new topic

tar · Post by **tar** » Mon Feb 02, 2004 3:37 pm

O.K.
remember space Zap?
bally midway
1980.
o.k.
I susspect that Satan's Hollow + Solar Fox (1981) Bally Midway .inp's
are outdated.
Too high of scores ,
improper emulation.

zlk · Post by **zlk** » Mon Feb 02, 2004 5:04 pm

Back when Satan's Hollow was in the arcades, there were people who could marathon the game. So I doubt the current scores at marp are too high.

diabolik · Post by **diabolik** » Mon Feb 02, 2004 5:59 pm

http://marp.retrogames.com/index.cgi?mo ... lines=9999

Bomb Bee scores are definitely outdated. There was a bug in emulation, which is now fixed - the game is much harder now.

You could get the ball easily "stuck" between the wall & top bumper for a while, thus gaining lots of easy extra points. This is no longer possible.

Watch any of the better Bomb Bee recordings to see what I mean.

I'm not sure when the bug was fixed - somewhere between mame70 and mame78 (I can check it out if needed). It's virtually impossible to reach scores like boxster's 400k now that the bug is fixed. I would imagine 100k would be a great score now.

LN2 · Post by **LN2** » Tue Feb 03, 2004 12:09 am

are you saying the ball no longer gets stuck for periods of time?

You think this changed somewhere between 0.70 and 0.78? Can you get more specific?

It's behaving the same way in 0.77u2 as it did in 0.66 and 0.71 etc. type versions. The game performs better in 0.77u2 versus 0.66 though so I can use a lesser frameskip.

Perhaps you are seeing similar and using automatic frameskip so auto-using a lower frameskip value which might influence how often the ball gets stuck versus it being a mame version issue.

I guess the easiest test of this is to test replays at various fixed frameskip values and see if it plays back successfully regardless of frameskip value. If it's dependent on frameskip value, then perhaps the ability of the ball to get stuck also is dependent on frameskip value.

More investigation needs to be done...

diabolik · Post by **diabolik** » Tue Feb 03, 2004 6:14 am

Thanks for your input Rick.

LN2 wrote:are you saying the ball no longer gets stuck for periods of time?

Yep.

You think this changed somewhere between 0.70 and 0.78? Can you get more specific?

Well the bug was obviously still there when boxster made his recording using 0.70, but when I played the game using 0.78 it wasn't there anymore.

Anyways: I've done some testing now and I'm 99% sure the bug was gone in 0.75, it was definitely still there in 0.74. I played the game with 0.75 almost 60 minutes and the ball didn't get stuck one single time. I played the game with wolf71-wolf78. I could get the ball easily stuck using versions wolf71-wolf74.

It's behaving the same way in 0.77u2 as it did in 0.66 and 0.71 etc. type versions. The game performs better in 0.77u2 versus 0.66 though so I can use a lesser frameskip.

So the ball still gets stuck in macmame 0.77u2? Is that what you are saying? Are you absolutely sure about that?

Perhaps you are seeing similar and using automatic frameskip so auto-using a lower frameskip value which might influence how often the ball gets stuck versus it being a mame version issue.

I guess the easiest test of this is to test replays at various fixed frameskip values and see if it plays back successfully regardless of frameskip value. If it's dependent on frameskip value, then perhaps the ability of the ball to get stuck also is dependent on frameskip value.

I used frameskip 0 with every version. I don't believe this is dependent on frameskip value.

tar · Post by **tar** » Tue Feb 03, 2004 9:53 am

Rick C.

More investigation needs to be done...

I watched BBH's 238,000 0n shollow2...(m35beta5)dos.
I am satisfied that the emulation is correct.
150k on first ship!
150k higher than other players' .inps
great game

LN2 · Post by **LN2** » Tue Feb 03, 2004 12:05 pm

tar wrote:Rick C.

More investigation needs to be done...
I watched BBH's 238,000 0n shollow2...(m35beta5)dos.
I am satisfied that the emulation is correct.
150k on first ship!
150k higher than other players' .inps
great game

tar, I was only talking about bombbee.

So the ball still gets stuck in macmame 0.77u2? Is that what you are saying? Are you absolutely sure about that?

I'll try a game in 0.77u2 again tonight to be sure....cuz I maybe was using 0.74 the last time I played it.

Given the ROM set wasn't changed and the MAME updates have no comments about bombbee it's interesting you have noted this change in how the game plays.

Something like ball physics would be within the game ROMs to an extent I wouldn't think it would just go away.

...so might be a frameskip issue. I'll do my test. Why don't you test playing it at frameskip 6 and see if you get the stuck issue?

diabolik · Post by **diabolik** » Tue Feb 03, 2004 1:42 pm

OK, tested with frameskip 6, still the same. Doesn't get stuck.

