exxos wrote: ↑Wed Apr 07, 2021 1:02 pm
Badwolf wrote: ↑Wed Apr 07, 2021 10:48 am
Yes! That's *got* to be the tipping point.
It seems TTRAM access for some reason happens too fast. When it's not working, TTRAM hits 903%, not 847%. So I think STERM is ever so slightly too fast. Inverting the 100mhz clock does the trick. What I may try is clocking STERM via a FF to delay it 10ns to see what happens. I assume worst case the score would drop. But if inverting the clock is enough to fix / break stuff, something is very borderline.
So what's next... 60mhz ...
Bear in mind the SDRAM is synchronous on the positive edge and the 030 on the negative. Inverted is probably the optimum, therefore as the data should be valid between the falling edge 1.5 cycles after CAS is latched and the falling edge 2.5 cycles after CAS is latched.
But AIUI your SDRAM is running at twice the CPU's frequency, so you're on your own!
BW
PS: my ceramic 030 (purchased just six months ago from your very boutique!) can't get beyond about 55MHz and I can barely keep the SDRAM in sync at 50!
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
DSTB1 Open source 16Mhz 68k and AltRAM accelerator for the ST
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark