17 Jan
2006
17 Jan
'06
9:13 p.m.
I've been on a little purchasing kick lately and I finally have MS MASM
6.1 with paper docs. I'd like to know what was fixed between 6.1 and
the later versions, such as 6.11 and 6.14. Can any of you who have
those later versions send me the 'readme' or scan any paper sheets that
might have described the differences?
When I grow up, I'm going to create BIOS ROM extensions for ancient
hardware ...
Thanks,
Mike
17 Jan
17 Jan
10:49 p.m.
On 1/17/2006 at 8:13 PM Michael B. Brutman wrote:
I've been on a little purchasing kick lately and I
finally have MS MASM
6.1 with paper docs. I'd like to know what was fixed between 6.1 and
the later versions, such as 6.11 and 6.14. Can any of you who have
those later versions send me the 'readme' or scan any paper sheets that
might have described the differences?
Buy? I picked up the Windows Server 2003 SP1 DDK back in December for the
price of a (200 MB) download from microsoft.com. It includes both ML
7.10.4035 and ML64 8.00.40310.39 for AMD IA64. It also includes C++
14.00.40310.41 for AMD64 and C++ 13.10.4035 for 80x86.
Get it here:
http://www.microsoft.com/whdc/devtools/ddk/orderddkcd.mspx
Such a deal!
Cheers,
Chuck
11:31 p.m.
Chuck Guzis wrote:
Buy? I picked up the Windows Server 2003 SP1 DDK back
in December for the
price of a (200 MB) download from microsoft.com. It includes both ML
7.10.4035 and ML64 8.00.40310.39 for AMD IA64. It also includes C++
14.00.40310.41 for AMD64 and C++ 13.10.4035 for 80x86.
Get it here:
http://www.microsoft.com/whdc/devtools/ddk/orderddkcd.mspx
Such a deal!
Cheers,
Chuck
Thanks for the tip. The nice thing about the old crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
Mike
11:44 p.m.
> Buy? I picked up the Windows Server 2003 SP1 DDK
back in December for the
> price of a (200 MB) download from microsoft.com. It includes both ML
> 7.10.4035 and ML64 8.00.40310.39 for AMD IA64. It also includes C++
> 14.00.40310.41 for AMD64 and C++ 13.10.4035 for 80x86.
> http://www.microsoft.com/whdc/devtools/ddk/orderddkcd.mspx
Thanks for the tip. The nice thing about the old
crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
Download option? It wants me to buy the CD (admittedly just for S&H, but I'd
rather just download it). Do you have a link to the download area?
--
--------------------------------- personal: http://www.armory.com/~spectre/ ---
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- When in doubt, take a pawn. -- Mission: Impossible ("Crack-Up") ------------
18 Jan
18 Jan
12:51 a.m.
On 1/17/2006 at 8:44 PM Cameron Kaiser wrote:
Download option? It wants me to buy the CD (admittedly
just for S&H, but
I'd
rather just download it). Do you have a link to the download area?
Looks like good ol' MS removed that at the beginning of the year (who knows
why).
But try the KMDF iso:
http://download.microsoft.com/download/3/b/5/3b519887-c4af-4ab3-9578-cec48cc
84d94/WDFv10.iso
That's free and supposedly contains the DDK tools, but I haven't verified
that it does (just what I heard). About a 200 MB download, but MS has lots
of bandwidth.
Cheers,
Chuck
17 Jan
17 Jan
11:44 p.m.
On Tue, 17 Jan 2006 22:31:41 -0600
"Michael B. Brutman" <mbbrutman-cctalk at brutman.com> wrote:
Chuck Guzis wrote:
I have a version of MASM old enough (4.0) that it fits, and runs,
on my HP95 lx palmtop. It only has a tiny ram drive (512K).
One of the cool things about the HP95 is that they include DEBUG
in the ROM.
Buy? I picked up the Windows Server 2003 SP1 DDK
back in December for the
price of a (200 MB) download from microsoft.com. It includes both ML
7.10.4035 and ML64 8.00.40310.39 for AMD IA64. It also includes C++
14.00.40310.41 for AMD64 and C++ 13.10.4035 for 80x86.
