Weird 80 Column Switching Behaviour

24 posts / 0 new
Last post
Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Weird 80 Column Switching Behaviour

I'm hoping someone can help me with the weird behaviour of the below clone 80 Column card on my Apple II+. This card has onboard switching, and a pass through for the composite output of the Apple II motherboard. But when I enter the command PR#3, the cursor disappears, but the output doesn't switch. Clearly the 80 column card is working, because the beeps become a lower tone and if I press "CTRL-Z,1,RETURN" the cursor returns to the 40 column screen.While in hidden 80 Column mode, if I type PR#6 to run a program from the floppy drive (I'm typing to the 80 column screen that I can't see) when the program starts the card then switches to 80 column mode even though the software is running on the now hidden 40 column screen.

It seems like the screen switching is backward to the way it's supposed to work. I've tested every IC on the 80 column card using my Minipro programmer, and I've tried swapping out the graphics chip for a new unit, so I feel like the problem might be with the Apple II+ motherboard. Does anyone know how the switching signal gets to the card so I can look for a problem on the motherboard? Obviously it comes through the 50pin edge connector, but what signal is used?

 

Online
Last seen: 21 min 56 sec ago
Joined: Jun 18 2010 - 13:54
Posts: 754
Why do you think it's just a

Why do you think it's just a switching problem? Seems like the 80 col mode is just not working. I'd start by looking at the crystal oscillator on the left. Perhaps it is not working.

S.Elliott's picture
Online
Last seen: 48 min 37 sec ago
Joined: Jun 23 2022 - 16:26
Posts: 211
More resources at APPLE2.ORG

When the card switches to 80 columns after booting a floppy disk, what do you see on the screen?

Does it show the characters you typed before booting the floppy, such as Applesoft's "]" prompt and your "PR#6" command?

 

Does anyone know how the switching signal gets to the card so I can look for a problem on the motherboard?

On 80-column cards that have onboard switching, they use one of the IO ports assigned to slot 3 to switch between Apple's composite video and the 80-column video.  But the specific port varies by brand:

  • Applied Engineering uses IO address $C0B2 in their ViewMaster 80.  So a BASIC program would switch to Apple video (40-col) with the command POKE49330,0
  • MicroMax uses IO address $C0B3 in their ViewMax 80.  So a BASIC program would switch to Apple video (40-col) with the command POKE49331,0
  • MicroTek uses IO address $C0b2 in their Magnum-80.  So a BASIC program would switch to Apple video (40-col) with the command POKE49330,0

 

Do you know which brand your card is?  If not, try searching in the 80 Column Cards section at APPLE2.ORG and see if you can identify it -- most of the entries are accompanied by photos, and many include manuals.  (Including those 3 links for ViewMaster 80, ViewMax 80, and Magnum-80 above.)

CVT
CVT's picture
Offline
Last seen: 1 day 4 hours ago
Joined: Aug 9 2022 - 00:48
Posts: 1049
I agree with Jeff that most

I agree with Jeff that most likely the card is simply not working and this is not a switching problem. The <Ctrl><G> beep is generated by the CPU of the Apple II and it will beep even if you remove the MC6845P chip from the card. The fact that it beeps simply means that the card's firmware stored in the ROM chip is getting mapped and executed correctly by the Apple II CPU.

 

I suggest to take a look at this very recent topic and start from there:

 

https://www.applefritter.com/content/videx-videoterm-clone-no-display-output-and-no-beep-help

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Thanks for the replies. I'm

Thanks for the replies. I'm sure the card is working in 80 Column mode, because it does display an 80 column screen when I boot a floppy if I've already initialised it by typing PR#3. As S.Elliott said, I see the commands I typed after starting the card but it didn't switch screens - PR#6 etc.

It seems like it switces at the opposite time to when it should - when it should switch back to 40col it goes to 80.

The card came in a box branded "Redstone" and is made in Taiwan, but I can't find any info searching that brand. It seems like it's a generic sort of card, I've seen virtually identical units on eBay and elsewhere but they all seem to be anonymous. I've read the firmware EPROM but there is no identifying info.

The command CTRL-Z,1,ENTER switches it off after initialisation, so the firmware might be similar to a Viewmax-80 card, although they don't look the same.

Something else strange just happened when I tried the card again. With or without it installed into the machine, when I type PR#3 the floppy drive boots (I'm using an Apple DiskII card in slot 6 with a floppy emulator). Maybe the motherboard has an issue with slot addressing?

Online
Last seen: 21 min 56 sec ago
Joined: Jun 18 2010 - 13:54
Posts: 754
Ah OK. I wasn't sure from

Ah OK. I wasn't sure from your original post that the 80 col display was actually working. Now it seems that your observation is correct. If it is indeed switching backward, I'd check out the transistors on the left edge of the card. They should be performing the actual video switching and if one of those is shorted it might cause exactly that.

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Thanks for the suggestion, I

Thanks for the suggestion, I'll check the transistors and maybe just replace the 3 of them.

