NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/sys/amiga/advocacy/32148

Internet Message Format | 1992-12-24 | 14.5 KB

Path: sparky!uunet!cis.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!gatech!swrinde!cs.utexas.edu!sun-barr!olivea!gossip.pyramid.com!pyramid!oracle!unrepliable!bounce
From: [e-mail] (David Navas)
Newsgroups: comp.sys.amiga.advocacy
Subject: Re: WOC Toronto: Amiga's Future
Message-ID: <[e-mail]>
Date: 24 Dec 92 17:43:48 GMT
References: <[e-mail]>
Sender: [e-mail] (Oracle News Poster)
Organization: Oracle Corporation, Redwood Shores CA
Lines: 334
Nntp-Posting-Host: mailseq.us.oracle.com
X-Disclaimer: This message was written by an unauthenticated user
              at Oracle Corporation.  The opinions expressed are those
              of the user and not necessarily those of Oracle.

In article <[e-mail]> [e-mail] (Shawn E Switenky) writes:
>Your welcome.

Well, thanks again for replying!

>I'd send you the video tape if I could.  A few hunderd words are much easier
>to send over the net.

Well, we'll all go out and buy NeXTs and send the video tape in our
NeXTMail :) :)

Better yet, post it here in csaa :)
As an MPEG movie :) :)

>that I took down as notes from both seminars and Q&A sessions.

That's cool.

>He only mentioned the 800x600 72Hz mode.  I think that 1024x768x60 falls in
>line with his comment about higher resolutions being available at lower
>refresh rates.

This is a good thing.  I would have bought an A4000, even WITH the IDE, if
only I didn't have to deal with those ridiculous video mode resolutions.
I'm not knocking them -- for other people they're VERY useful.  For me they're
a nuisance, I can't display them....  And, reminder, it was the FIRST
impressions that Dan left with after seeing the aging A1000.  Keep it, but
I don't want it to EVER boot into that mode -- and that includes those
boot menus too....

It would be good to never have to deal with 15kHz ever again....
I hope the OS folks decide to put that option in ROM....

>> >    - 8x Memory Bandwidth Increase

>Although, my notes do not explicitly say, I believe this is relative to
>the current AGA chipset, along with all the other increase references he
>made.  That was the imnpression I had from Lou's presentation.

This is very strange.  Consider that the blitter is only getting a 2x
increase in performance....  Hmm, and I do believe you could get
800x600x24 with 8x memory bandwidth increase....
?

8x is also not "incremental" over AGA.  It is a greater leap than ECS-->
AGA was, even....

>I think this means that, within the specifications of the project
>( ie # of transistors, time to develop, etc ) they can bring this extra
>features.  Maybe the design they had chosen was easier to implement with 24
>bits.

That does make sense.

>> About time.  Will this include DMA as well (like, say, the IIfx?).
>As far as I know, he did not say.

Probably not then.  Not that it's a big overhead, though....

>> >    - increased Chip Ram ( 8Mb )

>The details for that were worked out along time ago, and months ago I saw
>a post from Dave Haynie about where the the increased chip memory goes.
>Maybe he could restate this?

Hmm, I remember musings about putting it in some of the "high" ZIII memory?
Perhaps the 100% backwards compatibility will be restricted to the "shadow'd"
access of the lower 2megs?

>He did say that the new systems were at least a year away.

Sigh, I know.
I wish I cuold get the system part-by-part, though.  I'd have one HECK of
a good time putting it together and such!  :)

>> Pass on the DRAM please.  64bit VRAM seems like a minimum for work designed
>> to be one year in the future.  128bit would have been better....
>> [Hey, I can't afford an SGI, doesn't mean I don't want one :)]

>There might be other people who are willing to take a performane cut
>and still be able to afford a machine with these features.  The good thing
>here is DRAM is an alternative for lower priced machines.

Is high-speed DRAM really that much cheaper than VRAM?  I mean, 8x increase
in memory bandwidth is going to require some -really- -fast- DRAM....

>> Is this really 1280x1024x24?  Or is it simply a note that says -- hey,
>> we have a 24bit palette in our hi-res mode?  The former is impressive, the

>I believe that this means your former statement.

If true, then I will be a VERY happy person.
I will try not to raise my own expectations, as Cmdre (or anyone else for
that matter) doesn't seem to be able to live up to them :)

>Balanced was a word that he used.  My impression was that it was a new DMA
>management scheme that would reduce DMA contention problems.

Someone pointed out that DMA slots get scheduled in a static rather than
dynamic manner.