Get it here:
http://www.microsoft.com/whdc/devtools/ddk/orderddkcd.mspx
Such a deal!
Cheers,
Chuck
Thanks for the tip. The nice thing about the old crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
18 Jan
18 Jan
3:17 a.m.
On 1/17/2006 at 11:44 PM Scott Stevens wrote:
I have a version of MASM old enough (4.0) that it fits,
and runs,
on my HP95 lx palmtop. It only has a tiny ram drive (512K).
Most of my code nowadays is 32-bit, so the old MASM's don't work. I've got
MASM going back to the really terrible 1.0 and just about every major
version after that.
Cheers,
Chuck
6:22 p.m.
On Wed, 18 Jan 2006 00:17:14 -0800
"Chuck Guzis" <cclist at sydex.com> wrote:
On 1/17/2006 at 11:44 PM Scott Stevens wrote:
I won't be running any 32-bit code on the HP95LX ;-)
I have a version of MASM old enough (4.0) that it
fits, and runs,
on my HP95 lx palmtop. It only has a tiny ram drive (512K).
Most of my code nowadays is 32-bit, so the old MASM's don't work. I've got
MASM going back to the really terrible 1.0 and just about every major
version after that.
9 p.m.
Scott Stevens wrote:
I won't be running any 32-bit code on the HP95LX
;-)
Same here. My development environment is a 386-40, but the target
processor is an 8088. I'd like to develop natively on the target box (a
PCjr), but I fear I would grow old before getting anything done.
(It's more of a problem with the floppy shuffle and the slow access to
the fake hard drive that I'm using .. the machine is probably more than
fast enough for older versions of MASM and Turbo Pascal.)
9:57 p.m.
On 1/18/2006 at 8:00 PM Michael B. Brutman wrote:
(It's more of a problem with the floppy shuffle and
the slow access to
the fake hard drive that I'm using .. the machine is probably more than
fast enough for older versions of MASM and Turbo Pascal.)
Don't be too sure--MASM didn't get fast until after 1.0.
There were actually two IBM PC assemblers 1.0 that were sold as a bundle;
ASM, which IIRC, would run in 64K on a system with one diskette drive,
didn't have error messages (just numbers) and MASM, which was probably the
slowest assembler (per unit of processor speed) that I've ever seen.
Horribly buggy too--you'd think that an assembler would at least generate
the right code. Phase errors and the dreaded "Internal Error" were all too
common. The manual notes for "Internal Error":
"Usually caused by arithmetic check. If it occurs, notify your authorized
IBM Personal Computer dealer."
Uh huh. And my Authorized IBM Personal Computer Dealer will do what?
There were some very public rants about plunking down $99 for a miserable
piece of garbage.
Fortunately there were other alternatives, including 8086 cross assemblers
running CP/M 2.2. You used them if you were smart.
MASM 4.0 was like a breath of fresh air. Reasonably fast and not too
buggy.
Cheers,
Chuck
10:06 p.m.
On Wed, 18 Jan 2006, Chuck Guzis wrote:
There were actually two IBM PC assemblers 1.0 that
were sold as a bundle;
ASM, which IIRC, would run in 64K on a system with one diskette drive,
didn't have error messages (just numbers) and MASM, which was probably the
slowest assembler (per unit of processor speed) that I've ever seen.
Horribly buggy too--you'd think that an assembler would at least generate
the right code. Phase errors and the dreaded "Internal Error" were all too
common. The manual notes for "Internal Error":
Phase errors were usually caused by forward references,
MOV VAR2, 0
. . .
VAR2 DB ?
and would then generate the message for EVERY label from there to the end.
Avoiding forward references, or overiding the datat type would prevent
them
MOV BYTE PTR VAR2, 0
MASM 4.0 was like a breath of fresh air. Reasonably
fast and not too
buggy.
... and 5.0 was the first one with tolerable documentation provided.
10:26 p.m.
On 1/18/2006 at 7:06 PM Fred Cisin wrote:
Phase errors were usually caused by forward
references,
MOV VAR2, 0
. . .
