17 Bit Software 2: Collection B/17Bit_Collection_B.iso/files/1479.dms/1479.adf/Text/JayMinerConference

Text File | 1987-05-28 | 15.9 KB | 386 lines

Copyright 1987 by CompuServe AmigaForum and Jay Miner.

A formal conference with Jay Miner was held on AmigaForum CO channel 2 on
27-MAY-1987.  The topic was open.

The following is a heavily edited transcript of that conference.  Guest
speaker's and moderator's comments are identified with their initials, I.E.,
"JM:".  Forum members are designated with their names inside parenthesis,
I.E., (David Johnson).  Other editing included the unraveling of interleaved
comments, correction of obvious line noise problems, deletion of unrelated
remarks, and paragraph formatting.  Editorial comments added after the fact
(by me) are presented in square brackets.  The original, unedited conference
transcript will be available through the end of May; please contact me if
interested.

Following the transcript is a modified /UST listing of all who actively
participated in the conference.  In addition to the people listed, however,
there were many other forum members in att the floor
is now open for questions... you all know the protocol.

(David Johnson) ?
(Michael M. Merrow) ?
(John Little) ?

RR: David Johnson, you were first. Go ahead.

(David Johnson) Jay you have mentioned that the rgb output on the 2000 is
inferior to the 1000, any comments?

JM: David, no, the RGB output should be identical. The same chips are used,
they may have improved the output analog circuits some, but I doubt it.

RR: David, followup question on the same subject?

(David Johnson) Another question about the extra lines on the AGNUS chip...
will we see any upgrades in the future.  Perhaps like 1024 resolution non-
interlaced or 2 megabyte addressing for chip mem.

JM: Yes Dave, the revised Agnus chip I was working on recently was designed
to work with Video Ram to produce 1024x1024 displays and provide 2 meg chip
address, and larger blitter area moves.  It is just about finished... now
however who knows how long before they design a machine to actually use it.

RR: Jay, would you care to hazard a guess? Next generation? Two generations
from now?  No comment?

JM: My guess is the next generation will use the new Agnus and be at least
18 months more in coming.

RR: Michael Merrow, you were next.

(Michael M. Merrow) Jay, how easy (i.e. cheap, compatible) is it on the 2000
for some third party just to remove interlace.

JM: Mich, there is a way, I saw one working in Los Gatos about 3 months ago.
It was a frame buffer that had a full (640 by 480) non-interlaced display
and it was gorgeous!  It worked off the standard software (hi res) and
should be fairly easy to design.  I don't know what has happened to it but
someone at CBM is reputed to be working on it.

RR: Michael, any followup question on that subject?

(Marlene/Sysop) ?

(Michael M. Merrow) Just a guess please, but how soon might we see it after
the 2000?

JM: Mich, if someone doesn't come out with it in the next year, I will be
surprised.

RR: Jay, was this an external device, expansion buss card, or perhaps for
the video slot?

JM: It was an external device, but required some sync wires and connections
to the digital output before the color output ladder.

RR: How complex was this device; I.E., care to estimate how much one might
retail for?

(Mike Jeskin) ?

JM: The main complexity was the size of the memory buffer, and you can price
that.  The rest would probably fit in a couple of custom chips.

RR: John Little, you were next.

(Fadi Khulusi/MicroP) ?

(John Little) Jay, two questions: do you believe that with the new leaders
at Commodore they are capable of building a workstation around the Amiga now
or in the future; and what about the new long phosphor monitors, are they
worth the buy for CAD work?

JM: I'll talk phospors first.  I had a LP tube on my desk at LG [Los Gatos]
for the last year, and loved it. I've just been waiting for CBM to get the
price down so I can replace my Sony here at home.  The hi res flicker is
gone, and the trails are not bad enough to realy hurt animation.  About CBM
and workstations... CBM has some very good engineers, as does Germany; they
could do it if management will listen, but I doubt that they will... they
didn't listen to LG or the better engineers there.