>Again, I believe these references are compared to the AGA chip set.

20x AGA?  Sheeooot.  AGA is a 64bit, 7Mhz interface.  I -think- you get a 2x
increase just because of the different clock scheduling of VRAM over DRAM
(I am NO hardware expert, I am merely guessing, please feel free to correct).
Now they said this was a 64bit chipset (in some incarnations), so that makes
the bus interface at -70-Mhz?

Wow!  Kick me, I have a hard time believing this, but if it's true, we're
getting a really wicked-fast bus.

[Actually, every bone in my body want to believe this....]

If it isn't true, I hope they at least put some simple spline/matrix xform/
etc. type of options in a new blitter.  That would offset the diff. for
me :)
Personally, 8x AGA would work for me just fine.  5x AGA is, well, okay (it's
800x600x24 60Hz) -- that gets you a 'B-' in my gradebook :)

>What Lou had for a presentation was a slide show of Amiga pictures that
>was projected on to a screen.  All these things are copied, to the best
>of my abilities, right from the presentation slides.  The details you are
>asking for were not mentioned.

Hey, not picking on you or anything, I just have questions.  I didn't say
I had snide remarks :)

Too bad about the missing details though.  Someday we'll know.

>> >    - Video Upgrade module ( You can add more chips for parallel processing
>> >     chip set i.e. multiple blitters, and Higher Resolution Display Modes )

>> Like I really am going to believe this?

>Why not?  You are not a Commodore engineer, so you have no idea what they are
>working on.  Neither do I, but at least I figure that it would not be in their

Right on those three accounts, anyway.

>interest to tell us a lie.  If they are going to tell me things, I am going
>to listen.

"Unix is still supported"?  "We're going to get agressive [marketing] this
year"?

It doesn't have to be a lie, it could be that they are trying
to produce a system that, say, you could add more video -cards- to (like
our current video slot).  And sure, you could add a blitter there too.
I've talked to Ty a lot about this, and still remain unconvinced as to what
that would really give you in terms of performance, though.

>> It actually sounds like Lew is trying to raise market expectations for
>> RISC in the public, forcing Cmdre's upper management to actually commit
>> to such a product.

>Funny, to me it sounds like he is being realistic and preparing for the

Yes, it sounds like that to me AS WELL.  I was merely pointing out that
while the engineers were preparing for the future, upper management was
preparing to play some more golf in the Bahamas.  :) :)

Wow, that would make a great Gary Larson cartoon :)

>just keeping his options open.

Bingo.

>>This is not unheard of, though it's a really crappy way to run a company.
>I think that letting us know about the future is a very good way to run

Yes, I agree with this too.  I also agree that having developers/users in
on the ground floor of what is being developed is a GREAT way to run a
company.

What I'm lamenting is that the engineers aren't representative of what may
or may not be finally shipped from the company.  [Poor sentence, I hope the
meaning is clear, though].]

Or, at least they don't seem to be.  It will be interesting to see what
unfolds.

>> At anyrate, I'm all for RISC, so tell me the dweeb I have to shoot to make
>> this possible :)                                             ^^^^^
>Maybe you should just sit down and take a deep breath instead.

Perhaps.  Or perhaps I've seen the way Cmdre really operates.  Don't get me
wrong, the change in the hardware department is great.  Corresponding changes
in the software, the marketing, and the upper management needs to
happen.  In reverse order of importance.

>> >- There are no plans for virtual memory.
>> Bad, very bad.
>I have the opposite oppinion.  VM sucks.  On other operating systems, when

VM sucks for you.  Therefore, you have the option of running without it.
I hereby grant you, in accordance with the "it's a better world" concept,
the ability to run your machine without VM.

>real solution here is to install more RAM.

Yes, it is.  I can't do this while the machine is running, and until I can,
I want VM....

>This is what he said, and I agree.  The CDTV should be price competitive with
>console game machines.  I do not think there is a home market ( or could be )
>for a higher priced machine.

Well, I don't think there's a home market at all for the thing :)
If they honestly want to cost reduce it, it's going to have to decrease to
just below one of those console-CDROM additions.
The specs aren't really very competitive with the CDI/VIS market from what I
hear.  Of course, those markets don't exist uniformly arond the world, but
even still.  To stay ahead, you have to BE ahead....

In reality, I see the two goals (cost reduce, enhance) as going hand-in-hand.

>Why not, you have judged everthing else.  Seriously, this is a noble objective
>and one they are quite capable of carrying out.

