SNES Co-processor add-on
Daniel Carleton
I was in Software Etc. and heard from one of the clerks that a co-
processing unit was to be released soon for around $80. He said it would
eliminate slowdown on old games and enable much higher performance in new
ones. It is said to operate at 33Mhz. Does anyone have more information
about this device?
sea...@ac.dal.ca
be possible.
Grover Thomas
>$80 seems kind of cheap for something running at 33MHz. I doubt that this would
>be possible.
Not really, when you consider that it's 6502 technology (or should I say
65816). I wonder if it really operates at 33 MHz or just gives "33 MHz
performance."
--
__ __
(__) Grover Thomas InterNet: gro...@emunix.emich.edu (__©
(__) "[I support efforts] to limit the terms of members of Congress, esp- (__)
(__) ecially members of the House and members of the Senate" --Dan Quayle (__)
00d...@bsu-ucs.uucp
>> Date: 4 Oct 92 18:30:10 GMT
>> Organization: U.S. Army Topographic Engineering Center, Ft. Belvoir, VA.
>> Keywords: SNES graphics slowdown
>> Summary: Nintendo rumored to be releasing super chip add-on.
car...@tecsun1.stl.army.mil writes:
Since the SNES already runs at a slow speed 6 - 12 MHz (I don't remember the
specifications exactly), changing the speed dramatically would greatly affect
the existing games not set up for the faster speed and thus reducing the
playability.
Just because this person works at a software store doesn't necessarily mean
that they know what they are talking about. Remember they are salesmen not
hardware specialists.
Just my $0.02
=======================================
= Spyder Spyder Burning Bright, In the =
= Forest of the Night, What Immortal Hand or =
= Eye Could Frame Thy Fearful.....Symmetry? =
= -- Kraven's Last Hunt -- =
===========================================================================
>> Dave Bye | VAX <<
>> R.R. # 4 Box 202 | 00d...@bsu-ucs.bsu.edu <<
>> Portland IN 47371 | UNIX <<
>> (219) 726-2927 | da...@bsu-cs.bsu.edu <<
===========================================================================
= To the Winner goes the spoils..and sometimes a Genesis. =
= It's a Dog eat Dog world and i'm wearing Milkbone =
= Underware... -- Norm "Cheers" -- =
= What do you mean we're lost??? =
======================================
James Hague
>
>Not really, when you consider that it's 6502 technology (or should I say
>65816). I wonder if it really operates at 33 MHz or just gives "33 MHz
>performance."
Something to consider is that a 33MHz 65816 does not exist. I think
the fastest it can go is 14MHz. A 33MHz 65816 would be a *very*
formidable processor. That chip gets things done with very few
cycles.
--
James Hague
exu...@exu.ericsson.se
Gary Snow
>>> From: carl...@tecsun1.etl.army.mil (Daniel Carleton)
>>> Date: 4 Oct 92 18:30:10 GMT
>>> Organization: U.S. Army Topographic Engineering Center, Ft. Belvoir, VA.
>>> Keywords: SNES graphics slowdown
>>> Summary: Nintendo rumored to be releasing super chip add-on.
>
>
>car...@tecsun1.stl.army.mil writes:
>
>>> I was in Software Etc. and heard from one of the clerks that a co-
>>> processing unit was to be released soon for around $80. He said it would
>>> eliminate slowdown on old games and enable much higher performance in new
>>> ones. It is said to operate at 33Mhz. Does anyone have more information
>>> about this device?
>
>
>Since the SNES already runs at a slow speed 6 - 12 MHz (I don't remember the
>specifications exactly), changing the speed dramatically would greatly affect
>the existing games not set up for the faster speed and thus reducing the
>playability.
Not really. You see, its the same with computers, most games and such just
poll a timing sync register for their timing. Speeding up the computer just
gives you quicker screen redraws, and smoother/faster animation. It wouldn't
necessarily speed up current games beyond playability.
Also, have any of you considered that the co-processor chip might not be
a 65816 (or even from the 6502 family). That could explain the 33mhz
speed rating of the chip. Also, it might not even be a CPU chip, it could
just be an FPU chip.