VAR2 DB ?
and would then generate the message for EVERY label from there to the end.
Avoiding forward references, or overiding the datat type would prevent
them MOV BYTE PTR VAR2, 0
That would be too easy! MASM 1.0 would generate phase errors for seemingly
no paricular reason. Sometimes, just rearranging the order of instructions
(without changing the nature of the references would clear them up. As I
said, a miserable excuse for a product.
Cheers,
Chuck
MASM 4.0 was like a breath of fresh air.
Reasonably fast and not too
buggy.
... and 5.0 was the first one with tolerable documentation provided.
19 Jan
19 Jan
12:53 a.m.
When did MASM start optimizing forward jumps? (i.e. multiple passes if
necessary to make sure that 3-byte jumps were used only when necessary or
when specified.) Was it MASM 4--and was it a command-line option?
Cheers,
Chuck
1 a.m.
Chuck Guzis wrote:
MASM 4.0 was like a breath of fresh air. Reasonably
fast and not too
buggy.
Any TASM memories?
And what's everyone's opinion on IDEAL mode? I find people either love
macros/high-level structure stuff in an assembler, or completely hate it...
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Help our electronic games project: http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
A child borne of the home computer wars: http://trixter.wordpress.com/
2:14 a.m.
On 1/19/2006 at 12:00 AM Jim Leonard wrote:
Chuck Guzis wrote:
There were a mess of them, TASM, A86, ASM86, MAC, RMAC... Almost anything
would work if all you needed was a COM file. You could even use the macro
facility in a foreign assemlber like M80 to generate an executable. .EXEs
were a different matter, though.
Now, if you want a small assembler, I've got a copy of DRI RASM-86--the
executable is about 32K. It doesn't have true macros, but does implement
code macros (Remember Intel ASM86? You could add and remove instructions
at will).
(I couldn't remember the name of the little DRI assembler, so I had to take
a look on my bookshelf. Found a 1984 Comdex shuttle bus schedule with the
manual...).
Cheers,
CHuck
MASM 4.0 was like a breath of fresh air.
Reasonably fast and not too
buggy.
Any TASM memories?
4:42 p.m.
On 1/19/2006 at 12:00 AM Jim Leonard wrote:
And what's everyone's opinion on IDEAL mode? I
find people either love
macros/high-level structure stuff in an assembler, or completely hate
it...
It's like any preprocessor. You can obscure what you're trying to do with
too much content; on the other hand, a preprocessor can be a real godsend.
Before MASM incorporated things like CALL with argument-list and
procedure-local variables, I had a nice macro library that essentially did
the same thing. I still use macros to do things like maintain enum list
orderings and encapsulate little things like I/O delays. But by and large,
with the MASM 6 improvements, I use macros very sparingly.
But that's not to say that there's no use for them. Suppose I wanted to
write a cross-assembler for some alien CPU. If the restrictions on binary
format weren''t too onerous, I'd probably just write up a bunch of macros
and let MASM do the grunt work. Similarly, when programming Windoze DLL's
and device drivers, macros can come in very handy to encapsulate the
Windows-specific conventions.
Occasionally, you have the task of taking a list of data items and creating
a complex structure with them. Macros make the job easy.
I occasionally wish that C's preprocessor were more powerful, like that of
PL/I. But while the C preprocessor is standard, I've seen assembler macro
facilities that were so weak as to be almost useless and some that were
wonders (support for macro arrays, typed variables and so on).
Years ago, a friend wrote an assembler for the amZ8000. He wasn't much of
an assembly-language writer, but he was a real compiler guru. So the
result was an assembler with a syntax that resembled a HLL. I wasn't sure
that I liked the idea. Assembly is the first cousin to machine language
and ought not to overly obscure that relationship.
Cheers,
Chuck
18 Jan
18 Jan
10:58 p.m.
It was thus said that the Great Michael B. Brutman once stated:
Scott Stevens wrote:
I did a lot of development on a PCjr. I had quite a bit of memory in the
system, so I set up a 128k RAM disk, copied MASM (2.x I think), LINK.EXE,
the editor I used (PE.EXE 1.0---a whopping 40k executable) and the source
code I was working on. Just before running the program I was developing, I
would copy the source to floppy, just in case.
