Posts for adelikat

adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
Walkerboh: The more I think about it the more your analogy, just doesn't work. I think it more accuately reflects this situation: If person A is actually working on an improvement of a game and is posting his findings and then person B uses those findings to improve the game and submit a run before person A. In the situation with Phil and Sleepz, it is more like the waiter saw the money on the table and decided to leave it there until tomorrow when he comes back for his next shift. If it is not there when he comes back he doesn't have as much right to complain. Yes the money is his and the guy still took it, but the waiter just left it sitting there. The same is true here. Sleepz had no intention on working on this game again so I think it is fair game for anyone else to do so. Perhaps the situation would have been different if it took longer to obsolete sleepz run. If the movie had stood unchallenged for several years, I bet everyone would have been praising Phil. How incredible it was that he found a way to improve a movie that was thought to be perfect.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
DeHackEd wrote:
I can comment on a few of these. Doing side-by-side runs may be nice, but when games start being really really long and/or the time difference between them is much more significant, I don't think it'll be a good way to compare. If someone beat the current Super Mario RPG run by 4 minutes, I'm not going to do a side-by-side of it. Certainly not the whole way through.
Hmm, I don't think I was suggesting such an effort. In a situation like that you would only need to do a comparison on the small section that was improved. Also, I have no concept of how difficult it is to do an AVI like that. BTW, I check the 7-1 boss fight and Phil is actually does it 3 frames FASTER.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
(Voted yes) DeHacked: I love the side by side AVI's. It really helped me notice subtlties in Phil's movie that I wouldn't have noticed otherwise. I looked forward to more of these. I predict we will need this more as time goes on. Phil: I'd like to see some explanation on those small 1-2 frame improvements. In 4-1 you drop the shyguy and pick up the flurry. It seems this is because Toad can run faster with a flurry. Is this true? I was unaware that he ran at different speeds depending upon the item he was carrying In 6-3 on the 1st ladder, Instead of zipping up it with the Up+Down you come off of the ladder for a split second, land on the ground, and then zip back up. Interesting, it seems to save time but how? (and how much time). In 7-1 Boss fight, did you lose time (I trust that you were thorough and did not) but when side by side it seemed Sleepz caught up with you a little there. I could be just an illusion. Is seems to me that when Birdo appears again that is a sign that you went too far to the left which would cost you some frames. I know you have to wait for the mouth to open anway, but I don't it begins to open until it is on the screen. This is a very good quality movie, I hope everyone's reaction to it doesn't discourage you from submitting more runs in the future. Sleepz: When I watch Phil's movie, it seems like a Phil-type movie (even for reasons other than his trademark wobbling). He made several choices that differed from you (ones that saved time, but others out of personal taste). I don't blame you for being upset, but I think you may not be looking at this movie objectively because of it. Bisqwit: Your 1st you comment sounded like you didn't like Phil making this movie (or at least the way he went about it). After watching it you voted yes. Did you opinion then change? I was suprised at your 1st comment ("Ill will take my time publishing this movie" or something like that). All the Guidelines i've read seem to encourage exactly this type of movie and exactly this level of competition. In addition, nothing on your site encourages or suggests any type of edicate in how one must go about osoleting someone else is run. I don't recall reading anything about informing the current published author that you are working on the run nor anything that suggests you should tell them about your time saving ideas 1st and give them time to improve it themselves. It seems Phil's movie and actions are well within the spirit of these guidelines. Everyone: 1) At some point all NES games that are potentially time-attackable will be published here (if not already). Eventually all cool glitches, tricks, routes, etc. will be found. At that point do we just stop making movies? I think eventually 1 or 2 frames will be considered a big improvement for anygame (even long RPG"s). 2) Typically, when someone does post about small optimizations that can be made in a movie they get comments like "well why don't you just make the movie yourself" Most authors here seem to get annoyed by requests to redo their movies for small optimizations. This seems to be in conflict with the reactions to this movie.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
ok, I just tested the shortcut at the end of the silly maze and got it to work. I was unable to before due to a misunderstanding on my part. I forgot that gravity is reversed in this level and that you roll up instead of down. Compensating for that I managed to do it. So I am definately redo-ing this run now. I should be able to knock another second off of the time.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
unfortunately you are right about Name screen. I tried it and that would save an additional 32 frames. Doh! I don't know why I didn't see that. I probably won't redo it just for a faster title screen, at least not unless something else were discovered. As far as the Genesis version goes, none of the cool shortcuts and tricks can be done on the NES version. For one, you can't get as much distance when going off of a ramp or cliff. In the Nes version you immediately begin falling straight down. Also, you can't fall very far with out breaking the marble. In addition, the scrolling on the NES is very slow and if you get too far ahead you die. On several levels I had to slow down and let the screen catch up with me. On the Genesis they fixed the scrolling so that it is much faster.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
apparently I don't have the correct ROM, oh well Your last run worked fine though on it though
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
for me it desynched at level 80, but the part I saw was more than enough for a yes vote. excellent job the luck manipulation for the treasure room exits was done much better this time
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
OmnipotentEntity wrote:
I don't think the NES version is a hack, it's just the Japanese Version.