With what cash resources?  How are they going to compete with the Power PC?
I'm not being antagonistic (at least not intentionally), I'm just
trying to figure out how they could achieve such a goal.

>I guess I'm not DSP-ready, since I have no idea what AT&T has for a DSP
>processor line.  I guess I took that down wrong.

I believe there's a 1-800 number you can call.  Err, uh, if you lived in
the states, that is :)  [Does Canada have 1-800 number service too?]

>> Hence, I suppose, the phrase "takes one to know one."
>[RANTING AND RAVING MODE: ON]

I agree with everything you say about NewTek.
And, having said that, Cmdre has been every bit as difficult to work with
from my perspective.  Not the individuals, who have gone OUT of their
way to help.  The organization, once you get past the engineering
section, is chaos.  At least from my perspective....

>[RANTING AND RAVING MODE:OFF]

>The way I understand design cycles is that they have actually produced
>prototypes of the new chips.  Now they find what works and what does not.

Ooh, I'm hoping for a late January "Christmas"  :) :) :)
After which, of course, I won't be allowed to say anything :(
Which is why I want to get it all out of my system NOW, before it's too late :)

>before they can be sold in a machine.

Hmm, probably at least two more.  One to complete an alpha phase (hardware
freeze), then one to fix the bugs.  Probably at least three, though.  Of
course, I'm being silly and applying software design to hardware -- but
at least your info gives me a rough approximation.  Thanks.

>He never said that they had committed to a RISC Amiga.  What he did say was
>that they were now designing with RISC in mind.  They are doing this so that
>when the time comes to make the jump to RISC, they will have an easy time
>of it.  This is long range planning.

Aha!  So some folks around here have jumped to some conclusions about this,
eh?  :)
[Hello, Ty?  :(]

>I think that this is an excellent way to bring the Amiga some percieved
>legitimacy.  Many people declare the Amiga a toy since it does not run DOS

The best way to bring the Amiga some legitimacy is to sell it.  The best way
to sell it is to create some kind of software that:
    a) requires the specific hardware on this "AAA"  "AOOGA" chipset.
    b) can't be done any other way nearly as fast
    c) makes a lot of high-profile people's jobs FAR easier to do.

That assumes, of course, that marketing can take advantage of high-profile
folk using the Amiga.  Up to this point they've proven themselves utterly
incapable of this.  I would consider the video market to be just about as
high-profile as you can get.  More high-profile than, say, newspapers
(desktop publishing).

Ah well, I'll continue to dream.  Until then, I personally have no need for
feeling "legitimate" in any way shape or form.  I won't deny you your wish,
though.

>I'm no big fan of Windows.  It certainly is not any improvement over AmigaDOS
>or UNIX.  But if is sells more Amigas, and enables Commodore to bring out
>bigger, faster, stronger computers that I can afford then I see Windows NT on
>the Amiga as a good thing.

I agree.

>> Bingo -- a plan is good.
>> A plan is long overdue.

>I completely agree.  For years Commodore has kept some pretty tight wraps on
>the things that they do.  While they do have some valid reasons for this
>policy, I really do appreciate a VP from Commodore coming out and saying,
>"We are working to improve.  This is how we are doing it."  I think people

Well, the reasons are rather obvious ;)
If you had told me seven years ago that it would take seven years to get
a chipset that was 4x normal, I would have had second thoughts ;)

>spreading rumors of Commodore's death is far worse than having those people
>talk of the new Amigas CBM is working on.

rumors?  What they're not dead yet?  :^)
I agree with this too.  Which explains my rather long posts on this subject!
[And perhaps slightly explains my OTHER long ranting of this week, 'why
 personal computers have a long way to go to become useful'.]
I'd much rather talk about the future.

>Don't hold your breath for those features.  Commodore has some good reasons to

For memory protection -- there are several places where this could be added
IMMEDIATELY to the OS.  For VM, we have this from two separate companies.
And Cmdre's reason for not supporting it is...?

>do what they are doing.  If holding off on virtual memory means that I can buy
>a more powerful Amiga sooner and for less money, then I'm all for not having
>VM.  I would not use VM anyways.

The software people are not migrating to the hardware section and making
chips.  Many of them, it's true, are busy working on DIG.  Seems to me, though,
that these are (or should be) different people than the people who know how
to bash Exec and the MMU hardware....

>Shawn Switenky

Anyway, thanks, this is much more interesting than MB, death of CDTV, etc. ;)



David C. Navas                        [e-mail]
Working for, but not speaking on behalf of, Oracle Corp.