POLL: Who are the 5 best programmers.
Charles Ader
Tenie Remmel (tj...@mail.idt.net) wrote:
> This is an informal poll of people on the Internet. The results will
> be published in Programming Tips & Tricks Magazine next month. Names
> of people who respond will not be mentioned.
> ------------------------------------------------------------------------
> Who are the 5 best Assembly Language Programmers today? Not including
> yourself :) You don't have to rank them 1 through 5, just say which
> 5 people you think are the best assembler programmers.
It is obvious. Four of the five top slots belong to Mr. William Gates III,
the fifth slot must be exclusively reserved for S.N. (ok, it was a joke.)
(please please please don't flame me)
Simon Tatham
Colin Andrew Percival <cper...@sfu.ca> wrote:
> The first 5 people you can think of who DIDN'T get involved in the asm
> vs. C debate.
Damn. I knew I shouldn't have made that one post :-(
--
<^ I /\/\ O /\/ Simon Tatham <sg...@cam.ac.uk> <ana...@pobox.com>
_> ------------ Trinity College, Cambridge, CB2 1TQ, England.
Jean-Pierre Lebel
Tenie Remmel (tj...@mail.idt.net) wrote:
> This is an informal poll of people on the Internet. The results will
> be published in Programming Tips & Tricks Magazine next month. Names
> of people who respond will not be mentioned.
> ------------------------------------------------------------------------
> Who are the 5 best Assembly Language Programmers today? Not including
> yourself :) You don't have to rank them 1 through 5, just say which
> 5 people you think are the best assembler programmers.
Some of the stuff I've seen by Micheal Abrash impresses the hell out of
me, so I would put him somewhere in the list of five.
==============================================================
JEAN-PIERRE LEBEL | EMAIL: jpl...@sympatico.ca
C.E.T. Student | http://www3.sympatico.ca/jplebel
"If I was smart, I'd go back to using Deskmate."
==============================================================
Vision
On Sun, 26 Jan 1997 20:32:55 GMT, ca...@netcom.com (Charles Ader)
wrote:
>Tenie Remmel (tj...@mail.idt.net) wrote:
>> This is an informal poll of people on the Internet. The results will
>> be published in Programming Tips & Tricks Magazine next month. Names
>> of people who respond will not be mentioned.
>> ------------------------------------------------------------------------
>
>> Who are the 5 best Assembly Language Programmers today? Not including
>> yourself :) You don't have to rank them 1 through 5, just say which
>> 5 people you think are the best assembler programmers.
>
>It is obvious. Four of the five top slots belong to Mr. William Gates III,
>the fifth slot must be exclusively reserved for S.N. (ok, it was a joke.)
>(please please please don't flame me)
Wel Well Well.
Lest we forget Peter Norton...
after all hes the one that actually reveals secrets.. unlike Gates.
and another group would be Future Crew..
Vision...
Colin Andrew Percival
Tenie Remmel (tj...@mail.idt.net) wrote:
: Who are the 5 best Assembly Language Programmers today? Not including
: yourself :) You don't have to rank them 1 through 5, just say which
: 5 people you think are the best assembler programmers.
The first 5 people you can think of who DIDN'T get involved in the asm
vs. C debate.
Colin Percival
Terje Mathisen
Jean-Pierre Lebel wrote:
>
> Tenie Remmel (tj...@mail.idt.net) wrote:
> > This is an informal poll of people on the Internet. The results will
> > be published in Programming Tips & Tricks Magazine next month. Names
> > of people who respond will not be mentioned.
> > ------------------------------------------------------------------------
>
> > yourself :) You don't have to rank them 1 through 5, just say which
> > 5 people you think are the best assembler programmers.
>
> me, so I would put him somewhere in the list of five.
Mike is indeed a great guy, but I'd have to rate David Stafford above
him for pure asm programming:
David won both the "Annual Code Optimization Challenge"s that Mike
Abrash announced while writing his monthly column for PC Techniques.
Others, in no particular order:
Mike Schmit, Chris Hecker, Mike Cranford
Terje
--
- <Terje.M...@hda.hydro.com>
Using self-discipline, see http://www.eiffel.com/discipline
"almost all programming can be viewed as an exercise in caching"
John Harries
>Jean-Pierre Lebel wrote:
>>
>> Tenie Remmel (tj...@mail.idt.net) wrote:
>> > This is an informal poll of people on the Internet. The results will
>> > be published in Programming Tips & Tricks Magazine next month. Names
>> > of people who respond will not be mentioned.