(John Little) Thank you for your honesty.

(David Johnson) ?
(Vance) ?
(Claude J. Gosch) ?

RR: Maybe the change in management will have a positive effect on that.
Folks, since the queue is starting to fill up, I'm going to disallow
followups after the next questioner.  Marlene, you were next.

(Marlene/Sysop) Hi Jay, could you give us some insight into top management's
view about third party hardware developers and their role in the market
place?

JM: Marlene, I'm sure that by now they appreciate them more than in the
past, but since I haven't met the new CEO yet I can't say for sure.

(Marlene/Sysop) Well, I heard today from the CBM reps here in NY that the
1000+1000 upgrade is a reality... that doesn't sound like 3rd party support to
me?!!!

JM: I havent heard. What is a 1000+1000?

(Marlene/Sysop) The rumored CMB upgrade policy: your A1000 + $1000 gets you a
2000.

JM: Oh I see.  I suppose that would harm hardware developers?  If you have a
view on that I suggest sending it to Irving Gould at CBM Westchester.

RR: Marlene, this looks like an excellent topic for discussion; perhaps if
Jay is willing to stay around a few minutes after the CO we can chat about
it.

(Marlene/Sysop) ok Rick

RR: Mike Jeskin, go ahead.

(Mike Jeskin) Hi Jay!  Having just traveled for the first time in a year I am
wondering if there could be a briefcase Amiga in the works anywhere, and
also is there any chance of seeing some new printer drivers in the future?

JM: That is a good idea, also a real portable, but I haven't heard a thing
from anyone about such plans. I would like one too!!

RR: Jay, how about rack mount for control, scientific, and musical (road)
applications?

JM: Rick, the 1000 was designed with rack mount in mind, it would be easy.

RR: Are you suggesting that perhaps a simple adapter would do the trick?

JM: Yes a simple rack adaptor should do it.

RR: Okay.  I sort of jumped the gun... how about Mike's other question?
(Sorry Mike!)

JM: Mike, no I haven't heard a thing, I will ask one of the software guys
next time I see one, sorry.

RR: Fadi Khulusi, you were next.

(Fadi Khulusi/MicroP) Jay, are you now actively involved in the design
process of future generartion machines or just consulting part time?

JM: Fadi, I am on retainer to CBM as a consultant, and have assisted in the
revised Agnus chip over the last 4 months. That is finished now, and I don't
know what use CBM will be making of me if any.  I am also consulting on the
design of pacemaker chips for a small company called Ventritex here in
Sunnyvale and for a toy company.  Mostly I'm just resting.

(Fadi Khulusi/MicroP) Jay, good luck!

JM: thanks

RR: Jay, do you think at this time that they'll be calling you in soon for
another project?

JM: Somehow I doubt it. I told them lots of things they didn't believe, and
I doubt they will want my "told you so's".

(Ted Haskell) ??
(John/TCR) ??

RR: Hmm, not encouraging.  David Johnson, you were next.

(David Johnson) Jay, it was rumoured that the current agnus packet was
designed with a 2 meg upgraade in mind.  Is this the revision you were
working on?

JM: Depends on what you mean by "current Packet"

(David Johnson) The socket design.

JM: There is a new Agnus in the 500 (or supposed to be) that is called "Fat
Agnus".  That is just the old Agnus with some of the "Glue logic" pulled
inside of the chip package.  The "Super Hi-res Agnus" is the one with
1024x1024 and 2 meg address range, and is the one I helped with.

Vance, you were next.

(Michael M. Merrow) ?
(Claude Gosch) ?

(Vance) Are there any plans to increase the color palette on the next
generation chips and amount of colors displayed on screen without interupts?

JM: The colors could be increased to 64 but this would lock out the
processor completly from chip memory, but only in lo res mode. This however
was not done for reasons of time and chip area.

RR: Claude Gosch, you were next.

