[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Spectrum Internet Suite

On 6/9/2010 7:18 PM, geoff wrote:

   To: edepobay
Since I'm the other half who developed SIS, it should be noted that I haven't
given up on it.  But then I haven't touched the code since the last release
either.  You will find the requirements that SIS needs far exceeds what is
available with an 8-bit Contiki.  All of the open source components of SIS
scripts can be a good start for someone who wants to get an understanding what
it takes to put a Web Browser together.  To optimize memory, all of the
comments were pulled some time before 1.0 went into beta.  No effort was made
to maintain a commented version for primary development (I don't have any
commented version any more).


Touching on what was brought up:
* SIS was designed from the initial stages to support viewing of graphics,
   but not inline.  There is stub code in the parser that identifies an image
   and provides a message that it isn't supported.  All that is needed is
   for a picture viewer module to be written and accept the contents and
   provide a user interface to view/scroll the picture and pass return back to
   the standard SIS interface when the user is done looking at the picture.

   I don't share the same view as others when it comes to images.  IMO, it
   should be decoded on the local Apple because:
   * Who will maintain a 24/7 proxy server forever?
   * How will security be maintained of content that is being processed through
     a proxy?

   I believe it was a reported that a highly optimized JPEG decoder took 40
   seconds on an acceleratd IIGS for a 1024x768 24-bit picture when viewed on
   the 24-bit Second Sight display.  More time would be needed to support
   lower color bit-depth and larger picture.  It would be interesting to see
   how well a 30MHz accelerator performs here.


Since most pictures will be quite large compared to the IIgs screen,
with much deeper colors, it might make sense to reduce the resolution
by a factor of 8 and just display the "DC" terms of each 8x8 pixel DCT
block.

This would tremendously reduce the amount of processing, and would allow
most graphics to fit on a IIgs (or other Apple II) screen.  A further
simplification would be to render the pictures only in grayscale (or
something like it ;-), since color rendition will be terrible anyway.

Perhaps each 8x8 color block could be represented on the Apple screen
as a 2x2 monochrome block with five brightness levels...

-michael

NadaNet 3.1 for Apple II parallel computing!
Home page:  http://home.comcast.net/~mjmahon/

"The wastebasket is our most important design
tool--and it's seriously underused."