SwyftCard

[kelli hirsch]

Does anyone have a SwyftCard for sale? Thanks. kelli

[dwight elvey]

Hi
 If no one comes up with one, You can buy parts from
someplace like Jameco and wire one up. You need to
get some proto board. I hacked up an old PC proto board
as I recall to get the edge connector.
 I used point to point wiring but wire wrap is also
an option.
 I have PAL code and EPROM code. The rest is just following
the schematic in the manual.
Dwight

 

> Date: Wed, 15 Feb 2012 17:56:03 -0800
> Subject: [Canon Cat] SwyftCard
> From: kelli...@gmail.com
> To: cano...@googlegroups.com

>
> Does anyone have a SwyftCard for sale? Thanks. kelli
>

> --
> You received this message because you are subscribed to the Google Groups "Canon Cat" group.
> To post to this group, send email to cano...@googlegroups.com.
> To unsubscribe from this group, send email to canon-cat+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/canon-cat?hl=en.
>

[Charles Springer]

I have a stack of ancient JameCo Apple II proto-boards somewhere. I'll try to find them. They had gold plate, so they will still look new.

-- Charlie

[Sandy Bumgarner]

Hey Charlie, you should maybe kit the Swyftcard with the parts, a
protoboard, and a schematic. ???

Sandy
===

--
John O. 'Sandy' Bumgarner
KI6UOQ

[dwight elvey]

 Hi
 If Charlie can find them, I'd put together a number of kits.
I can make the PALs and program the EPROMs. All that would
be needed is to put them together.
 We'd need to determine if people want wire wrap or point to
point wire.
 Having done both, I prefer point to point.
Dwight
 

> Date: Mon, 20 Feb 2012 00:02:18 -0800
> Subject: Re: [Canon Cat] SwyftCard
> From: sand...@gmail.com
> To: cano...@googlegroups.com

[Russell Nelson]

Why not make PCBs? It's not that hard to do, nor very expensive.

[Sandy]

Correct on both counts.  I would do it if someone will buy them, then Dwight could kit that.  The reason I claim some degree of ownership for any new PCB is that my name as well as Mino's is on the 1.0 board, the one with 2 ROM.

OK, who wants to put a down payment on a PCB?  No gold fingers and ROHS compliant.  The schematic as published could only be the starting point as the board could easily be embellished.  Or do you want a near exact duplicate?  Prices depend on the quantity of boards ordered.  The down payment will about 50% of the cost of the board.

So, I repeat, who will verbally commit and then send a down payment so I could finish the design and order boards built?

Sandy

Sent from my iPhone

[Mike Willegal]

This is an old thread, but I was wondering if SWYFT card ROM files and PAL equations are posted somewhere?

Did anyone ever get those PCBs made?

If not, I think it may be relatively straight forward to convert my SUPERPROTO boards to SWYFT functionality.  It already has EEPROM and a GAL on board and has plenty of prototyping space for the LM311.  The GAL is a 22V10, but should be possible  be able to be reprogrammed to match SWYFT card functionality, since 6522 VIA support is not required, leaving room for anything not already present.  The PROM is an EEPROM, but this makes it possible to program/reprogram without use of a PROM programmer.  After programming, writes can be disabled.

www.willegal.net/superproto

any thoughts?

Regards,

Mike W.

[dwight elvey]

Hi
 The PAL code I have is hand generate JADEC code for which
fuses to blow. It was not generated with a compiler.
I don't fully recall how A12 acted but is was based on
a toggle function from the address decode, as I recall.
If you look at the schematic you'll note that it goes through
the PAL to select different banks to make full use of the
27128 that does not fit into the space allotted in the
Apple's memory window.
In any case, you need to, hard wire, remap A12 to
the EPROM.
I'd have to look it up.
I don't know what the SuperProto looks like but it sounds
like it can be used.
It may take me a day or so but I should be able to find the EPROM
dump and the PAL file I used.
Dwight

---

Date: Mon, 19 Aug 2013 10:56:53 -0700
From: mi...@willegal.net
To: cano...@googlegroups.com

To unsubscribe from this group and stop receiving emails from it, send an email to canon-cat+...@googlegroups.com.

