forums.marpirc.net

Special rules which probably should be mentioned.

1) circus. Your game is disqualified if on the first man your clown touches any ballons before the seesaw. DUe to bugs in emulation and/or original code, you can get a massive score boost by coining up before the clown leaves the platform on first boot.

Additionally i suspect there is an error in emulation that makes the game play easier than the real arcade, but this is unconfirmed.

2) Cadash: this game requires seperate scoreboards for each character. The arcade itself does this. The internal high score chart is scored by exp, with gold as a tiebreaker. Infinite leeching is impossible, because of the timer system, and the fact that the cost of extra time items and inns keeps rising. Non infinite leeching is part of the game, speaking as soeone who has one credit celeared the real arcade. The game is balanced for leeching, IMHO, as there are places where getting hit is unavoidable, and or uneconomical because of the timer. I think that speedrunning the game is the true test of play.

3) golden axe. This game scores by points, not by strength. I believe the maximum score is obatinable without dying, and probably with all characters. however some may fidn it easier. The game scores the thre characters seperately. SUggest useing strength as a tiebreaker.

4) playchoice games. I believe that they should be archived when a non playchoice version of the game exists. PC_10 SMB is one example. Another one would be DR Mario. And a third would be rush'n attack. A game that should NOT be archived would be PC_10 Pinbot, or perhaps pc 10 rad racer. Additionally, some that do not have other arcade versions, but were designed with console in mind, like PC_SMB3, should be archived. TO sum up, if an arcade version exists, or if the game is badly balanced for arcade play, it shouldn't be valid for a point based competition, though it' s fine for a speedrunning competition.

5) TMNT (and any other game with it's braindead scoring system). i vote archive. Even honest play incorporates leeching to a certain extent. This may in fact be true for all in the series, but i cannot confirm this. But is is absolutely confirmed for TMNT, and it sucks to lose because the person ahead of you got one more boomerang then you did. Either that, or implement an alternate score system, perhaps involving not getting hit? Also suggest only first round counts rule, as i belive some games like this can be marathoned forever.

6) suprmrio. bases almost covered, but note there are very few places you can turtle stair leech in this one (you really gottta go out of your way to find one!), and every single recollectable 1-up has been removed, and i believe it impossible to collect 100+ coins in a stage, die, then collect 100+ coins again, even with a checkpoint exploit by timeout in cloud heaven,pasageway,etc. Because of the huge extra life bonus at the end , i believe point pressing t the cost of lives to be counterproductive, as long as the turtle stairs trick is banned and the time bonus trick is banned.

_Zaphod_ wrote:2) Cadash: this game requires seperate scoreboards for each character. The arcade itself does this. The internal high score chart is scored by exp, with gold as a tiebreaker. Infinite leeching is impossible, because of the timer system, and the fact that the cost of extra time items and inns keeps rising. Non infinite leeching is part of the game, speaking as soeone who has one credit celeared the real arcade. The game is balanced for leeching, IMHO, as there are places where getting hit is unavoidable, and or uneconomical because of the timer. I think that speedrunning the game is the true test of play.

There is no discernable difference in gameplay between characters, it's really no different than any other game that has selectable characters.
As for the scoring, as it is, you rank higher when finishing the game if you have less experience. This is to cut down on leeching, and ties (as the game maxes out at 65535, from memory).

_Zaphod_ wrote:3) golden axe. This game scores by points, not by strength. I believe the maximum score is obatinable without dying, and probably with all characters. however some may fidn it easier. The game scores the thre characters seperately. SUggest useing strength as a tiebreaker.

What are you on about? Golden Axe is currently scored by the score, rather than strength. Strength is largely based on score, so it's not really suitable for a tiebreaker.

_Zaphod_ wrote:4) playchoice games. I believe that they should be archived when a non playchoice version of the game exists. PC_10 SMB is one example. Another one would be DR Mario. And a third would be rush'n attack. A game that should NOT be archived would be PC_10 Pinbot, or perhaps pc 10 rad racer. Additionally, some that do not have other arcade versions, but were designed with console in mind, like PC_SMB3, should be archived. TO sum up, if an arcade version exists, or if the game is badly balanced for arcade play, it shouldn't be valid for a point based competition, though it' s fine for a speedrunning competition.

PC-10 games are just NES games, running off a timer. It's a given that they aren't really suitable for arcade play. But this system was in arcades, so not much can be done there. Same goes for Mega-tech and Nintendo Super System - but not Megaplay or Vs. games.

