shmups.system11.org

Shmups Forum
 
* FAQ    * Search
 * Register  * Login 
It is currently Sun Jan 26, 2020 7:55 am View unanswered posts
View active topics



Post new topic Reply to topic  [ 201 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7
Author Message
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 12:05 am 


User avatar

Joined: 24 Jul 2007
Posts: 45
Location: https://irkenlabs.com/
Unless I fudged up the patching, no dice - getting tile corruption :(


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 8:37 am 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
Ha, it's too good to be true apparently. Especially considering that 3 people had the same idea literally at the same time :D

Toaplan lovers: hey look! we could simply... Fate: what were you thinking Image :mrgreen:

Yeah, I can confirm that with Dogyuun. It looks exactly the same as my previous video capture.
I guess to see a positive side of it, it's the same instead of (completely) different behavior. So the chance grows that it's not a continuum, i.e. if we nail it, we actually nail it without having to worry about arcane spillover effects that aren't immediately apparent.
If that's really the case of course remains to be seen.

TL;DR: The only thing left is to understand how these byte sequences work, or, less preferably, devising a way of efficiently testing many values. Then we've won.

@sergiopolog: do you have a Teki Paki by any chance? You could check if it works with the OSSC, it's the only PCB we don't have empirical info about so far :)


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 12:03 pm 


User avatar

Joined: 18 Nov 2016
Posts: 15
Location: Barcelona,Spain
@6t8k sorry, that one along with Ghox are the pcbs that I don't have yet for this system :cry:
_________________
My: VAPS Collection .


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 12:51 pm 


User avatar

Joined: 26 Jan 2005
Posts: 557
Location: Singapore
6t8k wrote:
Working patch for Knuckle Bash added and added a patch for Fixeight!
@parodius: can you try the latter? :)

Just did :
Code:
$ sha1sum.exe tp-026-1.fixeight
56017cc2adeb2fdfd25a992bbf31605b18581865 *tp-026-1.fixeight

Unfortunately, it seems to change nothing at all... There's no graphics glitch like on Dogyuun/Batsugun, and the sync signal is still messed up in the same way as before.


6t8k wrote:
Fixeight is an interesting special case, because it uses two different checksums (at 0xc0 and 0xc4), and it traverses the loop twice, one time checking against one checksum for even bytes and one time checking against the other checksum for odd bytes (ROM EVEN ERROR vs. ROM ODD ERROR). So basically everything is the same, it's just that you have to go through the process described above twice and change both checksums accordingly.

By the way, as per parodius' reportings, changing occurences #1 and #3 in Fixeight leads to a blank screen. I could confirm that these lock up the game even before the integrity check is called, so as suspected, #2 is the occurence to change.

Thanks for the explanations. Any idea what to try next for FixEight ? 8)
_________________
My sales thread : 2019/07/14..MASTER.VER.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 8:23 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
@parodius: my bad, somehow the impression snuck in that the #3 occurence can't be changed without provoking the black screen - however, in contrast to #1, it can! Could you re-download and re-apply? :)

---

6t8k wrote:
It's certainly possible that it would've been enough to change only one specific byte within that "configuration string" to achieve this.

Yes. I've tried a few variants for Dogyuun, not yet reaching both OSSC compatibility and sound graphics at the same time however. Here are the results:

Spoiler: show
Code:
(byte order on disk)

0202, 0302, 0040 <- first controller
0212, 0312, 0050 <- second controller
[A] (original values, but set the next-to rightmost bit to 0 akin to "OSSC compatible" values)
Results: OSSC recognizes the signal, but game has graphical glitches, the same as can be seen on the previous video capture

0302, 0202, 0042 <- first controller
0312, 0212, 0052 <- second controller
[B] (original values, but swapped the first two words each because in "OSSC compatible" values, the 3 comes first)
Results: OSSC does not recognize the signal, but game is fine (like unpatched ROM)

0302, 0202, 0040 <- first controller
0312, 0212, 0050 <- second controller
[C]  (A and B combined)
Results: did not try due to above results

0202, 0302, 0041 <- first controller
0212, 0312, 0051 <- second controller
[D] (original values, but set rightmost bit to 1 as a "compromise" between original values and A)
Results: OSSC recognizes the signal, but graphical glitches are much, much worse. almost everything is missing aside from backgrounds.


So the rightmost byte (order on disk) seems to be responsible...


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 14, 2020 11:03 pm 


User avatar

Joined: 26 Jan 2005
Posts: 557
Location: Singapore
6t8k wrote:
@parodius: my bad, somehow the impression snuck in that the #3 occurence can't be changed without provoking the black screen - however, in contrast to #1, it can! Could you re-download and re-apply? :)

Code:
sha1sum.exe tp-026-1.fixeight
3763350f56af46477bc30e60274cb277f47cb4a9 *tp-026-1.fixeight

No dice, still the same.
_________________
My sales thread : 2019/07/14..MASTER.VER.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Wed Jan 15, 2020 11:54 am 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
parodius: Hmm.. if it's the #1 occurence we may be out of luck for the time being. :?
Someone would have to step through the instructions to find out why changing the bytes there prevents the game from booting and see if something can be done to remedy that.