Something curious just happened when I tried the card again just now. I didn't realise I hadn't put it back into the machine after I took the photo and when I typed PR#3, the floppy emulator which is connected to a DiskII card in slot 6 booted. I put the 80-col card back and tried again and the same thing happened. Maybe there's a problem with one or more of the ICs on the motherboard causing slot addressing issues?

Online
Last seen: 21 min 56 sec ago
Joined: Jun 18 2010 - 13:54
Posts: 754
That's definitely possible.

That's definitely possible. Or it could also be the disk controller card. Try taking that out and putting the 80 col card back in. See if it works in that configuration. 

S.Elliott's picture
Online
Last seen: 48 min 37 sec ago
Joined: Jun 23 2022 - 16:26
Posts: 211
80-column cards initially

80-column cards initially used two 4013 chips; one 4013 managed the expansion ROM and cursor-size, and the other 4013 stored two bits worth of DEVSEL soft switches for accessing video memory pages.  Your card has one 4013 and one 4042, instead of the usual configuration with two 4013's.

It's likely they changed the second 4013 to a 4042 in order to store  four bits worth of DEVSEL softswitches -- one of those bits is probably the 40/80 display switch.  When the card detects the slot-3 DEVSEL signal (pin 41), it latches bits from the address bus in the 4013 or 4042.  (That's right, these switches are controlled entirely via the address bus, not the data bus.)

If you want to try troubleshooting the soft-switches, here's a BASIC command that will initialize the 80-column hardware but keep screen IO in 40-columns.  You must begin this command with a colon, exactly as shown, to ensure the command isn't parsed by DOS or ProDOS.  (Before using any DOS/ProDOS comands you'll need to RESET the computer but you won't need to reboot.)

:PR#3:?"HELLO":PR#0:IN#0

 

To test all sixteen possible soft-switches, go to the Monitor after the command above and try each of the possible soft-switch addresses one-by-one.  For example, type C0B0 and press Return.  If that has no effect, then type C0B1 and press Return.  Try all sixteen values C0B0 through C0BF if necessary.

 

PS: The screen capture includes a C058 soft-switch, the one used by Videx cards, but it probably won't have any effect with your 80-column card.

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Thanks so much for the

Thanks so much for the insight into how the switching works and how to test it. As it turns out, none of the addresses caused the card to switch, all just returned the memory value of that address.

But I had another thought - I have a replica Rev.0 motherboard that I recently built using Mike Willegal's PCB artwork, and when I tried the 80-Column card in this board it works as it should. The Rev.0 board has Integer Basic ROMs so it doesn't like your command beginning with the colon, but just typing PR#3 switches to an 80 Column screen and CTRL-Z,1,RETURN goes back to 40.

So I assume the problem is with the II+ motherboard, but I'm not sure where to look. I did try swapping out the 74LS138 ICs at H2 and H12, but that didn't help.

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
After doing some testing in

After doing some testing in the Rev.0 clone board where this card works, it seems the address C059 switches to 80 column mode and C058 switches to 40 column. However I can't get it to switch at all in my II+. What could cause the soft switching to work in one board and not another? 

CVT
CVT's picture
Offline
Last seen: 1 day 4 hours ago
Joined: Aug 9 2022 - 00:48
Posts: 1049
But you mentioned that your

But you mentioned that your Apple II+ has an issue with the slot addressing, like you would type PR#x and the device in SLOT y would boot, instead of the one in SLOT x. Is this still the case?

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
It only did this once and but

It only did this once and but hasn't done it again. Yesterday while trying to replicate it I tried slots with no card in them and instead of just hanging with no action, the machine seemed to crash with random characters on the screen. I'm thinking that there is something not quite right with the interface to the slots and that's why the switching signal is not getting to the 80 column card, but not sure what. The fact that unusual things seem to happen with the slots could be a clue.

 I've tried pulling every logic IC out of the board and testing them, but they all passed. The RAM and ROMs all pass with test software, and I've tried swapping out the processor but it made no difference. There aren't many other options, only a handful of ICs that I can't test like the 8304.

Offline
Last seen: 1 month 3 weeks ago
Joined: Jul 5 2018 - 09:44
Posts: 2587
I've had weird behavior like

I've had weird behavior like this when one of the 8T98 tri state bus drivers was bad or on motherboards where someone has substituted a 74LS367 for one or more of them.  The machine will often appear to work with no cards installed or even one card like a Disk ][ Controller, but add more cards and it starts getting flaky and weird things happen.  The chips are both tri-state bus drivers but the 74LS part can't handle the loads and maintain switching speed.

 

There are a number of other things that can cause it...  this is just one I've seen before.

 

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Solved!

Thanks for the suggestion! I just swapped the 3x 8T97s from my Rev.0 clone and the II+ board now works. The 80 Column card switches correctly and the all the slots are behaving themselves.

This was tricky to troubleshoot, because as you said, it can appear to work correctly but do weird things at times.

