6502B?

[dmu...@gmail.com]

I've got a weird Apple II+ here that appears to have been fiddled with quite a bit. The CPU appears to be labeled 6502B with a Synertek mark and a code of 7943D on it. Did Apple ever use Synertek as a source for the CPU in the II or II+? And even if they did, they didn't use the 6502B did they?

This machine isn't starting up properly -- I am not sure if it's because of the CPU, the ROMs (they were pretty much swapped around completely), the RAM (there are multiple manufacturers, even in the same row) or something else completely. But I was really surprised to see the 6502B in there.

Any ideas on what might've happened to this poor II+? I'm hoping I might be able to get it back to working condition someday, but I'm not really sure what the chances of that are at this point.

dennis

[Michael Black]

On Tue, 11 Nov 2014, dmu...@gmail.com wrote:

> I've got a weird Apple II+ here that appears to have been fiddled with
> quite a bit. The CPU appears to be labeled 6502B with a Synertek mark
> and a code of 7943D on it. Did Apple ever use Synertek as a source for
> the CPU in the II or II+? And even if they did, they didn't use the
> 6502B did they?
>

I don't know if Apple ever put it there, but I thought a B was only a 6502
rated for a higher clock frequency, I thought 2MHz. I don't see it being
a source of problem, and unless it's a CMOS 6502, which I don't think was
issued from Synertek, it's just a normal 6502, and will run the same as a
straight 6502 if the clock is only 1MHz. Being able to run faster means
nothing until the clock is raised.

Michael

[Michael J. Mahon]

Michael Black <et...@ncf.ca> wrote:
> On Tue, 11 Nov 2014, dmu...@gmail.com wrote:
>
>> I've got a weird Apple II+ here that appears to have been fiddled with >
>> quite a bit. The CPU appears to be labeled 6502B with a Synertek mark >
>> and a code of 7943D on it. Did Apple ever use Synertek as a source for
>> > the CPU in the II or II+? And even if they did, they didn't use the > 6502B did they?
>>
> I don't know if Apple ever put it there, but I thought a B was only a
> 6502 rated for a higher clock frequency, I thought 2MHz. I don't see it
> being a source of problem, and unless it's a CMOS 6502, which I don't
> think was issued from Synertek, it's just a normal 6502, and will run the
> same as a straight 6502 if the clock is only 1MHz. Being able to run
> faster means nothing until the clock is raised.

Well, there is one difference that might affect some things: the setup and
hold times will be different.
--
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon

[gid...@sasktel.net]

On Tuesday, 11 November 2014 21:41:46 UTC-6, Michael Black wrote:

The 6502B is an 65c02 enhanced upgrade. You can use the enhanced mnemonics, although you probably wont see it in a monitor listing.

Here is a quick way to confirm.

Type at the prompt:

CALL -151
300: 80 1 0 60
300G


if the code executes properly, you will not hear a beep and the break will not be executed.

The 80 mnemonic is a BRA (branch always) which is not available on a normal 6502.

Rob

[gid...@sasktel.net]

> The 80 mnemonic is a BRA (branch always) which is not available on a normal 6502.

Rather, the $80 byte is a BRA mnemonic. :)

[Steve Nickolas]

On Tue, 11 Nov 2014, gid...@sasktel.net wrote:

> The 6502B is an 65c02 enhanced upgrade. You can use the enhanced
> mnemonics, although you probably wont see it in a monitor listing.

I'm pretty sure that's not true, that the 6502B is just a faster rated
part.

-uso.

[gid...@sasktel.net]

On Tuesday, 11 November 2014 23:44:35 UTC-6, Steve Nickolas wrote:
> On Tue, 11 Nov 2014,
>

> > The 6502B is an 65c02 enhanced upgrade. You can use the enhanced
> > mnemonics, although you probably wont see it in a monitor listing.
>
> I'm pretty sure that's not true, that the 6502B is just a faster rated
> part.
>
> -uso.

Pretty sure isn't good enough. :)

I have 3 of these chips.