Just my $10's worth.....;-)
Gary
--
-----
Gary Snow
uunet!clark!gsnow or gs...@clark.edu
Ana M Rodriguez
carl...@tecsun1.etl.army.mil (Daniel Carleton) writes:
I believe we are talking about a chip that they are adding to the
software to boost the graphics processing power of the SNES. It will
appear in new titles coming in 93. Of course titles using this technology
will cost an additional amount for the added processor.
George Francis McBay
>Also, have any of you considered that the co-processor chip might not be
>a 65816 (or even from the 6502 family). That could explain the 33mhz
>speed rating of the chip. Also, it might not even be a CPU chip, it could
>just be an FPU chip.
Maybe I'm ignorant, but I just cannot see how adding a fast Floating
Point coprocessor would help game software on the SNES much. Do SNES games
make use of lots of floating point numbers?
Just wondering..
Gary Snow
>
>Maybe I'm ignorant, but I just cannot see how adding a fast Floating
>Point coprocessor would help game software on the SNES much. Do SNES games
>make use of lots of floating point numbers?
Speeding up floating point operations will invariably speed up graphics
handling......but I guess this is all for not, because I read that this new
chip is actually a 32bit RISC graphics accelerator to be added to cartridges
for very little money. Expect to see some of the first of these carts from
Nintendo this Christmas.
Doug Quarnstrom
>
> Since the SNES already runs at a slow speed 6 - 12 MHz (I don't remember the
> specifications exactly), changing the speed dramatically would greatly affect
> the existing games not set up for the faster speed and thus reducing the
> playability.
This depends entirely on whether this was planned by Nintendo.
Assume that they tell developers beforehand that they are going to
provide a collision detection coprocessor in the future. Maybe
it is interrupt driven and they provided a reference spec for it.
This would allow game companies to design for it, and would allow
Nintendo to gouge some extra bucks out of people who want the machine
to be faster. It almost makes sense, since it would allow them to
sell the basic unit cheaper (cheaper processor), and entry price
is pretty critical in this market. Once they have the coprocessor
interface set, they can introduce new products to get more money
from the consumer without driving up the base price. It MAY
be true. I hope so, because then I would be confident about the
potential future of the platform.
doug
Doug Quarnstrom
First, I guess I should point out that I am not sure that the
architecture I suggest above would be that useful for ganeral
video games. Not being a game programmer, I am unable to
know.
Second, it occurs to me that I DID read somewhere in a magazine
recently that Nintendo IS planning to introduce an accelerator
for 3D graphics so that they can do 3D games. This is much
different, because games would have to be written to use it,
and it would probably have no effect on games in existence.
doug
Adrian Havill
>Speeding up floating point operations will invariably speed up graphics
>handling......but I guess this is all for not, because I read that this new
>chip is actually a 32bit RISC graphics accelerator to be added to cartridges
>for very little money. Expect to see some of the first of these carts from
>Nintendo this Christmas.
True, but usually only for geometric type graphics: i.e. AutoCAD and the
like. The bitmap/sprite architecture that video games use seldom use floating
point. Neither the 68000 or the 68516 (whatever that overgrown 6502 is
called ;) ) have floating point capability in microcode. The stuff that DOES
require trig and the like (circular motion, rotations, etc) are usually left
to the on-board video/sprite/display coprocessor or are stored as
pre-compiled coordinate constants in an array.
There are always exceptions, of course... but even the big bad NEO GEO with a
12.5 Mhz 68000 would be prone to slowdown if it tried to move 380 sprites in
circular patterns via trig! (And it IS prone to slowdown-- try playing Blue's
Journey with two simultaneous players; most of the slowdown, like the early
SNES games--are due to bad programming rather than system limitations)
--
****************************** ----- | `` | | ------/ /
* Picard/Riker for President * | | | | | / \ /
* Vote Federation Party: * | ------- |\ | | | /
* The choice for the future. * | | \ / | /