Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!caen!spool.mu.edu!cs.umn.edu!kksys!edgar!orac!bret
From: bret@orac.UUCP (Bret Indrelee)
Newsgroups: comp.sys.amiga.advocacy
Subject: Re: The Amiga's Future
Summary: ECS is big engineering accomplishment
Message-ID: <362@orac.UUCP>
Date: 8 Jun 91 22:50:44 GMT
Article-I.D.: orac.362
References: <1991Jun4.003619.3661@news.iastate.edu> <1991Jun4.025024.823@mintaka.lcs.mit.edu> <1991Jun4.105736.15468@news.iastate.edu>
Organization: Technix Inc., Saint Paul MN, USA
Lines: 63

In article <1991Jun4.105736.15468@news.iastate.edu> taab5@isuvax.iastate.edu writes:
>In article <1991Jun4.025024.823@mintaka.lcs.mit.edu>, rjc@geech.gnu.ai.mit.edu (Ray Cromwell) writes:
>>In article <1991Jun4.003619.3661@news.iastate.edu> taab5@isuvax.iastate.edu writes:
>>
>>>   The ECS isn't going to do much for the Amiga, either, because it was
>>>obsolete before it even went into production.  The ECS is also 98% 
>>>identical to the very oldest Amiga chipset.  
>
>   With the ECS, Commodore basically took some of the registers that
>were hard-wired in the old chipset and made them programmable.  From 
>here, the new modes were achieved by using the new programmable 
>registers.  Overall, however, not all that many changes were made.
>
>   If you doubt my word, use common sense.  Most of the features of the
>old chipset are unchanged in the ECS.  Commodore made a few registers
>programmable, and as a result was able to add some resolution modes
>and other capabilities, but overall everything is the some.  Hell,
>Commodore did not even touch the Paula chip at all, which remains 
>exactly the same today as it was six years ago.  I can even take two
>of the custom chips from my old A1000 (Commodore's oldest machine) sitting
>here, put them into an A3000 (Commodore's newest and most advanced machine),
>and many people would not be able to tell the difference when using 
>the A3000.
>   
[Sorry for so much quoting, but I felt it important.]

The above argument can be though of in the following context:
1)  Original chip set had hard-coded values for some registers.
2)  ECS no longer hard-codes these values.
3)  Adding a register is no big deal.
4)  Therefore, ECS is no big deal.

The problem is that changing from a hardcoded value to a programable value
is *MUCH MORE* expensive in hardware than in software.  Each and every
register bit added to a chip takes a significant amount of space, and
space is limited.

A hardware designing can't just go into his favorite hardware editor
(usually called a CAD or Schematic Entry package) and globally replace
a constant with an expression.  These concepts are easy in software but
*Very expensive* in hardware.

I think that the ECS is a great piece of engineering.  It may not be the
best possible solution starting from scratch, but all of the major
engineering goals are met:
1)  Compatible with previous products
2)  Minimal changes to system (only replace one chip)
3)  Makes whole system more flexible
4)  Fits in original system design concept
5)  Allows field upgrade of existing systems

All these things are commendable.  They did not come without forethought
and hard engineering work.  Please quit assuming that because something
looks trivial to the outside interface, that it must have been easy to
do.


-Bret

-- 
------------------------------------------------------------------------------
Bret Indrelee		|	Our mail is still somewhat unreliable.  Sorry.
uunet.uu.net!cs.umn.edu!kksys!edgar!orac!bret		-And still trying