It is currently Fri Dec 15, 2017 4:49 pm

All times are UTC - 7 hours





Post new topic Reply to topic  [ 36 posts ]  Go to page 1, 2, 3  Next
Author Message
PostPosted: Thu Sep 22, 2016 3:18 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
Hi all,

I have recently been playing batman returns and i know there are three different endings depending on what you do during the game.

I have watched multiple videos and they all seem to show different endings for the game. However, on nestopia when you beat the game with the best ending, (this is the only one i have seen thus far) you get what appears to be some weird and odd behavior after penguin cutscene ends.

Ive watched multiple vids of this on youtube to try to see what is supposed to be the correct behavior.

https://www.youtube.com/watch?v=BgGdTHhbGcM

If you watch this one, the bat logo flashes and then cuts to what looks like some glitchy credits where the screen is flashing and showing batman running at the bottom of the screen. On nestopia it shows catwoman lying there with the bat logo flashing like a projector. Then it cuts to the scenes with the credits rolling with the names with backgrounds flashing together. Obviously, this is not correct to my eyes. If you hit start it shows batman running at the bottom of the screen, but only if you hit start.

Is this some sort of weird undocumented behavior with the ending for the mmc3 mapper. Does anyone else experience this with their emu or on other emus?

This seems like the correct behavior to me, but this is not what nestopia is doing. And im not sure what they were playing the game on. (Most likely an emulator)

https://www.youtube.com/watch?v=JqXx5ONFXjM

I want to state this is not a nestopia bug post topic, im clearly curious as to see if anyone else is experiencing this on other emus, if so, then they would share this issue with nestopia. Ive tried various versions of nestopia, and they ALL do this glitch with the best ending, and multiple vids all seem to show different degrees of behavior, so something has to be off somewhere.

EDIT** Doesnt matter what ending you get, they all show the credits with the background flashing, and you have to hit start to see the character running at the bottom of the screen. Weird.

Whats even weirder is this game seems to be ok, all the way until you get to the end! Then it gets screwy. :P


Last edited by *Spitfire_NES* on Thu Sep 22, 2016 7:57 pm, edited 2 times in total.

Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 4:24 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
1. There are several Batman Returns ROMs floating around; some are overdumps, others are beta versions, others are pirated/hacked versions. Unless the video author specifically states what ROM file they were using, with checksums..... you see where I'm going with this. (None of the YouTube video authors provided this data -- no surprise -- and they don't even disclose what emulators they're using (the 2nd video is obviously using an emulator, the first I'm not sure), or what settings (this matters too!)).

2. Secondly, what version of Nestopia? UE or Official?

3. Thirdly, regardless of Nestopia's behaviour, what *ultimately* matters is what the actual NES with a legit Batman Returns cartridge does. Can't immediately test using a PowerPak because it emulates mappers, thus which mapper pack you use could/would influence the resulting behaviour.

So unless you can find a video where someone played this out and got the ending in question, with a Batman Returns cartridge...

4. The fact it happens "at the end" of the game is, well, how do I phrase this.... what do you expect? :-) When emulator authors test games, they don't play the entire game through, testing all endings and scenarios. Problems that happen at the end of games are common because of that. There are several examples of this happening -- a recent one is my mention of VRC2 emulation oddities at the end of Wai Wai World, which turned up some beautiful emulation bugs.


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 4:43 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
koitsu wrote:
1. There are several Batman Returns ROMs floating around; some are overdumps, others are beta versions, others are pirated/hacked versions. Unless the video author specifically states what ROM file they were using, with checksums..... you see where I'm going with this. (None of the YouTube video authors provided this data -- no surprise -- and they don't even disclose what emulators they're using (the 2nd video is obviously using an emulator, the first I'm not sure), or what settings (this matters too!)).

2. Secondly, what version of Nestopia? UE or Official?

3. Thirdly, regardless of Nestopia's behaviour, what *ultimately* matters is what the actual NES with a legit Batman Returns cartridge does. Can't immediately test using a PowerPak because it emulates mappers, thus which mapper pack you use could/would influence the resulting behaviour.

