Posts for Meerkov

1 2 3 4
Posted: 12/29/2023 3:47 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
I kind of wish the commentary was still available. The version with the tas studio comments helps a little in understanding what's going on, but I feel like commentary explaining the routing would be interesting. Yeah I mean, it looks pretty suboptimal, and doesn't help it's several minutes slower than RTA, but still https://www.speedrun.com/neutopia
Posted: 11/19/2023 4:25 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Sure, please do
Posted: 11/19/2023 4:23 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
You can save yourself time just by trying every frame until you get the level you want. By your own analysis you shouldn't have to try more than 100 frames.
Posted: 11/10/2023 6:55 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
For what it's worth, these old comments about killing the boss faster are correct. Two missiles will kill the dragon head. But so will 1 missile, 1 spanner and 1 machine gun bullet. But the second method takes less frames. The head has 0x14 life (20), as does the Neck. Missile does 12 damage. Spanner does 6 damage. Machine Gun does 2 damage. The hit boxes are very precise, so I recommend monitoring the ram addresses to get a more accurate understanding of if you actually hit the right body part or not. I bet the later bosses could be optimized too, unfortunately I didn't want to try the shooting gallery. You can see my WIP on left 1 here. https://tasvideos.org/UserFiles/Game/1252
Posted: 10/25/2023 1:42 PM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
I notice there are differences between this and the Genesis version. In the Genesis version I was able to jump over the steam roller at the beginning of the game by bouncing on his head. Also the UI etc is different.
Posted: 10/10/2023 3:48 PM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
With modern technology we can do better. More precision in TAS STudio is attainable. The first screen is about ~17 seconds long. I was able to cut ~2 seconds off with more optimal movement (I didn't fully finish the level, just wanted to check if it was possible). Assuming the rest of the game is similarly suboptimal, a new run should come in around... 8 Minutes 19 seconds. Estimated available savings: 1 minute 20 seconds. New ideas: 1) Lag Management: Blowing up motorcycles etc causes a lot of lag. You can duck under the 1st motorcycle instead. In various areas the lag caused by explosions is noticeable. Jumping over enemies might be faster. 2) More carefully utilize jump/run in alternating fashion. The current movie is a little sloppy, and the speed boost is roughly 2x. 3) RNG manipulation on bosses - The box shuffling boss and slime boss seem like good candidates.
Posted: 9/27/2023 3:02 AM
(Edited: 9/27/2023 7:38 AM)
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Correct, level 14 is the max. I confirmed the level 14 play pattern works on the original japanese version. Edit: Also I am able to win in around 43 moves by concession, or fewer moves by score counting by abusing the AI more. Here's my example ending state. https://media.discordapp.net/attachments/280808167993245707/1156491241198985228/image.png?ex=651529e6&is=6513d866&hm=ddcef400e428e4ebbf461e51305896d1aa83781d6551742a2df5ae934b8bc18e&=&width=584&height=521
Posted: 9/25/2023 1:01 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
If anyone is curious, this run is not very optimized. There are two main issues: 1) Picking up strawberries only saves a few frames, but costs about 5 seconds per stage for the bonus scene (Total savings: ~30 seconds) 2) The original runner didn't optimize lag frames very well. A good example is how the runner attacks the first boss with balloons while it's invuln. This game is very poorly programmed. Most of the time it runs at around 45 FPS. But with as few as 2-3 sprites on screen, including your own projectiles, it'll drop to 40 FPS, or even 30 FPS. As such, by redoing the stages while paying careful attention to lag management, you can cut seconds off each level while still avoiding the bonus stage. I did a test run of the first two levels, and finished in a total time around 2:01. https://www.youtube.com/watch?v=QxwC37qf9g8 The current run reaches the level select screen at 2:14. The 13 seconds of savings is about ~10 seconds for skipping the bonus stages, and 3 seconds for optimizing lag. I estimate the true optimal time for this game is probably 12:10. I would suggest anyone considering this game check Kid Klown in Night Mayor World to see how it handles lag. Given that KKiNMW came out later, there is a good chance it's better optimized. The bosses are also more interesting, featuring new attacks (e.g. the owl boss now has projecile attacks, likely indicating that the game engine was optimized to not suffer the same lag issue).
