search

randyrossi / vicii-kawari

Commodore 64 VIC-II 6567/6569 Replacement Project
GNU General Public License v3.0
170 stars 21 forks source link

Orignal & clone REU fail testing even with NTSC motherboard clock. #26

Closed gillham closed 11 months ago

gillham commented 11 months ago

First let me say this is a great project/product and it definitely looks amazing on my (formerly) 6567R56A machine. Thanks!

I have a VICII-KAWARI Trion Mini in a 250407. It has firmware 1.16. I set it for motherboard clock on NTSC. A 1541 Ultimate 2+L acting as an REU seems to work fine. Neither an original REU nor a clone using an original REC work. They both appear to pass testing for the first bank and then immediately fail on the 2nd bank.

These REU units test good in other C64 / C64C machines. And I believe I have used them both in this same machine prior to adding the Kawari. I can certainly confirm that at some point by swapping back.

Anyway, I see the REU is listed as "untested" but thought to work with motherboard clock. It appears to almost work but isn't.
Is there anything diagnostics wise that I can collect or timing adjustments etc that I can make that might help?

randyrossi commented 11 months ago

I also thought it was working as some people reported it to work with the motherboard clock. But I think the problem with the 407 clock circuit is that it generates the wrong frequency for dot clock (12mhz instead of ~8) when a real VIC-II isn't plugged in. The 466 doesn't appear to have this problem. It's something strange with that clock circuit I never figured out. There is a way to get the Kawari to output a dot clock which you can feed back into the motherboard though. It's been done by some users and requires running a line from the kawari into the motherboard + firmware update. I will have to dig up the instructions though. What revision Mini do you have?

On Thu, Nov 16, 2023 at 12:01 PM gillham @.***> wrote:

First let me say this is a great project/product and it definitely looks amazing on my (formerly) 6567R56A machine. Thanks!

I have a VICII-KAWARI Trion Mini in a 250407. It has firmware 1.16. I set it for motherboard clock on NTSC. A 1541 Ultimate 2+L acting as an REU seems to work fine. Neither an original REU nor a clone using an original REC work. They both appear to pass testing for the first bank and then immediately fail on the 2nd bank.

These REU units test good in other C64 / C64C machines. And I believe I have used them both in this same machine prior to adding the Kawari. I can certainly confirm that at some point by swapping back.

Anyway, I see the REU is listed as "untested" but thought to work with motherboard clock. It appears to almost work but isn't. Is there anything diagnostics wise that I can collect or timing adjustments etc that I can make that might help?

— Reply to this email directly, view it on GitHub https://github.com/randyrossi/vicii-kawari/issues/26, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAI3HKEEDR2RBZWFUAWREFTYEZBFTAVCNFSM6AAAAAA7OPPG22VHI2DSMVQWIX3LMV43ASLTON2WKOZRHE4TOMZSGM2DINA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Randy Rossi

gillham commented 11 months ago

It says rev1.5 2023.

randyrossi commented 11 months ago

Okay, that makes things easier. Before I describe the mod, I should mention that I don't know for sure if the dot clock is responsible for your REU issue. You might want to first reach out to the discord channel to see if there are others who have an REU working with the Kawari without the mod. I don't have one myself and I never tested one. It does make sense that it's not entirely stable due to the dot clock though. In any case, if you do go through with the mod, you will get a machine that will have a working dot clock reaching pin 6 of the cartridge port for both NTSC and PAL. And it can always be un-done fairly easily.

On the Mini, there is a pad at the top labeled CLK. You can flash the special build for the dot clock mod (v1.5 link) available here to activate it:

https://github.com/randyrossi/vicii-kawari/blob/main/doc/FIRMWARE.md#specialty-builds

Then take a look at this video : https://www.youtube.com/watch?v=ONgFPn0k2sc

This describes where to wire that CLK pad to. In your case, since you have a more recent board rev, you don't have to take the signal off the 74LS06D like in the video. There is a dedicated pad that already has the 33ohm resistor so you don't have to add one like the video. So all you should have to do is add a wire from the CLK pad down to pin 21 of the kawari like he does in the video and remove FB17 (or lift one leg up).

On Thu, Nov 16, 2023 at 6:50 PM gillham @.***> wrote:

It says rev1.5 2023.

— Reply to this email directly, view it on GitHub https://github.com/randyrossi/vicii-kawari/issues/26#issuecomment-1815509782, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAI3HKA3A2UVL66POE5IQ4LYE2RETAVCNFSM6AAAAAA7OPPG22VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJVGUYDSNZYGI . You are receiving this because you commented.Message ID: @.***>

-- Randy Rossi

gillham commented 11 months ago

Thanks Randy! After a little struggling I got this working well. Basically the first time I flashed it, even though it said copy/verify/ok for each, it wouldn't boot. A few metal tools later I had a screwdriver that actually shorted JP2 for the fallback boot. I don't know if it was the SD2IEC or that I had it set for motherboard clock still when I first powered it up after flashing.

Anyway I flashed it again and ran the verify and then it booted. My clone & original REU both appear to work fine. The 1764 demos work and the test passes. C64 OS v1.05 worked great with the Fast App Switching (via the REU) as well. I tested it for 30-45 minutes. I need to test some games that use the REU like Sam's Journey or Sonic and leave it running in demo or similar for a number of hours.

Oh and it is pin 22 per the video and schematic. He says "sorry I kept saying pin 21 in my last video." 😀

Thanks for the help! I'll close this issue for now.