(Claude J. Gosch) A quick followup on Fat Agnus & a question.  Is the Fat
Agnus used in the 2000 - will it be used in the future production of the
1000?  Also, is slow hard drive data transfer caused in any way by hardware
limits, or is it software only?

JM: I don't think Fat Agnus is actually used yet in any machine.  And I
believe that the large package will prove uneconomical, and CBM will return
to the regular AGNUS AND a gate array chip that can replace Fat Agnus.  Just
my prediction.

RR: So, what you meant before was that Fat Agnus was _supposed_ to be in the
500?

JM: Yes.  Note, Fat Agnus is not the new super hi res Agnus.  I have no idea
about the hard drive problems mentioned.

RR: Okay. (We will have a future CO on that subject, folks; stay tuned.) Ted
Haskell, you were next.

(David Johnson) ??

(Ted Haskell) I hate asking questions like this but...
[1] Considering our throw-away and turn-key society do you feel that Amiga
owners and vendors will always be a minority sort of on their own?  And...
[2] What did you expect the user environment to be when the A1000 was being
developed?

JM: Ted, I'm afraid that CBM just can't out-advertise Apple and IBM now, if
they ever could (not just because of budget), and this dooms us to the fate
you suggest.  I have been consistently appalled at the ads and now I haven't
seen an ad since last summer.  On the user environment, I expected people to
be developing on the Amiga itself, instead of on a SUN or something, but the
software guys insisted on Suns.

RR: John at TCR, you were next.

(John/TCR) Jay, You *made* the Amiga chip set (as well as others); can you
give us an idea of what goes into designing a custom chipset for a computer
like the Amiga.  What is your educational background and emphasis??

JM: John, I got my BS (bull____) from UC berkeley in '58, and started
designing chips in '64, after first doing servo motors, transformers, and
exploding bolts.  Then in 1966 I started doing chips with GME... and I have
done chips for almost every semiconductor company in the valley.  I did
calculator chips until the 6502 came out.  Then I did video chips for Atari.
Then I did computer chips for pacemakers for Zymos.  Then in 82 we started
Amiga (whew), and I guess you know the rest.  About designing chips like
these, the active participation of good software guys and hardware guys is
what does the trick.  Also, running a good brainstorming session is not
easy; it is a fine line between fighting and brainstorming, and it must be
pushed just the right amount.  Also attention to detail, and most important
is to shoot high enough; thats the biggest problem, setting your sights high
enough.

(John/TCR) Thanks Jay!  (Whew, busy guy!!)

RR: Jay, a point worth highlighting...  it seems that every engineer and his
brother are now doing FLPA and ASIC design.  But what you are talking about
here is often radically different, and you've been doing it long before the
acronym "ASIC" was even used.  Correct, or all wet?

JM: Yes Rick, it doesn't matter really how you build the chips...  Gate
Array, standard cell, etc.  What counts is the specifications and
instructions, and designing them so the user (read programmer) can
understand them.

RR: Okay, I just wanted to point out that, although everyone is "designing
custom chips" these days, your phrase "I did calculator chips UNTIL the
6502" goes back quite a ways and indicates a great deal of experience in the
field.  Michael Merrow, you were next.

(Michael Merrow) Perhaps this is too broad a question, and it may not have
an answer, but there have been recent rumors about software compatibility
problems with the new machines.  If these rumors are true, can you give us
some insight as to what may be creating the problems?

JM: I haven't heard of these problems, so I'm no help.  Sorry.

RR: Claude Gosch, you are next.

(Claude J. Gosch) Jay, what do you see for the future of the 1000...
obsolescence or redesign for reduced cost?

JM: Claude, the cost reduced 1000 was the bigest battle I fought (and lost)
at Commodore.  It still hurts.  I hope they see the error of their ways
soon, but I doubt it.  They put a godawful lot of money into the 500 and
2000, and could have had a cost reduced 1000 to sell for $30 more than the
500 with 1/10 the budget, and shown that, CBM could support a machine for
more than one year.  It makes me sad.

