exxos's Atari, Amiga & retro forum

@mikro it's more to show the stability of the clock patch, since joy is famous for being a good tester for that.

Steve wrote: 09 Jul 2024 19:21 @mikro it's more to show the stability of the clock patch, since joy is famous for being a good tester for that.

Yep that was my acid test for my bird.

Getting a decent Oscilloscope also helped me understand what was going on.

In my case replacing original (piggy backed chip) clock patch with a phantom clock patch (as similar to Exxos V2/V3 as @Steve mentioned previously).

Signals all look alot better - Joy Demo works beautifully.

One thing i didn't check is if input voltage makes any different, under load my bird reads 4.92v but if input voltage were to be set to say 5.1v under load would that make any difference, i only say that as i'm sure i read on the DFBX1 (or was it CT63!) thread that 5.1v on the falcons being used there was optimal and anything lower could cause issues, again, adds to the 'each Falcon is unique' argument.

jwd wrote: 10 Jul 2024 13:21 One thing i didn't check is if input voltage makes any different, under load my bird reads 4.92v but if input voltage were to be set to say 5.1v under load would that make any difference, i only say that as i'm sure i read on the DFBX1 (or was it CT63!) thread that 5.1v on the falcons being used there was optimal and anything lower could cause issues, again, adds to the 'each Falcon is unique' argument.

Your probably thinking of my known issues bit on the DFB1X https://exxosforum.co.uk/atari/last/DFB1X/index.htm

Voltage less than 5.00 can cause issues on like 50% of CPUs . But that's not know for stock falcons. But I can say , that higher or lower voltages can make or break a machine in general. If a machine malfunctions with variations in voltages, then something is clearly wrong with the machine.

exxos wrote: 08 Jul 2024 16:19 Across the 47pF cap.
...
So for some reason there has to be like 100pF extra somewhere to get those results.

Odd indeed since I don't even have that 47p cap!

That said, my grounding lead was connected all the way up at SCSI port. I'm afraid I've no idea where my little short ground springs are.

BW

Badwolf wrote: 10 Jul 2024 14:10 Odd indeed since I don't even have that 47p cap!

You have a V2 though ? There was a 100k pulldown and maybe a series resistor, but I didn't take things any further as I moved to the V4. For the V2 something like 22pf may be better.

exxos wrote: 09 Jul 2024 18:45 Atari Falcon running JOY demo with exxos clock patch V4 and DFB1X accelerator.

Nice! This will be my test too. I'm wondering something: it's said that it's because of the intensive access to the Blitter that the Joy demo is the ultimate test to check whether a clockpatch (or a better clock patch) should be installed or not. But the Blitter is integrated into the Combel, and the Combel generates the clock signal. How can it be corrupted?

About the V4 clock patch installation, I have a small question, as it is not addressed in the installation instruction:

My Falcon had the first modification made by Atari: remove C208 from the back of the motherboard and put a 0 Ohm on R222. My question is: should I put C208 back?

Looks like that capacitor and resistor were put there to try to stop ringing.

In fact, I think it's the opposite: C208 was present from the start,
and Atari's first attempt to correct it was to remove it. ¨


Now the question is: with the V4, does I have to fit C208 back?

C208 is fitted on my Falcon. Would seem a bit odd to remove it really. Though it's only 10pF and the V4 has 22pF on its input anyway. So it's probably better to have it yes, but it's probably not going to matter much on the V4 anyway, as its got its own input buffer to cope with the crap clock input anyway.

Perfect

Many thanks!