_Zaphod_ wrote:5) TMNT (and any other game with it's braindead scoring system). i vote archive. Even honest play incorporates leeching to a certain extent. This may in fact be true for all in the series, but i cannot confirm this. But is is absolutely confirmed for TMNT, and it sucks to lose because the person ahead of you got one more boomerang then you did. Either that, or implement an alternate score system, perhaps involving not getting hit? Also suggest only first round counts rule, as i belive some games like this can be marathoned forever.

Stupid scoring it may be, but that's what it is. Blame Konami.

_Zaphod_ wrote:2) Cadash: this game requires seperate scoreboards for each character. The arcade itself does this. The internal high score chart is scored by exp, with gold as a tiebreaker. Infinite leeching is impossible, because of the timer system, and the fact that the cost of extra time items and inns keeps rising. Non infinite leeching is part of the game, speaking as soeone who has one credit celeared the real arcade. The game is balanced for leeching, IMHO, as there are places where getting hit is unavoidable, and or uneconomical because of the timer. I think that speedrunning the game is the true test of play.

Just because a game has separated scores by character does not necessarily mean it should be split by character on MARP. Gauntlet has separate character scores but none of the romsets on MARP are split. Splits should only be done when a different character has a huge impact on the game, such as if it makes playing an entirely different set of levels possible. The characters in Cadash aren't different enough to warrant a split.

The game used to be scored by experience, but it maxes out at 60,000 (or was it 65,535? whatever), and tie scores are never good. If you don't finish the game, you use your experience count, but if you do finish, it's 150,000 - whatever your experience was at the end of the game. This encourages people to NOT max out the score, and finish the game with less experience, which is harder... that's sort of like speedrunning, right? So what's the problem?

4) playchoice games. I believe that they should be archived when a non playchoice version of the game exists. PC_10 SMB is one example. Another one would be DR Mario. And a third would be rush'n attack. A game that should NOT be archived would be PC_10 Pinbot, or perhaps pc 10 rad racer. Additionally, some that do not have other arcade versions, but were designed with console in mind, like PC_SMB3, should be archived. TO sum up, if an arcade version exists, or if the game is badly balanced for arcade play, it shouldn't be valid for a point based competition, though it' s fine for a speedrunning competition.

Playchoice-10 SMB and Vs. SMB aren't identical games though... there's already many clones in MAME with no discernible gameplay differences as far as anyone can tell, so why remove PC-10 versions just for being on the Playchoice-10?

Sure some games are unsuitable for serious competition, but I believe there are at least a few games that have been changed to the speedrun scoring type (SMB2, Power Blade?)

5) TMNT (and any other game with it's braindead scoring system). i vote archive. Even honest play incorporates leeching to a certain extent. This may in fact be true for all in the series, but i cannot confirm this. But is is absolutely confirmed for TMNT, and it sucks to lose because the person ahead of you got one more boomerang then you did. Either that, or implement an alternate score system, perhaps involving not getting hit? Also suggest only first round counts rule, as i belive some games like this can be marathoned forever.

I don't remember TMNT ever looping. The scoring system sucks, but yeah, blame Konami for that one. A "not getting hit" scoring system doesn't sound any better to me.

6) suprmrio. bases almost covered, but note there are very few places you can turtle stair leech in this one (you really gottta go out of your way to find one!), and every single recollectable 1-up has been removed, and i believe it impossible to collect 100+ coins in a stage, die, then collect 100+ coins again, even with a checkpoint exploit by timeout in cloud heaven,pasageway,etc. Because of the huge extra life bonus at the end , i believe point pressing t the cost of lives to be counterproductive, as long as the turtle stairs trick is banned and the time bonus trick is banned.

Turtle stair leeching is already banned, so what's the problem here...?

BBH wrote:
Just because a game has separated scores by character does not necessarily mean it should be split by character on MARP. Gauntlet has separate character scores but none of the romsets on MARP are split. Splits should only be done when a different character has a huge impact on the game, such as if it makes playing an entirely different set of levels possible. The characters in Cadash aren't different enough to warrant a split.

I think they are different enough to warrant a split, especially for scoring purposes.

Fir example, early on the wizard and the priest can strike the kraken boss directly, while the warrior and the ninja have to strike at the arms. The game itself doesn't change, but your strategy changes significantly with the characters. The priest has much more difficulty with tthhe crawling kelp boss then the other thhree. The wizard has toleech to get the best spells, etc.

