I mixed a few styles up together. I just don't know what style is the most popular with most people.
The Only Improvement
Not lowering flags
I once got this idea, and I made it the first time after hundreds of failures. I don't know how it works exactly, but I am sure that you have to somehow get into the block under the pole and then jump up. You may need to steer or NOT to steer to SOME directions when you are inside the block. If you make it, you'll skip the flag-lowering animation.
Note: I didn't use this trick in stage 8-3 otherwise I would have had to decide whether to waste 42 frames in watching three fireworks or not to by wasting 19+ frames somewhere else to miss them....
Special Thanks
Thanks to Phil, genisto and pom for their well-optimized movies that I studied and JXQ for his FCMFix tool.
NesVideoAgent: Hi! I am a robot. I took a few screenshots
of this movie and placed them here.
Oh! I also corrected the ROM name.
You indicated Super Mario Bros. (JU) [!]
I updated it to Super Mario Bros. (JU) [!].nes
adelikat: I guess no movie is unimprovable :P Accepting the movie due to awesomeness, and encoding.
Well, I am currently working on SMB1 warpless, SMB1 FDS and those SMB2j runs as mentionned in my homepage 30 minutes after this get submitted. But if someone is already working on some runs, I would like to be informed so I and he won't lose free time for nothing. Also, Bionic Commado project is paused for that, don't worry, I'll continue working on it after those runs get done.
How did I forget that! I shouldn't have sumitted this until I had TASed ALL the other possible runs...just kidding.
Well, seriously, I had been working on SMB2j warped run for 4 days before I decided to TAS SMB warped run first. What is that bad is that I think that it is that this new trick that the only thing that will make that new movie look that different from that old one.
EDIT: I like "that".
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Joined: 12/22/2006
Posts: 193
Location: Flowood, MS
My hypothesis on why this trick happens is based on the following:
1. If Mario grabs the flagpole, the flag lowers, and then something happens that stops him from making it to a solid block to run into (which triggers the score tally and loading of the next stage), then mario can go back and grab the flagpole again. However, the game recognizes the flag is down and thus doesn't play that animation again.
2. In this situation, Mario grabs the flagpole while he's intersecting the block. On the very next frame, the block ejects him, but Mario is still holding onto the flagpole.
Since this happens within two frames, the game has time to register that the pole has been grabbed, but not enough time to play the flag animation before the block ejects him, so when the block ejects him I think Mario is actually grabbing the pole a second time. Given that 1 is true, I think the game doesn't play the animation because it thinks the flag is already lowered.
That's my thoughts on it, anyway.
Also, jumping through the stairs in 4-2 isn't possible because there's no ceiling to further push Mario into the wall (so if he jumps, he just jumps out of the block).
<adelikat> tony hawk is porn for me
<Comicalflop>my mom is hot
I'll be darned if this new trick isn't applicable to Super Mario Bros. 2.
But aren't they on completely different engines?
I think i've watched about 70% of the TAS's on this site... It's become one of the first sites I check daily for updates.
I'm going to start posting more.
This trick is the same to do with big Mario.
And this page needs to be updated.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Any working theory on the trick?
Such as "if you grab the flagpole beneath its bottom (by entering the block it is standing on), the game will skip the flag lowering animation, giving a faster completion time"?
Any working theory on the trick?
Such as "if you grab the flagpole beneath its bottom (by entering the block it is standing on), the game will skip the flag lowering animation, giving a faster completion time"?
Also, how does this effect the 21 frame rule and fireworks? I'm not clear on either, so it may have zero effect.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Also, how does this effect the 21 frame rule and fireworks?
The 21 frame rule and fireworks work just fine without the trick and aren't improved by the trick, so it can't be said that the trick effects them.
You probably tried to use the word "affect", though.
The fireworks are subject to a timer ending digit rule, as always.
The 21 frame rule is subject to the screen blanking as always.
My theory about that bug is that, if you check with an editor or you already toy with the game, you know that the flagpole is anchored in the ground. By ground, I meant the lowest part of the screen( Maybe even below that) not the floor where Mario walk. So, Mario touches the flagpole in the ground.
My theory of why the flagpole doesn't fall is simple. From the top of the block, the one that graphically makes look the flagpole is on, is Y = 0. And top of flagpole is Y = 0 + a. But since Mario hits the flagpole under point zero, Y = 0 - b. Then, the flagpole doesn't drop because it's not programmed to activate something with these values but you still got 100 points as it is the minimum points set that the game gives you.
Any working theory on the trick?
Such as "if you grab the flagpole beneath its bottom (by entering the block it is standing on), the game will skip the flag lowering animation, giving a faster completion time"?
Add "you have to grab the pole during a small jump, but not to jump too high or too low" to it. Better than nothing, at least.
EDIT: I agree with Phil.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Great find! I was entertained. Very yes vote.
BTW, is this an 84 frame improvement, (17996-17912) as stated in the NES forum topic, or a 62 (17974-17912) frame improvement, due to Pom ending input at the frame of the axe hit? Just from my preference I like the "earliest time input can be ended and still complete the game" option, but I know some people like to keep prior timing from previous runs for continuity.
IMHO, ending on the earliest possible input should have long since been established as the standard for all tool-assisted speedruns.
It really is interesting that the flag wouldn't drop when it was touched.
Since the game is so short, I was not expecting there to ever be and improvement of this run.
Does anyone think it could be improved even more? I wouldn't imagine there wouldn't be anything else to improve except if you could skip the fireworks animation in a similar way or something in level 4-2.
Super Mario Bros. console speedrunner
- Andrew Gardikis
Horray for new world record!
According to doppelganger's SMB disassembly the trick works because Mario Y coordinate > 0xA2 at the moment of touching the flagpole.
lda Enemy_Y_Position,x ;check flagpole flag's vertical coordinate
cmp #$aa ;if flagpole flag down to a certain point,
bcs GiveFPScr ;branch to end the level
lda Player_Y_Position ;check player's vertical coordinate
cmp #$a2 ;if player down to a certain point,
bcs GiveFPScr ;branch to end the level
IMHO, ending on the earliest possible input should have long since been established as the standard for all tool-assisted speedruns.
That's not so simple as it may sound.
Ending the movie on the earliest possible input doesn't necessarily mean that the *game* is ended at the earliest possible point. On the contrary, ending the movie at the earliest possible opportunity can cause the game ending to be delayed by a significant amount, even to the point where entertainment suffers considerably. A recent RCR submission was an extreme example of this: The movie was ended some seconds before a normal "end at the moment the boss dies" movie would have, and that caused the game ending to be delayed by something like 2 minutes, IIRC. While technically interesting and amusing, from a viewer's point of view it was an entertainment-killer: Having to wait for 2 minutes for the final boss to die was quite a let-down.
Personally I agree with Bisqwit here: http://tasvideos.org/Rules.html#3TheExactTerminationPointIsSubjectToDebate
"Ending the movie at the moment when no further input can cause the game to end faster" is a rather sound rule of thumb.
Joined: 11/18/2006
Posts: 2426
Location: Back where I belong
You are correct, Warp, and I remember watching that RCR by Chamale and wondering what the hell was going on at the end. So I agree that there are certainly exceptions. For the majority of games, however, it is a rule that works well, and provides a very definite way of comparing movies. I understand the points made in the rules section, and agree with them to some point, but for general continuity I think it's the easiest way to compare movies. But as long as there are rules, there will be situations where those rules don't apply.
