Posts for hopper


1 2
7 8 9 32 33
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
BoltR wrote:
That's what the publication process/the front page is for. Quality control for the people who only want to watch the good videos.
That's not what I want, though. I want to watch the unpublished videos for a number of reasons. I want to watch some of the videos that have mostly No votes so that I can provide constructive criticism and encourage the author to try again. I also like to watch videos that have a lot of Yes votes to see if I can catch something that others have missed. Also, if there's no consensus, sometimes I like to watch a video to see if I can offer an insight that might lead to a resolution. That may seem like it covers every possible scenario, but I don't want to just watch every single submission, and I certainly don't have time to do so. No, I use the votes to decide how I want to spend my time. I choose to watch this video or that video based on criteria such as how much time I have, what game it is, and how the voting is going. I'm just saying that the polls help me decide what I want to watch. Maybe they don't help you, but they help me. And if they help me, they may help others.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
BoltR, I don't think you've understood what I was trying to say.
BoltR wrote:
hopper wrote:
While it seems justifiable to ask judges to read through an entire thread, most Mario/Zelda/Metroid runs draw 10 or more pages of comments, and a lot of bitter debate. I can't stand reading page after page of people sniping at each other, even if I had the time to read through that many responses.
Having to sift through garbage is part of the job description of being a judge.
What I meant was, while it seems justifiable to ask judges to read through 10 pages, it's not reasonable to expect everyone else to do so. I can't stand reading page after page of people sniping at each other, even if I had the time to read through that many responses.
BoltR wrote:
hopper wrote:
The concern I have about not having polls, or hiding poll results from non-judges, is this. If a run is obviously exceptional, the first page may be all that you have to read to get a sense for how the run is regarded.
This is exactly why you aren't suppose to use the votes to decide if you are going to publish it or not.
Again, I'm not talking about the judges. Of course the judges aren't supposed to use the votes to decide if they're going to publish it or not. The regular visitors do use the votes to decide which videos they're going to watch. I'm not a judge, so I'm not obliged to read 10 pages of discussion. I just want to watch the video and then post my opinions, with or without reading the opinions of others first. The poll results help me to do that. Without that resource, I'm less inclined to participate, and others may be discouraged from participating as well.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Bisqwit wrote:
Another idea: Re-enable voting, but make poll results viewable only to judges.
That sounds like it would be easier for the judges. While it seems justifiable to ask judges to read through an entire thread, most Mario/Zelda/Metroid runs draw 10 or more pages of comments, and a lot of bitter debate. I can't stand reading page after page of people sniping at each other, even if I had the time to read through that many responses. The concern I have about not having polls, or hiding poll results from non-judges, is this. If a run is obviously exceptional, the first page may be all that you have to read to get a sense for how the run is regarded. Likewise, a video that clearly doesn't meet our standards will immediately be obvious when it receives comments like, "way too many mistakes", or "it looks like you just played the game in slow motion instead of using re-records to make everything perfect." But when a run's merits require debate, it's impossible to tell if there are more Yes votes or No votes without actually reading every post. Runs that are clearly exceptional, or clearly crap, don't truly require debate: any judge could instantly accept for being an obvious improvement, or instantly reject for not being a proper TAS. The runs that require feedback are runs of games that have never been published before, where there's the possibility that time-saving strategies have been overlooked, or where there is no clear consensus on whether a run meets all of the requirements for entertainment and perfection. Those are the runs that people need to be encouraged to watch, but the lack of a running poll may discourage people from watching. My time is limited, and I like to know what people think of a movie before I watch it. The poll results gave me instant insight into how a movie is regarded, but now the only way to know is to read the discussion thread, and if the thread has several pages of responses, lazy... I mean, busy people like me may not bother.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