>> > ------------------------------------------------------------------------
>>
>> > Who are the 5 best Assembly Language Programmers today? Not including
>> > yourself :) You don't have to rank them 1 through 5, just say which
>> > 5 people you think are the best assembler programmers.
>>
>> Some of the stuff I've seen by Micheal Abrash impresses the hell out of
>> me, so I would put him somewhere in the list of five.
>
>Mike is indeed a great guy, but I'd have to rate David Stafford above
>him for pure asm programming:
>
>David won both the "Annual Code Optimization Challenge"s that Mike
>Abrash announced while writing his monthly column for PC Techniques.
>
>Others, in no particular order:
>
>Mike Schmit, Chris Hecker, Mike Cranford
>
>Terje
Good grief, what a lot of Mikes.
Maybe it is important what you call your kids.
How about changing this to most self-effacing demo coders.
[Damn, I can't think of anyone to vote for!]
Hugs,
Johnny
jo...@curved-logic.com Falling with style
Phil Carmody
Terje Mathisen wrote:
> Jean-Pierre Lebel wrote:
> > Tenie Remmel (tj...@mail.idt.net) wrote:
> > > Who are the 5 best Assembly Language Programmers today?
> > Some of the stuff I've seen by Micheal Abrash impresses the hell out of
> > me, so I would put him somewhere in the list of five.
> Mike is indeed a great guy, but I'd have to rate David Stafford above
> him for pure asm programming:
> David won both the "Annual Code Optimization Challenge"s that Mike
> Abrash announced while writing his monthly column for PC Techniques.
> Others, in no particular order:
> Mike Schmit, Chris Hecker, Mike Cranford
> Terje
Hmm, I'm glad to see you've stuck to the 'not including yourself' bit.
Otherwise you'd have had to stick yourself alongside David Stafford.
Some of the stuff D.S. has come up with has a kind of 'I got 128K of
code in a 64K segment by using both 0000 and 0001 as entry points'
perversity to it, it makes me whimper and cry seeing how smart he's
been. (the technique I mention is based on a hack once used in the days
of program memory storage on solid cylinders)
Some of the stuff you come out with only makes me swear at how clever
you've been.
I personally place swearing below whimpering and crying...
My mate Mats (with a bit of help from me) came up with a phong shading
texture mapper which (reproducably) clocks at 1.5 pentium ticks per
pixel inner loop. He looks up to many of the other Demo writers, so
there must a plentiful supply of incredibly good coders out there. Most
of which do not have names, just handles. The Demo scene is a little too
detached from newsgroups like this - they could contribute (and learn) a
lot.
--
Do they sterilise needles for lethal injections?
Phil Carmody being uncontroversial = p...@scigen.co.uk
But Fat Phil being controversial = fat...@freenet.hut.fi
(In their eyes, I guess everything I say is controversial)
Terje Mathisen
Phil Carmody wrote:
>
> Terje Mathisen wrote:
> > Jean-Pierre Lebel wrote:
> > > Tenie Remmel (tj...@mail.idt.net) wrote:
> > > > Who are the 5 best Assembly Language Programmers today?
> /SNIPS/
> > > Some of the stuff I've seen by Micheal Abrash impresses the hell out of
> > > me, so I would put him somewhere in the list of five.
>
> > Mike is indeed a great guy, but I'd have to rate David Stafford above
> > him for pure asm programming:
> > David won both the "Annual Code Optimization Challenge"s that Mike
> > Abrash announced while writing his monthly column for PC Techniques.
> > Others, in no particular order:
> > Mike Schmit, Chris Hecker, Mike Cranford
> > Terje
>
> Hmm, I'm glad to see you've stuck to the 'not including yourself' bit.
> Otherwise you'd have had to stick yourself alongside David Stafford.
Thanks, but David definitely stands head and shoulders above me:
For the last Code Opt challenge (Conway's Game of Life) I did all the
math, proving to my own satisfaction that my approach was the fastest
possible, i.e. the least # of operations/cell.
I honestly believed that nobody could beat me by more than a few
percent, by writing better outer loop code.
Well, see below.
