Is microprocessor an integrated circuit???

[Rich Grise]

On Wed, 26 Jan 2005 18:47:32 -0500, Spehro Pefhany wrote:

On Wed, 26 Jan 2005 23:11:23 GMT, the renowned "Bradley1234"
someone@yahoo.com> wrote:

Can you show links? drawings/pdfs or ?

http://www.findarticles.com/p/articles/mi_m0EIN/is_1999_April_26/ai_54477202

OK, who knows how to read a die? This should settle it! (NOT!)
http://content.scu.edu/cgi-bin/viewer.exe?CISOROOT=/svhocdm&CISOPTR=632&CISORESTMP=&CISOVIEWTMP=

That's an 8008, and actually, it does look a lot like a micromemory array
in that corner.

So, it apparently is microcoded. Silly me!

Thank
Rich

 

[MikeMandaville]

Bradley1234 wrote:

Some microprocessors are made with discrete components, but most today
are integrated circuits, meaning the components have been built onto

one circuit

Microprocessor means: micro programmed processor such that
instructions are processed by referencing internal memory locations and

there is a processor within a processor.

Most commercial microprocessors are general purpose, meaning the
instruction set gives developers methods to move/control data, but some

uPs are specific and have a limited/specific instruction set

The stored program computer has been standard since the Edvac, but the
advantage of the Edvac was one of programming speed, not physical size.
The word "microprocessor" entered the language when Intel introduced
the 4004 in 1971. That the word "micro" means "small" demonstrates
that the term "microprocessor" was never meant to be a retroactice one.
After all, to refer to a room-sized computer as being "small" would
seem, well, rather humorous, one would think, in this age of the
personal computer.

 

[Andrew Holme]

Rich Grise wrote:

On Wed, 26 Jan 2005 23:20:46 +0000, Ken Smith wrote:

In article <KlTJd.90$lg5.2@trnddc06>, Bradley1234
someone@yahoo.com> wrote: [...]
This is simply not accurate. Microprocessors typically operate with
instruction and data fetch modes. When an instruction fetch takes
place, the op code causes a branch to an internal memory of the
internal cpu state

Specifically not true in the 4004, 4040, CD1802, or Z80.

In the 8008, the basic machine language was almost microcode. The top
two bits selected the function, the next three bits selected an
address register, and the next three bits selected the other address
register.

The instruction set of the Novix NC4016 RISC processor, which executes FORTH
primitives in a single clock cycle, has been dubbed "external microcode"
because it is so low-level: it resembles microcode, and even permits a
degree of parallel execution to be programmed. The NC4016 is not a
microcoded processor, however, because it has no microprogram counter and no
microcode ROM.

 

[Keith Williams]

In article <slrncvfvpv.jqf.The-Central-
Scrutinizer@linux.client.comcast.net>, The-Central-
Scrutinizer@p.o.b.o.x.com says...

On Wed, 26 Jan 2005 18:59:32 GMT, Bradley1234 <someone@yahoo.com> wrote:

"TCS" <The-Central-Scrutinizer@p.o.b.o.x.com> wrote in message
news:slrncvfmeu.jqf.The-Central-Scrutinizer@linux.client.comcast.net...
On Wed, 26 Jan 2005 16:47:31 GMT, Bradley1234 <someone@yahoo.com> wrote:
Answer: it can be. Some microprocessors are made with discrete
components,
but most today are integrated circuits, meaning the components have been
built onto one circuit

Microprocessor means: micro programmed processor such that instructions
are
processed by referencing internal memory locations and there is a
processor
within a processor.

No it doesn't. It means micro-sized processor. What you're describing
also
includes the vax-11/780 whose processor board was hardly a microprocessor.


Yes, the 11/780 cpu was a microprocessor. Who on earth came up with micro
sized processor? Why not micro stereotypical processor?


BULLSHIT. The 11/780 was a minicomputer.

PDP/11s were minicomputers. VAXen were larger than a breadbox. At the
time many called them Midi-computers. Some weird ones even called them
mainframes.