The game used to be scored by experience, but it maxes out at 60,000 (or was it 65,535? whatever), and tie scores are never good. If you don't finish the game, you use your experience count, but if you do finish, it's 150,000 - whatever your experience was at the end of the game. This encourages people to NOT max out the score, and finish the game with less experience, which is harder... that's sort of like speedrunning, right? So what's the problem?

Mainly that the low exp game is not necesarilly the fastest, nor the most skillful. It is sas that the exp on the xcoreboard can be maxed out though... I think because of this scores from incompleted games probably shouldn't even count, if it is possible to max the exp without completion. A low exp game will probably have it's owntype of leeching, killing low exp but higher gold monsters, and lettign high exp ones survive, and doing enough leeching to afford the time and equipmen tot survive. it's just my opinion. but i still stand by my statement that the characters play very different, much more so than gauntlet.

And i never maxed out my exp when I beat the game. ever. i simply leeched enough at the rock monsters to buy a truckload of bells at the bell shop, to give me enough hitpoints to allow me room for error when playing through the rest of the game, ad enough levels to not suffer too badly at the graveyard zombies. (both went hand in hand). i abuse the ninjas penetrating shot on the way to the castle, and the diagonal shot onthe boss first forn, and would always sit and TANK inside the final form of the last boss and jam on the fire button till it died. And i abused his fast walking speed to get in more leeching. The main reason i think seperate boards for each character is so we can get players of all characters on the leaderboard. As it stands, with gauntlet 2, the elf will take the top scores because it a) had good magic, b) can shot thru cracks, and doesn't have the bad luck penalty of the wizard. Because the game rans up based on score, the elf has the best chance at taking little damage, and o f getting food when the gaem starts sarving you once you scored past the magic X million point mark. This game remains non slipt, bcause picking a weaker character gimps you. But each character in cadash is capable of beating the game, when played the correct (and different for every vcharacter) way! but they will not score equally.

Playchoice-10 SMB and Vs. SMB aren't identical games though... there's already many clones in MAME with no discernible gameplay differences as far as anyone can tell, so why remove PC-10 versions just for being on the Playchoice-10?

That's my point thy are console games, an offten not balanced properly for scoring. and i'm not talking about SMB2 or the like. I'm talking about games that throw out extra lives liek candy and make it impossibly NOT to accidentlly leech. SMB1 fpr PC-10 is leech central, and if I really wanted competition on this game, i'd implement it thusly.
1) no turtle stair.
2) one life only. after you die once, your game is scored. Use any other scoring trick you want.

But why bother when there is a version that is balanced for arcade play? Why play Mario Bros on pc-10 when you can play the original? Why play the famous gun games on pc-10 when they are better for he unisystem? Vs. Gradius should be played instead of PC10 version. The PC-10 rules under certan circumstances encorage dieing to squeeze out more points, an in other cocumstances encourage leeching (when leeching gets points faster than not)

Sure some games are unsuitable for serious competition, but I believe there are at least a few games that have been changed to the speedrun scoring type (SMB2, Power Blade?)

They woud have to. but many games that DO have scores are just as unsuitable for arcade play. I'd put SMB1 into that category, without the special rule i suggested. When an honest attempt with less skill scores more then a more skilled attempt simply by fact of the less skilled person dieing. we have a problem.

I don't remember TMNT ever looping. The scoring system sucks, but yeah, blame Konami for that one. A "not getting hit" scoring system doesn't sound any better to me.

More than any other game, this demostrates the problem with leeching. there is a fixed number of points available to the player without leeching, regardless of player skill. All of them are collected when you beat the game. only thru leeching can you score more. Every enemy is worth one point. there are a fixed number of enemys. therefore we have a maximm score when beating the game. except that every shot cancelled is another point. Do we place limits on the number of projectiles for any given enemy you are allowed to strike? Clear scores without leeching are very similar, and differences between them are not inicative of skill. in fact, a lower clear score is indicative of more non leeching skill. of course this means there will be ties. Make it a speedrun or archive.

As for golden axe. the game has a theoretical max (knock every enemy it is possible to knock off the edge with a killing blow.) It is posible to reach this with and without dieing. but dieing results in a lower strength. Hence strength makes a good tiebreaker, because ties can easily happen in all clear scores for this reason. I have not reached this theoretical max, but i have 1 life cleared the game. And i usully get the same score when I do.

_Zaphod_ wrote: I think they are different enough to warrant a split, especially for scoring purposes.

Fir example, early on the wizard and the priest can strike the kraken boss directly, while the warrior and the ninja have to strike at the arms. The game itself doesn't change, but your strategy changes significantly with the characters. The priest has much more difficulty with tthhe crawling kelp boss then the other thhree. The wizard has toleech to get the best spells, etc.