BRA, INC and DEC all work but not TRB and TSB.

haven't tried any other 65c02 codes yet

[Michael Black]

That's what I remember. It's the 6809 that puts the clock speed letter in
the middle, hence 68B09 is faster than a 6809.

Bsides, I thought it was other companies than Synertek that came out with
a CMOS/enhanced 6502. I remember an article in Micro about some potential
upgrade (I can't remember if it was someone hoping for things, or a paper
from Synertek or the like) but which never happened.

Michael

[Michael Black]

No. Unless I'm wrong, and the B was an actual improvement, it's just a
6502 that's been tested to run at 2MHz. Just like logic ICs, there's no
difference if the clock speed remains the same. But if you speed up the
clock, the 6502B will run fine at that speed.

Actually, I'm sure I had a straight 1MHz 6502 in my OSI SUperboard, and
when I changed the clock to 2MHz, I never had problems with the "slower"
CPU running at twice its rated speed.

Michael

[Patrick Schaefer]

Am 12.11.2014 00:29 schrieb dmu...@gmail.com:

> Any ideas on what might've happened to this poor II+?

This is the processor from an Apple ///. Due to the extra delay caused
by the memory mapping circuit Apple needed 6502s rated for 2 MHz
operation. These were marked SY6502B (a SY6502A is for 1.5 MHz, a plain
SY6502 for 1 MHz).

You can use it as replacement for the plain SY6502. It is the same chip,
only tested to work with a higher frequency.


Patrick

[Patrick Schaefer]

Am 12.11.2014 19:51 schrieb Patrick Schaefer:

> This is the processor from an Apple ///. Due to the extra delay caused
> by the memory mapping circuit Apple needed 6502s rated for 2 MHz
> operation. These were marked SY6502B (a SY6502A is for 1.5 MHz, a plain
> SY6502 for 1 MHz).

Even better: -A is for 2 MHz, -B for 3 MHz. Here is the datasheet:
http://www.classiccmp.org/cini/pdf/SYM/SYM1_REF05.pdf


Patrick

[dmu...@gmail.com]

On Wednesday, November 12, 2014 1:51:06 PM UTC-5, Patrick Schaefer wrote:

> Am 12.11.2014 00:29 schrieb <HIDDEN>@gmail.com:
>
> > Any ideas on what might've happened to this poor II+?
>
> This is the processor from an Apple ///.

Were there any reports of Apple using leftover stock from the /// in the II or II+? The date code on my motherboard is 47 81, so I'm guessing that's the 47th week of 1981? Definitely within the span of time when the /// was being made and sold.

Just curious if this was the stock CPU or if someone replaced it at a later point.

[D Finnigan]

On 11/11/14 5:29 PM, dmu...@gmail.com wrote:
> I've got a weird Apple II+ here that appears to have been fiddled with quite a bit. The CPU appears to be labeled 6502B with a Synertek mark and a code of 7943D on it. Did Apple ever use Synertek as a source for the CPU in the II or II+? And even if they did, they didn't use the 6502B did they?

I have a II Plus with a Synertek CPU in it.

The brown Apple IIe Technical Reference manual says that Apple put the
6502B in this machine. Now it's been a few weeks since I last looked,
but I think the white Apple II reference manual says that Apple used a
6502A in these machines.

[Michael Black]

On Wed, 12 Nov 2014, dmu...@gmail.com wrote:

> On Wednesday, November 12, 2014 1:51:06 PM UTC-5, Patrick Schaefer wrote:
>> Am 12.11.2014 00:29 schrieb <HIDDEN>@gmail.com:
>>
>>> Any ideas on what might've happened to this poor II+?
>>
>> This is the processor from an Apple ///.
>
> Were there any reports of Apple using leftover stock from the /// in the
> II or II+? The date code on my motherboard is 47 81, so I'm guessing
> that's the 47th week of 1981? Definitely within the span of time when
> the /// was being made and sold.
>

I don't know.

Initially, faster parts carry a premium price. With time, that may
decrease. So at some point, it may not matter, and since the faster CPU
doesn't mean anything without a faster clock, there perhaps was a point
where Apple used what it could get. But that's just an idea.