Did that for several years.
-spc (I think I had 384k of RAM in my PCjr ... )
I won't be running any 32-bit code on the
HP95LX ;-)
Same here. My development environment is a 386-40, but the target
processor is an 8088. I'd like to develop natively on the target box (a
PCjr), but I fear I would grow old before getting anything done.
19 Jan
19 Jan
12:57 a.m.
Michael B. Brutman wrote:
Same here. My development environment is a 386-40,
but the target
processor is an 8088. I'd like to develop natively on the target box (a
PCjr), but I fear I would grow old before getting anything done.
(Hi Mike!) I am probably the only person crazy enough to still develop
on the actual hardware itself (in my case 5150/CGA). The keyboard and
ST-225 seek noise are comforting and nostalgic, but the real benefits
are: I get to think a little more as something is compiling, and
thinking is the real work; also, I have immediate confirmation that my
program runs on the intended hardware.
Of course, I'm doing CGA fiddling, so I *must* test on the hardware, and
most times it's just easier to code the whole thing on it.
I get around the floppy shuffle problem by using a serial cable and
transmitting changed files once a night using FastLynx.
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Help our electronic games project: http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
A child borne of the home computer wars: http://trixter.wordpress.com/
12:51 p.m.
You can develop on the original hardware, and I did a lot of work on my Jr
during the 8 years that I had it, but the floppy shuffle got very very old.
Back then I had a 256K system. After creating the requisite RAM disk to fill
the low memory hole left by the bass-ackwards video buffer there wasn't much
left. I was using the Zbasic 3 and 4 series compilers back then, and made
extensive use of overlays. (I wrote a BBS that ran on an XT for two years in
NYC.) If I really need to max out the memory I would cut the video buffer to
4k (text only), skip the RAM disk, and then have about 220K left.
I guess I'm spoiled now. My 'daily driver' Jr has 640K, full sized PC 5150
keyboard, DOS 5, an a SCSI drive (bi-di parallel port with a parallel-to-scsi
adapter). That's quite a better system than before and I wouldn't have to
shuffle anything, but my requirements have grown too. MASM 6.x might have a
chance of running on it, but Turbo C++ 3.0 won't, and most of the stuff
available from 1988 on up needs more horsepower than the Jr has.
Right next to it is the 386-40 with the much faster disk, networking to my other
machines for file transfer, etc. The 386-40 is there for data transfer anyway,
and it makes a good 'mothership' for the Jr. (It has a Central Point Option
Board, which is nice too.)
18 Jan
18 Jan
5:21 a.m.
----- Original Message -----
From: "Michael B. Brutman" <mbbrutman-cctalk at brutman.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Tuesday, January 17, 2006 11:31 PM
Subject: Re: MASM 6.1 vs 6.11
Thanks for the tip. The nice thing about the old crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
Mike
That's the same reason I recently purchased MS Visual Basic 3.0 Pro, a nice
stack of paper manuals (compared to the 5.0 I purchased years ago which was
mostly just a cdrom and a small booklet). For some reason I feel like
messing around with 16 bit Windows 3.1 programming lately.
6:27 p.m.
On Wed, 18 Jan 2006 05:21:07 -0500
"Teo Zenios" <teoz at neo.rr.com> wrote:
----- Original Message -----
From: "Michael B. Brutman" <mbbrutman-cctalk at brutman.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Tuesday, January 17, 2006 11:31 PM
Subject: Re: MASM 6.1 vs 6.11
16-bit Windows runs exceptionally well[1] on modern hardware, as
long as you can get driver support. And if you are doing
low-level programming on older processors, the tools are all DOS
based anyway. Virtual DOS sessions run well enough inside
Windows 3.11. And of course, you can run DOS and Windows 3
inside a virtual machine, i.e. Bochs on your favorite free 'unix'
variant.
[1]As well as 16 bit Windows ever _could_, I should qualify.