LN2 · Post by **LN2** » Tue Feb 03, 2004 5:20 pm

ok, I just tried a couple games of bombbee in macmame 0.77u2. It does appear as if the STUCK issue on those bumpers is fixed!!!

Just the way the ball moves around definitely has a different feel to it....more realistic.

I guess the last time I played it was in 0.74.

...so we need to specifically narrow down when this was fixed so for bombbee a special rule can be added that at least that version or later of MAME must be used.

..then remove all scores using older versions.

..so it doesn't get stuck in 0.77u2...so that version is ok.

I just retried bombbee in 0.74u2 and the ball DOES get stuck in macmame 0.74u2....so for macmame 0.77u2 or later must be used.

There were no official versions of macmame between 0.74u2 and 0.77 so that's as far as I can narrow it down.

So...try the same for PC versions to find specifically which version between 0.74 and 0.77 fixed it. Odds are it's the same as macmame would be if we had all those versions.

Test bombbee in 0.75 and 0.76.

Then we are in business. One of the scores removed will be mine....

but I will play it some to set a new score.

I agree I think 100k would be a quite good score now. It's really important to get those middle targets now so you get that 1000pt bumper and use that all up as often as possible.

tar · Post by **tar** » Tue Feb 03, 2004 5:43 pm

LN2

I was only talking about bombbee.

Well tell SJV to start his own topic.
.
Solar Fox .inp has been looked at ,
I find it to be excellent. 1,200,000 score.
.
Diabolik , great Space Trek .inp 1,000,000 score.
I viewed it.
wrx2 got a little more ,
but then we understand.

JTC

It's just before dawn on a cold Korean morning. The WRX2 headquarters compound is dark and silent. A recording is submitted to MARP beating one of the WRX2 league's first place scores. This automatically generates a message picked up by the sentry currently on watch who leaps into activity. He madly cranks the siren which sends the alert sound wailing through the previously silent grounds. Suddenly all hell breaks loose. The elite corps of Asian video game aces scramble like fighter-pilots in an air raid, throwing on their clothing and sliding down firehouse poles to their computer stations where they begin the desperate charge to regain the high score.

lol..!! . funny
tar

Francois Daniel · Post by **Francois Daniel** » Tue Feb 03, 2004 6:10 pm

LN2 wrote:...so we need to specifically narrow down when this was fixed so for bombbee a special rule can be added that at least that version or later of MAME must be used.

I agree with you Rick, but if we made a special rule for this game, why we don't make a special rule for cps1 games who have difficulty and lives setting changed lately. These games (1941, 3 Wonders, etc) are more difficult since 0.73 or 0.74 and marp records are quite impossible to beat with these new settings ?

Francois

diabolik · Post by **diabolik** » Tue Feb 03, 2004 6:44 pm

LN2 wrote: So...try the same for PC versions to find specifically which version between 0.74 and 0.77 fixed it. Odds are it's the same as macmame would be if we had all those versions.

Please read my previous post.

I already tested it and came to conclusion that it was fixed in 0.75. So it's now up to editors to do their job... Any of you editors reading this?

tar wrote: Diabolik , great Space Trek .inp 1,000,000 score.
I viewed it.
wrx2 got a little more ,
but then we understand.

Thanks.

Francois Daniel wrote: why we don't make a special rule for cps1 games who have difficulty and lives setting changed lately. These games (1941, 3 Wonders, etc) are more difficult since 0.73 or 0.74 and marp records are quite impossible to beat with these new settings ?

I'm not familiar with those changes, but wouldn't it be possible to make special rules for these games which give permission to use those old, easier settings?

LN2 · Post by **LN2** » Wed Feb 04, 2004 12:51 am

Francois, if just a difference in default dip switch/game settings then those games could still use the settings the older versions of MAME used if there are lots of scores for the game using those rules.

That would mean adding special rules specifying what settings you can use...instead of it just being the mame default settings for the game.

That's different than the bombbee situation, where evidently the emulation of the game is now different and improved...likely more accurately than before.

Think of it like if the games you listed had some bug in an old mame that allowed you to easily score quite high...and the bug was fixed in a later version of mame so you can't score nearly as high...I'm sure you would agree the best thing to do there is remove all those older scores and require at least that version be used.

I remember seeing a few other games with this situation also here at MARP...think was forum discussion on them...no special rules added for a couple I think. Sorry, I can't remember offhand what the others were...would know when I happen to see the game again or a new upload for it etc.

----for the editors---to put it in a nutshell

for bombbee, we need special rules requiring mame 0.75 or later be used for the game...and remove all scores for it that are using older versions...which currently might be all of them for bombbee.