Offline
Last seen: 1 month 3 weeks ago
Joined: Jul 5 2018 - 09:44
Posts: 2587
Glad that helped.  I remember

Glad that helped.  I remember the first time I had trouble with those was with a Rev 4 (if I remember right) motherboard that I got at a swap meet that had at least one 74LS367 in it intstead of the proper 8T98 chips and it was a royal pain to diagnose because of the weird behaviors.  The next time I ran into the probem with a bad 8T98 it was much easier to figure out.  I did find that one of the other 74 family chips would work reliably, but definitely not S or LS.  I think it was 74H367 or maybe 74F367.  Those versions are actually usually harder to find than 8T98 though, especially these days.  Back then you could get just about any of those chips because they were either still in production or only recently discontinued.  That's not the world we live in anymore.

 

CVT
CVT's picture
Offline
Last seen: 1 day 4 hours ago
Joined: Aug 9 2022 - 00:48
Posts: 1049
softwarejanitor wrote:...
softwarejanitor wrote:

... 

That's not the world we live in anymore.

 

If that is indeed the case, I would like to propose the bulgarian made MC6887 as another alternative. It was licensed from Motorola and mass produced in the 80s for the Pravetz 82 (a 1:1 clone of the Apple II+) and they must still have a huge stockpile, since I see it all the time in electronics stores in Bulgaria for the equivalent of 25 US cents: https://vikiwat.com/integrated-circuit-cm6887.html

 

Offline
Last seen: 1 month 3 weeks ago
Joined: Jul 5 2018 - 09:44
Posts: 2587
Here in the US, a lot of the

Here in the US, a lot of the less common logic chips are often hard to find and extremely expensive if you can.  8T98 for example on eBay are $2-$10 each, and they're relatively easy to find compared to some other chips.  But if you go to the mainstream electronics parts vendors like Mouser or Digi-Key they usually don't have most of those kind of parts at all anymore.  That's the world we live in.  Unfortunately due to historical trade issues between the west and former Soviet oppressed countries like Bulgaria the chips you speak of that were made over there are basically unheard of and not sold over here.  I know that they made a lot of clones and equivalents of US parts but there isn't a great source of information that I know of about what the equivaliencies are or how to source them outside of eastern Europe.  It might be a business opportunity for someone if they could put that together.

 

Online
Last seen: 10 min 45 sec ago
Joined: Feb 27 2021 - 18:59
Posts: 521
Jameco

Mouser and Digikey would not be my first stops for replacing chips in 40 year old equipment. There are suppliers like Jameco and Unicorn who specialize in this kind of obsolete component.

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
Fortunately here in Melbourne

Fortunately here in Melbourne, Australia there is an electronic parts dealer that's been around since the 80s, and they have lots of old stock parts that you can't easily get anymore. I've bought ICs from them that had mid-1970s date codes.

CVT
CVT's picture
Offline
Last seen: 1 day 4 hours ago
Joined: Aug 9 2022 - 00:48
Posts: 1049
softwarejanitor wrote:...I
softwarejanitor wrote:

...

I know that they made a lot of clones and equivalents of US parts but there isn't a great source of information that I know of about what the equivaliencies are or how to source them outside of eastern Europe.  It might be a business opportunity for someone if they could put that together.

 

I think just the awareness that an entire mirror electronics universe existed might be enough for someone who needs a particular chip to find it and source it. Shipping from countries from the former eastern bloc to anywhere in the world is not a problem and it's actually super cheap due to a 150 year old international agreement. Here is another example, already sold out though: https://www.ebay.at/itm/310218207518

 

Offline
Last seen: 22 hours 37 min ago
Joined: Apr 26 2016 - 08:36
Posts: 699
Silicon-Surfer wrote:Thanks
Silicon-Surfer wrote:

Thanks for the suggestion! I just swapped the 3x 8T97s from my Rev.0 clone and the II+ board now works. The 80 Column card switches correctly and the all the slots are behaving themselves.

This was tricky to troubleshoot, because as you said, it can appear to work correctly but do weird things at times.

 

You've also already discovered that the MiniPro II is not a reliable way to test ICs.

 

 

Offline
Last seen: 1 month 3 weeks ago
Joined: Jul 5 2018 - 09:44
Posts: 2587
baldrick wrote:Silicon-Surfer
baldrick wrote:
Silicon-Surfer wrote:

Thanks for the suggestion! I just swapped the 3x 8T97s from my Rev.0 clone and the II+ board now works. The 80 Column card switches correctly and the all the slots are behaving themselves.

This was tricky to troubleshoot, because as you said, it can appear to work correctly but do weird things at times.

 

You'

 

 

I'm one of the (apparently few) people who love the MiniPro TL866 (I have both the original and the II version) for programming chips...  But I don't even try to use it for testing chips at all.

 

Offline
Last seen: 2 months 2 days ago
Joined: Apr 8 2022 - 22:25
Posts: 20
I'm sure the MiniPro isn't

I'm sure the MiniPro isn't the best way to test ICs, although I hadn't tested the 8T97s with it. They're not on the list of chips it can test, but I did check all the 74xxx logic ICs on the board and they do all seem to be OK. However after removing the 8T97s because they were obviously flaky, I tested them as 74367s and they passed, so it can't pickup a chip that's not completely failed obviously.

Log in or register to post comments