This wasn't as bad as some comments had led me to expect. It certainly didn't show off any astonishing combos, huge smashes or awesome fighting. It aimed to be fast, and it was. I liked Samus' death. It didn't really last long enough to become boring, nor was it especially entertaining. I probably should have voted Meh, but it met its goal of being fast, so I voted Yes without really considering the Meh option.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I forgot to mention that you took a lot of shots at the boss on the right in Floor 50 after it was already dead. That probably cost about a minute and, being at the end, it would have been easy to fix. Nevertheless, you should start over using two players, and skip the cutscene at the beginning. This is a really entertaining game, and no one else has seemed interested in making a TAS of the Genesis version. If you can redo this run with no obvious errors, it's guaranteed to be published.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Okay, I've seen the whole thing now. The self-imposed "kill everyone simultaneously" goal is a really bad idea in a one player game. Getting everything set up to make that happen takes forever and is incredibly boring. That's something that should never be said about Snow Bros. When two reasonably skilled players play this game, it can be enormous fun to watch! Some of these stages drag on and on. It's not entertaining, so I vote No. The music in the last level (Floors 41-49) is actually one of my favorite pieces of video game music, but it actually sounds a lot better in the NES version than the Genesis version. That's the only thing I prefer about the NES version of this game.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
First of all, I'm glad that someone has finally made a TAS of the Genesis version of Snow Bros. I agree with everything that GeminiSaint said about this game's superiority to the NES version. I've been saying as much in the NES version threads for... years? This was one of my favorite arcade games when I was young, and I still enjoy playing the arcade game today. That said, you wrote right in your submission text that this run is improvable. That's a no no right off the bat. I could have guessed that it was improvable just from reading the 34 minute completion time in the topic title. I'll watch the run anyway before voting, but it's late and I don't have half an hour to spare right now. It doesn't look like this is going to be published, but thanks for trying, and please try again. Edit: Okay, I just watched the first 20 levels, and I agree that it doesn't look very TAS-like. For one thing, you could freeze bad guys and kill the first boss much faster if you manipulated the game into giving you the blue power-up sooner, instead of getting the red one every time. You didn't have the more powerful snowball power until the end of Floor 13!
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
klmz wrote:
Strangely, most of the time during watching I was missing the old camero positioning. What's making me feeling that way? EDIT: Maybe it was the low frame rate of the game combining with no interframe motion-blur. I even felt slightly a bit dizzy(really?).
I tend to not like it when the camera is zoomed in too close to Mario, which gives me a headache, but that didn't happen long enough in this run to give me a headache. I didn't care for the zoom level/camera choices for a few seconds near the beginning of Bowser1, but I've generally felt that none of the runs have had camera work as good as FODA's original 120 star run. I don't have a problem with the camera work in this run. I don't really care for Mario taking unnecessary damage in BitFS, and I have no idea what he was doing after he tossed Bowser. It looked like he was having a seizure for a few seconds there. I really like the new route in BitFS, though. That time saver more than makes up for Mario's strange behavior. Yes vote.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I'm sure the issue of obsoletion has been discussed to death already, but I'm not going to read 6 pages to find out. To me, there are three obvious reasons why this should be listed as a "glitched" run like SNES "glitched" Zelda: A Link to the Past, and should not obsolete non-glitched runs. 1. Like the stair glitch in Zelda: ALttP, Samus glitches the game so that her position doesn't match the picture on the screen, allowing her to be out of phase with reality. She jumps on things that aren't there, but do exist in the room that her location is at. This game is glitched at a very low level. 2. This glitch is even more extreme than the simple opposite direction glitch that allowed Link to walk through walls in ALttP. In order to screw things up enough, it requires a saved game to be created, switches between saved games, and even dies in the alternate game in order to get garbage data that looks like a door loaded into memory. This is way fucked upper than ALttP. If reason 1 is good enough for SNES Zelda, than reason 2 is even more compelling for Super Metroid. 3. X-Ray glitching up 11 rooms for the better part of 5 minutes is boring!!! We always talk about pure frames vs. entertainment, and no matter which side of that you're on, you have to agree that a huge entertainment sacrifice is made in order to show off this amazing discovery. To obsolete the accomplishments of the existing non-glitched runs, which are jawdroppingly entertaining, and leave only this in its place would be a tragedy. This run does something amazing and deserves to be published. But you can't use a 22 minute run (of which at least 5 minutes are glitching or messing around with saved games) to obsolete movies that are nothing but entertainment for 40 minutes. The incredible accomplishments that made those runs possible deserve to be published. This is the opinion of someone who understands very little about how this run was actually accomplished. Rather than invalidate my opinion, I think it represents the opinion of the average viewer. It's obvious to anyone that Samus is screwing with the game map by glitching off the screen on both occasions, because it causes the tiles to be all screwed up. There is a vertical door floating in mid-air! Obviously Samus is out of sync with the game map, and that's all a layman needs to know. This run obviously messes with the game in a way that the other runs don't.