Hey guys,

I managed to hit the 41 MHz mark right this minute with my J64C project!

The VIA is definitely a massive choking point I found. Previously maxing out at a bit over 25 MHz, with the VIA installed. I have removed it to see what I could get out of this board.

Unfortunately this kills off any real use for my little computer as it now has no access to the outside world, but I thought it would be fun still to see how far we can push the W65C02.

I am running a small loop, which is writing to an address in RAM, and running forever. So it isn't an extensive test by any stretch, but seems stable (barely!!) doing what it's doing. Look at it sideways though and it will crash at this speed.

Next test will be to over volt things a little and see what that does and also to replicate some of the functionality of the W65C22 VIA in to a CPLD, to regain some usability.

My poor oscilloscope is being driven to the edge too. Attempting to clock much higher than this makes the poor thing go nuts making it harder to perform measurements.

Amazing! I’d suggest running the Dormann test suite in a loop as that will exercise a rich instruction mix.

That’s a new record. I have replaced a 65C22 with CPLD so it can have serial port and CF disk interface. It is able to run reliably at 36Mhz and passed Dormann instruction test. I can also run at 40 Mhz, but I have to raise supply voltage to 5.3V

viewtopic.php?f=4&t=7433
Bill

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Not a bad idea. Maybe we should start a thread, listing the requirements of what it takes to qualify as a pass, ie. minimum up time, Dormann test, pass etc.

Mine ‘got’ to 41 MHz and appeared ok on the oscilloscope, but I wouldn’t be surprised at all if it failed an extensive test, so take the 41 with a grain of salt.

I’d like to redesign my board again too. I have learned a lot over the past few months and could hopefully make some decent improvements to signal integrity.

That's legendary!

I'd say that you still hold the crown, as I can't confirm yet if mine would pass extensive testing, due to no I/O. So, I was probably a bit quick with the 41 MHz claim. The excitement got to me seeing a 4 at the start of the clock speed.

How stable, remains to be seen.