Re: MISC-d Digest V99 #70
- To: MISC
- Subject: Re: MISC-d Digest V99 #70
- From: theFox@xxxxxxxxxxxxxxxxxxx (Jeff Fox)
- Date: Mon, 12 Jul 1999 11:35:19 -0500 (CDT)
Dear MISC readers:
>Date: Sat, 10 Jul 1999 21:04:34 -0400
>From: "vic plichota" <atsvap@cgo.wave.ca>
>Jeff, either cite specific examples, or shut up. I am sick of your
>vague allusions.
Vic, Please be polite. You will get more cooperation that way. I can
do neither, either, or both. But consider branching.
Word vs byte branching: We have word based branching only on the 5-bit
instruction machines while ShBoom II with 8-bit opcodes opted for byte
based branching so that you can branch to an arbitrary opcode. However
extra cycles are added to branching. This is how I would understand
the memory to work and this is what I heard. If someone knows more
they are welcome to say so.
I know it is considered important in certain compiler environments and
that it doesn't matter much in Machine Forth. It sounded nice to me
but was called brain damage by someone who knows more then I.
>From: "Lloyd R. Prentice" <pai@tiac.net>
Lloyd has had lots of great ideas and I am sure will continue to
have them. I could only contribute so much as he says. ;-)
>Yet in no way were we smart enough to roll
>Forth-based JPEG encoders/decoders;
You know I love to use the two JPEG decode and display program examples done
at iTV as examples of differences in Forth coding style. :-)
>From a software point of view, what does a state-of-the-art, commercial
>MISC OS, web-browser, word processor, relational data base, spread sheet,
>graphics editor, action-adventure game, etc., etc., etc. look like? Or is
>MISC irrelevant to the desk top?
This reminds me of Chuck's comments about a small team writing
this in 1/10 of 1% of MS's code size etc. First you have to decide
what you would plan to write.