*** M E M O R U M ***

Don’t forget that the RAM bank switching needs the I/O system to be working okay. Although if BASIC is working, the bank switching for the ROMs must be working okay.

Before changing anything, I suggest a short machine code program be run that can actually test if the correct RAM banks can be switched into and out of the memory map.

Mark

STAR.COM works fine.

SPECTRON.{BAS,MTX,RUN} don't. They mostly get to the screen asking for which game mode to play and don't respond to the key press. They're also reluctant to acknowledge the key press at the "Press any key" stage.

Yes, works for me too.

I have not got Star.mtx on the CF that I am using at the moment. It was Star.run, running from SDX BASIC that has the problem for me

Could it be a CPU problem? The CPU is the latest generation of Z80 and isn't designated "A" or "B"

If it’s a genuine part, then it will likely be CMOS, but that’s not a problem.

We have seen lots of fake, allegedly high speed CMOS, parts that are not what they purport to be. But they are usually remarked NMOS parts, provided they started life as at least “A”s (4MHz), they should be OK too

These are all from Farnell.

Anyway, things have got interesting...

Replacing the CPU with one of the scavenged Z80Bs I friend gave me and CP/M was a little bit more stable. Key presses in Spectron.com weren't always picked up though. Also, sometimes CP/M wasn't able to read the CF reliably when doing a DIR, sometimes getting corrupt entries and sometimes saying incorrect filenames etc.

I then replaced the brand new Z80CTC with the old one and things got a lot better. Keyboard input happened without any problem. Spectron works fine from CP/M.

CP/M itself is a lot more stable but still often locks up after a couple of minutes.

The new CPU and CTCs are rated at 6MHz. Underclocking shouldn't be a problem should it?

I've used the 10mhz CMOS Z80 (Z84C0010PEG) in my MTX's for years, with no issues. In fact we found out it works better with the CFX2 than the NMOS version. The only reason one of them has the NMOS chip back in it is for testing.

As Dave says we've found a lot of fakes marked up as the 20mhz part (Z80C0020PEG). For safety I got both the 10mhz and 20mhz parts from an official distributor (Mouser) and no issues at all.

Though for the MTX there's no point in paying extra the faster part. For the adventurous, there is some headroom for over clocking the MTX but not enough to need the faster part. I ran one at 4.9152mhz for a while, even that was enough to upset the VDP with some games. Which reminds me, you could try under clocking yours to see if that helps the stability. If you have the 74S04 at 9D socketed you could do that with a daughter board putting a flip flop in between the oscillator circuit and the vavious phi drivers.

The CMOS version of the CTC also works as an NMOS replacement with no apparent issues.

Dave wrote: ↑14 May 2019 23:19 I wonder if there might be a problem with the DRAM timing circuit?

Assuming that all the chips in the circuit are working, it might be worth checking C5 and C6 for damage. With all the work on the machine, they might have taken a knock? I had a weird RAM fault that turned out to be a cracked capacitor

Both the capacitors C5 and C6 are in pristine condition.

Mmm, interesting, but no, underclocking from the rated max is not a problem.

stephen_usher wrote: ↑16 May 2019 20:40

Dave wrote: ↑14 May 2019 23:19 I wonder if there might be a problem with the DRAM timing circuit?

Assuming that all the chips in the circuit are working, it might be worth checking C5 and C6 for damage. With all the work on the machine, they might have taken a knock? I had a weird RAM fault that turned out to be a cracked capacitor

Both the capacitors C5 and C6 are in pristine condition.

Can you read the values on them?