Maker Pro
Maker Pro

Interesting Motorola 68K-like IC?

M

mario

Jan 1, 1970
0
Dear all,


I acquired recently a board that looks like an S-100 card, except it has 80
connectors instead of 100, and was made by tandy.

So, this board has, among other things, an IC that looks a lot like
theMotorola 68000 (64 CERDIP, 900 mils wide) and in fact, it sports the
Motorola logo, but this is what is written on it:

8040000
SC88019L
GN78302

I also got another board which appears to be some kind of memory card, but
again, the RAM ICs are some really strange beast: 16 pin CERDIP parts from
Motorola, with the following text:

8040665
BQD8346 (this number varies, as well as the first letter can be an "A")

These are intermingled with similar-looking ICs which however have the text:

MCM6665AL20
FQD8416 (this number varies)


I don't know what kind of bus do these cards plug into. Maybe a TRS-80?


many thanks in advance for any hint!


mario
 
M

Mark (UK)

Jan 1, 1970
0
Hi!

Well, MCM6665 is a DRAM, 4164 equivalent I think (or maybe 4116, my head
is spinning now...)

Yours, Mark.
 
F

Franc Zabkar

Jan 1, 1970
0
Dear all,


I acquired recently a board that looks like an S-100 card, except it has 80
connectors instead of 100, and was made by tandy.

So, this board has, among other things, an IC that looks a lot like
theMotorola 68000 (64 CERDIP, 900 mils wide) and in fact, it sports the
Motorola logo, but this is what is written on it:

8040000
SC88019L
GN78302

I also got another board which appears to be some kind of memory card, but
again, the RAM ICs are some really strange beast: 16 pin CERDIP parts from
Motorola, with the following text:

8040665

This site ...

http://www.vintagefunworld.com/items.asp?ID=14

.... claims this part is equivalent to a standard 4164 DRAM.
BQD8346 (this number varies, as well as the first letter can be an "A")

These are intermingled with similar-looking ICs which however have the text:

MCM6665AL20

My Motorola Master Selection Guide (1984) lists this as a 65Kx1,
200nsec DRAM, probably equivalent to a 4164. Yes, it does say "65K",
but I guess that's a typo.
FQD8416 (this number varies)

That's a YYWW date code.
I don't know what kind of bus do these cards plug into. Maybe a TRS-80?

"8040665" appears to be a Tandy part number. I suspect 8040000 is
also.

At least one web site states that the 8040665 was a "TANDY Original
Dynamic Ram Memory IC used in the earlier Tandy TRS-80 and Color
Computers (CoCo)".


- Franc Zabkar
 
M

mario

Jan 1, 1970
0
Dear Franc,


read answers below:

Franc Zabkar said:
This site ...

http://www.vintagefunworld.com/items.asp?ID=14

... claims this part is equivalent to a standard 4164 DRAM.
Yes, so it seems. I guess it would help if I coud find a really good Tandy
website/resource.
My Motorola Master Selection Guide (1984) lists this as a 65Kx1,
200nsec DRAM, probably equivalent to a 4164. Yes, it does say "65K",
but I guess that's a typo.
It should be equivalent to a 4164. The two ICs were randomly "mixed"
together, serving apparently identical functions. The reason I listed them
both is, I was hoping someone would recognize at least one of them. Tactic
which apparently worked.
That's a YYWW date code.
I knew that, I just wanted to specify it for completeness (avoid questions
like "are you sure you didn't just copy the date of production?").
"8040665" appears to be a Tandy part number. I suspect 8040000 is
also.
I am more inclined now than before reading your post, to think the 8040000
is a Moto 68K, under another name.

Now, it remains to be seen which TRS computer was so advanced to accept such
a CPU.


Many thanks for your help!!!


mario
 
M

Michael Black

Jan 1, 1970
0
mario" ([email protected]) said:
Dear Franc,


read answers below:


Yes, so it seems. I guess it would help if I coud find a really good Tandy
website/resource.
It should be equivalent to a 4164. The two ICs were randomly "mixed"
together, serving apparently identical functions. The reason I listed them
both is, I was hoping someone would recognize at least one of them. Tactic
which apparently worked.