We have had policy that splits go with different "tracks" not with different characters. Imagine how many splits there would be at some fighting games. Street fighterII has already over 20 clones...and to split them more...

That's my point thy are console games, an offten not balanced properly for scoring. and i'm not talking about SMB2 or the like. I'm talking about games that throw out extra lives liek candy and make it impossibly NOT to accidentlly leech. SMB1 fpr PC-10 is leech central, and if I really wanted competition on this game, i'd implement it thusly.
1) no turtle stair.
2) one life only. after you die once, your game is scored. Use any other scoring trick you want.

Yes, I agree, one life only would have been a good rule for pc_smb, and would have made competition more interesting, less leeching.
Also no "turtle stair" sounds good.
It would be somewhat awkward to change the rules now, when it has been competed with these rules so long time. It always sucks to delete all old recordings, while other option being to try playback all old recordings and look for 1man score. First option is harsh, second is difficult(to playback old recordings sometimes is) and worky.

However I don't rule out the possibility changing pc_smb rules. In fact that sounds very reasonable.
Cheers,
TJT

_Zaphod_ wrote:I think they are different enough to warrant a split, especially for scoring purposes.

Mainly that the low exp game is not necesarilly the fastest, nor the most skillful.

2) one life only. after you die once, your game is scored. Use any other scoring trick you want.

You have some good points but (as others have said) the basic rules of split for different tracts only (not different characters) keep a lot of intensivley numerous splits away and keeps things simple. We'll just sweep golden tee under the carpet on that one... heh. Those interested in each game would like to see different ways to play not just the designated marp highscore way, but you can also take that energy to play several different characters and focus it into playing for lowest points, or lowest speed. At http://www.homeactionreplay.org , there is no hard line rule of keeping splits with in the boundaries of tracts although we try to follow marp as much as possible, but there isn't any cadash clones there yet. For mario, i do like the one life only instead of trying to watch for leeching, makes things slightly easier on the confirms, then again we'd have to nuke recordings, or make a scoring split (something that hasn't been done often at marp.)

_Zaphod_ wrote: But why bother when there is a version that is balanced for arcade play? Why play Mario Bros on pc-10 when you can play the original? Why play the famous gun games on pc-10 when they are better for he unisystem? Vs. Gradius should be played instead of PC10 version.

you come to a retrogaming site, you know mame's main purpose, and ask these questions? To preserve the history, and to potentially find interesting differences even though they look alike.

The different characters are effectively different tracks, even though they go thru the same screens.

If the GAME doesn't split up the char scores, then the arcade gods have spoken, and there is to be no character split. But if the game has seen fit to put in different score tables, then sometimes a split is justified. But not all the time. In gauntlet, every character moves in 8 dirs, shoots, fights, and uses magic. Cadash the chars are incredibly different and play the same game very different ways. With figuting games, it's just silly to split. There's so much flexibility that by any reasonabel scoring metric, everyone can outdo everyone esle by playing better. Well there is ONE excption. World Heroes. The "Brocken character can speed thru the game liek NO other, because of a very annoying feature whereby if he pummels you with weak atacks, and knocks you dwn with a stringof them, it's INSTANT KO! this makes him king at speedrunning the game. and it's not a bug, as near as I can tell. the game never was well balnced. sequels improved in that regard.

EDIT: Never mind. i CANNOT reproduce this in mame! I did it MANY times on the arcade hardware, but this does NOT occur in either of the two mame sets now!

In general when a game cannot be fairly scorred by the games own rules. i think the default special rule should be time taken to beat the game is score, ad low score is better. When this doesn't work (SMB1 pc-10 is a good example here) then other special rules must be made to fix it. This one life rule should be applied to all games that allow infinite leech thru extra lives that can't be fixed by dipswitch adjustment.

_Zaphod_ wrote:The different characters are effectively different tracks, even though they go thru the same screens.

This is your choice. Choose character best for score.

There aer a number of bugs in SMB2 that can be abused in speedrunning.

1) If you jump, and graze the top of an enemy without getting hit on the way up, you can jump a second time if you press jupm before you start falling.
2) you can jump off of ladders by jumping the instant you leave the ladder. this can be used for mondo shortcuts in th e final world.
3) numerous other glitches can be abused.

This may not be considered a problem, but you probably shoudl know.

Chad wrote:We'll just sweep golden tee under the carpet on that one... heh.

Exactly how was that mistake made and what can be done to rectify it?