Thanks for the tip. The nice thing about the old crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
Mike
That's the same reason I recently purchased MS Visual Basic 3.0 Pro, a nice
stack of paper manuals (compared to the 5.0 I purchased years ago which was
mostly just a cdrom and a small booklet). For some reason I feel like
messing around with 16 bit Windows 3.1 programming lately.
7:16 p.m.
On 1/18/2006 at 6:27 PM Scott Stevens wrote:
16-bit Windows runs exceptionally well[1] on modern
hardware, as
long as you can get driver support. And if you are doing
low-level programming on older processors, the tools are all DOS
based anyway. Virtual DOS sessions run well enough inside
Windows 3.11. And of course, you can run DOS and Windows 3
inside a virtual machine, i.e. Bochs on your favorite free 'unix'
variant.
I use DOS, but I run with a 32-bit DPMI server. I find the 16-bit
performance of Pentium-class systems to be very much inferior to the native
32-bit mode.
http://www.thefreecountry.com/programming/dosextenders.shtml
Cheers,
Chuck
7:34 p.m.
----- Original Message -----
From: "Scott Stevens" <chenmel at earthlink.net>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Wednesday, January 18, 2006 6:27 PM
Subject: Re: MASM 6.1 vs 6.11
16-bit Windows runs exceptionally well[1] on modern
hardware, as
long as you can get driver support. And if you are doing
low-level programming on older processors, the tools are all DOS
based anyway. Virtual DOS sessions run well enough inside
Windows 3.11. And of course, you can run DOS and Windows 3
inside a virtual machine, i.e. Bochs on your favorite free 'unix'
variant.
[1]As well as 16 bit Windows ever _could_, I should qualify.
I am either going to use my 386DX/40 or 486/133 VLB system for it, both run
Windows 3.11 (WFW)
7:15 p.m.
At 05:21 AM 1/18/2006 -0500, you wrote:
----- Original Message -----
From: "Michael B. Brutman" <mbbrutman-cctalk at brutman.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Tuesday, January 17, 2006 11:31 PM
Subject: Re: MASM 6.1 vs 6.11
Massive resources for current work in MASM:
Link: assembly language resources
http://www.grc.com/smgassembly.htm
Includes current Windows versions (sorry, that's off topic).
"So tell me, just how long have you had this feeling that no one is
watching you?" (Christopher Locke: Entropy Gradient Reversals)
--... ...-- -.. . -. ----. --.- --.- -...
tpeters at nospam.mixcom.com (remove "nospam") N9QQB (amateur radio)
"HEY YOU" (loud shouting) WEB ADDRESS http//www.mixweb.com/tpeters
43? 7' 17.2" N by 88? 6' 28.9" W, Elevation 815', Grid Square
EN53wc
WAN/LAN/Telcom Analyst, Tech Writer, MCP, CCNA, Registered Linux User 385531
Thanks for the tip. The nice thing about the old crusty version that I
picked up is that it runs on an old DOS PC and it has the paper docs.
I'm not ancient, but online docs suck compared to paper.
Mike
That's the same reason I recently purchased MS Visual Basic 3.0 Pro, a nice
stack of paper manuals (compared to the 5.0 I purchased years ago which was
mostly just a cdrom and a small booklet). For some reason I feel like
messing around with 16 bit Windows 3.1 programming lately. 
10:07 p.m.
That's the same reason I recently purchased MS Visual Basic 3.0 Pro, a nice
stack of paper manuals (compared to the 5.0 I purchased years ago which was
mostly just a cdrom and a small booklet). For some reason I feel like
messing around with 16 bit Windows 3.1 programming lately.
Use Delphi 1.0. It's muuuuch better.
g.
--
--
"I'm not crazy, I'm plausibly off-nominal!"
Proud owner of 80-0007
http://www.f15sim.com - The only one of its kind.
6793
days inactive
6794
days old
24 comments
10 participants
participants (10)
-
cclist＠sydex.com -
chenmel＠earthlink.net -
cisin＠xenosoft.com -
geneb＠deltasoft.com -
mbbrutman-cctalk＠brutman.com -
spc＠conman.org -
spectre＠floodgap.com -
teoz＠neo.rr.com -
tpeters＠mixcom.com -
trixter＠oldskool.org