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I've never played this game before, but I think I know what's going on. In Level 1, Kyle Reese breaks into the Skynet compound and sends himself back to 1984. In Level 2, Kyle has just stolen some clothes from the department store and is being attacked by a whole bunch of police helicopters. That's not in the movie, but whatever. So, he's killing gang members, who presumably have it coming for living a life of crime, but why is he shooting police officers? Then he rescues Sarah from Arnie in the bar. In Level 3, he's breaking out of a huge police station, which for some reason has a lot of criminals running around in it, and again shooting police officers. I guess it's worth taking a few innocent lives to save the savior of humanity, but couldn't he try to avoid it? How many games are there where you're supposed to kill cops? Okay, so once again he rescues Sarah before Arnie can get to her. Why is this game all about Kyle? Okay, so Level 4 skips the exciting car chase (what?) and goes right to Arnie having his flesh barbecued off and he's chased them into the factory. Surely at this point the game is finally going to be about Sarah Connor after Kyle sacrifices himself. Nope. Kyle runs through the factory, and the limping Terminator keeps popping up everywhere. Why are there so many freaking Terminators? Okay, so there's a big exciting boss fight now, right? Nope, just avoid Arnie until you can find the helpless and cowering Sarah, and the press crushes the Terminator, presumably activated by Buddha. It wasn't Sarah, so some higher power must have decided to save her. So... did I enjoy watching this movie? Well, that's an obvious no. Sarah is an NPC in a game about a movie that is kind of about her. The levels are linear and seem to be very short, and it looks like you're just supposed to follow the path and occasionally stop to shoot bad guys, some of whom are good guys. Kyle never fights the Terminator at any point. I know he mostly fled from the Terminator in the movie -- firing shots to slow him down and doing enough damage to force Arnie to perform some home surgery on his eye -- but have you ever heard of an action video game without a boss fight? Other than the first level, every level ends when Kyle reaches Sarah. No big boss fight with Skynet or a Terminator in 2029, no big gun fight in the bar, no manufactured boss fight in the police station, and no duking it out in the factory. Just get from Point A to Point B and you win. Not even a boss fight in the final level. I'm just amazed by this design decision. This game is sometimes too faithful to the movie (no boss fights), and sometimes not faithful enough (Kyle wiping out the LAPD instead of Arnie). I didn't care for it. I think this game failed as a Terminator game, and the people who designed it should have to sit in a corner and think about what they've done.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I have v2.0, but I stuck with v1.5 for a couple of reasons. First, v1.5 is Win16, so I was able to record the video with DOSBox. Good old Ctrl+Alt+F5. I would have to use something else to record v2.0. I tried Snaggit but I could never get it to record sound for some reason. The other reason is that v2.0 is too freaking easy! I'll probably record a speedrun of v2.0 anyway, but I'm used to the tool-assisted videos around here where all games have to be played on their hardest difficulty level. Version 1.5 makes only one of my jumps easier compared to v1.0, but it can play multiple sound effects simultaneously, it has background music, and I don't have jump constantly from beginning to end! Version 2.0 adds a ladder to the room with the invisible moving platforms, allowing me to skip them completely, and it adds a sold platform to the room before the final boss, eliminating the need to time the jump from the springboard. Plus there's that second Super Jump in the second Boss tank fight, though I wouldn't need or use it in the run. Anyway, if I find a decent (hopefully free) way to record it, I'll do a v2.0 speedrun, but I don't like the fact that it's so much easier than v1.5. It feels like I'm not playing on the highest difficulty level. I couldn't agree with you more about the Save game thing. I just don't get it. It was a bummer getting trapped in v1.0 and having to start over, but v1.5 added Save points. Why make the game easier if you're adding Save points? If anything, he should have made the game harder!
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I did my best to explain what I think is going on in the glitch description. The collision detection in MicroMan is a lot like the collision detection in Super Mario Bros. or any other platform game. MicroMan can sometimes get partially inside a block that disappears and reappears, and the block pushes him out. Normally you could never get deep enough into a wall to skip a room. In this one case, the left side of 3-11 doesn't have a wall on the left side, but it leads into a room that used to have a wall on the right side. Loading a saved game resets all of the enemies, power-ups, and walls, which means that the wall in 3-10 is back, and so is the Boss tank. When I walk left in 3-11, I'm walking into a wall, and the game tries to push me out. For some reason, it can push me into the floor, and the game is presumably still trying to push me out of the floor. When I press right, that seems to help it push me out, and I get back into 3-10. The same glitch can be used in one other place in v2.0 by saving in 8-16 while MicroMan has Super Jump. Loading the saved game and try to get back into 8-15 and you can re-fight the second Boss tank. The only other place where there's a room that doesn't have a wall next to a room that sometimes has a wall is after the final boss fight, but there are no save points after the fight. If there was, you could walk left from 8-19 back into 8-18 and re-fight the last boss.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Sorry, I screwed up the URL. Try again. Also, I discovered a new glitch that lets me walk through a wall and kill a dead enemy!