Posted: 3/28/2023 5:11 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Looks to me like there are still frames to save. I toyed around on level 1, and you can gain frames by holding the jump longer around frame 700, but letting go of Right for 1 frame to avoid clinging to the rope (you don't have to stop the jump short). And since jump speed is much faster than walking, this probably adds up to several seconds over the run.
Posted: 2/18/2023 8:26 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
A couple things: 1) I can't file issues on Henry's github fork, so there is no way to directly communicate there. 2) There is a lot of talk here about brute force but I don't get it. Reverse engineering is likely a more fruitful path. This is 1989 cryptography on a toy computer. It almost certainly has weaknesses that can be exploited, including the ability to just run the algorithm in reverse, from the desired plaintext message. Consider: the developers would have developed this game already knowing the 8 byte machine code they wanted, and then they would run that through an encryption algorithm. Playing the game runs the decryption algorithm. As long as you know the 8 byte machine code you want to generate, it should be as simple as running the decryption algorithm in reverse. i.e. easy. The fact that the level can by loaded, even if the exit appeared in a suspiciously wrong place... that implies something about the machine code. I.e. that it can't be that complicated. It's bootstrapping the game, and maybe is starting at a slightly wrong offset, or something else like that. The fact you can load garbage levels implies the same thing. Ultimately what those 8 bytes are doing are likely just triggering a level load at an offset in memory or something else straight forward. I would consider looking at the plaintext machine code for the first prize world. It's almost certainly going to be nearly identical to the second. If I was designing the game, why wouldn't it be? Maybe it would be helpful for someone to post the plaintext machine codes for both Prize world 1 (official), and prize world 2 (suspected). Now, mind you, I'm posting this in 2023, and a lot of the historical documents, like the reddit thread, don't seem to exist any more.
Posted: 12/25/2022 6:49 PM
(Edited: 12/25/2022 7:08 PM)
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
https://youtu.be/RcWKfIzSaeo?t=685 It looks like there is a glitch that lets you walk on air. I didn't watch the full 48 minute video, but I didn't see this described in the writeup for this game, so, is this a new glitch? Edit: I tried to reproduce, and it seems like there is just an invisible walkway here? Really weird. Some kind of programmer's mistake due to the moving platforms?
Posted: 12/25/2022 6:46 PM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
So, there is a glitch that lets you walk on air. I have no idea how it works. Does this maybe enable some new skips? https://youtu.be/RcWKfIzSaeo?t=685
Posted: 9/5/2022 2:51 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
I didn't see the script linked in the description, but to me it sounds like this used a breadth first search strategy. While RAM has certainly gotten cheaper over the years, my bet is you could also use a more efficient algorithm. 3-SAT, sounds honestly more complicated than necessary. I'd suggest something like A-star rather than BFS.
Posted: 8/19/2022 5:13 PM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Nice run :) I'm not sure if you know, but you can use the R and L bumpers to tilt the table. This might let you make some shots that you had to take a few tries to line up.
Posted: 4/3/2022 8:19 PM
(Edited: 4/21/2022 10:50 PM)
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
On level 1 it looks like you go backwards at the end to get the pin. Does that cost time? In level 2, when you hit the top of the finish line, does that cost time? Might be nice to add a sentence about that in the description because I wasn't sure. Neat video
Posted: 1/3/2022 3:01 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
While I agree the other movie is entertaining, I'm not sure we should be comparing a playaround run to a simple any% run... I don't really think that's fair. The two categories optimize different things. While the playarounds are cool, they often are *really long and honestly kind of boring after the first couple of images*. A pure any% speedrun without trying to draw silly pictures will tickle a different kind of entertainment factor. That said: Spikestuff is correct, that it looks like this is not yet approaching optimal speed. My gut feeling is that I'd rather see a suboptimal run accepted, and later beaten, than to see no run at all. But I think the precedent tends to be: if the staff beat your run, they reject it out of hand and suggest coming back with a more optimized run. I'd rather the staff could give a clear "if you optimize better, we won't reject you for game choice", which I think is a bit unclear after the last message whether the game would be rejected out of hand or not.
