🚀 go-pugleaf

In article <1992Dec16.183721.17939@netcom.com> keving@netcom.com (Kevin Gwarada) writes:

>   I've used Including my Amiga, and IBM.  I heard somewhere that there is an
>   excellerated 6502 for industrial purposes (somewhere around 10 Mhz)-- If
>   its true, for fun, I'd seriously concider trying to implement it in an old
	There is a product called the Zip chip, made by Zip
Technologies.. I don't have their # handy, but they have accelerated
65c02's that you can swap your regular 65c02 with..(Note: this is made
specifically for the II's, so it will come with instructions and all)
>   apple (I realize what this implicates); but it would be great to see it
>   functioning.  Will we ever return to the days of mass-produced compact code??
	Umm.. Yeah, I suppose those of you off in the land of the I*M
don't realize this, but Mass-produced compact code still exists...
Ohh.. so do Apple II programmers.. On a side note, compact code would
be my explaination why all my AppleIIgs programs run as fast (from a
general users standpoint) as equivalent pc clone programs run on clone
machines that have a mhz rating 5 times as high as my GS.. Of course
the pipelining can take some credit.

	Summary: It's hard to return to those days when they are still
here..... :)
>   -K
			Richard..

--
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=++=+==+=+=+=+=+=+=+=+=+=+
|Richard Wifall - rwifall@nmsu.edu - IRCnick: TheJazz - Member GS<>IRC        |
|-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-/-\-|
|Ask me about DeskTracker, BorderBlinker, and a demo that Dave hasn't finished|
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+

In article <RWIFALL.92Dec16131603@dante.nmsu.edu> rwifall@nmsu.edu (Richard J Wifall) writes:
	   Umm.. Yeah, I suppose those of you off in the land of the I*M
   don't realize this, but Mass-produced compact code still exists...
   Ohh.. so do Apple II programmers.. On a side note, compact code would
   be my explaination why all my AppleIIgs programs run as fast (from a
   general users standpoint) as equivalent pc clone programs run on clone
   machines that have a mhz rating 5 times as high as my GS.. Of course
   the pipelining can take some credit.

At least part of the reason is 65816 instructions on average take fewer
cycles than 8086 instructions.  It could be argued that the 65816
instructions get less done, but for tightly coded stuff the
instructions are about equal.  

There's a nice version of this seeming paradox for microcontrollers.
A 2MHz 68HC11 is roughly the same speed as a 16MHz Z8.

Jay Carlson 
nop@theory.cs.mankato.msus.edu

In article <NOP.92Dec18034213@theory.Mankato.MSUS.EDU> nop@theory.cs.mankato.msus.edu (misinstalled gnus and now has to pay for it) writes:
>In article <RWIFALL.92Dec16131603@dante.nmsu.edu> rwifall@nmsu.edu (Richard J Wifall) writes:
>	   Umm.. Yeah, I suppose those of you off in the land of the I*M
>   don't realize this, but Mass-produced compact code still exists...
>   Ohh.. so do Apple II programmers.. On a side note, compact code would
>   be my explaination why all my AppleIIgs programs run as fast (from a
>   general users standpoint) as equivalent pc clone programs run on clone
>   machines that have a mhz rating 5 times as high as my GS.. Of course
>   the pipelining can take some credit.
>
>At least part of the reason is 65816 instructions on average take fewer
>cycles than 8086 instructions.  It could be argued that the 65816
>instructions get less done, but for tightly coded stuff the
>instructions are about equal.  
>
>There's a nice version of this seeming paradox for microcontrollers.
>A 2MHz 68HC11 is roughly the same speed as a 16MHz Z8.
>

I hope this isn't getting too off topic, but I suppose in general you're
right.  A 6502 may take less cyles to do an add for instance, than an
equivalanet 8086 or V20.  But as Intel progresses, they get better at 
reducing the time for clock cycles.  A '286 was faster yet, a 386 could
do an instruction in 2 cycles, and a '486 could do 'em in 1 cycle.  Meanwhile
the Pentium is capable of fetching one instruction while simultaneously
decoding yet another, while also executing a third.  A 66MHz Pentium is
rated (maximum) to be 100Mips.

I don't claim to know about how many cyles and what not a 6502 and
the Motorola 68xxx series is capable of, and if anyone knows, please tell me.

The difference of course relates to the two main theories of microcode 
optimization.  CISC (complex instruction set) vs. RISC (reduced instruction set)meaning CISC has more complex instructions that take longer, whereas RISC has
less instructions that perform faster.  Of course, you'd have to use more RISC
instructions to perform an equivalent CISC routine, but you get the point.

MPB

In article <1992Dec16.183721.17939@netcom.com> Kevin Gwarada,
keving@netcom.com writes:
>I believe its great to revive the apple (so to speak) on my IBM.  I believe
>I've had more fun with that machine than any other (mostly because all the
>concepts are new, everything now is just based on what was done before)
computer
>I've used Including my Amiga, and IBM.  I heard somewhere that there is an
>excellerated 6502 for industrial purposes (somewhere around 10 Mhz)-- If

Besides high speed 6502s (rumor from years ago is that it's the only
CPU certified to be implanted into the human body), Sanyo re-engineered
the 65816 to 14 MHz, with normal power specifications.

There were a couple college kids who a few years ago tried to redesign the
65816
and from their calculations (I don't know how fast samples got.. I had heard
around
13-15 MHz), they could have gotten up to 20-25 MHz. Hmm.. wonder if they ever
got
in touch with Sanyo, since they seem interested in using the 65816 for a PDA.