> Just curious if this was the stock CPU or if someone replaced it at a
> later point.
>

I thought from the rest of your description, it was sounded like this had
leftover parts generally. Which sounds like someone was testing something
else, and just never put the parts back in the originals.

Michael

[Michael J. Mahon]

Actually, if the chip runs correctly at 2MHz, its setup and hold times are
bound to be different. ;-)

[Michael J. Mahon]

And the faster processor was needed in order to obtain reliable setup and
hold times...

[Michael Black]

Yes. I'm just saying that at the same clock speed, that shouldn't be a
factor. If it was a redesign, there might be changes. But these are just
6502s that have been tested to run at a higher speed.

Michael

[Michael J. Mahon]

I agree. I was only pointing out that a processor capable of a higher speed
is 1) sometimes required to obtain reliable operation in a circuit, and 2)
sometimes the faster processor reveals marginal timing in other
circuits--e.g.: the original Z80 SoftCard, which does not operate reliably
with a //e.

[gid...@sasktel.net]

On Tuesday, 11 November 2014 23:44:35 UTC-6, Steve Nickolas wrote:

> On Tue, 11 Nov 2014,

> > The 6502B is an 65c02 enhanced upgrade. You can use the enhanced
> > mnemonics, although you probably wont see it in a monitor listing.
>
> I'm pretty sure that's not true, that the 6502B is just a faster rated
> part.
>
> -uso.

Ok! I was able to get around to checking the chips I have. It turns out your "Pretty Sure" is "Pretty Good". I admit it. I was wrong.

I have 3x 6503 chips, made in Mexico, that give the extra instructions. No other name on them. And they are all installed in IIe's.

And I have one 6502B in one of my II+'s. I was sure I was testing on this computer when I was doing the instruction tests.

[Steve Nickolas]

On Mon, 17 Nov 2014, gid...@sasktel.net wrote:

> I have 3x 6503 chips, made in Mexico, that give the extra instructions.
> No other name on them. And they are all installed in IIe's.

I believe the 6503 is a Rockwell-manufactured 65C02?

-uso.

[Michael Black]

That seems odd.

I thought there was a 6503, and it had something like a smaller subset of
address lines (so it could only address a subset of the 64K address
space). There were one or two of those in the line. It's not unlike the
6510, used in the Commodore 64, which included some I/O in the CPU (and
wsa used to control what RAM lay in the address space). Still a basic
6502, but with some variation.

The CMOS version of the 6809 is called the 6309.

Michael

[Steve Nickolas]

On Tue, 18 Nov 2014, Michael Black wrote:

> I thought there was a 6503, and it had something like a smaller subset of
> address lines (so it could only address a subset of the 64K address space).
> There were one or two of those in the line. It's not unlike the 6510, used
> in the Commodore 64, which included some I/O in the CPU (and wsa used to
> control what RAM lay in the address space). Still a basic 6502, but with
> some variation.

Isn't that the 6507 you're thinking of?

-uso.

[Patrick Schaefer]

6503 is a 28 pin device with 4k address space, /IRQ, /NMI and internal
clock generator. 6507 has /RDY instead of /IRQ and /NMI, and can access
8kB memory.

All devices share the same die with the 6502, but the bonding options
are different.


Patrick

[Michael Black]

Maybe, I can't remember specifics. There was a whole line, though I'm not
sure how available some of them were.

I was just wondering if they'd denote CMOS that way, or some other way.

Michael

[Steven Hirsch]

On 11/18/2014 12:02 PM, Michael Black wrote:

> I thought there was a 6503, and it had something like a smaller subset of
> address lines (so it could only address a subset of the 64K address space).
> There were one or two of those in the line. It's not unlike the 6510, used in
> the Commodore 64, which included some I/O in the CPU (and wsa used to control
> what RAM lay in the address space). Still a basic 6502, but with some variation.

Are you thinking of a 6504? The Apple Lisa floppy controller is built around
this part.

[Michael Black]

I was pointing out that there were variants of the 6502, some that likely
never saw much use (and perhaps never came to market). I can't remember
what was what, just that there was a range of CPUs in the line, with
varying number of address lines and such.