--
Keith

 

[microx]

Hmm.. this is interesting.

I found this at:
http://www.microcomputerhistory.com/f14patterson.htm

(The same Dave Patterson of the book I quoted above).

<QUOTE corrected_spelling>
The word "microprocessor" was used before the 4004 to mean the
processor in a microprogrammed computer by some people of the time, so
it's not shocking to see the word microprocessor in a document of that
era. I kind of used the term ambiguously, but it was a single chip CPU.
</QUOTE>

So it seems the word can take both meanings, depending on what era you
are talking about.

 

[Keith Williams]

In article <ct95lr$kc6$1$830fa79d@news.demon.co.uk>, andrew@nospam.com
says...

Bradley1234 wrote:
PLEASE show an example of a microprocessor that doesnt use microcode

Broadly speaking, it depends whether the microprocessor is RISC or CISC:
RISC (Reduced Instruction Set) processors don't use microcode - they have
"hard-wired" logic; CISC (Complex Instruction Set) processors generally do
have microcode.

This is certainly *not* true. The issue of microcode is orthogonal to
the term "RISC". Early RISC processors typically weren't microcoded,
but "RISC" and "CISC" have grown towards each other.

The exact definitions of RISC and CISC are somewhat
contentious. The 6502, which is often said to be the first RISC processor,
was not microcoded.

The first "RISC" processor I know of that was given the title was the
IBM 801, which never saw the light of day (at least as a CPU).

--
Keith

 

[Keith Williams]

In article <ct9g52$89h$1@theodyn.ncf.ca>, et472@FreeNet.Carleton.CA
says...

"Andrew Holme" (andrew@nospam.com) writes:
Bradley1234 wrote:
PLEASE show an example of a microprocessor that doesnt use microcode

Broadly speaking, it depends whether the microprocessor is RISC or CISC:
RISC (Reduced Instruction Set) processors don't use microcode - they have
"hard-wired" logic; CISC (Complex Instruction Set) processors generally do
have microcode. The exact definitions of RISC and CISC are somewhat
contentious. The 6502, which is often said to be the first RISC processor,
was not microcoded.

The 6502 was never RISC.

It's only after RISC came along that some people thought it just meant
"few instructions" and then decided the 6502 fit that definition.

Right, RISC == "Reduced Instruction Set Complexity", not "Reduced
Instruction Set Computer". There is a difference.

--
Keith

 

[Keith Williams]

In article <ed4gv0df61qb52cvopcsh982fci0au04rb@4ax.com>,
speffSNIP@interlogDOTyou.knowwhat says...

On Wed, 26 Jan 2005 21:30:02 GMT, the renowned "Bradley1234"
someone@yahoo.com> wrote:

You can post that it is a can of grapes, but that doesn't make it so.

The vax-11/780 was not a microprocessor. Using microcode does not make a
microprocessor.

Well okay, if you say the definition is micro sized processor, then it
depends on your definition of what the word IS is.

Its a definition that I thought wasnt challenged, micro programmed processor




There are many examples of microprocessors that didn't use microcode.

Hey cool, Im going to learn something new, I like to learn. Even though we
have an arbitrary definition going, where microprocessor might also mean
"purple monkey dishwasher"

PLEASE show an example of a microprocessor that doesnt use microcode

If Im wrong, Ill take back what I said

I know there are "processors" that are mechanical, but strictly a
microprocessor

Scenix (now Ubicom) for one. Microchip sued them and Micon Design
Technology in Munich District Court, claiming copyright infringement
of their microcode. Their defense was that they didn't have any.

And a startup company named TeraGen apparently made a version of the
venerable 8051 without microcode.

And, apparently, the Alpha.

But I think most microprocessor cores do use microcode. CISC ones
certainly do, RISC ones may not.

It depends. The RISC/CISC issue is orthogonal to microcode/hard-wired.
I can name processors in all four quadrants. Indeed they have in this
thread.