> Some of the stuff D.S. has come up with has a kind of 'I got 128K of
> code in a 64K segment by using both 0000 and 0001 as entry points'
> perversity to it, it makes me whimper and cry seeing how smart he's
> been.
Exactly.
His Game of Life code did a constant factor _more_ operations/generation
than mine, however his code did those operations on a much smaller
working set, thereby managing to keep most/all of it within the 8K L1
cache.
The result was not just a few percent faster code, his version was a
full 100% faster! :-(
> (the technique I mention is based on a hack once used in the days
> of program memory storage on solid cylinders)
This hack is still at least slightly useful: :-)
I have an x86 strcpy() library replacement which is faster than inline
code as soon as the average string length exceed 3 or 4 bytes. One of
the tricks it uses is very much like the one you mentioned above:
; setup code here, then I need to jump halfway into the first
iteration:
test eax,12345678h ; 5-byte opcode, with 4-byte immediate value
org $-2 ; Step two bytes back into the immediate part
next_iteration:
mov [esi+ebx],eax ; Two-byte opcode: Store previous dword
; This is where we start the first iteration, start by loading some
data
mov eax,[esi]
add esi,4 ; Update both source and destination pointers
...
jcc next_iteration
The problem with this kind of code, is that it takes one more iteration
before you get proper instruction pairing, since the initial trick loads
the instructions and marks instruction boundaries, but then the second
iteration has to this all over again.
Terje
PS. The only hack I've written which can come close to normal Stafford
standards is the following ascii email bootstrap code: Using only the
70+ characters defined in the Internet MIME standard, write an
executable program.
This one has the added benefit of being able to survive most forms of
reformatting, like having the CRLF line terminators replaced with just
LF (Unix), CR (Mac) or nothing (Word processors writing one
line/paragraph).
Here is a simple 64-bit capable Unix-style DiskUse utility encoded this
way:
Run it (under a debugger if you don't trust me) as DU /? to get a syntax
description.
-----------------------------------------------------------------
This is a program encoded as a text file.
To use the program, remove all extra lines before and after
the text executable and save the remainder to disk as DU.COM.
The result is a normal executable .COM file!
*********** Remove this line and all lines above it! ***********
ZRYPQIQDYLRQRQRRAQX,2,NPPa,R0Gc,.0Gd,PPu.F2,QX=0+r+E=0=tG0-Ju E=
EE(-(-GNEEEEEEEEEEEEEEEF 5BBEEYQEEEE=DU.COM=======(c)TMathisen95
&&P4Vw4+Vww7AjAyoAwAAzJzAEAJP7JzBaANQ+wAIxJzAeAJX9JzIxANJ0JzKzAK
WwwAP4JzPwAJi8JzXxANA3wAa4JzavAKA9wAX8JzXyALy8au1+Wwe7CA0/G7LBW7
Www7v7h8CAi8A9A6x8CAP7G7AK0VKASzAza6vAAaLAh90TI9H9LAx90QH8I8Vww7
Agavy9AmAj6+B7V7gA6PAzG72+H9f7AtIxAvCzAkCzAgAkQ/AiYxXzYxQ/U+DaT+
Q/AkCZAMAUCZQ/AtBaAKAWAXANCZAQQ/YxPwPwSzXzQ/GwAXAQBaAMAHQ/AKARQ/
AkCzAgAkXzFwArAtQ/DaARAQAICZAMAKCZAMA4A5Q+=6C8E/L6BAOjyATwND4SzA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*********** Remove this line and any lines below it! ***********
Simon Hosie
Tenie Remmel (tj...@mail.idt.net) wrote:
> Who are the 5 best Assembly Language Programmers today? Not including
> yourself :) You don't have to rank them 1 through 5, just say which
> 5 people you think are the best assembler programmers.
Colin Andrew Percival:
> The first 5 people you can think of who DIDN'T get involved in the asm
> vs. C debate.
Ooh ooh! Me me!
Terje Mathisen
Simon Hosie wrote:
>
> Terje Mathisen:
> > Not at all!
> >
> > How do you propose to bootstrap someone without a useful email program?
> >
> > How do you think LapLink-style transfer of the executable to the target
> > machine happens? (They can admittedly use a much wider character set,
> > but my program can transfer itself over _any_ kind of connection.)
>
> If you hadn't emphasised "any" then I wouldn't reply, but I'm afraid I
> can't help myself now. <ahem> ...