So unless you can find a video where someone played this out and got the ending in question, with a Batman Returns cartridge...

4. The fact it happens "at the end" of the game is, well, how do I phrase this.... what do you expect? :-) When emulator authors test games, they don't play the entire game through, testing all endings and scenarios. Problems that happen at the end of games are common because of that. There are several examples of this happening -- a recent one is my mention of VRC2 emulation oddities at the end of Wai Wai World, which turned up some beautiful emulation bugs.


For sure i thank you for responding koitsu. Here is another vid of seemingly correct behavior:

https://www.youtube.com/watch?v=Pz3hOIkEqYs

and another of the diff endings with no shaking:

https://www.youtube.com/watch?v=M3wy1QSbVro

It does this on UE and Martins last 1.40 as far as the glitchy behavior goes and all signs point to this being inaccurate behavior, obviously im not 100% though.

I do not have the real cart to test at this moment but this seems like pretty screwy behavior. I will see if i can track down some footage on the actual nes of this. I found one vid but he cuts it off at the end. lol.

Yes i completely understand an author is not going to test until the end of a game. :lol:


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 4:58 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
Do you have a way (save states close to the end, etc.) to test behaviour on NestopiaUE 1.47? https://sourceforge.net/projects/nestopiaue/

These videos are all on emulators. It's easy to tell because of 1) the recording dates (prior to RGB NES capability), and 2) the precision of the graphics (no NTSC effects). However, let's just pretend they were recorded from actual NESes for the sake of argument.

The first video even says in its description that you _must_ play it at 60fps else you miss visuals due to 30fps vs. interlacing. That brings into question whether or not the other videos were recorded at 30fps, and if the capture devices were doing proper deinterlacing (many do not, particularly because the NES does its weird 240p thing), which means what you'd end up seeing in the video wouldn't necessarily match what the person saw on their TV set.

Starting to understand why linking to YouTube videos recorded by random (non-technical) gamers doesn't really help narrow down what the "proper" behaviour is? (Again: I say/mean that respectfully) I know it's all you've got, but it's not particularly helpful.

What WOULD be helpful are save states for both ending types (assuming this game does have different endings). :-)

Edit: looks like rainwarrior thought the same thing, re: 30fps and interlacing.


Last edited by koitsu on Thu Sep 22, 2016 5:05 pm, edited 2 times in total.

Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 5:02 pm 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 5898
Location: Canada
I think this is just a 60fps vs 30fps issue.

Through the game's ending it alternates frames. 1 frame of background, 1 frame of text. It's supposed to appear like a "half dark" background with overlaid text. The original Batman game did this in its intro too.

It used to be that if you uploaded 60fps video to YouTube it would discard every second frame and you'd be left with 30fps. With an effect like this that alternates frames, you only see half of what you're supposed to. Same deal when Mega Man takes damage and disappears in a lot of YouTube clips, it's an artifact of this video process, nothing to do with emulation or mappers or bad dumps or any of that.

If you're not seeing it in your emulator correctly, maybe you've got frameskip on or something? You didn't really describe what you think is wrong with your version of it.


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 5:53 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
Ok, first things first, here is a savestate for the end of the game.

I am running the game in 60fps with no frameskip on. I used this first on nestopia for pc, it gets all weird at the end.

Second, i decided to try this out on my tv with the xbox nestopia port, which is pretty much up to date with UE, and in 480p. It STILL does this in hd, at 60fps. So maybe i am confused here? :mrgreen: :lol:

Im not sure what the issue is, I thought id ask here to see if anyone else has this issue...


Attachments:
batmanreturns.zip [2.27 KiB]
Downloaded 46 times
Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 5:56 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
Can you please provide details about the ROM file you're using? DO NOT upload the ROM here! Since you're using Nestopia, **all** output from View -> Image Info would be helpful. This can sometimes affect state state compatibility (i.e. you shouldn't use a save state from a different ROM if you can avoid it). Thanks.


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:00 pm 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 5898
Location: Canada
*Spitfire_NES* wrote:
it gets all weird at the end.

