SpritesMind.Net

I basically just rearranged the VRAM code according to the tech memo: plenty of time between the last read and next setting the control reg. Didn't do much more than reorder the instructions, really.

I made sure to spend at least 116 CPU clocks following each read sequence (consecutive reads are ...

One last shot: hackpatch v3. Only fixes for hardware in this one.

Sorry, couldn't help myself.

I believe the tearing evildragon pointed out is caused by updating the scroll values midscreen. I didn't bother myself with that.

Where was that memo hiding? :P

So it seems the delay belongs after the last read. If you reorganize the code, you probably don't need any artificial delays (NOPs or such) -- just move anything you can between the last read and setting the address for the next write. But, using a buffer is a ...

OK, I gave it a shot: hackpatch v2. Let me know if it's better.

It's probably the VRAM reads/writes. The code does a bit of read and writeback. I'm thinking that setting the address and reading immediately may fail occasionally. I'm not sure about writing immediately. Anyway, it might be enough to have a NOP before the first read or write.

I'm also thinking ...

The problem now is that inputs seems not to be read properly: my controller does not respond and I can't leave the menu screen , it's probably something related to timings when programming the controller.
It's much more obvious than that. (hint: something is missing.)

For anyone still "testing ...

Unfortunately, MEFAS has no email address mentioned in their website, so it's up to somebody located in US to call them up to avoid international call fee hell burden.
You should know that MEFAS is owned by EAG (apparently since November 2007!), so I suggest you check http://www.eaglabs.com/ . EAG ...

For the record, I actually found one game (Bugs Bunny Double Trouble, 2nd level with the bull) that on some frames will change the display width (from H32 to H40) during vertical blanking to gain some extra DMA speed then wait for the VBLANK flag to be cleared before switching back to H32 mode.
To ...

Make it static.
*or*

If you want to do non-const stuff, write your own memcpy function (or steal one).

Just curious, are you testing this with Genesis Plus? I did a lot of research to make sure the leftmost column scrolled like it does on the real hardware for 40 cell mode. But now that I think about it, I may have not tested 32-cell mode and assumed the same rules apply...
I checked. Genesis Plus ...

Many games use hint on each line without problems, and i'm pretty sure i've seen games enabling or disabling hint mid-frame on purpose...
Have you seen any that does both? :D Well, whatever it is, it's not usual. That much is clear.

the official doc somehow says the M3 bit stops HV Counter when ...

That doesn't sound right. From what I understand, the counter is frozen when the VDP's /HL pin (input from the I/O chip) is held low. Not to say that it can't freeze otherwise, but under normal conditions, it should keep on running.

Assuming there's no issue with the counter, I'm thinking that the ...

Here's a hack-fix for Fusion (v3.64, currently) (works with Gens, too):

Sunset Riders (E) [!]
===============
00018E:4F06 (checksum fix)
0004CA:8016

As for the technicals:

In the H-INT routine,
...
@0004BA: MOVE.B $00FFB062,D0
@0004C0: CMP.B $C00008,D0
@0004C6: BNE @0004D0 ; if V-counter == byte ...

Still it's rather strange that this is called during the main initialization (not the main loop)
I don't consider that the main initialization, but that doesn't really matter. The thread for that intro sequence contains its own initialization, as do other threads. The routines merely provide a ...

when sprite layer is disabled, mickey's foot is still visible and animating around (how come?)
That's what the game does. It takes tiles from the foot and moves them from the sprite plane to the scroll plane. Don't mind it; there are a lot of weird things going on behind the scenes.

EDIT: more ...