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Apparently a sub-9 minute run is possible. I've gotten it down to 9:03. It turns out that you can Super Jump off of the springboard and directly onto a ladder in one of the rooms where I used to wait for about half a minute for the moving platform to get into position. Saving that much time, I still have Super Jump when I get to the magnet room, so I can jump directly to the next ladder, saving another half a minute. Having so much extra health from not going through the magnet allows me to walk through one of the enemies instead of stopping to kill it, saving another few seconds. The end result was a 62 second improvement. There's also apparently a weird glitch that causes the final boss to get stuck in the middle of the room and stop shooting. Since he's completely defenseless, you can just move to the uppermost block and shoot directly into him continuously. The strange thing is that the glitch seems to make him stronger! Despite a continuous barrage of fire, it took 10 seconds longer to beat him that way than with the old jump-up-and-down technique. I posted a video of the glitch anyway.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
[Edit] Okay, forget what I said about the 11:06 time. I've gotten it down to 10:05. Both Ins and Z are used for fire in this game, and I figured out that I could shoot twice as fast by mashing them both at the same time. Here's a breakdown of where I gained or lost time compared to the 11:11 run. Entering Boss 1 fight: No difference. I kneel so that all three of my shots hit the Boss tank each time (saved 7 seconds in my 11:06 run), and use Ins+Z to fire twice as fast, which might have saved all or part of a second, which I may have lost to unfavorable randomness when it took the Boss tank longer to come down to the lower level. After the Boss 1 fight: 7 seconds faster Looks like I lost a second from sub-optimal jumping before I got to the room with the two moving platforms and a wall gun. Before jump room: 6 seconds faster I realized that I still have Super Jump when I enter that room, so I skip the platforms and jump up to the ladder, saving 6 seconds. After jump room: 12 seconds faster Double firing saved some time in the room with the RoboDog and two caterpillars. After RoboDog room: 14 seconds faster Less waiting in the room with the Robot and taking less damage, plus faster room re-entry before the second Boss tank fight save 6 seconds. Entering Boss 2 fight: 20 seconds faster Double firing at the second Boss tank saves 10 seconds. After Boss fight 2: 30 seconds faster I skip the teleporter and energy in the room before the big jump, saving 5 seconds. Entering final battle: 35 seconds faster Double firing at the Bubble Ship Boss kills him so quickly that I don't even to go down and get an energy recharge! It used to take me three recharges to beat this guy! This saves 31 seconds. After final battle: 66 seconds faster Final time: 10:05. I think a sub-10 minute run might be possible with flawless play and optimal randomness. If anyone wants to take a shot at it, here are the rooms where you need good luck. Room 1: at least 50% of the time, the platform won't be there when you need it. Room 4: Drop from the ladder too soon and you teleport into the robot. Sometimes, despite flawless ladder climbing, I hit him anyway. Room 6: At least 1 out of 20 times the Flying Disc shoots me. Room 9: It's possible to get through this room without being hit by the Robot, but it partially depends on when it starts shooting and whether the wavy shots are high or low as they pass over you. Room 10: Very random. The block might not be there when you drop from above, and you might get hit in the face or the back by a missile. The direction and timing of the missile are random. Room 12: The Boss tank will always start on top, but he might not start shooting missiles right away. How soon he teleports down, how close he is to you, how many missiles he shoots, and whether or not he stays on the lower platform long enough to kill him in one round are all random. Room 13: The magnets are random. Sometimes they'll pull you a long way, sometimes they won't. Room 14: Sometimes the Wall Gun hits, you sometimes it doesn't. The timing and direction of the shots is random. Room 15: There are actually at least two invisible moving platforms here. You can tell which one you're on based on where MicroMan's feet are. The higher platform doesn't get as close to where you need to jump as the lower platform. If you jump too far right before hitting the invisible springboard, you'll hit the higher moving platform. Aim for the middle of the rectangles on the wall to hit the lower one. You're on the right one if MicroMan's feet are right on top of the rectangles in the middle of the rectangles in the background. Room 19: The timing and direction of the Wall Gun's shots determines how soon you can jump up there. If you jump too late, you'll lose time brushing up against the wall. Ideally, the disappearing platform should disappear when you get there to make the jump to the ladder easier. Room 25: It's possible to avoid being hit by the Robot's shots, but it takes some luck. Room 28: The boss fight is always random. For best odds of beating the Boss tank in only two rounds, you want him to spend as much time on the lower platform as possible. How long he stays there is random. Room 31: I'm not sure how important randomness is here, but a good fight is critical. It's extremely difficult -- but not impossible -- to beat the Bubble Ship Boss with only one energy recharge. Unfavorable randomness in rooms 1, 6, 10, 14, 19, 28 and 31 can ruin a speedrun. I soldiered on despite bad luck in room 10 because I was tired of starting over and I thought I could still beat my old time, which I did. To beat my new time will require me to start over any time anything goes wrong, which could take hundreds of attempts. I'm not sure if I'm still up to it. Feel free to record your own run. If you can beat me, I'll host the video.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