Posted: 1/2/2022 9:04 AM
(Edited: 1/2/2022 9:10 AM)
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
So, firstly, I'd like to say: I think this should be published (to the vault). I don't like the policy of rejecting games just because the aren't interesting. That's a total conflation of audience reception into the judging of the vault. The vault is not supposed to care about audience reception. Barney H&S is one of only ~900 games published for the genesis. These games were bought and played by many of the same people who would have owned star-tier games like Sonic. While the movie may not be entertaining, the game certain deserves a place among other ho-hum platformers. However, I would go a step farther and suggest that this needs a branch of "No-Autoplay", and we should also accept the null-input run as the any% record. While the "null input" has been rejected as trivial (something I disagree with...), this branch (whichever it is accepted to) must implement the no-autoplay rule or else the game will end up in a pickle. Consider this: you can walk right on the last level, and simply end input 1 frame earlier. Barney will do a little jig and then walk the rest of the way by himself. This adds ~25 seconds of video, but wins the any% category against this video. Or, you could end the input 3000 frames early, and get a similar additional amount of seconds of video after end of input... You can see how this would inevitably push people to simply end input earlier and earlier, until we get back to the null input submission (which again: I think we should revisit and accept). However, if the judges are still not convinced about allowing a null-input run, then the rule for this game must simply be that *any* amount of auto-playing is illegal. Which is a weird constraint on any%, but it's better than no submissions accepted at all.
Posted: 1/2/2022 8:20 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
One problem with writing the entire writeup as an extended series of jokes is that it becomes hard to understand what actually is occurring here... If I remember correctly, can't you simply not push any buttons and the game plays itself? What is the reasoning behind not letting the game play itself? Is this intended as a different branch or something?
Posted: 12/31/2021 12:00 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Kind of an interesting run. Message for future runners: the optimization here isn't perfect. The "door trick" is often not good, as it results in losing momentum. Unfortunately, at least the places I looked, I wasn't able to use the momentum to gain any frames, due to other issues with having to wait for enemies or traps to get out of the way. But I think there are still frames to be saved in this game.
Posted: 12/22/2021 4:03 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
It's difficult to resync the movie to mGBA, but from what I can tell, I think the first level can still be faster.
Posted: 12/20/2021 3:02 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Stupendous work 100th_Coin. Great write up as well! I feel like most people, even trained programmers, struggle with assembly. But your writeup hopefully will make this awesome glitch a little easier to understand, or at least appreciate, to the rest of us :)
Posted: 11/24/2021 8:30 PM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Looks good. One question: At 3:45, there is a long hallway and you go diagonal towards the wall before strafe-boosting. Would it be faster to go directly to the wall (left), instead of diagonal? Nice run
Posted: 10/1/2021 2:28 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
Nice run. If you can get the sub, you can save like 15 seconds. That's probably enough time that doing some kind of modification to the route is possible. You can see my TAS of the boss fight with the sub here https://www.youtube.com/watch?v=TTVxBbBU7sU Sync the videos so they both start at "POWER LEVEL RAISED", and you can see that my TAS using the sub gains about 15 seconds on this run. (unfortunately, I started about a minute behind because I had to do the bonus game).
Posted: 4/25/2021 4:35 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
I don't agree with this one's acceptance 8 years later. Another vote for Bizhawk.
Posted: 4/20/2021 1:53 AM
Quote
Meerkov
Experienced Forum User, Published Author, Active player (398)
Joined: 10/4/2015
Posts: 98
For example, there are other similar games. In FILA Decathalon, in the discus throw, the "fastest" completion arguably would be just dropping the discus. That's not very interesting. https://www.youtube.com/watch?v=lLte-pQZ0-0 So for games like that (and this one) going for highest score is probably a better goal.
1 2 3 4