To post to this group, send email to cano...@googlegroups.com.

Visit this group at http://groups.google.com/group/canon-cat.
For more options, visit https://groups.google.com/groups/opt_out.

[Vitorio]

ROM files are around, MESS can apparently emulate the SwyftCard as of last year.

Are those dumps the same you'd use for burning to hardware?

[Jonathan Gevaryahu]

I believe mess has both the rom and .jed you'd need to make a card, yes.

-- 
Jonathan Gevaryahu
jgeva...@gmail.com
jgeva...@hotmail.com

[Charles Springer]

A PCB would be simple. It is hard to make a simpler Apple II PCB. I just found a couple Swyftcards in my storage container.

I had thought of an ARM9 or A8 stuck in an Apple IIe case. Now I wonder about a single board with an ARM on it emulating Apple IIe and open up the memory in the CAT code to use an SD card for mass storage and the 40 or more MBytes (cheap ARM9) to 500 MBytes RAM we would have available after emulation. Then all the keyboard and everything still works.

This might be the easiest way to get full functionality and I like the Apple sculpted keyboard.

-- Charlie

[rad...@gmail.com]

You need to have your own ROM files to make mess work. They are about, but hard to track down. I'd have to check what version of them I have, not sure off the top of my head.

-David Mercer

David Mercer
Portland, OR

---

From: Jonathan Gevaryahu <jgeva...@gmail.com>

Sender: cano...@googlegroups.com

Date: Mon, 19 Aug 2013 17:40:34 -0400

To: <cano...@googlegroups.com>

ReplyTo: cano...@googlegroups.com

[dwight elvey]

Hi
 The JEDEC file is what a programmer takes. It is what is
usually compiled, as output, by a compiler. There is a binary
and ASCII format that is valid. I just did the ASCII.
The one I have is unlikely to match the original PAL since
I most likely treated unused fuses differently than might
have been originally done. This is usually a background pattern
that minimizes power in circuit.
Still, it is functionally the same.
The JEDEC file that MESS has may be from me. I've sent them
such information on a number of different computers over time.
It may be extracted from a Swyft cards as well.
The file is just a string of 1s and 0s that are the fuse pattern
to blow. I created it from the TI data spec sheet for that PAL.
It was interesting that I was once told that such was too complicated
for a human to do. This might be true of the newer parts but
I suspect that a skilled human can produce a better optimized
circuit but most likely, not fast enough to be practical.
On the simple PAL, it was just looking at the logic and making it
do what one wanted. I made errors but they'd have been the
same if I were to use a logic compiler. My third chip worked.
I did a fourth chip because I was unsure about the reset state
of the A12 signal. Nothing in the documents determines what it should be.
Dwight

---

Date: Mon, 19 Aug 2013 14:57:39 -0700

Subject: Re: [Canon Cat] SwyftCard

From: charles....@gmail.com
To: cano...@googlegroups.com

[dwight elvey]

I thought I'd note:
I have a neat card from SCRG that works
like a ROM replacement for disk.
It is cool in that with it, I can be ProDOS or DOS4.0 ( 4.0?? )
as well as a number of utility programs in EPROMs.
It is interesting that it takes over the Apples bus the same
way that the SWYFT card does, making these two cards
incompatible in the same system at the same time.
The SWYFT care cares which device socket it is in but
this card doesn't.
You can guess how I figured out they were incompatible.
Dwight

---

From: dke...@hotmail.com
To: cano...@googlegroups.com
Subject: RE: [Canon Cat] SwyftCard
Date: Mon, 19 Aug 2013 16:04:27 -0700

[Jonathan Gevaryahu]

Really not that hard to find. attached.

[dwight elvey]

Yep, that was my JEDEC file.
Dwight

 

---

Date: Mon, 19 Aug 2013 19:36:51 -0400
From: jgeva...@gmail.com

[Mike Willegal]

Thanks all -  this card is indeed, very simple, even by  Apple II card standards.   I'll convert the JEDEC to PALASM since I need to do a bit of tweaking to adapt it to the 22V10 on my SUPERPROTO card.   One question:  I see that system reset goes to the PAL separately from the on card power on reset circuit.  The document describes power on reset behavior.  Does anyone understand if the card is supposed to reset either A12 or enable states when system reset occurs?  I'm thinking that system reset just disables the board until reset goes away, leaving current A12 and enable states, but I might have missed something.

