Should Pauline scores count for the Main HSL?

[xelnia]

A DK score has been submitted that uses the Pauline sprite-swap hack and I can't remember if we ever decided as a community if that version was acceptable for the main HSL. So, here's the poll.

The submission is here: https://donkeykongforum.net/index.php?topic=364.msg44309#msg44309

A thread discussing how Pauline INPs won't play back for regular DK is here: https://donkeykongforum.net/index.php?topic=411.0

[Lyriell]

So Tanya is a legend and amazing and I could go on for days.

But if Pauline mame inps don't work, that breaks the mame/arcade thing.

Theoretically if the game logic is the same, then all is good in the universe.

Maybe a separate leader board like remix and stuff? 

I'm also of the opinion that there should be a Deranged leader board as I would enjoy playing that.

How do games with pallets swaps work elsewhere?  Is green beret separate to rush n attack or are they the same?

[Lyriell]

Maybe a better example is Contra vs Probotector?

[bonerforce]

Why is there not a PE HSL? It has been around for about 10yrs... I watched ACU-AZU on youtube get a killscreen on Pauline Edition, he ran boards.... like a boss

[80sArcadeKid]

So I decided to run the test myself to see what differences there are if any between the dkong and dkongpe (pauline) romset.

YouTube Unlisted Link (you will hear audio from both the original recording and the playback).

If you playback the Pualine recorded INP with the Pauline romet, everything works fine, but... as in the other thread linked, if you play it back in the DKONG romset it goes weird.
There is a definite change going on after the cutscene at the end of level one, but the VERY START of L2 barrels looks like it's working up until the second girder ladder. It feels like a killscreen positional type of change (that happens at L22) but without the instant death, the controls go wonky. You will see that the INP is actually still playing back all my inputs, but the game freaks out and Jumpman dies multiplr times.

I'm personally gutted that Tanya's PB will probably not get accepted into the HSL, but like other romhacks it should probably go in it's own Pauline HSL. That said I personally believe it's just a syncronous mismatch between the romsets and Pauline from L2 is just the same in difficulty and play as the original romset, it's just got a quirk and that is only displayed on the MAME version. There is no way to playback a hardward version.

[80sArcadeKid]

Interestingly the change happens at exactly 14,900 just before it registers the 100 points to score 15,000.
The 4th girder barrel doesnt steer down but goes on ahead in Pauline version. From that point Jumpan is probably trying to mimic the inputs but as it's out of sync with Pauline the jumps are in different places. Notice at 3:36, Jumpman moves and jumps in the same directions as Pauline but is on the bottom girder, so the inputs are still going through.

[Rev John]

The 4th girder barrel doesnt steer down but goes on ahead in Pauline version.

Let me piece together what I think I know about this but am probably wrong anyway.  Whether barrels go down ladders depend on where the joystick is pointing plus a random factor.
As far as I can make out the relevant bit of code starts at about #2172 (thank you to whoever did the disassembly!) where we have a barrel on top of a ladder.  A few bytes later is the bit where the code always sends a barrel down a ladder if the oilcan isn't lit.  But then everyone knows that yeah?   
At #218C register A (which I think of as the 'accumulator' whether it's called that or not) is loaded with the contents of #6018 which the dissembly describes as "#6018 = constantly changing ... timer of some sort?"  Note that #6019 is also used for random numbers in DK ("#6019 - constantly changing timer - very fast")
I seem to recall reading that the Z80 chip has a memory location that increments with *every* instruction parsed.  Which I reckon must be a lot of overhead.  I figure this must be #6018 or #6019, which would make the random number generation sensitive to small changes in the code ... possibly even to the processing of two different sprites?

My understanding is most likely completely wrong  but I'm not sure anyone has ever nailed down exactly how #6018 and #6019 (which control a LOT of DK's randomness work).  Anyway, that's my 3 cents.  I like cows.