>
> Even a piece of string?
Ouch!
You got me. :-)
Actually, if you tie an empty soup can to each end of the string, you
can use it to spell out the text/code program, i.e.
-Hello, are you ready?
= Sure, go ahead.
-The code is: Zulu, Romeo, Yankee, Papa, Quibec, India, Quibec, Delta,
Yankee, Lima
... etc.
After about 30 minutes of this, the receiver would have entered the
characters in his text editor and saved them, whereupon he could run the
code. (Modulo any transmission channel errors) :-)
The initial bootstrap code looks like this:
1FEF:0100 5A POP DX
1FEF:0101 52 PUSH DX
1FEF:0102 59 POP CX
1FEF:0103 50 PUSH AX
1FEF:0104 51 PUSH CX
1FEF:0105 49 DEC CX
1FEF:0106 51 PUSH CX
1FEF:0107 44 INC SP
1FEF:0108 59 POP CX
1FEF:0109 4C DEC SP
1FEF:010A 52 PUSH DX
1FEF:010B 51 PUSH CX
1FEF:010C 52 PUSH DX
1FEF:010D 51 PUSH CX
1FEF:010E 52 PUSH DX
1FEF:010F 52 PUSH DX
1FEF:0110 41 INC CX
1FEF:0111 51 PUSH CX
1FEF:0112 58 POP AX
1FEF:0113 2C32 SUB AL,32
1FEF:0115 2C4E SUB AL,4E
1FEF:0117 50 PUSH AX
1FEF:0118 50 PUSH AX
1FEF:0119 61 POPA ; Crucial opcode: No 808x need apply
1FEF:011A 2C52 SUB AL,52
1FEF:011C 304763 XOR [BX+63],AL ; Modify code below, to
1FEF:011F 2C2E SUB AL,2E
1FEF:0121 304764 XOR [BX+64],AL ; generate a JMP xxx opcode
1FEF:0124 2C50 SUB AL,50
1FEF:0126 50 PUSH AX
1FEF:0127 752E JNZ 0157 ; Forward jump to flush prefetch
queue!
Terje
Simon Hosie
Simon Hosie
Phil Carmody:
> Some of the stuff D.S. has come up with has a kind of 'I got 128K of
> code in a 64K segment by using both 0000 and 0001 as entry points'
> perversity to it,
That reminds me of those programs on the C64 that started with SYS2051
(BASIC RAM started at 2049, and the most compact stub should have been
SYS2061). I don't know why the hell they did that, though, there never
seemed to be any useful code there, typically a BNE or something, from
memory (BNE is a two byte instruction so it's not clever at all because
those two bytes store the line number which can be anything you like).
David Torrez
Tenie Remmel (tj...@mail.idt.net) wrote:
:
: Who are the 5 best Assembly Language Programmers today? Not including
: yourself :) You don't have to rank them 1 through 5, just say which
: 5 people you think are the best assembler programmers.
The Best Assembly Language Programmer (#1) is easy:
SCOTT NUDDS
Simon Hosie
Terje Mathisen:
> but my program can transfer itself over _any_ kind of connection.
> Even a piece of string?
> It probably depends on how many bytes long the string is... :-)
Damn! That was a perfectly stupid question before you made sense of it.
Simon Hosie
David Torrez:
> The Best Assembly Language Programmer (#1) is easy:
> SCOTT NUDDS
And an extremely close second place... My cat, Claude!
Scott Nudds
Simon Hosie (gum...@airdmhor.gen.nz) wrote:
: David Torrez:
: > The Best Assembly Language Programmer (#1) is easy:
: > SCOTT NUDDS
: And an extremely close second place... My cat, Claude!
Can you post some assembler code that your cat has produced? Perhaps we
should check your claim.
Better yet. Why not give your cat the Blitscale challenge and see if it
can beat me. So far no one has beaten my code.
Simon Hosie
David Torrez:
> The Best Assembly Language Programmer (#1) is easy:
> SCOTT NUDDS
Simon Hosie (gum...@airdmhor.gen.nz) wrote:
> And an extremely close second place... My cat, Claude!
Scott Nudds:
> Can you post some assembler code that your cat has produced? Perhaps we
> should check your claim.
Uhh.. he won't let me use his computer, sorry, and whenever I ask him he
just stares at me blankly.