There have been three MicroMan games in total. The first one, called "MicroMan", was a playable demo to show off Brian Goble's Windows Animation Package, which he hoped to license. I made a speedrun of myself beating that game in 2:57 in the other MicroMan shrine. Then there was The Adventures of MicroMan, which had two episodes. Brian Goble no longer sells the registered version, so I can't show off Episode 2: Savage Stones. I'm talking to Brian Goble by email right now and asking him to declare The Adventures of MicroMan freeware. Brian Goble later co-founded a company called HipSoft, which now sells a game called MicroMan's Crazy Computers, which is a totally new game based on MicroMan, which contains many familiar enemies and gameplay elements. In the Bubble Ship Boss fight, if I stand on the two higher platforms (one of which disappears), he'll run into me, which hurts a lot. But yes, I could drop down when he comes over. Here's why I don't think it would be faster. First, I would have to get all the way over there, which would take several seconds. Then I would have stop shooting at him much sooner and wait for him to go away when he got to those platforms. I would get to shoot at him less times per pass, and there would already be the time I lost getting there. More of my shots would hit him from there, but in such a short fight, I doubt I could make up the time. Normally it takes two or three energy recharges to beat him, but somehow I managed to beat him with only one, so this fight is over pretty quickly. I just need to be more careful about not falling down after I get the energy recharge. Anyway, I'm determined to show off the two time savers that I came up with that save 14 seconds. Unfavorable randomness and slower boss fights ruined every attempt yesterday, but I'm not done trying.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I'm glad somebody likes it. :) As far as the speedrun, I've found a way to save 14 seconds but I keep losing all of it the second Boss tank fight or the Bubble Ship Boss fight (came within one volley of beating him without having to get a second energy recharge). Roughly 50 failed attempts later, ranging from 2 to 10 minutes per failure, and I've just about had enough for today. Well, maybe just a few more hours.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Maybe. I figured it was DOSBox related. I've since beaten that time, but it didn't look as good. The luck in the first boss fight was just unbelievably good. Now I know what I'm doing, but I can't reproduce with skill what I achieved in the first attempt with luck!
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Post subject: Adventures of MicroMan speedrun
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Most people think of DOSBox as being a DOS emulator, but it's actually a PC emulator that's geared towards playing DOS games. That means that it can also run a copy of Windows 3.1 so that you can run Win16 games. (I wrote a tutorial about how to run Windows 3.1 in DOSBox here. It also links to all of the drivers you'll need.) It seems that I'm not the only one who has fond memories of playing a game called The Adventures of MicroMan. I decided to do a single segment speedrun in DOSBox, and I recorded the AVI. It's not tool assisted because DOSBox doesn't support save states yet, though Qbix discussed his desire to add them in an interview for the site which you can read here. It's definitely possible to beat this time through better luck manipulation, but the randomness was quite favorable in this particular run. I wasn't able to completely avoid taking damage, but in some cases no time is lost when I do (like if I'm on a moving platform). There is one room where I pause for a second in order to avoid teleporting into an enemy. There are times when you see me enter a room, then immediately leave and re-enter. This is to make an enemy I call "the Leader" leave the room so that he doesn't shoot guided missiles at me. You can't tell from watching it, but this game is actually pretty hard. Most people played v1.0, which was very challenging. MicroMan is constantly running out of energy, and there are rooms that have bottomless pits that have some really difficult jumps. Many rooms require you to find invisible platforms, invisible springboards, and invisible moving platforms. I know where they are, so I making finding them look easy. If you don't know the game inside and out like I do, it's time consuming and really a matter of trial and error to figure out how a room is supposed to be completed. The hardest part was that if you died in