MAMEWorld >> News
View all threads Index   Threaded Mode Threaded  

Pages: 1

SmitdoggAdministrator
Reged: 09/18/03
Posts: 16377
Send PM


DU: Blood Bros.
#371376 - 11/26/17 08:19 PM Attachment: bb.jpg 271 KB (0 downloads)


Bill D. dumped an alt USA version of Blood Bros. licensed to Fabtek.

[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



SmitdoggAdministrator
Reged: 09/18/03
Posts: 16377
Send PM


Update new [Re: Smitdogg]
#371378 - 11/26/17 09:11 PM Attachment: DSCN2473.JPG 509 KB (0 downloads)


caius and Osso got West Story, a bootleg of Blood Bros. It fixed 2 bad roms in the dump in mame and is also a new set (different program).



[ATTACHED IMAGE - CLICK FOR FULL SIZE]

Attachment



Kitsune Sniper
Furry guy
Reged: 01/30/04
Posts: 2257
Send PM


Re: Update new [Re: Smitdogg]
#371379 - 11/26/17 09:19 PM


> caius and Osso got West Story, a bootleg of Blood Bros.

Blood Bros. was one of my childhood favorites. Yay for new dumps! Thanks all!

Edit: Wow, West Story seems to be running slower than what it should be. Does the PCB have inadequate hardware, or was that just a result of the hackers messing with the code?



AJR Hacker
MAME Fan
Reged: 02/01/16
Posts: 62
Send PM


Re: Update new [Re: Kitsune Sniper]
#371380 - 11/26/17 11:15 PM


> Edit: Wow, West Story seems to be running slower than what it should be. Does the PCB
> have inadequate hardware, or was that just a result of the hackers messing with the
> code?

The bootleg hardware is indeed less powerful. The original Seibu hardware apparently used DMA to buffer the video RAMs (as all later Seibu games do). The bootleggers didn't implement that and instead hacked up the 68000 code to laboriously copy out the video memory in software.



Kitsune Sniper
Furry guy
Reged: 01/30/04
Posts: 2257
Send PM


Re: Update new [Re: AJR Hacker]
#371382 - 11/26/17 11:33 PM


> > Edit: Wow, West Story seems to be running slower than what it should be. Does the
> PCB
> > have inadequate hardware, or was that just a result of the hackers messing with the
> > code?
>
> The bootleg hardware is indeed less powerful. The original Seibu hardware apparently
> used DMA to buffer the video RAMs (as all later Seibu games do). The bootleggers
> didn't implement that and instead hacked up the 68000 code to laboriously copy out
> the video memory in software.

That's crazy. Will MAME be able to reproduce this behavior correctly?



Haze
Reged: 09/23/03
Posts: 4638
Send PM


Re: Update new [Re: Kitsune Sniper]
#371385 - 11/27/17 12:11 AM


> > > Edit: Wow, West Story seems to be running slower than what it should be. Does the
> > PCB
> > > have inadequate hardware, or was that just a result of the hackers messing with
> the
> > > code?
> >
> > The bootleg hardware is indeed less powerful. The original Seibu hardware
> apparently
> > used DMA to buffer the video RAMs (as all later Seibu games do). The bootleggers
> > didn't implement that and instead hacked up the 68000 code to laboriously copy out
> > the video memory in software.
>
> That's crazy. Will MAME be able to reproduce this behavior correctly?

It's not really complex, except for the possible waitstates / delay the DMA might cause periodically if you want to be 100% spot on with the timing.