This is not a description of what happens.

What do you see?
What did you expect to see instead, and why?

(In Nestopia 1.36 I see exactly what I described in my previous post. Is this not what you see? Why do you think this is incorrect?)


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:02 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
koitsu wrote:
Can you please provide details about the ROM file you're using? DO NOT upload the ROM here! Since you're using Nestopia, **all** output from View -> Image Info would be helpful. This can sometimes affect state state compatibility (i.e. you shouldn't use a save state from a different ROM if you can avoid it). Thanks.


yes, give me a moment and ill put this up here:

File: Batman Returns (USA).zip <Batman Returns (USA).nes>
Directory: C:\Documents and Settings\Owner\Desktop\
Soft-patched: No
CRC: C247A23D
SHA-1: C3615A7DBF65D6B6AF607D5ACD9919AF48AF0DD3
System: NES-NTSC
Board: NES-TLROM, Mapper 4
PRG-ROM: 128k
CHR-ROM: 256k
Chips: MMC3C
Battery: No
Dump: OK


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:05 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
Pending which ROM he's using, I tested the save state using the following:

Code:
File:         Batman Returns (U) [!].nes
Soft-patched: No
CRC:          C247A23D
SHA-1:        C3615A7DBF65D6B6AF607D5ACD9919AF48AF0DD3
System:       NES-NTSC
Board:        NES-TLROM, Mapper 4
PRG-ROM:      128k
CHR-ROM:      256k
Chips:        MMC3C
Battery:      No
Dump:         OK


This game has a Nestopia database entry that forces chip type. I'll play around with that and report back.

Code:
    <game>
        <cartridge system="NES-NTSC" dump="ok" crc="C247A23D" sha1="C3615A7DBF65D6B6AF607D5ACD9919AF48AF0DD3">
            <board type="NES-TLROM" mapper="4">
                <prg size="128k" />
                <chr size="256k" />
                <chip type="MMC3C" />
            </board>
        </cartridge>
    </game>


During the credits roll, there definitely appears to be a visual oddity -- I now understand why the OP had trouble describing it.

What it looks like to me is that the game is switching nametables every alternate frame: one nametable has background graphics, the other has nothing but text, and they're being rapidly alternated. I'll see if I can make an animated GIF depicting the problem (expect a post edit).

After the credits, there's no catwoman with a bat signal, only a penguin telling you that you took too much time (so there appear to be multiple endings). I mention this because one of the videos had catwoman with a bat signal, and the bat signal was "flickery", looking almost intentional.


Last edited by koitsu on Thu Sep 22, 2016 6:06 pm, edited 1 time in total.

Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:06 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
rainwarrior wrote:
*Spitfire_NES* wrote:
it gets all weird at the end.

This is not a description of what happens.

What do you see?
What did you expect to see instead, and why?

(In Nestopia 1.36 I see exactly what I described in my previous post. Is this not what you see? Why do you think this is incorrect?)


Yes, rainwarrior i could have responded more appropriately. :lol:

The text and the scenes at the end are flashing together when they should be alternating, like you said. Its like they are appearing together and flashing like a projector.

However, this IS in 60 fps. I can confirm that because it does this on the xbox ports on my tv, and also because i can display the fps.


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:07 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
koitsu wrote:
Pending which ROM he's using, I tested the save state using the following:

Code:
File:         Batman Returns (U) [!].nes
Soft-patched: No
CRC:          C247A23D
SHA-1:        C3615A7DBF65D6B6AF607D5ACD9919AF48AF0DD3
System:       NES-NTSC
Board:        NES-TLROM, Mapper 4
PRG-ROM:      128k
CHR-ROM:      256k
Chips:        MMC3C
Battery:      No
Dump:         OK


This game has a Nestopia database entry that forces chip type. I'll play around with that and report back.

