Offline
NUMBSKULL

Which link cable are you using?

Did you try the MIDI out link I posted?

Last edited by catskull (Nov 12, 2015 4:09 am)

Offline
Montreal, Quebec

Just tried it. Everything seems to check out OK.

Offline
Montreal, Quebec

also using a Pokemon Yellow Link cable, cut in half. Splits to a Y to fit DMG and GBC ports. GBC side has all 6 conductors, Dmg side has conductors at pins 2,3,5 and 6.

| 6 4 2 |
\ 5 3 1/  <-----a diagram looking at the link cable

Offline
Montreal, Quebec

thinking of installing the Aref potentiometer on the freeduino, manual says it can increase analog sensitivity. good/bad idea? only remaining options are trying the DMG with brand spanking new batteries, with a DC adapter, or a different DMG altogether.

Offline
Montreal, Quebec

just finished another round of testing. still on the fritz. at this point I've tried two different DMGs, three different Arduinos, made sure the Link cable is working, ran the MIDI test sketch with no problems, replaced all my LEDs and resistors, what else could possibly be wrong? running notes in LSDJ on MIDIOUT mode with N command still spits out nonsense. it's almost as if the bits are being masked or that there's a sync problem. it's hard to tell whether this is the case, I guess I could check every byte received and see if they share a set or clear bit in common. it just seems like a huge pain in the ass right now, and I 'm not sure what that would accomplish. I'm almost convinced it's the way the arduino sketch is interpreting the signals from the DMG.

and now I have no idea what to try next. somebody. please help.

Offline
Toronto, Ontario, Canada

The remaining things I could imagine it being:

-Link cable (even if you tested it with a multi-meter it's possible there's enough cross-talk to ruin the signal)
-The arduino sketch you're using (ensure it's the correct, up-to-date version)
-Your Arduino Software (ensure it's the correct, up-to-date version)
-The cartridge itself (drawing too much power or somehow causing some form of crosstalk?)
-The MIDI gear (if the MIDI output is slightly out of spec it could explain why it works for many people but not you)

Those are all sort of long shots, though.

Offline
NUMBSKULL

Are you absolutely sure your resistors are the correct values? I recently accidentally used some wrong values and I had erratic behavior. Midi out worked, but it sent nonsense.

I agree it's a long shot.

Offline
Montreal, Quebec

the resistors idea is interesting, but I've pretty much confirmed that the arduinoboy hardware is fine because the MIDI test sketch you sent me worked properly. I've looked up crosstalk and it seems arduino  occasionally has this problem. why it wold consistently happen to me and no one else I can't figure out but it seems that these inaccuracies can be solved by going into the sketch and taking multiple samples at Analog In and then averaging them, which I'm gussing the arduino can do even many times while the GB is sending one serial bit. this is going to be Fun, as I have very limited experience programming the arduino. but the plan is simple. find all instances in the sketch of analog reads, and apply some kind of multiple checking and average the results. but if that's not the problem  I'll have done it for nothing and still not know what the problem is...or why it went wrong for me in the first place


one thing I did notice is that one of the conductors of the Link cable measures 1~2 ohms while the rest read zero.  I want to try this out with a different link cable but I'm hesitant to cut up my only other good one so I'm looking for an F-1 Race Quad splitter to cannibalize for the port. does that port have all the conductors I need?

also I was thinking what if the analog ins were made #1, #3 and #5 or #1, #4 and #6 so the inputs were more spread apart but they still run pretty close on the PCB so that probably wouldn't make much o a difference.

Last edited by Roni (Dec 11, 2015 6:32 pm)