--
Keith

 

[TCS]

On Thu, 27 Jan 2005 08:43:08 -0500, Keith Williams <krw@att.bizzzz> wrote:

In article <slrncvfvpv.jqf.The-Central-
Scrutinizer@linux.client.comcast.net>, The-Central-
Scrutinizer@p.o.b.o.x.com says...
BULLSHIT. The 11/780 was a minicomputer.

PDP/11s were minicomputers. VAXen were larger than a breadbox. At the
time many called them Midi-computers. Some weird ones even called them
mainframes.

VAXexn were frequently called "superminis". Nobody in their right mind
called them microprocessors until they were integrated to a small chipset
in the form of the microvax.

 

[Keith Williams]

In article <1106835288.952612.300580@f14g2000cwb.googlegroups.com>,
microx@ieee.org says...

Hmm.. this is interesting.

I found this at:
http://www.microcomputerhistory.com/f14patterson.htm

(The same Dave Patterson of the book I quoted above).

QUOTE corrected_spelling
The word "microprocessor" was used before the 4004 to mean the
processor in a microprogrammed computer by some people of the time, so
it's not shocking to see the word microprocessor in a document of that
era. I kind of used the term ambiguously, but it was a single chip CPU.
/QUOTE

So it seems the word can take both meanings, depending on what era you
are talking about.

I've been around processors (all flavors) for over 30 years and have
never heard a micro-programmed processor called a microprocessor. When
it mattered they were called microcoded, but never microprocessors.

--
Keith

 

[microx]

Well, no offense, but 2005 - 30 = 1975 > 1972 (the year of the 4004
processor), so maybe you weren't around before the change of meaning.
;-)

 

[Bradley1234]

PDP/11s were minicomputers. VAXen were larger than a breadbox. At the
time many called them Midi-computers. Some weird ones even called them
mainframes.

VAXexn were frequently called "superminis". Nobody in their right mind
called them microprocessors until they were integrated to a small chipset
in the form of the microvax.

I dont recall saying the PDP 11 or 11/780 were microprocessors, I said
their respective cpu boards were. They sure werent analog.

the PDP11 was a mini computer, the 11/780 was much larger and more
expensive, and was a super-mini.

The best of the time was the Control Data Cyber 170, a giant, water cooled
monster with vector scan graphics crt. That was the epitome of mainframe.

 

[Bradley1234]

Youre not talking about the TB-216 are you?

But there is at least bit-slice processor called a "microprogrammable
processor" where the programmer actually writes a microprogram. It was
used in the "enhanced controller-tester", and they used it to test big ol'
disk drives - like, 600 MB in a unit the size of a desk. It had 8 KB of
TTL SRAM. They could make it emulate whatever kind of computer the drive
was destined to be plugged into.

"Microprogrammed processor" would be appropriate for a bit-slice where
the microprogram is nonvolatile, but if it's all on a chip, it's still
a microprocessor.

I think "the whole processor on one chip" is the definitive difference.

Thanks,
Rich

 

[Bradley1234]

See what I mean?

A byte is 8 bits, except maybe in a parity memory system when its 9 bits

So in the C language you can say "a literal byte" is other than 8 bits?

Are you sure thats not a Char? which can be 16 bits?


We wouldnt disagree that a

byte is 8 bits

Is this a troll? Of course a byte (as defined in C, anyway) can be
other than 8 bits.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers:
http://www.trexon.com
Embedded software/hardware/analog Info for designers:
http://www.speff.com

 

[John Larkin]

On Thu, 27 Jan 2005 20:58:23 GMT, "Bradley1234" <someone@yahoo.com>
wrote:

Sure send me $450 to purchase the spec, and Ill felch thru it, that seems
like a cromulent arrangement to me

In hardware, a byte is 8 bits, period, case closed

In the C language it has to represent -127 to + 127, hey thats a coincidance
8 bits gives you that.

Well Ive never seen anywhere (in reality) where people in the business use a
byte for more than 8 bits.

Your limited experience does not make the case. Bytes are usually 8
bits but the accepted definition of 'byte' allows other sizes. An
8-bit byte is an 'octet'.

http://en.wikipedia.org/wiki/Byte

John

 

[Keith Williams]

In article <hr9Kd.1067$lg5.202@trnddc06>, someone@yahoo.com says...