Code: [Select]

; called from #2167

216D  CD6E23    CALL    #236E ; check for ladder.  if no ladders, RET to higher sub.  if at top of ladder, A := 1
2170  3D        DEC     A ; is there a ladder to go down?
2171  C0        RET     NZ ; no, return

2172  78        LD      A,B ; yes, load A with B which has the value of the ladder from the check ??
2173  D605      SUB     #05 ; subtract 5
2175  DD7717    LD      (IX+#17),A ; store into +17 to indicate which ladder we might be going down ???
2178  3A4863    LD      A,(#6348) ; get status of the oil can fire
217B  A7        AND     A ; is the fire lit ?
217C  CAB221    JP      Z,#21B2 ; no, always take ladders before oil is lit

217F  3A0562    LD      A,(#6205) ; else load A with mario's Y position + 5
2182  D604      SUB     #04 ; subtract 4
2184  BA        CP      D ; is the barrel already below mario  ?
2185  D8        RET     C ; yes, return without taking ladder

2186  3A8063    LD      A,(#6380) ; else load A with difficutly from 1 to 5.  usually the level but increases during play
2189  1F        RRA      ; roll right (div 2) .  now can be 0, 1, or 2
218A  3C        INC     A ; increment.  result is now 1, 2, or 3 based on skill level
218B  47        LD      B,A ; store into B
218C  3A1860    LD      A,(#6018) ; load A with random timer ?
2
2190  E603      AND     #03 ; mask bits.   result now random number between 0 and 3
2192  B8        CP      B ; compare with value compted above based on skill
2193  D0        RET     NC ; return if greater.  on highest skill this works 75% of time, only returns on 3

2194  211060    LD      HL,#6010 ; load HL with player input. 

; #6010 - copy of input (see #6011). except when jump pressed, bit 7 is set momentarily.
; #6011 - input.  right sets bit 0, left sets bit 1, up sets bit 2, down sets bit 3, jump sets bit 4

2197  3A0362    LD      A,(#6203) ; load A with mario's X position
219A  BB        CP      E ; compare with barrels x position
219B  CAB221    JP      Z,#21B2 ; if equal, then go down ladder

219E  D2A921    JP      NC,#21A9 ; if barrel is to right of mario, then check for moving to left

21A1  CB46      BIT     0,(HL) ; else is mario trying to move right ?
21A3  CAAE21    JP      Z,#21AE ; no, skip ahead and return without going down ladder

21A6  C3B221    JP      #21B2 ; yes, make barrel go down ladder

21A9  CB4E      BIT     1,(HL) ; is mario trying to move left ?
21AB  C2B221    JP      NZ,#21B2 ; yes, make barrel go down ladder

21AE  79        LD      A,C ; else load A with random timer computed above
21AF  E618      AND     #18 ; mask with #18.    25% chance of being zero?
21B1  C0        RET     NZ ; else return without going down ladder.  If zero then go down the ladder anyway

21B2  DD3407    INC     (IX+#07) ; increase Barrel's deployment/animation status
21B5  DDCB02C6  SET     0,(IX+#02) ; set barrel to go down the ladder
21B9  C9        RET      ; return


[homerwannabee]

I voted yes, but I would hope there would be a note of some sort explaining it was with the Pauline hack.

[Dallas Taylor]

My first KS was not accepted because I did not record an .inp file. I live-streamed the game and had 13 people (I think) watching it happen in real-time. I believe I was number 3 on the DK Junior list at the time.

My understanding was, it did not matter that I was a proven top-tier player with the skillset to accomplish what I claimed. What mattered was that there was no .inp file to verify the validity of my claim.

I would go for a separate leader board for the Pauline version, but if the .inp file cannot be read properly, then I would argue that there should be no exceptions. Or, rework the rules to make exceptions, but that would open multiple cans of worms.

[80sArcadeKid]

I would go for a separate leader board for the Pauline version, but if the .inp file cannot be read properly, then I would argue that there should be no exceptions. Or, rework the rules to make exceptions, but that would open multiple cans of worms.

