Config Menu

[Atari8guy]

Upon installing this device, I got into the config menu once...since then if I type the SYS 54301 I just get a blank screen and the READY prompt... is that expected?

[frntc]

No, of course not :-) How did you install it (additional address lines?)? Can you hear the SID when you start a program (never entering the config menu)?

[Atari8guy]

no additional address lines. I just plugged it in. the sound works and the mouse works in GEOS, which was my reason for getting it to begin with, so I'm happy and it's working fine as far as I can tell, but I can't get back into the config menu...I downloaded the firmware from here this morning, and as I said, it worked the first time

[frntc]

No real idea what went wrong. Have you tried flashing the firmware again? (maybe with slightly lower/higher bus timings, via SKpicopatch)

[Atari8guy]

I tried several different timings, i was able to get into the config menu once, but not repeatedly. The menu was somewhat garbled but I was able to work through it? I'm also sort of at a loss....I checked all the solder points and reflowed a couple of them, even though they looked okay. No difference.

[frntc]

Usually such problems only occur when the timings are bad (the next firmware will include some minimal diagnostics...).

If your SKpico-PCB is ok, and other timings did not help... do you have another Pico to test? Maybe writing to the flash doesn't work? (that's what the config menu does)

[Atari8guy]

I have another pico w but it has no headers on it. I suppose I could source another one. In the mean time I'll give a few more variations on the timing a try - I read through your stuff, you think lower is better?

[frntc]

I made a brute force timing diagnostic some time ago, which was inconclusive. But people who ran into timing problems reported that lower values solved their issues.

[Atari8guy]

Just to give the last bit of information I can share. None of the timing changes, up or down, really made a difference in behaviour that I could tell. I was able after a reflash to get back into the menu one more time, it was garbled again, but I was able to set the SID to 6581, and leave the second one disabled. I was NOT able to change the timing to NTSC (my machine is NTSC) it stayed on DREAN (I don't even know what that is). However, once I rebooted the machine I again had no access to the config menu, but outside of that, everything works, SID tunes, demos, and the mouse in GEOS. I'm not likely to play with it again for a while as aside from the menu access it seems to work and my vacation time is over, I will watch though, in case you do track down the bug or release new firmware. Thanks for quick responses, even though we didn't get it quite solved.

[frntc]

The just updated firmware includes a small diagnostic routine for timings -- you might want to give it a try. Apart from that, it might still be the case that it's a hardware issue (flash etc.).

[W0RMinator]

I just wanted to mention that I had the same issue with one of the sidkick pico's that I made. Was not able to change much in the settings and it was stuck on DREAN. Some text was garbled. Like Atari8guy mentioned, sound tests worked fine. I tried reinstalling firmware multiple times and no changes. I've narrowed it down to a hardware issue with the raspberry pi pico board (official board). I desoldered it from the sidkick board and soldered another pi pico board (official) and it worked. Not quite sure what's wrong with the pi pico board but it definitely has some sort of failure. Just posting this in case anyone else has the same issue.

[SX64man]

I have seen the same exact timing issue (thinking it is an Argentinian Drean C64 instead of NTSC) on two of my 5 NTSC machines that have SKPicos with the stereo setup. My several PAL machines appear to be more forgiving and have no issues with the default 15,11 f/w timing settings. However, there appear to be subtle variations of the combined hardware which create a borderline timing condition on some NTSC machines. I have tried to pick up these subtle variations on my scope but it has been inconclusive so far. All SK boards were made from scratch with identical components. The only workaround I have found is to reflash f/w with various combinations of 15/11/100, 13/9/100, 11/7/100, 10/6/100 in the batch file until one appears to work without channel clicks or dropouts, menu corruption, or error message. The 100 setting is to pass the diag test (a setting of 20 offset did not work for me with my test harness). The 100 appears to work on several machines so far and I plan to flash the other 5 to see if all the diag runs pass. All my boards are h/w rev 0.1 with FW0.211
FRNTC, I put the SKpicotiming.prg util as one of the PRG entries in the F7 menu and that comes in handy. It would be awesome if this utility had the ability to read the current value in the firmware, display it, and then give you the option to write new values, if possible. It would make it easier to test and document the best timing values for a specific C64. Thanks for the great work!

[MarkMccabe]

I have the same issues on some of the ones I built and tested in my NTSC 326298 5 pin video mainboard. It comes up with the menu garbled, thinks it a Drean machine (which is a PAL system), and it will not save changes made. Funny thing is it functions fine, plays music and passes the diagnostic cartridge test (both the SID, sound, and control port checks for the paddles) perfectly.
It's funny because on the ones I built where the menu comes up fine (sometimes after some real time timing tweaks, I hav
e yet to get the paddles tp pass the diagnostic tests... Confusing...

[frntc]

The Drean vs. NTSC detection might be an issue of the VIC-type-detection routine. As I don't have a 326298-NTSC nor a Drean machine the routine might have a bug (which I either did not test in Vice, or didn't pop up there).

The garbled/non-starting menu is a matter of bus timing values.