Re: Accessing Command-line text

On Sat, 24 Feb 2007 +0100, Herbert Kleebauer <klee@xxxxxxxxx> wrote:
"¬a\\/b" wrote:
On Thu, 22 Feb 2007 09:57:32 +0100, Herbert Kleebauer <klee@xxxxxxxxx>
But if you are interested in multi-precision programming, take a look at

http://www.informatik.uni-bonn.de/~schoe/tp/TPpage.html

It is a simulated Turing Machine for multi-precision arithmetic. It is
a nice example that speed is made by the algorithm and not by the hardware.
If you will be able to get only one percent of this speed (for real long
numbers) with optimized assembly programs, then you are very good.

my routine that find factorial (fatt()) should be faster because use 1
int array only; so cpu should have that memory in some cache

Surely your program to calculate n! is faster than the one I posted
(but mine calculated the correct result). But that has nothing to
do with the fastest known algorithm to multiply large numbers
(Schönhage-Strassen algorithm):

http://en.wikipedia.org/wiki/Sch%C3%B6nhage-Strassen_algorithm

my routine seems ok. there was something to repair in the output
routine and all was done. where are my errors in the last post?

than i doubt that "Schönhage-Strassen" multiplication is better in the
particular multiplication
digit0 * (digit0, digit1, digit2, ..., digitn)
as here. So my routine could be the faster

My code was just copied from a program to display an ellipse
segment which was only optimized for size and not speed. Because
my assembler doesn't support floating point instructions, I had
to do it with multi precision integers. But I'm not sure if it's
easier at all with floating point instructions.


my geeks language is better than all, because has 2 dimensions (like C
language etc etc) and show there is nothing wrong to write loops using
jumps, ("goto"s) but in this way i can write something that all you
professional and experienced programmers all togheter, can't write;
i alone a poor hobbyist programmer that fun in programming in his free
time with "ne arte, ne parte"

To sad that nobody is smart enough to recognize that your
"geeks language is better than all". But at least you are not
alone: there is a language which is the best for teaching
assembly programming without being an assembly language and
there is a fully integrated assembly package which is better
than any HLL for writing applications, but nobody is smart
enough to recognize it.

so you think too your assembly languge is better than C?
yes for you

this show that my language is superior too

you wrong on
0) the metal cpu has more power than an assembly immaterial language
1) assembly not need indentation
2) assembly not need reduce instructions length in txt format
3) assembly not is the language of choice for write
fast mathematical routines, input-output routines
arrays routines, all that need to be fast and small

Somebody must be wrong, but the question is: who?

i think "not i"

But if you are so good at implementing an algorithm, write
a program which can display an segment of an ellipse.
Here the specification:

Well, you are very good in doing all that
in this example, yes you seem not exceeded in that

too it seems to me cpu run >80% when wait
after show the picture

To draw a path use the commands l,e and E:
------------------------------------------

x1 y1 x2 y2 l : draw a line from x1,y1 to x2,y2

x2 y2 l : the last point of the previous line
is used as the first point

x1 y1 m1 x2 y2 x3 y3 m3 e : draw a segment of an ellipse from
the start point x1,y1 through the
point x2,y2 to the endpoint x3,y3
m1 (m3) is the angle between the curve
and the y-axis in the point x1,y1
(x3,y3). 0-> 0 degree,
128-> 90 degree
255->180 degree

x2 y2 x3 y3 m3 e : the last point of the previous curve
is used as the first point

If E is used instead of e, the
complete ellipse is drawn


To stroke the path, use the s command:
-------------------------------------

width color s : stroke current path with width "width"
and color "color"; clears the current path

To fill an area use:
--------------------

x1 y1 x2 y2 .... xn yn color f : x1,y1 - xn,y2 are used as starting
points for filling with color "color"

Other commands:
---------------

color c : clear screen and fill it with color "color"

count w : wait "count" * 1/18.2 seconds
press any key to immediately exit the
waiting loop

b15 g15 r15 b14 g14 r14 ... b0 g0 r0 p
: set color palette with the 16 given
RGB values

Space and "," are delimiters for the numbers and anything
after a ; till the EOL is ignored.

========================================================================

Here a sample input file (_.dat) and my implementation of the
program (elli.com):


echo >>_.dat ;;;;;;;;;;;;;;;;;;;;; picture 4 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;
echo >>_.dat 0 c 162,290 80,155 l 80,140 l 180,140 l 180,155 l 97,155 l
echo >>_.dat 173,280 l 80,290,230 130,340 173,280,58 e 95,290,230
echo >>_.dat 130,325 162,290,58 e 80,290 95,290 l 200,240,0 250,340
echo >>_.dat 300,240,0 E 215,240,0 250,325 285,240,0 E 320,240,0
echo >>_.dat 370,340 420,240,0 E 335,240,0 370,325 405,240,0 E
echo >>_.dat 455,255 440,255 l 440,340 l 540,340 l 540,325 l 455,325 l
echo >>_.dat 455,270 l 455,270,128 540,215 440,150,158 e 455,255,128
echo >>_.dat 525,215 440,165,158 e 440,150 440,165 l 140,240 2 f
echo >>_.dat 207,240 12 f 327,240 9 f 447,260 13 f 2 15 s
echo >>_.dat ;;;;;;;;;;;;;;;;;;;;; picture 4 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;
echo >>_.dat 90 w

yes: love and peace for 2005

"picture 4" show 2005; is it not better 2007?

0 c
162,290 80,155 l 80,140 l 180,140 l 180,155 l 97,155 l
173,280 l
80,290,230 130,340 173,280,58 e
95,290,230 130,325 162,290,58 e
80,290 95,290 l

200,240,0 250,340 300,240,0 E
215,240,0 250,325 285,240,0 E
320,240,0 370,340 420,240,0 E
335,240,0 370,325 405,240,0 E

450, 340, 540, 340 l
485, 140 l
470, 140 l
520, 325 l
450, 325 l
450, 340 l

140,240 2 f
207,240 12 f
327,240 9 f
451, 339 1 f
2 15 s
90 w

elli.com <_.dat
del elli.com
del _.dat
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

it seems here if i copy "the above" in a "e.bat" file
if i double click above that "e.bat", than "elli.com" is deleted
but "_.dat" exist

if i execute that in a dos cmd windows than both "_.dat" and
"elli.com" are deleted
.

Relevant Pages

  • Re: Optimization help please
    ... few people that can improve such a routine using assembly. ... Have the compiler output assembler code for that routine and see if any ... be read then the CRC can be computed on each 'block' and the algorithm could ... > checksum is cpu and disk intensive, ...
    (comp.lang.asm.x86)
  • Re: Thinking assembly?
    ... This is one part of thinking in assembler to me. ... > algorithm and implement it in the higher-level language. ... you can't merely go back the HLL code to obtain ...
    (alt.lang.asm)
  • Re: Asm For Nerds
    ... matching "with the programming patterns that i use". ... assembler versions simultaneously while developing *new* functionality. ... When that is understood, the language ... Language loses relevance quickly if the algorithm is just plain crap. ...
    (alt.lang.asm)
  • Re: Accessing Command-line text
    ... a nice example that speed is made by the algorithm and not by the hardware. ... my assembler doesn't support floating point instructions, ... "geeks language is better than all". ... @echo off ...
    (alt.lang.asm)
  • Re: Survey
    ... >> and way more readable than any HLL. ... > that bug in the symbol table management code that causes your assembler ... > implement the same algorithm with far fewer lines of code in many ... terrified by the incredible complexity of this Language, ...
    (alt.lang.asm)