Thanks and regards,

Mike W.

On Monday, August 19, 2013 9:51:12 PM UTC-4, dke...@hotmail.com wrote:

Yep, that was my JEDEC file.
Dwight

 

---

Date: Mon, 19 Aug 2013 19:36:51 -0400
From: jgeva...@gmail.com
To: cano...@googlegroups.com
Subject: Re: [Canon Cat] SwyftCard

Really not that hard to find. attached.

On 8/19/2013 6:59 PM, rad...@gmail.com wrote:

You need to have your own ROM files to make mess work. They are about, but hard to track down. I'd have to check what version of them I have, not sure off the top of my head.

-David Mercer

David Mercer
Portland, OR

---

From: Jonathan Gevaryahu <jgev...@gmail.com>

[dwight elvey]

I believe it should reset BANK but hold INH until RES is done.
BANK effects the mapping of A12. A12 should just be a decode
and not hold state. Only BANK is a state.
I think that is what I did in my PAL but would have to look at the
bits for that.

---

Date: Tue, 20 Aug 2013 08:36:49 -0700

[Mike Willegal]

OK - Here is what I have for now - tested only in simulation.  There are several syntax's for PALASM type tools.  This is built using ATMEL's WINCUPL.

Name     SWYFT ;

PartNo   A ;

Date     8/20/2013 ;

Revision 01 ;

Designer Engineer ;

Company  willegal.net ;

Assembly None ;

Location  ;

Device   16r4;

/* *************** INPUT PINS *********************/

PIN 1    =  Q3CLK   ;

PIN 2    =  !DEVSEL ;

PIN 3    =  !LRST   ;/*local on board reset*/

PIN [4,5] =  [A1,A0] ;       

PIN [6..9] =  [A15,A14,A13,A12]  ; 

PIN 12   = !SRST           ;/* system reset */

/* *************** OUTPUT PINS *********************/

PIN 13     =   !INH      ;/*inhibit to motherboard */      

PIN 14    =    ACT     ;/*active      */

PIN 17   =   !BANK    ;/*bank selected  */   

PIN 18   =  !OE       ;/*output enable  */   

PIN 19    =  !OA12      ;/*bank sel address bit  */   

OE = !SRST & !LRST & ( A15 & A14 & (A13 # A12)) & ACT;

INH = OE;

OA12 = BANK;

BANK.D = (DEVSEL & !A1 # (!DEVSEL & BANK)) # LRST # SRST;

ACT.D = (DEVSEL & !A0 # (!DEVSEL & ACT)) # LRST;

[dwight elvey]

Hi Mike
 I don't think you have the reset right. There might be other issues
but I'm not familiar with thei PALASM.
I found a file that I used to back check JEDEC file.
see attached.
The on board drives reset circuit holds the system reset
until it has ensured that the system reset has completed. I believe
it otherwise ignores the system reset as an input.
In my pal, I treat the pin 12, system reset, as an O.C. pin output,
not an input.

---

Date: Wed, 21 Aug 2013 07:20:06 -0700

[dwight elvey]

Also, INH is a O.C. signal as well.
I don't think the equation looks right for !OE but
I've not demorgan it.
Dwight

---

From: dke...@hotmail.com
To: cano...@googlegroups.com
Subject: RE: [Canon Cat] SwyftCard

Date: Wed, 21 Aug 2013 11:25:46 -0700

[Jonathan Gevaryahu]

Does anyone have an actual real IAI swyftcard I could borrow to attempt to dump the pal using a bpmicro programmer (assuming the security fuse on the pal is not blown)?
If it turns out to be secured I'd also like to lend it to a friend who has a behavior-level pal-analyzer board to try to suss out the exact details.

[dwight elvey]

I was looking at my Pal.txt file to compare the OE signal.
There was a mistake in the text on the side for this signal.
(A15*A14*/ACT) should have been (A15*A14*A13*/ACT).
It was an error in the text, not the fuses.
This PAL code has been confirmed to work.  The original
PAL is soldered to the board as I recall.