Code:
    <game>
        <cartridge system="NES-NTSC" dump="ok" crc="C247A23D" sha1="C3615A7DBF65D6B6AF607D5ACD9919AF48AF0DD3">
            <board type="NES-TLROM" mapper="4">
                <prg size="128k" />
                <chr size="256k" />
                <chip type="MMC3C" />
            </board>
        </cartridge>
    </game>


During the credits roll, there definitely appears to be a visual oddity -- I now understand why the OP had trouble describing it.

What it looks like to me is that the game is switching nametables every alternate frame: one nametable has background graphics, the other has nothing but text, and they're being rapidly alternated. I'll see if I can make an animated GIF depicting the problem (expect a post edit).

After the credits, there's no catwoman with a bat signal, only a penguin telling you that you took too much time (so there appear to be multiple endings). I mention this because one of the videos had catwoman with a bat signal, and the bat signal was "flickery", looking almost intentional.


Yes i made a quick savestate and this is one of the endings, all 3 endings have the same flashing going on. Im glad you can see what i mean koitsu. Its very ODD to say the least.


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:15 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
It's not really "odd", it's just that the behaviour has to be figured out. Again, I will repeat myself (for a third time): it would be helpful to see what the actual behaviour is on a real NES, not an emulator. I have this weird feeling it might be intentional, maybe to simulate 24fps film or give that kind of visual effect (an interlaced image with text "overlaying" atop background graphics, but with a flicker to it?).

BootGod's DB confirms the game does in fact use MMC3C (not B or A), and the checksums are right: http://bootgod.dyndns.org:7777/profile.php?id=582

Disabling the Nestopia internal DB has no effect on behaviour.

Debugging this is going to be painful because Nestopia has no real debugging support; it would be interesting to see if FCEUX 2.2.3 behaves similarly, and what is going on with some of the PPU control bits during rendering of a frame as well as during NMI/VBlank. And, again: this is complicated by the fact that it happens at the end of the game, so someone gets to play these through all the way to make save states... :P


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:19 pm 
Offline
User avatar

Joined: Fri May 21, 2010 4:10 pm
Posts: 279
koitsu wrote:
It's not really "odd", it's just that the behaviour has to be figured out. Again, I will repeat myself (for a third time): it would be helpful to see what the actual behaviour is on a real NES, not an emulator. I have this weird feeling it might be intentional, maybe to simulate 24fps film or give that kind of visual effect (an interlaced image with text "overlaying" atop background graphics, but with a flicker to it?).

BootGod's DB confirms the game does in fact use MMC3C (not B or A), and the checksums are right: http://bootgod.dyndns.org:7777/profile.php?id=582

Disabling the Nestopia internal DB has no effect on behaviour.

Debugging this is going to be painful because Nestopia has no real debugging support; it would be interesting to see if FCEUX 2.2.3 behaves similarly, and what is going on with some of the PPU control bits during rendering of a frame as well as during NMI/VBlank. And, again: this is complicated by the fact that it happens at the end of the game, so someone gets to play these through all the way to make save states... :P


I certainly understand. With the way its flashing i just can't see the text and level graphics being displayed that way. Thats why it automatically rose a red flag upon playing through and seeing the ending.

I am glad to know i am at least not crazy, and it does beg to differ what this does on other emus.

Also another oddity, in the nst database the mirroring is listed as H, on this website it is listed as V. Perhaps an error? I tried changing the mirroring to V, but i think i got the same results IIRC.

http://tuxnes.sourceforge.net/nesmapper.txt


Top
 Profile  
 
PostPosted: Thu Sep 22, 2016 6:26 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3192
Location: Mountain View, CA, USA
MMC3 supports real-time changing of H/V mirroring, so changing it in the header won't do you much good: https://wiki.nesdev.com/w/index.php/MMC ... 2C_even.29

It doesn't matter what the mirroring is in NstDatabase.xml. And Bootgod's DB is authoritative at this point; it clearly says "Mapper Ctrl" (real-time controller mirroring per MMC3). It's done in code, not with a solder dot.


Last edited by koitsu on Thu Sep 22, 2016 6:27 pm, edited 1 time in total.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 36 posts ]  Go to page 1, 2, 3  Next

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 10 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
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group