Processor Speed upgrade (accelerator)

[Rubywand <rubywand@swbell.net>]

Below are responses on the "Processor Speed upgrade" topic on
Sandy's Upgrade the GS site ... 





      I've heard some things about an ARM emulator. The ARM is a RISC
processor that runs at HIGH speeds and is very power efficient.

      Lets investigate the ARM idea!

-----


      I'm not into ARM ... but maybe something like that. Basically,
what you could do is:

      - Get a fast (100MHz+) processor
      - Get some on-board memory
      - Emulate a 65SC816. Not a GS, mind you, just the bare CPU.

      Sounds silly at first ... but it's actually not a bad concept. The
65SC816 on the accelerator has to go through an FPGA (nowadays, at least
... the ZIP used ASIC, but now speeds of programmable devices are a lot
better) anyway, so why not use a "pretend" CPU. Probably cheaper, and
faster.

      The difficulties I see are:

      - Making a stable board, speed-wise (100MHz parts are not be
trifled with)

      - Getting the whole GS-to-accelerator interface sussed out (look
how long it took the ZIP guys to come up with their board)

      - And, of course, restricting the features. There's a certain
appeal to sticking a video-card on too ... and how about an expansion
port? And while we're at it ... you see where this leads.

      So there. Now I've posted something, too :)


      Zonk 

-----


      I say stick with a genuine 65816. Although, not the fastest chip
in the world, it is much easier to design and build for 14MHz then it is
for 100MHz.

      Perhaps a proof-of-concept could be achieved, by using descrete
logic at first with 2.8 MHz and slowly raising the speed until we reach
the target of 14MHz (with an option of overclocking at 16).

      cfs 

      Charles Stephens 

-----


So, what processor should we use? A 65816 seems ridiculous (TO **ME**)
due to the fact that it is outdated technology, and extrememly slow
compared to todays standards. I think we should use a very high speed
chip which will emulate a 65816 through software. I am not talking about
emulation of the
      _entire_ IIGS....ONLY emulate the processor.

      One thing I do not want to do is deal with the 1Mhz bus speed on
the IIgs. Any accelerator produced *today* should include ONBOARD RAM.

      What processors could we use? 
      Is there 65816 emulator already coded for that Processor?

      ARM---What speeds can you get these chips? Are there any 65816
emulators available for them? How much do they cost? What would the
highest emulated 65816 speed be in MHz?

      Pentium or 80X86 chip--Speeds up to 233Mhz. Would it be possible
to buy SMALL quantites of these chipse? 65816 emulators have ALREADY
BEEN CREATED. We could interface the chip to some RAM, write the code
that interfaces the Accelerator to the GS.

      Macintosh Processor chips: I don't know what chips the PowerMac
use, but would a 68k processor be a feasible processor? How fast do they
run? Can you get them in small quantities? As far as I know, there is a
65816 emulator being created for the 68k processor.


      What other processors could we use? A Z-80 chip perhaps? I don't
know if these chips were ever made to run at high speeds.


      Are there OLD computers/Electronic products that good processors
can be scavenged out of?

-----

      It is much simpler, cheaper and easier to design something that
uses 65816 than having to work with a faster processor that a) eats
power and b) requires significant emulation effort.

      The I/O bus has nothing to do with memory access. You could couple
the memory with the CPU and have a PGA to controll access to either on
the accel RAM or the on board system RAM. So programs will really fly.

      65816 would offer 100% compatability over anything an emulator
could produce.

      68K's are cheap, but overkill, same with ARM (and ARMs are more
expensive).

      Zilog did make several sucessors to the Z-80 that are quite faster
(Z-180, 280, 380). But we are talking about a different universe
altogether. 

      Charles Stephens 

-----

Some observations on the suggested accelerator design:

      > Accelerator:
      > 
      > A IIgs with a Transwarp installed is faster than stock, but it
is 
      > still slow. The Zip and Transwarp boards have lots of chips to 
      > interface the cards with the 1mhz bus speed. 
      > 
      > It would be difficult to interface a card with the slow bus, and 
      > the resulting acceleration isn't that great, and over a certain
point
      > (about 12 Mhz??), there is **NO** speed increase, because the
bus
      > just cannot transmit data that fast. I boosted my Transwarp from 
      > 7Mhz to 12.5Mhz, and the speed increase was noticeable, but I
want
      > more!
      > 
      > 
      > Thats why I think any NEW accelerator made TODAY should have the 
      > following:
      > 
      > Fast 65816 emulator (40Mhz for the IIgs would *FLY*)

      Even using some of the fastest processors around (which are VERY
EXPENSIVE) you'll be lucky to achieve 40 MHz emulation of the 65816. A
400 MHz processor might get you close, though I suspect you'd be nearer
to 30 MHz than 40, and the processor itself will cost hundreds of
dollars. I doubt such an approach has much of a chance at being
commercially viable.

      Using a standard 65816 at 20 MHz is probably about the best you'll
do at this point in time... you might push it a little higher, but it
won't be easy.

      When I met with Bill Mensch in February of 1997, he indicated at
that time that another die shrink (for the 65816) may not be far off,
and that the processor should be capable of 40 MHz at that point.

      However, our own experience indicates that getting other
components to work with the processor at that speed is by no means
simple, and may very well be near impossible. The speed of most RAM
chips, for example, is typically rated at 2 vdc. The 65816's
architecture calls for much higher voltages to establish a 'high'
condition, and we find that the speed of RAM chips falls off
dramatically above 2 volts.

      If a lot of components are attached to the bus lines, the added
capacitance adversely affects performance even more. We had great
difficulty getting even 20 MHz operation reliably with these factors in
our C64/128 accelerator design. I doubt very much that the speed of any
accelerator using the 65816 will go much beyond the 20 MHz mark.


      > RAM on card
      > ROM is copied into Accelerator RAM
      > Interface to the 1Mhz bus to access Slots.
      > 
      > This has the following advantages:
      > 
      > You don't have to access the RAM through the slow bus. 
      > ROM does not have to be accessed through the slow bus either.
      > All existing peripheral cards can be used (although accessed at
1Mhz)

      These things will get you fairly close to true 20 MHz performance.
Our own C-64/128 accelorator is based on this type of operation.


      > This creates the following problems:
      > 
      > Video RAM is on Card. The IIgs motherboard cannot display the
      > video, due
      > to the fact the Accelerator is running at high speeds. 
      > A 65816 emulator has to be coded for the Fast processor on the
      > Accelerator.

      Some problems need not occur. Again, drawing on our C-64/128
accelerator design, the video RAM is merely shadowed into the computer.
Reading from RAM always comes from the high-speed RAM on the
accelerator, while writing goes to both the accelerator RAM and the host
computer RAM.

      If video RAM is in a fixed location, the range can be detected in
real-time by the hardware to determine if writing through to the
computer RAM is necessary. Such a 'write-through' can further be
enhanced by cacheing the byte to be written through, allowing the
processor itself to continue to process the next instruction without
waiting until synchonization with the host bus occurs.

      While you can only squeeze one byte through into the host in a
single cycle at the host's maximum speed, this is often ample to avoid
slowdowns for all but the tightest loops writing to video RAM.

      I/O needs to be handled somewhat differently... both reads and
writes need to be limited to the speed of the host I/O. However, since
there is often much other processing time involved in working with the
data to be read or written, again you'll tend to get pretty good
performance for such operations.

      Doug Cotton
      Creative Micro Designs, Inc.



  
Rubywand