the same thing, it's time to synchronise to it

> Ok. You somehow felt that my intention was sarcasm. <irony>I
> understand</irony>.

Even now you try to be sarcastic..., but you couldn't, because you lack any sense of mature humour. Your jokes are high-school quality. I'm sorry for you.

> Ok, you spoke for others without mandate. That's impolite in any case and
> yes, you should consider to be ashamed. It doesn't matter to what extend
> you regard your opinion as a matter of course.

I present my excuses to all who impolitely were made "thoughtful" by me.

> If 3 people asynchronously say you the same thing, it's time to
> synchronise to it

No. You might know the saying about the 1 million flies who cannot be wrong when ... ?

---
MS-DOS forever!

Lucho and Khusraw, it is perhaps time for the three of us to leave Joseph
Heller's "General Scheisskopf" alone, in his rantings at others, and just
proceed with more productive things ...

I now have it confirmed, by Lucho and another associate, that the "model"
for the next UIDE driver works fine. While I consider its next upgrade,
it COULD be released now ...

Is anyone interested in a UIDEJR, "junior UIDE", which caches up to 200-MB
using HMA space (the current driver goes up to 255-MB), BUT which requires
only 1728 bytes of upper-memory for such a cache?? All other features of
UIDEJR would be the same, including up to a 1-GB cache using upper-memory.

It may be nice to learn if a "1.75K Caching Driver" can benefit DOS users.

> > If 3 people asynchronously say you the same thing, it's time to synchronise to it
>
> No. You might know the saying about the 1 million flies who cannot be wrong when ... ?

Yes, I know this German proverb from Tom Ehlert (but only in English, so it may be interesting to read it in original - I mean in German). However I'm sure that you understood me, even if I may not have said it in the best possible way