"Keith Williams" <krw@att.bizzzz> wrote in message
news:MPG.1c62cb9ccf5ad60298989e@news.individual.net...
In article <1106835288.952612.300580@f14g2000cwb.googlegroups.com>,
microx@ieee.org says...
Hmm.. this is interesting.

I found this at:
http://www.microcomputerhistory.com/f14patterson.htm

(The same Dave Patterson of the book I quoted above).

QUOTE corrected_spelling
The word "microprocessor" was used before the 4004 to mean the
processor in a microprogrammed computer by some people of the time, so
it's not shocking to see the word microprocessor in a document of that
era. I kind of used the term ambiguously, but it was a single chip CPU.
/QUOTE

So it seems the word can take both meanings, depending on what era you
are talking about.

I've been around processors (all flavors) for over 30 years and have
never heard a micro-programmed processor called a microprocessor. When
it mattered they were called microcoded, but never microprocessors.

--
Keith

This is why any one person does not know everything. Its far too big of a
technology to have seen and done it all. A person could go thru life only
knowing the Mac OS. Or mechanics who have never seen the automatic manual
clutch on an old VW bug and cant believe it exists

Ah, I haven't but you have. I see.

I suppose we can be sure that we all have applied a loose definition and are
upset when someone else calls it another thing. We wouldnt disagree that a
byte is 8 bits

No, we most certainly *would* disagree, because you're flat-assed wrong
(again).

Ive heard people call the TRS-80 a mainframe, because in support manuals of
old, it would say connect the cable to the port of the main frame or
whatever;

Oh, "I've heard..." Such a strong strawman you have there. I'm
impressed!

Mechanics sometimes call the 4 wheel drive transfer as a "transfer case" but
its called the transfer.

<shrug> It's not a microprocessor and this isn't
sci.mechanics.transmissions. Words mean things. Technical words have
technical meanings, not necessarily the same meanings as the same words
in common speech.

You're wrong, learn and get over it.

--
Keith

 

[Bradley1234]

It's eighteen bucks, cheapskate.

Oh, the one I found was $399.00 eighteen bucks, thats not bad.

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers:
http://www.trexon.com
Embedded software/hardware/analog Info for designers:
http://www.speff.com

 

[John Larkin]

On Thu, 27 Jan 2005 21:29:49 GMT, "Bradley1234" <someone@yahoo.com>
wrote:

Now here is the sure-fire way to discern the incompetent, unqualified,
pretentious individual anywhere in the industry, anyone who would start out
by saying something to the effect of "Your limited experience..." or "Since
you are so inexperienced, but me, Im so worldly and all knowing..." or
"You should leave the high tech concepts to people who are more
experienced..."

No, you should just do a tiny bit of research before posting
authoritative statements in an established newsgroup full of
engineers. You are wrong about "microprocessor" and you are wrong
about "byte" and you seem to want to argue about both. There's nothing
wrong with being wrong now and then, but you won't be popular if you
keep insisting you're right when you're not.


John

 

[Bradley1234]

"Keith Williams" <krw@att.bizzzz> wrote in message
news:MPG.1c632a0bb56a43259898a4@news.individual.net...

In article <PBcKd.1366$lg5.500@trnddc06>, someone@yahoo.com says...
Sure send me $450 to purchase the spec, and Ill felch thru it, that
seems
like a cromulent arrangement to me

You think you know the spec well enough to state what it states, yet
have never read the spec.

You think I havent read the spec before? I dont have the thing sitting here
because its not needed for SW dev.

Ive seen it, there are lots of essay style sections, sub sections, legal
looking and not many pictures, I like books with pictures in them

In hardware, a byte is 8 bits, period, case closed

Wrong. Get a refund on your tuition.

Im working on that, easier to squeeze water from a rock, those crooks.

In the C language it has to represent -127 to + 127, hey thats a
coincidance
8 bits gives you that.