The only other possibility I can think of right now would be that Fixeight could (for some strange reason) use the values that are normally fed to the second GP9001, which are always stored adjacently even for games that have only one GP9001 like Fixeight does.
I'll create a corresponding patch shortly that you can try if it's not too much trouble at this point so we can rule that possibility out.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Wed Jan 15, 2020 12:17 pm 


User avatar

Joined: 26 Jan 2005
Posts: 557
Location: Singapore
No trouble at all.
_________________
My sales thread : 2019/07/14..MASTER.VER.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Wed Jan 15, 2020 8:53 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
Revised the Fixeight patch, please retry.

I'm now quite sure though that the first occurence is the relevant one..

A few observations in regards to the black screen boot failure: it seems like there is another integrity check or something.
The failure is in fact triggered when any address between 0x100..0x70EE is modified. All bytes within that range are processed and afterwards, an infinite loop waits until 0xFF appears at 0x28F001, which is memory shared with the V25 audio CPU I suppose (?). This eventually happens (not from within that loop), but when any byte within the aforementioned range is changed, 0xFD appears instead.
If I manually set that value to 0xFF via the memory viewer, the game boots with upside down graphics, in Taiwanese region, has no audio and crashes when a credit is started. Interesting.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Wed Jan 15, 2020 10:50 pm 


User avatar

Joined: 26 Jan 2005
Posts: 557
Location: Singapore
Code:
sha1sum.exe tp-026-1.fixeight
be33802651f983ce081d203715a7e3c8ddcc9d4b *tp-026-1.fixeight

Unfortunately, I couldn't see any difference with this patch either.
_________________
My sales thread : 2019/07/14..MASTER.VER.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Thu Jan 16, 2020 8:20 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
Yeah, they seem to have done things differently with Fixeight for some reason..
Removed the patch until there's a way to work around the above-mentioned issue. I'm sure it's possible somehow.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Sun Jan 19, 2020 5:09 pm 



Joined: 20 Apr 2018
Posts: 92
Your efforts on this are awesome, keep up the good work!

That being said, would the OSSC Pro, just announced, have the ability to fix the sync on these?


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Sun Jan 19, 2020 6:15 pm 



Joined: 20 Apr 2018
Posts: 92
The following PCBs work with a HAS > RetroTINK SCART2COMP > CRT TV (no OSSC in the chain).

Here are the results with HAS > OSSC > Sharp LCD TV. OSSC FW .85a reset settings.

Japanese Knuckle Bash:

Default: Red and Green LED on, screen black.
Line3x 320x240 default: Red and Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1250.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1400.00: Green LED on, screen black.

OSSC LCD Screen:
AV1: RGBS 263p
15.60kHz 59.34Hz

--------------------------------------

Korean Knuckle Bash:

Default: Red and Green LED on, screen black.
Line3x 320x240 default: Red and Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1250.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1400.00: Green LED on, screen black.

OSSC LCD Screen:
AV1: RGBS 263p
15.61kHz 59.37Hz

--------------------------------------

Snow Brothers 2:

Default: Red and Green LED on, screen black.
Line3x 320x240 default: Red and Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 501.00 - ~680.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to ~681.00: Green LED on, Red LED comes on intermittently, with the Red LED coming on more often as the H. samplerate is set higher, screen black.
Line3x 320x240 with H. samplerate set to 703.00: Red and Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1250.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1400.00: Green LED on, screen black.

OSSC LCD Screen:
AV1: RGBS 263p
15.61kHz 59.37Hz

--------------------------------------

Fixeight: I KNEW I WASN'T CRAZY!!! MY FIXEIGHT WORKS ON DEFAULT SETTINGS!

Default: Red and Green LED on, screen black.
Line3x Generic default: Green LED on, game works but top edge (left if screen rotate, player 1 side) has slight distortion. Very slight, almost not noticeable.
Line3x 320x240 default: Red and Green LED on, screen black, I can hear the game intermittenly, TV says incompatible resolution.
Line3x 320x240 with H. samplerate set to 750.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1250.00: Green LED on, screen black, I can hear the game, TV says incompatible resolution.
Line3x 320x240 with H. samplerate set to 1400.00: Green LED on, screen black, I can hear the game, TV says incompatible resolution.

OSSC LCD Screen:
AV1: RGBS 263p
15.62kHz 59.41Hz


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Tue Jan 21, 2020 12:15 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
XtraSmiley wrote:
Fixeight:
[...]
Line3x Generic default: Green LED on, game works but top edge (left if screen rotate, player 1 side) has slight distortion. Very slight, almost not noticeable.

Thanks for the tests, that Fixeight result was unexpected! Well, it fits the mould of it having additional idiosyncracies I guess. Would be interesting to have a look at the sync waveform, but I don't have the PCB at this time..

XtraSmiley wrote:
would the OSSC Pro, just announced, have the ability to fix the sync on these?

Since it'll have a Si5351C configurable clock generator, I assume we'll at least be able to achieve results akin to what you can see in the "before" photos that parodius posted.