Adam Osborne's "An Introduction to Microcomputers: Volume 2, Some Real
Products" covers this. In the 1976 edition that I have there are:

6502
6502
6504
6505
6506
6512
6513
6514
6515

and I'm not going to bother figuring out what's different. They are all
6502s with a subset of the pins available to the outside. This is too
early for CMOS.

Remember, once they had the design, they could make variations. When we
first heard about it, it was going to be the 6500, which was
pin-compatible with the 6800 (so you could drop in the 6500 and have the
6500 set of instructions and architecture). It was never clear if any of
those were delivered, but soon Motorola objected, so we got the 6502,
which wsa the same CPU but the pinout not compatible with the 6800.

So Mos Technology was able to rearrange things pretty fast.

I thought there was supposed to be a 6501 too, but I can't remember what
that was.

And of course, later there were variants, as I mentioned the 6510 as used
in the Commodore 64, and Rockwell (if not them, someone else) had a 6502
with a FORTH interpreter inside the actual IC.

I gather there are still descendants of the 6502 used in some places, for
a while in game consoles (before those moved to 16bit). We see the same
practice today with ARM processors, a company designs the basic CPU, but
just licenses it out, so other companies actually manufacture it, and add
whatever other things on-chip that cellphone or whatever needs.

Michael

[Scott Alfter]

In article <alpine.LNX.2.02.1...@darkstar.example.org>,

Michael Black <et...@ncf.ca> wrote:
>On Mon, 17 Nov 2014, Steve Nickolas wrote:
>
>> On Mon, 17 Nov 2014, gid...@sasktel.net wrote:
>>
>>> I have 3x 6503 chips, made in Mexico, that give the extra instructions. No
>>> other name on them. And they are all installed in IIe's.
>>
>> I believe the 6503 is a Rockwell-manufactured 65C02?
>>
>That seems odd.
>
>I thought there was a 6503, and it had something like a smaller subset of
>address lines (so it could only address a subset of the 64K address
>space).

There's the 6507, which is missing the top three address lines (among some
other things). As far as Wikipedia knows, it was only used by Atari, and
only in three products: the 2600 console and the 810 and 1050 floppy drives.

My IIe has a Rockwell 65C02 as part of a DIY enhancement kit. It's labeled
"R65C02," not a 6503. AFAIK, Apple usually got its 65C02s from other
sources. It implements four new instructions, but as these are only shared
with WDC 65C02s (which have two more instructions on top of that) and they
conflict with the added instructions of the 65816, their usage in software
targeting the Apple II is probably not a good idea.

_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( http://alfter.us/ Top-posting!
\_^_/ >What's the most annoying thing on Usenet?

[Michael Black]

On Wed, 19 Nov 2014, Scott Alfter wrote:

> In article <alpine.LNX.2.02.1...@darkstar.example.org>,
> Michael Black <et...@ncf.ca> wrote:
>> On Mon, 17 Nov 2014, Steve Nickolas wrote:
>>
>>> On Mon, 17 Nov 2014, gid...@sasktel.net wrote:
>>>
>>>> I have 3x 6503 chips, made in Mexico, that give the extra instructions. No
>>>> other name on them. And they are all installed in IIe's.
>>>
>>> I believe the 6503 is a Rockwell-manufactured 65C02?
>>>
>> That seems odd.
>>
>> I thought there was a 6503, and it had something like a smaller subset of
>> address lines (so it could only address a subset of the 64K address
>> space).
>
> There's the 6507, which is missing the top three address lines (among some
> other things). As far as Wikipedia knows, it was only used by Atari, and
> only in three products: the 2600 console and the 810 and 1050 floppy drives.
>
> My IIe has a Rockwell 65C02 as part of a DIY enhancement kit. It's labeled
> "R65C02," not a 6503. AFAIK, Apple usually got its 65C02s from other
> sources. It implements four new instructions, but as these are only shared
> with WDC 65C02s (which have two more instructions on top of that) and they
> conflict with the added instructions of the 65816, their usage in software
> targeting the Apple II is probably not a good idea.
>

I always thought that was weird, two CMOS versions of the 6502, and they
don't have the same extras. And if I recall, the Western Design 16bit CPU
has a different set of improvements to the 6502 than the CMOS (or maybe
it's a mix?)

