Looking at assembly with 80c49 instructions from datasheet, it seems this 80c49 sends data towards that second 80c49 even in this cut down version c9d. So the 80c49 rom content is likely same as in c9c.
outl p1,a
outl p2,a # this is towards that second 80c49
Now I find out that some Japanese guy has made windows pc side emulation software+patch for Thunder storm+ some kind of vhd bios to run the game in real victor hc95 computer (but no real vhd player or vhd interface card, just RS232 connection towards windows pc). http://d4.princess.ne.jp/msx/gomi/thunderstorm.htm https://www.youtube.com/watch?v=bBMQJyDl9No
Only the VHD video and game cassette must be dumped itself, it is not in that web page.
I now made minor correction to c9d circuit diagram and also added the c9c specific circuits to schematics (of course resistor and capacitor values are not know what those would be). The components market as Kxxx in real pcb are missing from schematics. Every vhd pin comes through this Kxxx (three pin component where middle pin is in ground and signal goes through from other end to other end, no resistor value measured, so coil and capacitor to ground in one package?). I do not know where to get those symbols for kicad.
Also added pcb footprints, but no final pcb layout yet. https://github.com/popolonx/MSX_STUFF/blob/main/vhd_c9d_card...
I have tried to look at the asm code for 8049, and also found out simulator for it https://www.intel-vintage.info/intelsofware.htm https://docs.google.com/open?id=0B9rh9tVI0J5mYmNhZGM4ZTgtNzl...
But simulator needs asm as input, not binary file. And asm shuold be correctly reproduced. One problem is that Dasmx does not know to properly disassemble the code. With threading it skips many code parts and without it finds too much.
By the way, does somebody know how to read the movx command usage?
mov r1, #3Eh
mov a,@r1
...
movx a, @r1
As that movx should be reading data from bus (data from 8255 PA lines), but how does this command works?
What is this address 3E content used for? Does it have any effect on bus read, so value in 3E can have any value? Or how does it work? In 8048 manual this was not explained properly and now signal flow was shown.
Anyway it seems commands get from msx side are read during timer interrupts(but only if address 3E does not have bit 7 set, but bit 7 is set during startup) and thus also sent towards vhd player. So maybe the port datalines connected to vhd pins are broken and that is why no activity is seen for output pin(dsub15-pin4). Or that also the interrupt line shuold be connected to right vhd pin to get proper data, but that is not changing address 3E data(but all vhd lines data is read).
I tried card without 8049 and _VP commands game i/o error after some seconds, so the 8049 was working towards 8255. I have some used 8749 coming from ebay, so can then test if the port was broken or not.
I received some 8749 chips, and programmed one. Then soldered socket to c9d card and put this 8749 into socket. At least the _VP command now behaves same as before. Next i will logic analyse the vhd lines to check if anything changed for given _VP basic command variants.
Ok, no activity in vhd lines, so it seems that it really needs data from vhd player to be able to send commands toward vhd player (for interrupt INT_ line and maybe other input lines). Testing physically all wiring combinations between card and the vhd player is however too much, as there are very many possible combinations.
I tried now logically connecting pins
Dsub15-pin 3 to vhd din13-pin3,
Dsub15-pin 4 to vhd din13-pin4,
Dsub15-pin 8 to vhd din13-pin8,
Dsub15-pin 10 to vhd din13-pin10,
Dsub15-pin 12 to vhd din13-pin12
Dsub15-pin 14/15 to vhd din13-pin1 (gnd)
And because vhd pins 6,7,9, seemed to have hsync type signals based on oscilloscope and c9c part has those handling ( that japanese guy also mentions hsync in his page).
Anyway now _VP commands seems to command player!!!! At least g command works to change and inquire frame. Need to test more.
Have to next check if the thunder game also plays fine.
