[ANNOUNCE] Asciiizer 1.0 Released

                                                                       
                                gg    gg    gg                     
                                 `     `     `                     
  _g@@@_      ____        __   __    __    __   ________,      __  
  '`   Mg    #*`"*    _a#*%@j  *@    *@    *@   ##**M#@#    _wM*M@g
        $   $L       g`     `   @     @     @        gP    #`     @
     __g@=   M@@g_  (`          $     $     $       gP    g#******P
   _#`  @       "@k $           $     $     $     _#P     M        
  g`    @         # 7g          $     $     $    _@#      @_       
  @g___#@   __  _#   M@g_____  _@    _@    _@   g@E__     "W@g____g
   `**  $r  *#M**      "*****  **`   **`   **`  ******#M@    `*****
        @`                                                              
               asciiize -- an image to ascii converter                  

"What is asciiize?

    Asciiizer is an image to ascii converter, written by James
    Mitchelhill It takes an image and converts it to text, the hope
    being that the text, viewed as a whole in a monospace font, will
    look like the image. Asciiizer was written as a response to a
    perceived flaw of available ascii converters, namely that they
    concentrated only on gradients and ignored the "shape" of the image.
    By combining both types of converter, asciiizer hopes to produce an
    image that retains both shading information and the "shape" of the
    image.

    In simple terms, you give asciiizer a picture and it gives you back
    some text that looks like the image.


Why should I use asciiize?

    Currently, so far as I'm aware, it's the only image to ascii
    converter that implements an edge detection algorithm. In my
    opinion, it produces better conversions without any manual
    preprocessing than other software. This is just my opinion, however
    so if you prefer something else, don't get angry. At the time of
    writing, the "Image 2 ASCII 2001" competition is being held,
    asciiize is an entrant and it'll be interesting to see how it fares.
    One other advantage is that asciiize should be cross-platform.

    It's also GPL'd


1.3 Why shouldn't I use asciiize?

    It's possibly the most difficult image to ascii converter there is
    to both install and use. The reason for this is that asciiize is
    written in  python and, since I'm too lazy and/or stupid to go
    through the rather torturous process of packing asciiize in an easy
    installer (maybe some nice person will do it in the future), you'll
    need to download and install both Python and PIL (the Pythonware
    Imaging Library) before you can use asciiize.

    The user interface is also a little lacking. In short, rather than
    being a nice graphical application, PIL is run from the command
    line. This means that, if you're using Windows, you'll need to use
    an MS-DOS Prompt window to run asciiize. This may present a problem
    to some people. If you don't use Windows, presumably you already
    know how to use the command line.

    The method of selecting what font to optimise for is, to be honest,
    horrible. There are reasons for this, which, if you're truly
    interested, you can find in the source. Perhaps someone else can
    improve on this?

    The name is designed to provoke misspellings. That's three I's.
    Why? It amuses me. 

    In addition to these woes, asciiize in edge and combined modes runs
    very slowly, taking minutes to convert a moderately sized image to
    a moderately sized block of text. My computer is an 866Mhz P3 with
    64MB of RAM. Running asciiize on anything slower would be painful.

    It's also GPL'd."
       ---excerpt from the README


I wrote asciiizer for my own amusement before I was aware of the 
competition. But since there's one going on, I've entered some images 
into the Image 2 ASCII Competition, so you should be able to see what 
asciiizer can do. All the images there were optimised for andale mono 
8pt, simply because I like that font. Some used gradient mode, others 
combined mode.

So, you probably don't want to download asciiizer after that long list of 
why you won't want to use it. If you still do, everything's available 
from:

<http://www.dreamingbutterflies.com/james/projects/asciiizeR1.0/>

-- 
James Mitchelhill
jan...@mailandnews.com

In alt.ascii-art, Harry Mason <spa...@no.spam> wrote:
> Of course, pnmtools can't do it in greyscale:

Try this:

http://groups.google.com/groups?as_umsgid=eli$010...@qz.little-neck.ny.us

As someone recently commented, it is optimized for white on black.

Sample usage:

anytopnm $file | pnmscale -yscale 0.5| pnmscale -width 75 | ppmtopgm \
	pgmnorm 2>/dev/null | pnminvert | pgmtoascii > $out

Elijah
------
will try out that python thing

On 25 Jun 2001 23:53:26 GMT, Eli the Bearded <*@qz.little-neck.ny.us> wrote:
> In alt.ascii-art, Harry Mason <spa...@no.spam> wrote:
> > Of course, pnmtools can't do it in greyscale:
> anytopnm $file | pnmscale -yscale 0.5| pnmscale -width 75 | ppmtopgm \
> 	pgmnorm 2>/dev/null | pnminvert | pgmtoascii > $out

Ah, I don't have pgmtoascii, just pbmtoascii. I'll have to upgrade pnmtools.

-- 
Harry Mason ("hjm200.ecs@soton@ac@uk" =~ tr/@./.@/)

In alt.ascii-art, Harry Mason <spa...@no.spam> wrote:
> On 25 Jun 2001 23:53:26 GMT, Eli the Bearded <*@qz.little-neck.ny.us> wrote:
> > anytopnm $file | pnmscale -yscale 0.5| pnmscale -width 75 | ppmtopgm \
> > 	pgmnorm 2>/dev/null | pnminvert | pgmtoascii > $out
> Ah, I don't have pgmtoascii, just pbmtoascii. I'll have to upgrade pnmtools.

You missed my point -- get pgmtoascii from this post:
http://groups.google.com/groups?as_umsgid=eli$010...@qz.little-neck.ny.us

I wrote it for my own use.

Elijah
------
used the conversion for a webcam to ascii art CGI