Ok so Tanya's version is not an .inp file, it's the Pauline rom set in an original PCB. So all the .inp conversation is only relevant to my test or any .inp submissions. (I've attached it if anyone wants to play it back)

The .inp file can be read just fine using the Pauline rom, and plays until the end, so there is no question of it not reading correctly for a Pauline leaderboard, hell I'll submit it and more if we make one! .
The issue is only when playing back with the dkong rom set.

The conversation is more about 'what's goin on here?' considering the hack is just suppose to be a sprite swap so in theory should run just fine in the dkong rom set.

Rev John's insight is helpful but only to someone who knows what's down. Does the original creator of the Pauline rom hang about here? They could probably nail the issue straight away.

[Dallas Taylor]

Ok so Tanya's version is not an .inp file, it's the Pauline rom set in an original PCB. So all the .inp conversation is only relevant to my test or any .inp submissions. (I've attached it if anyone wants to play it back)

The .inp file can be read just fine using the Pauline rom, and plays until the end, so there is no question of it not reading correctly for a Pauline leaderboard, hell I'll submit it and more if we make one! .
The issue is only when playing back with the dkong rom set.

The conversation is more about 'what's goin on here?' considering the hack is just suppose to be a sprite swap so in theory should run just fine in the dkong rom set.
[/quote]

Obviously I do not have the understanding of how these things work that you do. I have a hell of a time getting my own .inp files to playback  I can read what you and Rev John wrote, but for the most part, it's like trying to understand a foreign language.

That being said, I suppose if I were playing a "hacked" version of a game, it would be for my enjoyment, not expecting to have a high score recognized unless a HSL already existed. For me, the best example would probably be the "fixed" version of Junior. I might play it, but if I were to PB, I would not submit the score.

So it's really up to much smarter people than me to decide, but maybe it's time to create a HSL specifically for the Pauline version of the game, or perhaps add it to the existing list with an asterisk type of designator just to denote that it was the Pauline edition of the game.

[sapphire_drag0n]

For the record, I had no idea that there was any difference (other than the images) between Pauline and DK. I'm having a beautiful Pauline cab made up and was testing out the roms for the first time. If Pauline is easier, harder or different to DK in terms of game play, then it really doesn't help my game play and practice. I never, ever wanted to cause or be involved in any kind of controversy or sides on anything that could have any kind of bad impact on an amazing community that I have enjoyed being part of and still enjoy being part of. For the record, I withdraw my submission for my new PB, and will do my best to get it or a better one again on my DK cab. 

Kind regards,
Tanya.

[Rev John]

I fully suspect that Pauline is the exact same difficulty as regular DK.  I'm guessing that the inp de-synced because of a different stream of random numbers being used, but I'm not aware of anything that would make them less random in a way that would affect difficulty.  I think the game play experience would be the same.

[Dallas Taylor]

For the record, I had no idea that there was any difference (other than the images) between Pauline and DK. I'm having a beautiful Pauline cab made up and was testing out the roms for the first time. If Pauline is easier, harder or different to DK in terms of game play, then it really doesn't help my game play and practice. I never, ever wanted to cause or be involved in any kind of controversy or sides on anything that could have any kind of bad impact on an amazing community that I have enjoyed being part of and still enjoy being part of. For the record, I withdraw my submission for my new PB, and will do my best to get it or a better one again on my DK cab. 

Kind regards,
Tanya.

No controversy, for my two cents anyway. Just a discussion among the folks who know how things work and some of us who do not If we can't have a spirited discussion without encountering problems, then we're not much of a community.

I would absolutely NOT withdraw your submission. Xelnia created this discussion and the related poll to hear our input. I say let this thing run its' course, let the poll and the discussion lead the powers that be to come to a final decision and see where it lands.

And for the record, please understand that my opinions here are those of an absolute layman when it comes to understanding code and how the games actually work. That's why I say let smarter people than me figure it out.

[Les.coffman]

I agree that allowing the score opens things for a lot of different issues or conversations about the other versions of DK.  I do agree there should probably be another HSL for those in some shape or fashion, though that then becomes more work for those maintaining the lists.