Michael

[Steve Nickolas]

I believe the 65SC02 was the original 65C02, then Rockwell made their
extensions which made it into later WDC issues?

-uso.

[Michael Black]

Something like that. But didn't Western Design do one of those CMOS
versions? I thought the 65SC02, but it's been so long I'm not certain.

Here was some inconsistency, that never seemed to be accounted for, but
obviously a result of multiple companies having rights to the 6502 so they
did as they chose to do.

Michael

[mdj]

On Thursday, 20 November 2014 06:53:49 UTC+10, Michael Black wrote:

> Here was some inconsistency, that never seemed to be accounted for, but
> obviously a result of multiple companies having rights to the 6502 so they
> did as they chose to do.

Yeah the second sourcing arrangements make it really difficult to figure out which 65c02's do what. The original was done by Bill Mensch, and that's the 'classic' 65c02 instruction set which is implemented both on the 65c02 and the 65c816 emulation mode.

The Rockwell extensions give the convenient but ultimately unusable on an Apple II bit operations, and the 65C02S is the current WDC model that has both the Rockwell extensions, and the STP and WAI instructions as well, which make it very useful for low power operations, in addition to making the already legendary interrupt latency even better.

Once you leave out the extensions above the only really useful extras are being able to increment and decrement the accumulator, and push and pull X and Y directly, though I have on occasion found BIT immediate kinda handy.

The other changes, being zero page indirect addressing and the JMP(abs,X) are most useful for improving the efficiency of compilers and interpreters. Worthy changes, but of limited extra utility to someone writing assembler by hand.

Matt

[MarkO_]

> That seems odd.
>
> I thought there was a 6503, and it had something like a smaller subset of
> address lines (so it could only address a subset of the 64K address

> space). There were one or two of those in the line. It's not unlike the
> 6510, used in the Commodore 64, which included some I/O in the CPU (and
> wsa used to control what RAM lay in the address space). Still a basic
> 6502, but with some variation.
>

I have some Rockwell 6503's in some enhanced Apple //e's They seem to be a 65C02, with a different name.

> The CMOS version of the 6809 is called the 6309.
>
> Michael

The 6309 is a Hitachi licensed Motorola 6809, with quite a few "cool additions". It is a Drop In Replacement for the 6809. I am in the processes of replacing the 6809's in my Tandy CoCo 2 and 3 with the 6309's.

MarkO

[David Empson]

MarkO_ <app...@markoverholser.com> wrote:

> > That seems odd.
> >
> > I thought there was a 6503, and it had something like a smaller subset of
> > address lines (so it could only address a subset of the 64K address
> > space). There were one or two of those in the line. It's not unlike the
> > 6510, used in the Commodore 64, which included some I/O in the CPU (and
> > wsa used to control what RAM lay in the address space). Still a basic
> > 6502, but with some variation.
>
> I have some Rockwell 6503's in some enhanced Apple //e's They seem to be
> a 65C02, with a different name.

Pure speculation without checking details, but it wouldn't surprise me
if Rockwell used "6503" to indicate a second source 65C02 based on the
WDC architecture, to distinguish it from their own R65C02, which had
additional instructions (most of which were for manipulating and testing
bits in memory).

I can't see any mention of that "6503" in my 1987 Rockwell data book.
The book does mention an R6503, which is an NMOS processor based on the
original 6502, but with 28 pins and a 4KB address space.
--
David Empson
dem...@actrix.gen.nz

[D Finnigan]

I just looked in my platinum IIe, built 1989. It has a Rockwell 6503.

[Steve Nickolas]

On Thu, 18 Dec 2014, D Finnigan wrote:

> I just looked in my platinum IIe, built 1989. It has a Rockwell 6503.

Funny, mine (though I think it was built in 1987) has a GTE 65SC02.

-uso.