---

Date: Wed, 21 Aug 2013 15:41:34 -0400

[Jonathan Gevaryahu]

I have the necessary tools to desolder the pal without damaging the board.

[dwight elvey]

Mike is using a different programmable part.
All he needs is working equations.
I don't see the need to desolder a good board.
I'm equipped to desolder chips and have years
of experience doing it. I would not say I could
do it without possible damage to the board.
Dwight

 

---

Date: Wed, 21 Aug 2013 23:26:21 -0400

[Russell Nelson]

I've decided hundreds of chips from boards. You almost always have to sacrifice one to save the other.

[Jonathan Gevaryahu]

If you do it right, you can almost always get a chip off the board without damaging either the chip or the board.
4-layer boards with big ground/vcc planes make getting chips out a bit more of a pain, but it can still be done. I've done it.
The tools I use here are a soldapult, extra solder, extra flux, and a good iron. I'm eventually going to get a Hakko or Tenma desoldering gun which is even better for this sort of thing.

[Jonathan Gevaryahu]

I'd like to at least TRY to read out the original chip, is all.

[Sandy Bumgarner]

Through hole parts are a problem.  I have had a wee bit of experience here also and with through hole I often just clip the chip's leads and then pull the pins from the board one by one hence choosing to be easy on the board.  You can put pins back on the chip so it may be ugly but pretty safe.

But I like Charlie's idea: put the whole tham ding into an ARM inside a keyboard that has LEAP keys.

Sandy
===

[Mike Willegal]

Hmm,

regarding reset.

The manual indicates that the swyft card LM311 circuit reset,  resets the PAL, which means that active is set and bank is set to bank 1, since they are the only two parts of the PAL that maintain state.  It says A2e reset is used by other software for warm start.  I was confused by the need to extend system reset onto the card, thinking it did something to the card, but I agree with you that it is used to make sure that swyft card reset is not held longer than system reset during power up, which could cause some start up issues.  I'll fix that.

And yes,  inhibit and reset should be either OC or tristated when off.

I'll have to review the A12 selection logic, it looks like I may have oversimplified.

regards,

Mike W.

[dwight elvey]

The SWYFT cards reset by the comparator chip is the ensure
that it holds the system reset longer than the system does so that
it can take over the processor when the reset is finally released.
My equation for BLANK should be correct for the data input of the
flop.
Also OE in my text had an error as I noted. It should have read:

/OE = /((A15*A14*/ACT*/A13*A12)
         | (A15*A14*A13*/ACT))
This could be optimized and simpler but was written to match
the PAL's logic diagram.
OE should be invalid for 0CXXXh addresses and
valid for address higher with ACT active.
A simpler equation would be to reject just 0CXXXh
giving it a 12K window that it could bank switch A12
in.
Dwight

---

Date: Thu, 22 Aug 2013 16:18:57 -0700

[dwight elvey]

I may not be clear on how it should treat the
system reset. As I recall it operated as expected
when I tried it.
I may only reset the BANK flop if that.
Dwight

 

---

Date: Thu, 22 Aug 2013 18:06:23 -0700

Subject: Re: [Canon Cat] SwyftCard

From: russn...@gmail.com
To: cano...@googlegroups.com

Yep, and there's already capable hardware: the Raspberry Pi. All we really need is a keyboard with leap keys.

On Aug 23, 2013 12:13 AM, "Sandy Bumgarner" <sand...@gmail.com> wrote:

Through hole parts are a problem.  I have had a wee bit of experience here also and with through hole I often just clip the chip's leads and then pull the pins from the board one by one hence choosing to be easy on the board.  You can put pins back on the chip so it may be ugly but pretty safe.

But I like Charlie's idea: put the whole tham ding into an ARM inside a keyboard that has LEAP keys.

Sandy
===
On Thursday, August 22, 2013, Russell Nelson wrote:

--
John O. 'Sandy' Bumgarner
KI6UOQ

--
You received this message because you are subscribed to the Google Groups "Canon Cat" group.

