exxos's Atari, Amiga & retro forum

The chips have arrived... I have a display... It says LYNX and has a flashing cursor...

I had a while of it not working and a contended data bus. That was due to the BOM saying that there should be a 74LS273 in one socket and the schematic saying that it should be a 74LS373.

When I type PRINT it crashes.

I need to retest all the RAM just in case.

stephen_usher wrote: 23 Sep 2026 11:05 The chips have arrived... I have a display... It says LYNX and has a flashing cursor...

I had a while of it not working and a contended data bus. That was due to the BOM saying that there should be a 74LS273 in one socket and the schematic saying that it should be a 74LS373.

When I type PRINT it crashes.

I need to retest all the RAM just in case.

It wouldn't be fun if it was all plug and play :lol:

Hehe... Yeah.

Well, I'm back to the same RAM refresh issue I had on the 128K. It's not reliable enough for the first bank to reliably hold its data. I may have to look into the same sort of fix I did on the 128K.

Well, the RAM refresh has been "solved" by jumpering the main RAM to only 16K. It's likely that the system is relying upon the Z80 refresh which only does 128 rows.

The keyboard is still not responsive. You have to use ZX81 typing speed to get an accurate input.

OK, I'm stuck now without a reference system to compare to.

1. The video settings only ever work if jumpered for a 48K machine.
2. The machine only works semi-reliably if configured as a 48K machine (16K + 16K +16K) or an unsupported hybrid 96K (16K + 64K + 16K)
3. If the RAM is set to proper 96K mode (64K + 16K + 16K) the machine beeps twice on boot, still runs the video in 48K mode and crashes soon after as the RAM isn't being refreshed correctly. I have no idea if it should beep twice or not without a reference machine.
4. There's almost zero information on-line to work from.

*sigh*

Re: Refresh. If I can inject the Z80 /RFSH signal into /CAS before the system's /RAS goes low then the RAM should do a CAS-Before-RAS refresh. Hmmm...

When I upgraded mine to 96K, it was the left hand side bank that I replaced with the 64K.

I had to change two jumpers IIRC, one to remove the 12v and one to add the extra address line.

I'll try and set up mine over the next few days.

I have a plan!

Inject /RFSH into the /TCAS signal "directly" and use some unused gates on chips to do it.

Lift IC74 pin 3 (input from the /CAS generation stream)

Generate a combination of /CAS and CPU /RFSH

Wire IC49 pin 10 (/RFSH) -> IC57 pin 9 (74LS04 inverter)
Wire IC57 pin 8 (RFSH) -> IC 41 pin 12 (IC41 is a 74LS32, OR gates)
Wire IC76 pin 6 (CAS) -> IC 41 pin 13
Wire IC41 pin 11 (CAS or RFSH) -> IC57 pin13
Wire IC57 pin 12 -> IC74 pin 3 (/(CAS or RFSH))

Thopught up a simpler one using a piggybacked '08 but it didn't work. Probably too slow.

Tried injecting it right at the end of the chain but no go.

The keyboard problem is weird. If a key is pressed and it doesn't show as being pressed but it's held down the previous key press is repeated. This is even if the key is on a different scan row.