RR: David Johnson, you were next.

(David Johnson) Jay, would you care to comment about the differences between
the Westchester and the German designs of the 2000.  Are there any advantages
of one over the other.

JM: Dave, Westchester didn't have a 2000, Los Gatos had the 2000 that was
competing with Germany.  It had Zorro style Amiga expansion slots only, but
we could squeeze in 2 IBM slots with a revision in time for Christmas if
they wanted.  They said this wasn't enough IBM slots, and Germany would
build it by Christmas. We all know what happened to that (last estimate
July?).  The main difference (other than more slots) was the shape of the
slots.  The LG engineers felt that IBM slots would be hard to shield and
hard to cool, and wanted CBM to support the Zorro card shape. CBM picked the
German 2000 instead.  And that's a quick summary.

(David Johnson) Thanks Jay !

RR: Okay folks, it's time to wind this down.  I want to thank Jay
for taking the time to be with us here tonight and as always, thanks to you
members for putting up with your fumble-fingered moderator!  Jay, if you
could stay around a few minutes (no obligation, though), I'm sure our members
would enjoy chatting with you.

JM: Sure

Okay, folks... <banging gavel>... it's all yours!

[At this point the formal conference ended, but Jay and several members
stayed around and chatted for p 8  t"*( t 8( kut*((*"8*"t0& t" " &0)^ ,2 &$  !t   " 8**"t0*!t& $(8 "j__|0"8!t0("  +xtb(t<0"8*"t$&*6( ("t0!t* 't"8  *t$&*  2 8*+t"   8*u problem.  I
didn't know if you had ever seen this before.

RR: Jay, do you feel that _any_ package has yet come near using Ami's
full audio capabilities?  (I have an opinion, I'm curious about yours)

JM: Rick, I was very impressed when Bob Hoover [Mimetics] came over and did
a SoundScape demo.  I have a CZ230S and it works great.

RR: Do you feel then that SoundScape fully utilizes the Amiga's sound
capabilities?

JM: Rick, can't think of anything it doesn't use.

RR: Understand, I don't think there's anything wrong with SoundScape Pro
MIDI; it's one of the most versatile packages out there, and I use it
myself... but I don't think it stretches the Amiga as far as sound
generation is concerned.  Nobody has gone beyond the basic "what it says in
the Amiga manuals" approach yet.  I think the Amiga is capable of quite a
bit more, and it'll just take time.

(Vance) Are there any plans to make the other chips you developed more
powerful, I.E. a palette of 256 luminences of red blue green and 8 bit
planes?

JM: I have some ideas, but CBM hasn't expressed any interest yet.  They do
have some good engineers, and I'm sure they also have lots of good ideas.
Generally higher resolution means fewer colors however, or more cost.

(Scott S.) Jay,  What about the release date of the 2000?

JM: I can't be sure, but I feel they are having tough FCC noise problems,
and we will see delay after delay.  Just guessing.

(Scott S.) Shouldn't sell my 1000 just yet eh?

JM: No, it might even appreciate.

----------------------------------------------------------------------------

User  User ID       Nod  Rm   Name / Forum Area
----  ------------  ---  ---  -----------------
 17   76703,4253    RIC   2   Rick/SYSOP
 39   76515,3476    DNC   2   Scott S.
 44   76515,3055    EVI   2   Dave Ramsden
 51   73577,123     BOO   2   Ted Haskell
 75   73250,12      HAR   2   Michael M. Merrow
 81   73200,3553    DET   2   David Kieltyka
 82   73557,1223    TTO   2   David Johnson
 88   76046,3160    SMO   2   Jay G. Miner
 92   73017,3224    TTO   2   Peter Hodgins
 94   76406,25      WPK  18   Marlene/Sysop
110   72447,2161    NYY   2   Vance