To unsubscribe from this group and stop receiving emails from it, send an email to canon-cat+...@googlegroups.com.
To post to this group, send email to cano...@googlegroups.com.
Visit this group at http://groups.google.com/group/canon-cat.
For more options, visit https://groups.google.com/groups/opt_out.

[Russell Nelson]

Yep, and there's already capable hardware: the Raspberry Pi. All we really need is a keyboard with leap keys.

On Aug 23, 2013 12:13 AM, "Sandy Bumgarner" <sand...@gmail.com> wrote:

--

[dwight elvey]

After rereading the manual, I believe I treated the
reset correctly.
Dwight

 

---

From: dke...@hotmail.com
To: cano...@googlegroups.com
Subject: RE: [Canon Cat] SwyftCard

Date: Thu, 22 Aug 2013 18:53:12 -0700

[Mike Willegal]

Hi,

I just finished turning one of my SUPERPROTO prototyping cards into a Swyft card. So far, it seems to work fine.   I didn't explore functionality that much, but the software looks pretty clever, despite the lack of a GUI (I know - don't flame me).  :-)

The equations that work in the GAL22V10 on my hacked SUPERPROTO follow.  I use an EEPROM in the SUPERPROTO, so if you can load some kind of monitor or programming routines into RAM, you should be able to program the EEPROM in circuit.  I haven't tried this, but it should work.

Once I get the rework instructions up on my Wiki, I'll offer SUPERPROTO swyft card kits, with the components necessary to build your own SWYFT card, including preprogrammed GAL and EEPROM at the usual SUPERPROTO price of $65.  The conversion of a SUPERPROTO to SWYFT card does require good soldering/rework skills, as you have to cut some traces and solder a few wires directly to pins on the back of the card.  If there is enough interest, I could possible make dedicated reproduction swyft cards.

Name     SWYFT22V10 ;

PartNo   A ;

Date     8/20/2013 ;

Revision 01 ;

Designer Engineer ;

Company  willegal.net ;

Assembly None ;

Location  ;

Device   g22v10;

/****************************************************************/

/*                                                              */

/* Apple 2 I/O board support chip                               */

/*                                                              */

/****************************************************************/

/* Allowable Target Device Types :  GAL22V10                   */

/****************************************************************/

/**  Inputs  **/

PIN 1           = 2MHZclock;     /* O3 Clock               */

PIN 2  = A0;     /* address line 0              */

PIN 3           = A1;     /* address line 1               */

PIN 4  = rw;             /* read write (write low)      */

PIN [5..7]      = [A13,A14,A15]; /* Address lines A13-A15           */

PIN 8 = prom_write_dis; /* disable writes to eeprom */

PIN 9           = A12;     /* Address lines A12            */

PIN 10  = !devsel;        /* device select               */

PIN 13  = !lreset; /* local reset */

/**  Outputs  **/

PIN 14 = AO12; /* address A12 output                */

PIN 17         = active;     /* board active  */

PIN 18         = !rd_oe;          /* read/output enable (28F256 pin 22)      */

PIN 20 = bank; /* bank selected */

PIN 21         = !rom_en;     /* rom enable  */

PIN 22 = !inhibit; /* inhibits motherboard ROMS */

PIN 23         = !board_en ; /* controls 74LS245 out en */

PIN 19  = !sreset;         /* system reset                       */

/** Declarations and Intermediate Variable Definitions **/

/** Logic Equations **/

active.D = (devsel & !A0 # (!devsel & active)) # lreset;

rd_oe = rw & rom_en;

rom_en = board_en & (rw # !prom_write_dis);

board_en = active & !lreset & ( A15 & A14 & (A13 # A12));

inhibit.d = board_en;

inhibit.oe = board_en;

bank.D = !((devsel & !A1 # (!devsel & !bank)) # lreset);

AO12 = (A12 & A13) # (!A13 & bank);

sreset.oe = lreset;

sreset = lreset;

[dwight elvey]

Looks like it should work. The only thing
you might want to change is that the driven data
for the pins sreset and inhibit should be static values
with only the .oe pin driven dynamically.
You can get momentary contention during switching
time on these pins otherwise, with a driven signal from
the system. You don't want to drive both the data
and the .oe with the same data on a O.C. pin.
Dwight

---

Date: Tue, 27 Aug 2013 19:28:18 -0700

From: mi...@willegal.net
To: cano...@googlegroups.com
Subject: Re: [Canon Cat] SwyftCard

[Mike Willegal]

From the images I've found on the web, it looks like the chips in the Swyft card are socketed.  Does someone have different information?

