Category Archives: Retrochallenge 2017/4

Retrochallenge 2017/4 – Interak CF card – Part 4

We are well into the second half of Retrochallenge now and it fair to say that I’m not where I wanted to be. However I am learning a great deal about Z80s and CF cards are their respective wily ways.

Things continue to be confusing.

A quick re-cap of my previous posts…

It started off with the status of the CF card never showing that the read transfer was ready. However, using the ROM monitor on the Interak rather than the program showed that it was. There were various trials and hair being pulled out but what it boils down to is this.

If I use…

    ld        BC, (0x0067)
    in        a, (C)

I get one value but if I use…

    ld        a, 0x00
    in        a, (0x67)

I get a different answer.

Note 0x67 is the CF card status port.

According to the Z80 manual from Zilog, in a, (C), puts B on to the top of the address bus, C on to the bottom and reads from the addressed I/O port. So that’s 0x0067.

Now, in a, 0x67, puts A on to the top of the bus (previously loaded with 0) and 0x67 on to the bottom and reads the addressed I/O port.

These two are functionally the same but give different results.

One suggestion that has been made though the VCfed forums is that it’s a timing issue as the in a, (C) version takes a little longer.

I have a few things to try. I have a couple of other CF cards on order and they should arrive soon. I’ll try them to see if there is some issue with the timing of this card. I’ll also look to see if I can get a wait state in to the IN operation.

More news when I have it.


Retrochallenge 2017/4 – Interak CF card – Part 3

Things are a little weird as I suggested in my previous post.

I have the CF card installed and if I use the ROM monitor in the Interak to access it through a series of Port commands, I can see what appears to be sensible data. If, however I use a piece of Z80 assembly language, I never see the CF card become ready.

I have tried my own code and two other variants found on the web.

Let me show you what I mean. This snippet came from Scott Baker’s website…

1069:         	WAITDRQ:
1069: F5      	        PUSH    AF
106A:         	WAITDRQLP:
106A: DB67    	        in 	A,(I7)
106C: E608    	        AND 	08H
106E: FE08    	        cp 	08H
1070: 20F8    	        JR	NZ, WAITDRQLP
1072: F1      	        POP 	AF
1073: C9      	        RET

To use this, I have set up the sector and all other parameters needed for a disk read. If I run this routine, it never finishes. breaking the code at 0x106C shows that the accumulator is either 0x40 or 0x41. Which means “Drive ready” or “Drive ready + error”.

However, if I use “P 67” at the monitor, I get “0x58” which means “Drive ready + Drive seek complete + Data request ready”.

I can run from 0x106A to 106C again and again I see 0x40 or 0x41 in the accumulator. If I use “P67” again, again I see 0x58.

Can anyone see the deliberate mistake because I can’t 🙁

Retrochallenge 2017/4 – Interak CF card – Part 2

In part 1 I was able to build a CF card adapter for the Interak. Mostly this involved using RC2014 parts.

Now it’s time to tell you about the testing and things are pretty weird.

A bit of background.

At the moment, my Interak-1 has no mass storage at all. No hard drive or floppy. Nor does it have a serial port that I can connect to my PC.

It only has cassette, screen and keyboard. This is making testing challenging.

In an earlier post, I described a USB keyboard adapter that I had built during last year’s Retrochallenge. This gives me a USB keyboard. At the end of the post I mentioned adding USB serial in so that I could use a serial terminal emulator to “squirt” data at the Interak as if it had been typed in.

I did that mod and this is the method I am using for testing because no one wants to type in a couple of hundred bytes of hex manually that may well crash.

It has no handshaking and so I have to be careful not to send too much at once but it’s working well.

First steps.

I am following in Scott Baker’s footsteps a bit here. Scott has added CF to the RC2014 and it’s his board design I am using. He describes the process of setting up the card and reading a sector as follows…

read register 7 until the busy bit (0×80) is unset
write 1 to register 1
write 0xEF to register 7
write 0×82 to register 1
write 0xEF to register 7
0xEF is the “set feature” command. Feature 1 enables 8-bit mode. Feature 0×82 disables any write caching. You should do this whenever the compactflash is reset or power cycled.
Read a sector
read register 7 until the busy bit (0×80) is unset and the ready bit (0×40) is set
write 1 to register 2, to set the sector count to one sector
write bits 0..7 of the block address to register 3
write bits 8..15 of the block address to register 4
write bits 16..23 of the block address to register 5
take bits 24..27 of the block address, or them with 0xE0, and write to register 6
write 0×20 to register 7. This is the “read sectors” command.
read port I7 until the busy bit (0×80) is unset and the DRQ bit (0×08) is set
read 512 bytes from register 0

By using the PORT command (P) of Zymon 2 (the Interak’s ROM monitor) I can carry out the above instructions and it seems to work well.

Just a note. Zymon is a little quirky and it’s not the easiest thing to follow. I can’t put blank lines in for clarity, they get ignored. Also, the P command reads from (no parameter) or writes to (one parameter) an I/O port. However, if I type, for example “P 67” to read the status byte on the CF card, Zymon puts the byte on the same line making it impossible to see whether or not I wrote to the port or read from it.

In this photo I set up the card ending with the line “P 67 EF”. I then do the read (up to line “P 67 20”). I get the status with “P 67”, the result is “58” – Drive ready + Drive seek complete + Data request read.

The, the following “P 60″s show data being read from the first sector of the drive.


Now the weird. I have tried doing this using a bit of assembler code and it doesn’t work.

I’ll put the full details in my next post.

Retrochallenge 2017/4 – Interak CF card reader revisited – Part 1

Retrochallenge is well under way now and so it’s time I made a start and got something blogged about. Although my posts here don’t show a lot of activity, I do some work on one or other of my retro systems most weeks throughout the year but Retrochallenge is a good reminder to blog about some of this activity in the hope it may be useful to someone else or help to promote interest in our hobby.

Catching up.

A little while ago I tried to get a CF card adapter to work on my splendid little Interak-1. That didn’t work properly but an email conversation with a friend of mine who also has an Interak suggested that my cables were too long and that I needed to get the CF card as close to the CPU as possible.

So, my first Retrochallenge project this time around is to re-visit that project and try and get it to work.

In the previous attempt I used a CF card adapter intended for the RC2014 (itself a Retrochallenge project). In order to get that to talk to the Interak I needed to make a bus adapter and to do that I modified an Interak prototype board.

As I said  above, the suspicion is that the CF card was too far away from the CPU.

Lets build a Mezzanine.

One way to get really close to the CPU is to slip a PCB between the CPU and its original socket and use this new PCB to link to the CF card adapter. This is the approach I’m using here.

I still want to use the CF adapter for the RC2014 and so I still need to convert from my Z80 CPU to the RC2014 bus. What better way to do that than to use an RC2014 CPU card?

Spencer Owen of RC2014 had a small number of prototype CPU cards in his Tindie store and bought one of those, soldered a wire-wrap socket to it, plugged that into a conventional socket and on into the socket my CPU originally sat in. (This is going on a bit, see the pictures).

Now the Z80 sits on the RC2014 CPU board and the Interak CPU board with the CF adapter over to the side.