Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!uwm.edu!ux1.cso.uiuc.edu!news.iastate.edu!vaxf.iastate.edu!XGR39
From: xgr39@isuvax.iastate.edu (Marc Barrett)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Graphics Chips
Message-ID: <1991May4.183924.8432@news.iastate.edu>
Date: 4 May 91 18:39:24 GMT
References: <1991May2.030706.30302@kuhub.cc.ukans.edu>,<21173@cbmvax.commodore.com>
Sender: news@news.iastate.edu (USENET News System)
Reply-To: xgr39@isuvax.iastate.edu
Organization: Iowa State University, Ames, IA.
Lines: 72

In article <21173@cbmvax.commodore.com>, daveh@cbmvax.commodore.com (Dave Haynie) writes:
>In article <1991May2.030706.30302@kuhub.cc.ukans.edu> 2fmlhiccup@kuhub.cc.ukans.edu writes:
>>Here is a quick question...
>
>>   Why is it so difficult to make display chips for the Amiga that will 
>>allow it to do resolutions/colors that i see on VGA?  What are the hardware
>>and software limitations?
>
>VGA chips are very simple chips.  They were originally simple enough for IBM
>to build as a gate array (VGA == Video Gate Array).  Amiga chips, on the 
>other hand, are considerably more flexible, but also much more complex,
>and could only be built as full custom devices.  I would guess that an 
>implementation of the Amiga chips that does, more or less, the same resolutions
>as the basic VGA, 640x480x8 with 24 bit CLUT, would wind up more complex than
>IBM's latest XGA chips.  You're certainly getting something for all that extra
>complexity, but it's not as easy to crank out.  And of course, those several
>hundred Amiga chip registers should stay as compatible as possible in any 
>such new design, or all those idiots bypassing the software interfaces will see
>their software break.
>
>To give you an idea of the relative timing of gate arrays vs. full custom, 
>consider the A3000.  I got seven revisions of the Buster chip between December
>'89 and May '90 (the last few revs were relatively minor compatibility tweaks).
>The last time I was waiting on a full custom part, the Chip Guys gave me two
>revisions in about the same time frame.  The difference between 5,000-10,000
>transistors and 100,000 transistors can be a significant one.

   I agree that a project like the updating of the custom chipset is not
an easy one.  This is only made worse by the inadequate funding of research
and development by Commodore's management.

   I have never said that, by spending more on R&D, Commodore's 
semiconductor group could get the chipset out tomorrow.  In all things
there is a point of diminishing return, beyond which you get fewer
and fewer results for the money that you spend.  However, I do not
think Commodore has reached the point of diminishing return with the
amount that they are currently investing in R&D.

   In short, I believe that better R&D funding could help.  Again, I
do not think the chipset could be out tomorrow if Commodore spent their
entire net worth on R&D this year.  But I do think that better funding
could shave as much as 20% off of the development time of the chipset. 
With a long-term project like this one, adequate funding from the 
beginning could have saved as much as a full year in the devlopment
time.

   It is likely too late for the chipset that is in devlopment now,
but more R&D funding could help in future chipsets.  There will be a
day when the 32-bit chipset will be as out-of-date and inadequate
as the current one.  If Commodore would increase their R&D investing
today, this will save years in the devlopment time of that chipset,
in the distant future.

   Unfortunately, I doubt their is anybody in Commodore's management
who is as capable of thinking in the long-term as I am.  Commodore's
managers do almost everything in the short-term, and cannot maintain
long-term projects.  This is expecially apparent in Commodore's
marketing practices.  Commodore attempted a short spurt of image
advertizing at many universities last year.  However, when the 
advertizing did not show absolutely immediate results, the head of
Commodore's educational-institution efforts was forced out.  As a
result, I really doubt that there will be another Commodore 
advertizement at many of these institutions ever again.

>
>>Jason Wilson
>
>
>-- 
>Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
>   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
>      "That's me in the corner, that's me in the spotlight" -R.E.M.