Regards,

Mike W.

[Charles Springer]

I found a couple of cards last week in storage. I don't need them (keep one for a NIB Apple IIe Platinum in my climate controlled shipping container). If anyone gets the PAL equations figured out I will have some boards made. Or just dup the PAL fuse map. Can we still get 16R4's or L8's or whatever it is?

-- Charlie

On Wed, Feb 15, 2012 at 5:56 PM, kelli hirsch <kelli...@gmail.com> wrote:

Does anyone have a SwyftCard for sale? Thanks. kelli

--
You received this message because you are subscribed to the Google Groups "Canon Cat" group.

To post to this group, send email to cano...@googlegroups.com.

To unsubscribe from this group, send email to canon-cat+...@googlegroups.com.

[Mike Willegal]

The PAL equations have been reverse engineered by DKE - but it wouldn't hurt to get confirmation by reading or testing a "real" Swyft card PAL.   Parts are available.  The original used a 16R4, but I think it should be possible to use 16V8 GALs, which are a bit more modern, reprogrammable and should be a bit easier to find.  The LM311 and 27C128 are also readily available.  It might be cheaper to substitute a 27C256 for 27C128, but you would have to check the PCB layout to make sure the extra address bit doesn't cause issues.

Would you be duplicating the PCB layout yourself?

regards,

Mike W.

[Jonathan Gevaryahu]

I'd be happy to borrow(or buy) one to try to dump the PAL.

To unsubscribe from this group and stop receiving emails from it, send an email to canon-cat+...@googlegroups.com.

To post to this group, send email to cano...@googlegroups.com.

Visit this group at http://groups.google.com/group/canon-cat.
For more options, visit https://groups.google.com/groups/opt_out.

[Sandy Bumgarner]

I have two unpopulated Swyft cards.  An early one and a production one.  Any interest in HD pictures of them?  The early one has Mino and my name on it!

Sandy
===

--
You received this message because you are subscribed to the Google Groups "Canon Cat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to canon-cat+...@googlegroups.com.
To post to this group, send email to cano...@googlegroups.com.
Visit this group at http://groups.google.com/group/canon-cat.
For more options, visit https://groups.google.com/groups/opt_out.

--

[Jonathan Gevaryahu]

Sure! If you've got a flatbed scanner, can you scan both sides of the pcb?
If not, just good high-res pictures of both sides are fine too.

-- 
Jonathan Gevaryahu
jgeva...@gmail.com
jgeva...@hotmail.com

[Sandy Bumgarner]

I can do both.  The picture is easiest so that will happen first.  Where can I put it and the forthcoming scan file?

Sandy
===

[Charles Springer]

The PCB is trivial. I'm sure GALs would be fine. We switched to them for prototyping as soon as they became available.  Saved me a pile on debugging the ImageWorks frame grabber for Apple II, which took a lot of experimenting to get the timing right.

-- Charlie

To unsubscribe from this group and stop receiving emails from it, send an email to canon-cat+...@googlegroups.com.

To post to this group, send email to cano...@googlegroups.com.

[Jonathan Gevaryahu]

dropbox perhaps?

[Sandy Bumgarner]

Dropbox for what?

I think gmail will do 10MB or so.

But, I can unzip however the Linux compression defeats me, even using something like 'Total Commander'.  Must be due to my wonderfully addled chemo brain! )-:

[dwight elvey]

Here are some pictures. The first two are my home
made SWYFT card. The one on the left it a picture
of a real SWYFT card( source unknown ).
Dwight

 

---

Date: Wed, 4 Sep 2013 17:29:14 -0700

Subject: Re: [Canon Cat] SwyftCard
From: sand...@gmail.com
To: cano...@googlegroups.com