Nope (understand that I'm not a 'C' programmer, only a processor
developer). From: http://www.comeaucomputing.com/techtalk/

So you would say that in 8 bits (aka a "byte") you cannot represent the
range of -127 to +127? What would your math teacher say?

You are a processor developer? Are you familiar with the use of
signed/unsigned numbers?

How many bits are in a byte?
Although it's common that the number of bits in a byte is 8,
this is not so for every system. That's right, a byte is not
always 8 bits. A byte is one of those terms which has an
interesting history and ends up meaning different things
to different people. For instance, there are some computers
where it is 6, 7, 8, 9, 32-bits, and so on.

Oh suuuuure there are, and in South American jungles a byte can be 3 bamboo
sticks? You may be confusing byte with word, a word is a variable width
unit that is relative to the architecture of the processor or application.
A byte is 8 bits today, even when its 9 bits.

In C (or C++) you can tell what it is for your system by looking
at limits.h (known as climits in C++) where the macro CHAR_BIT
is defined. It represents the "number of bits for the smallest
object that is not a bit-field", in other words, a byte. Note
that it must be at least 8 (which mean that strictly speaking,
a CPU that supports a 6 bit byte has a problem with C or C++).
Also note that sizeof(char) is defined as 1 by C++ and C
(ditto for the sizeof unsigned char, signed char, and their
const and volatile permutations).

It might be helpful to show a quote from Standard C:

* byte: "addressable unit of data storage large enough to hold
any member of the basic character set of the execution environment.
NOTE 1 It is possible to express the address of each individual
byte of an object uniquely. NOTE 2 A byte is composed of a
contiguous sequence of bits, the number of which is implementation
defined. The least significant bit is called the low-order bit;
the most significant bit is called the high-order bit."

* character: "bit representation that fits in a byte"

Thus by the 'C' standard the byte is *not* fixed at 8-bits.

No, it leaves the definition open. In the typical limits.h, the word byte
is 8 bits, its like saying an inch can be whatever length you want it to be.
To machinists? they stick to a set standard.

Well Ive never seen anywhere (in reality) where people in the business
use a
byte for more than 8 bits. There can be a byte embedded within a 32 bit
word, if the bus width is 32b, since when would that be called a byte?

...and you're telling me that I don't know everything. Well, I've been
in the business long enough to know that you're zero for two today.

Yet you havent shown where people, today, in business, use the byte for MORE
than 8 bits? Not in some novelty scenario 40 years ago thats long gone,
does Intel use a non 8-bit byte? Xilinx? Altera? Microsoft?

Aside from pointing to data on the surface of the moon we cannot read
here
online, like specs costing $$ can you show where a byte is used for more
than 8 bits in the software industry?

Several systems had byte sizes other than 8-bits. If you want
references that aren't on "on the surface of the moon", even though
they are the *standards* which you're mistakenly using, try a search on
"6-bit byte" and report back. You gotta promise to report back though
or I won't do any more work for you.

Yes Ive heard of Octal, yes some people for a particular era/industry
adopted the word byte for 6 bits. The originator made it 8 bits which has
lasted and is the standard. In hardware, are there any non 8 bit byte
devices?

Do you use 2629 code? Hollerith punched cards? if not, why not? I want, no
I demand all computers today be equipped with a Hollerith punched card
reader that uses 2629 code.

--
Keith

 

[Rich Grise]

On Thu, 27 Jan 2005 17:13:41 +0000, Bradley1234 wrote:

That is a microCODED processor. The specific term, "microprocessor", was
coined to mean "a whole processor on a chip"

coined? seems like it was just adopted to refer to the single chip thing.
Someone was saying the 11/780 was not microcoded?

Yes. Coined. Made up. Invented. Pulled out of the air. A brand new word,
never even conceived of before that fateful day.

No one has claimed that the 11/780 was not microcoded.

What we are trying to inject through your headbone is that microcoding
a processor does NOT make a microprocessor.

Putting the Program counter, ALU, and accumulator on ONE CHIP is what
makes a microprocessor.

Thanks,
Rich