I knew that, I just wanted to specify it for completeness (avoid questions
like "are you sure you didn't just copy the date of production?").

I am more inclined now than before reading your post, to think the 8040000
is a Moto 68K, under another name.

Now, it remains to be seen which TRS computer was so advanced to accept such
a CPU.
If it's really a Tandy board, given the date codes it likely is a 68000.
(My first thought was that it wsa Motorola's first attempt at something
better than the 68000, which was the 88000 series, but that came later.)

Radio Shack had a couple of 68000 based computers in the early eighties.
They weren't intended for the home user, too expensive, but they did
run Xenix, a Unix variant. One model was like a retrofit on one of the Z-80
based models, while the other was a standalone.

Ask in comp.sys.tandy

Michael
 
M

mario

Jan 1, 1970
0
Radio Shack had a couple of 68000 based computers in the early eighties.
They weren't intended for the home user, too expensive, but they did
run Xenix, a Unix variant. One model was like a retrofit on one of the Z-80
based models, while the other was a standalone.
I think you are talking about the Model 16 and 6000.
I would really like it if I could find a place that documents the bus into
which these cards plugged in, as well as the cards.
Ask in comp.sys.tandy
I will. I hope it's still "alive".

Thank you, Michael, for confirming the IC model.
 
E

Eric Smith

Jan 1, 1970
0
mario said:
So, this board has, among other things, an IC that looks a lot like
theMotorola 68000 (64 CERDIP, 900 mils wide) and in fact, it sports the
Motorola logo, but this is what is written on it:

8040000
SC88019L
GN78302

There's a high probability that it's a house-numbered MC68000.
GN7 is an MC68000 mask code, although it's remotely possible that
the same mask code could have been used for some other part.
 
T

Tim Wescott

Jan 1, 1970
0
Eric said:
There's a high probability that it's a house-numbered MC68000.
GN7 is an MC68000 mask code, although it's remotely possible that
the same mask code could have been used for some other part.

That 88xxx makes me suspicious -- are you sure it isn't an 88000 part of
some sort?
 
M

Michael Black

Jan 1, 1970
0
Tim said:
That 88xxx makes me suspicious -- are you sure it isn't an 88000 part of
some sort?
I wondered too. I thought their first RISC, before they got together
with Apple, was the 88000 series.

Michael
 
T

Tim Wescott

Jan 1, 1970
0
Michael said:
I wondered too. I thought their first RISC, before they got together
with Apple, was the 88000 series.

Michael
Yes it was. I've been told it's a nice processor, but rather a flop on
the market -- they came out after ARM and MIPS. The Apple/Motorola/IBM
(and HP IIRC) team was necessary to crash the market.
 
E

Eric Smith

Jan 1, 1970
0
Tim Wescott said:
That 88xxx makes me suspicious -- are you sure it isn't an 88000 part
of some sort?

Yes, I'm sure. Those have too many I/O signals for a 64-pin DIP package.
 
E

Eric Smith

Jan 1, 1970
0
Tim said:
That 88xxx makes me suspicious -- are you sure it isn't an 88000 part
of some sort?
Yes, I'm sure. Those have too many I/O signals for a 64-pin DIP package.

I forgot to mention that SCxxxxx numbers (such as SC88019 on this part)
are what Motorola (now Freescale) uses for house-numbered or custom
parts, and the number generally has NO relation to a normal MCxxxxx
number. Whereas XCxxxxx numbers are used on pre-qualification parts and
are normally the same as the eventual MCxxxxx part number, other than
the prefix.

Since the part in question has the GN7 mask code, it's probably either
just a house-numbered standard MC68000, or an MC68000 that has been
tested to slighly different specifications than a standard one.
Motorola did make some special MC68000-based parts with custom microcode
for IBM for the XT/370 and AT/370 products, but those had unique mask
codes.

At one time, the data sheets on the MC68000 actually listed some minor
differences between mask codes. For instance, on early mask codes, if a
word write was performed to an odd address, both data strobes were
asserted for the access even though it was automatically aborted and
resulted in an Address Error exception. This resulted in memory
corruption, which wasn't usually important because an Address Error was
normally an unrecoverable error anyhow. But they fixed this on later
mask codes so that the data strobes are not asserted in those
circumstances.
 
Top