However, considering that the Framemeister can digest these PCBs with odd sync just fine without any artifacts in the resulting image, and that I can't tell what crucial component the OSSC Pro would be lacking in comparison to the Framemeister, I see no reason why the Pro wouldn't be able to completely cancel out the problem in principle (also refer to this post). Whether the Pro will be able to compensate for it automatically or if it'll make more sense to readjust some parameters manually remains to be seen. marqs can probably give a better assessment though (feel free to correct me). :)


Either way I'd still like to finalize these patches. As for Fixeight, I've got something going, but before I release the patch I want to see if I can optimize it.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Wed Jan 22, 2020 8:56 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
New Fixeight patch, parodius, the stage is yours :)

It changes two additional bytes compared to all previous patches in order to appease another integrity check performed by the NEC V25 that would otherwise prevent the game from booting.
These bytes are in the 0x40100..0x470ee range. As I want to avoid changing the gameplay behavior in any way, I set a corresponding memory watchpoint and credit-fed my way through the game (while cycling through the different characters along the way for what it's worth) to see if the game accesses that range - it didn't. A simple smoke test like this might not mean much, but as far as I can tell, that specific range is only accessed by the various integrity checks when the game boots.


Yesterday I listened to this interesting conversation with Artemio and SmokeMonster about some technical aspects of video game preservation and the MiSTer project. The GP9001 needs to be delidded and analyzed already!


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Thu Jan 23, 2020 8:17 am 



Joined: 22 Jun 2005
Posts: 212
Location: Communism's New Home (CA, USA)
So at this stage, V-V works but Batsugun does not?
_________________
I enjoy jamma games more than anything on console.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Thu Jan 23, 2020 2:55 pm 


User avatar

Joined: 26 Jan 2005
Posts: 557
Location: Singapore
6t8k wrote:
New Fixeight patch, parodius, the stage is yours :)

It changes two additional bytes compared to all previous patches in order to appease another integrity check performed by the NEC V25 that would otherwise prevent the game from booting.
These bytes are in the 0x40100..0x470ee range. As I want to avoid changing the gameplay behavior in any way, I set a corresponding memory watchpoint and credit-fed my way through the game (while cycling through the different characters along the way for what it's worth) to see if the game accesses that range - it didn't. A simple smoke test like this might not mean much, but as far as I can tell, that specific range is only accessed by the various integrity checks when the game boots.


Hats off, it works !
I couldn't spot any graphical glitches during attract mode.
Excellent reverse engineering/hacking work, man.

Spoiler: show
ImageImage
_________________
My sales thread : 2019/07/14..MASTER.VER.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Thu Jan 23, 2020 7:21 pm 



Joined: 20 Apr 2018
Posts: 92
Awesome work man!!!!!

I suggest updating the first post with the new info, and adding third column with working/not working patches.

Amazing..!


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Thu Jan 23, 2020 10:35 pm 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
Glad it works!

thchardcore: Dogyuun and Batsugun don't have a working patch yet because they each have two GP9001 chips which makes it more complicated. At this time it's a bit like a combination lock, and for each try an EPROM has to be burnt anew.

XtraSmiley: For now I think it'd be best to simply link to the post with the patches. Separation of concerns.


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Fri Jan 24, 2020 8:13 am 


User avatar

Joined: 19 Jan 2016
Posts: 307
Location: Sweden
XtraSmiley wrote:
Line3x 320x240 with H. samplerate set to 1250.00: Green LED on, screen black.
Line3x 320x240 with H. samplerate set to 1400.00: Green LED on, screen black.


Bit off tangent here, but just to save you some time in future testing, those settings will never produce a valid signal.

I saw 6t8k recommended using 320x240 opt with high sample rates earlier in the thread, but I think that’s a red herring. The resulting output total frame is just wildly off-spec, and actually exceeds max pixel clock of OSSC: 3*1250*4/3= 5000; 5000*263*3*59.4= 234 MHz!
Don’t know how ossc handles it, maybe it just doesn’t increase actual h.samplerate once pixelrate limit is reach.

If you wanna test for compatibility, simply generic mode is enough.
_________________
OSSC TV Compatibility report thread


Top
 Offline Profile  
 
 Post subject: Re: Dogyuun, Batsugun, Fixeight, V-V incompatible with OSSC
PostPosted: Fri Jan 24, 2020 9:45 am 


User avatar

Joined: 14 Aug 2019
Posts: 246
Location: BW, Germany
^ Well aware, we were not expecting to suddenly get a viable image this way!

My idea was that, for people who don't have a scope, comparing the specific artifacts in the image I coaxed out this way could be used to identify if other PCBs suffer from the very same problem while excluding other possibilities. I don't think it's useless in that regard.
Granted though as I said earlier checking if it works in generic mode with default samplerate is enough for compatibility itself, besides, many displays, especially TVs, presumably won't display the off-spec signal anyway, so that analysis is not always possible.


Top
 Offline Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 201 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7

All times are UTC


Who is online

Users browsing this forum: DewmanSNK, SamIAm and 7 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Space Pilot 3K template by Jakob Persson
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group