v1.0, you had to start over. Version 1.5 added save points and made three of the rooms easier. The big jump off of the springboard onto a disappearing platform in the room before the final boss is frustratingly difficult and ended a lot of games in v1.0 because of the lack of save points. I elected to use v1.5 for this speedrun, even though it made one of my jumps easier, because v1.5 added background music, which hopefully improves the entertainment value. If you didn't play this game back in the day, you might not be very impressed by this video. If you did play the game and remember how freaking frustrating it was, you may find this enjoyable. In fact, no one has told me that they've ever beaten the game, so if you never did, you can finally see the ending. The AVI can be downloaded here. There's also a really bizarre glitch that I wasn't able to use in the speedrun, which you can watch here. I apologize for not linking directly to the AVIs, but these are large files and I'm a bit paranoid about spiders/crawlers/grabbers snatching them from the links in this forum. Edit: Now 27 seconds faster due to better Leader avoision, much better boss fights, and better play.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Running backwards definitely got old after a while, but at least the monotony is broken up by a lot of bad guys getting owned by fireballs. That's one of the things I like about the warpless runs. We all knew that the warpless run had been sadly neglected. It's nice to see someone apply the new tricks, like the flagpole glitch, to improve this important run. Voting Yes.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
It certainly was repetitive in the early going. You used the same route with the glitch through the pink ghost four levels in a row. It got a lot better in the later levels when the ghosts got more aggressive. I own this GBA cartridge because I like those games but, let's face it, Ms. Pac-man doesn't look great on the GBA. Eating ghosts forces a long pause, so ghost avoision needs to be a top priority. Maybe it can't be improved, but I just didn't get much out of it. As for deciding when to end the game, it looked like the level after the cut-scene might have been a new map. Maybe it should have played until the end of that map? Are there still new fruit items at that point, or do they start repeating? Yes, I could look it up, but I just can't find the motivation.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
I still have the same concerns that I had with the last one: you set off a firework in 4-1, which seems like it would have to be slower than waiting for the next clock tick (since no one has told me otherwise), and I'm less impressed by the route you took in 1-2 where Bisqwit uses the platform with the two Goombas to jump on top of the wall before the three pipes, whereas you stay on the ground. Had you gained time in a way that made that jump impossible?
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
Good job, but there are a couple of issues. First, you set off a firework, which has to be slower than waiting for the next clock tick. Also, in Bisqwit's submission, he bounces off a Goomba in 1-2 to get on top of the bricks right before the three consecutive pipes (you can go down the first one), whereas you go under, which is less impressive, and thus boring. Otherwise, great job.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Post subject: Re: #1917: Zowayix's NES Super Mario Bros. 3 "lost levels" i
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
NesVideoAgent wrote:
http://themushroomkingdom.net/smb3_lost.shtml
I find this page fascinating. The discovery of 15 unused levels is kind of neat, but finding two unused enemies is really cool. The two extra mini games is also an interesting find. I love stuff like this. This is right up there with the Arwing in Zelda OoT and the Citadel in Goldeneye. I wonder if any unused stuff made it into any other Mario games. The code for SMB1 has been analyzed to death -- partially thanks to efforts to determine if the level fragment that makes up the Minus World was deliberate or a bug -- so I assume that sprites for any characters that didn't make it into the game would have been discovered by now. There could be cool unused stuff in SMB2 and SMW. Anyway, let's talk about the run. I like it. It has some merit as a demonstration of a hacked ROM, in that it shows off a cool piece of SMB3 history that can't otherwise be seen. I even support your choice in Level 1, which allows the entire level to be seen. The swimming in Level 8 didn't look very optimized, though. I like the stylistic choices you made, as far as playing with Koopa shells and messing around during auto-scrolling levels, and I like how you dealt with the fact that some levels have no exit. If you can submit this run "fully optimized" (i.e. without any known improvements), I'll vote Yes to this.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Post subject: Re: #1899: Bisqwit's NES Super Mario Bros in 06:54.85
Experienced Forum User
Joined: 11/15/2004
Posts: 804
Location: Canada
NesVideoAgent wrote:
This is a pretty decent screenshot, but I think the most appropriate screenshot would be one of Mario walljumping during one of the really huge jumps in 8-1.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
1 2
7 8 9 32 33