exxos's Atari, Amiga & retro forum

@stephen_usher Good luck for today,I hope it all goes great :)

@rubber_jonnie that looks interesting but I'd not have a clue what to do with it. But ott just gor a clock.

BTW, apologies I didn't make it yesterday, glad to see you had a good day though.

Today went well.

This was the Atari display:


Lots of interest about the TT and Spectre GCR and a few about the CPU switcher/TF536.

This was yesterday's display.


Now I'm going to have a relaxing evening and an early night in the hotel and drive home after rush hour tomorrow.

Home from Cambridge and everything dumped out of the car and into the living room.

Time to have a little sit down before the big push to move stuff into better locations. Knackered.

stephen_usher wrote: 10 Oct 2021 17:40 Now I'm going to have a relaxing evening and an early night in the hotel and drive home after rush hour tomorrow.

Very good, Stephen.

Good security arrangements too: if anyone looks like they might pinch any of your gear, you just beat them to a pulp with the MTX500! :lol:

BW

Hehe.

Well, tonight's escapade has been finishing putting all the exhibition stuff away and then looking at the Spectre GCR cartridge to see why it doesn't work on the TT.

Well, there were a few very poorly done bodges on the board, reverting it to stock from what looks like some sort of test using a 74LS04 piggy backed on another of the chips but only connected by VCC and GND with a resistor and capacitor connected to various legs. Three traces on the board were cut and there are blobs on some other chip legs. I'm wondering if it was a delay for a pulse. The bodge wires on top were great loops of wire, one soldered to a pin on a chip and the other end to the cut trace. I had to fix it as one wire had come free. The bodge wire on the underneath was less poorly done but it was still very long.

Anyway, I removed the two top bodge wires and replaced them with a couple of neat ones underneath. I also noticed that the ground plane around the outside of the PCB wasn't actually connected to ground, so I fixed that.

Still, it still doesn't work on the TT. The pulses from the chips look to be doing their thing. I'm wondering if the Mac ROMs are too slow and aren't being read correctly so the "Mac" is crashing with corrupt images in memory.

P.S. I did notice that the two Mac ROMs I have seem mismatched. One looks to be a rev B and the other a rev C. Maybe this is causing a crash on the TT as it's using code that is changed which the ST doesn't use.

I had a look at the Spectre GCR cartridge again this evening, more out of interest with regards to what the TT's address bus looks like at the cartridge port and other voltages.

Well, the +5V is flat with just a small amount of spiky noise, so that's one thing ticked off any possible problem with this cartridge and the NetUSBee. The voltage is very slightly high at 5.04V but what's 4 hundreths of a volt between friends?

However, the address bus is more interesting, and seeing all cartridge stuff uses the "write using reads" trick this may be playing a part in the problems. This is a typical trace from any of the address lines:


As you can see, the address lines are ringing quite badly. The bus really could do with some terminating resistors to damp the oscillation, possibly in combination with some small value capacitance to kill the high frequencies.

Interesting it's like there is a extra length of wire on the bus


Tom

The data lines are worse...

Is this at the cart port or on the cart itself then? Does it work in an ST? I'm not familiar with the TT, is it like the Mega STE in that clock speed is adjustable, and if so does a slower clock speed make a difference?