A lot of other bootlegs have a 2nd copy of sprite ram and code to transform sprite lists from one format to another too, obviously this takes CPU cycles to do. Then you have bootlegs of bootlegs which do that, but where it's been confirmed on the PCB that the 2nd copy of spriteram isn't even the one that gets used (and infact, the ram doesn't exist, the code copies to nowhere)

All good fun



Kitsune Sniper
Furry guy
Reged: 01/30/04
Posts: 2257
Send PM


Re: Update new [Re: Haze]
#371386 - 11/27/17 12:27 AM


> > > > Edit: Wow, West Story seems to be running slower than what it should be. Does
> the
> > > PCB
> > > > have inadequate hardware, or was that just a result of the hackers messing with
> > the
> > > > code?
> > >
> > > The bootleg hardware is indeed less powerful. The original Seibu hardware
> > apparently
> > > used DMA to buffer the video RAMs (as all later Seibu games do). The bootleggers
> > > didn't implement that and instead hacked up the 68000 code to laboriously copy
> out
> > > the video memory in software.
> >
> > That's crazy. Will MAME be able to reproduce this behavior correctly?
>
> It's not really complex, except for the possible waitstates / delay the DMA might
> cause periodically if you want to be 100% spot on with the timing.
>
> A lot of other bootlegs have a 2nd copy of sprite ram and code to transform sprite
> lists from one format to another too, obviously this takes CPU cycles to do. Then you
> have bootlegs of bootlegs which do that, but where it's been confirmed on the PCB
> that the 2nd copy of spriteram isn't even the one that gets used (and infact, the ram
> doesn't exist, the code copies to nowhere)
>
> All good fun

I didn't mean because it was complicated; I meant because of the different hardware and how MAME's implementations can sometimes run better than the actual machine. Basically, "Will you be able to make sure MAME runs the game as crappily as the actual hardware does."



BrianT
MAME Fan
Reged: 08/07/04
Posts: 124
Send PM


Re: Update new [Re: Smitdogg]
#371388 - 11/27/17 01:02 AM


The commit to the source has just been made. Turns out the 2 "bad" roms mentioned in the source are actually the same on this PCB, so they didn't change.

However caius did measurements from the PCB so the bootlegs are going to sound radically different in terms of speed then before. But again this is using actual measurements from the PCB plugged into the driver.

As mentioned, Bill's Blood Bros. is the US Fabtek license so it's good to have that finally dumped!

BrianT



SmitdoggAdministrator
Reged: 09/18/03
Posts: 16377
Send PM


Re: Update new [Re: BrianT]
#371389 - 11/27/17 01:05 AM


Yeah Blood Bros. is a classic.



Haze
Reged: 09/23/03
Posts: 4638
Send PM


Re: Update new [Re: BrianT]
#371390 - 11/27/17 01:11 AM


> The commit to the source has just been made. Turns out the 2 "bad" roms mentioned in
> the source are actually the same on this PCB, so they didn't change.
>

this surprises me greatly, there are obvious 'bad dump' type problems with the explosions etc.

if the PCB caius has exhibits the same problem it's possible it's been repaired with bad roms at some point, this is always a possibility.

if it doesn't, then it could be some encryption, but last time I looked I concluded it was more likely they were bad roms.



AJR Hacker
MAME Fan
Reged: 02/01/16
Posts: 62
Send PM


Re: Update new [Re: BrianT]
#371397 - 11/27/17 03:03 AM


Could you also possibly check what code the bootleg Z80 sound program is executing at 3000?



Haze
Reged: 09/23/03
Posts: 4638
Send PM


Re: Update new [Re: Haze]
#371417 - 11/27/17 07:45 PM


> > The commit to the source has just been made. Turns out the 2 "bad" roms mentioned
> in
> > the source are actually the same on this PCB, so they didn't change.
> >
>
> this surprises me greatly, there are obvious 'bad dump' type problems with the
> explosions etc.
>
> if the PCB caius has exhibits the same problem it's possible it's been repaired with
> bad roms at some point, this is always a possibility.
>
> if it doesn't, then it could be some encryption, but last time I looked I concluded
> it was more likely they were bad roms.

ok, it's a stupid encryption
https://github.com/mamedev/mame/pull/2861

actually makes me think the original MASK dump should be verified tho as there's one byte that differs still.


Pages: 1

MAMEWorld >> News
View all threads Index   Threaded Mode Threaded  

Extra information Permissions
Moderator:  John IV, Tafoid 
0 registered and 41 anonymous users are browsing this forum.
You cannot start new topics
You cannot reply to topics
HTML is enabled
UBBCode is enabled
Thread views: 961