The hack that allows acces to worlds A-D without completing the game 8 times
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
7/15 no SNES unless it can be done just as fast (or faster) than the nes hack
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
added PRG1 to the submission text
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
as far as I know, your speed has nothing to do with the targets. Also, by missing the 1st one it makes it more obvious that I'm doing it on purpose and not just missing due to sloppiness Also with the saving the children levels, when I did the calculations, it takes a full 2 seconds more to save all 8 (due to the score racking up at the end). Since there are two of these levels, that 4 seconds. I felt that was a bit much to sacrifice for a marginal entertainment value. Also the second save the children level breaks up and glitches badly (even on the actual console). By driving around with one of them it fixed the graphics problems and made the level a lot cleaner looking. [edit] I checked, the frame count is identical when I shot all the target and when I shot none of them. So the targets definately do not affect speed.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
good job
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
I totally agree with Phil, I would vote yes on even a 1 frame improvement. I agree that it would suck to be obsoleted by only a few frames, but that should inspire you to try harder. quote from the guidelines to this page: "Competition breeds perfection"
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
good job, look foward to seeing the rest
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
okay here are som frame counts: 1st frame of boss exploding until no pieces are remaining killing boss at near extreme edge of screen = 380 frames killing boss at center: = 276 104 frame difference megaman at extreme edge after boss explosion = 532 megaman at center after boss explosion = 388 potential difference (sum of the two extremes) = 248 frames In this megaman you can move around after making the boss explode, so is would be best to destroy the boss closest to the center as possible regardles of megaman's position, then have him quickly run to the center. If you notice in kopernical's run he always has megaman dead center and the bosses usually explode relatively close to it (dustman is a notable exception to this) Of course this makes it more complicated because it may be faster to destroy the boss off center as long as you saved enough frames to compensate for the added explosion animation
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
still, it is an improvement. I would go ahead and finish it and submit it. Faster is Faster even if only 1 second
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
starting with less energy against don 2 causes him to come back with less energy after the 2nd knockdown
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
Truncated wrote I watched both now. The non-cs run is definately a lot more entertaining Thanks, I sped up those fights considerably (250 frames each on average) But I remember the enthusiasm for the previous CS glitch video was very low Doesnt seem that either are going to stir up much attention. Phil and Genisto's didn't either. Phil Wrote: We have tested a lot of things but not enough apparently Well you certainly were very thorough. Just to improve Sandman by one notch reflects at least 5 hrs of work including spreadsheets of frame counts and damage charts and many many calculations. I spent at least that much time on Bull 2 only to come up empty handed.
It's hard to look this good. My TAS projects
Post subject: fmv to fcm conversion
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
how would i go about converting a .fmv to .fcm file to be played in FCEU?
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
I'm in the last stages of two new runs for this game The WIP's can be found here: http://www.adelikat.com/Punchout.html Two runs because one exploits the clock stopping glitch and the other does not. The main reason I am posting here is to get feedback on whether I should submit one or both (i'll finish both either way). The cs run reflects the fastest possible times according to the game clock but makes the run about a minute and a half longer in real time. Here are the improved times for each run compared to my previous attempt and to the current published movie: Ver 2 Cs Ver 1 (cs) Phil & Genisto's Published Movie Piston 1 :40.25 "" :40.82 :42.82 Don 1 :14.97 :11.97cs "" :14.97 Soda :47.00 :31.82cs :32.48cs :47.00 Don 2 :58.99 :44.48cs :44.48cs 1:00.48 Sandman 2:19.00 "" 2:19.25 2:19.25 Macho :46.25 :35.82cs :35.97cs :46.48 (note: macho no cs has not yet been recorded, i've got this time during the testing). In addition I have improved ver 2 from my previous run by manipulating the opponents counts so they get up as soon as possible, making both runs as short as possible in real time.
It's hard to look this good. My TAS projects
Post subject: How do a calculate frames per second?
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
I'm using FCEU if that makes a difference
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
The only thing that affects new subscribers is delivering papers to existing subscribers. You definately smash some non-subsriber windows. Unfortunately, even in a perfect run you can't have everyone be a subscriber (since you only get one per day). As far as completing the bonus levels, It would look better except for the fact that you would have to watch the same one 7 times which might get pretty redundant. I might lean more towards an absolute best time since there more that enough opportunity to rack up points breaking windows and knocking down things. Also, if I remember correctly, you can deliver several papers to the same house and get the bonus points everytime.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
unfortunately it desynced for me. Even though it is slower don't scrap it yet! Press backspace to show the frame count, then compare the two movies to find where you lost time. Go through the original movie and find several land marks and jot down the frame count. Then run your movie and see how those points compare with the original.
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
excellent job! In your submission you mention that you manipulate luck to get the exits in the treasure rooms to be the closest possible. In level 1 you went to two treasure room and went into a different exit both times. Shouldn't you have found the fastest exit and made sure you got it both times? either way, definate yes vote
It's hard to look this good. My TAS projects
adelikat
He/Him
Emulator Coder, Published Author, Site Developer, Site Owner, Expert player (3602)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
The castle is really insane, you have to fight 4 bosses at once! Also, you are submerged in water. In addition if you defeat one of them they drop the question mark thing, but if you get it you stop but the others still move around and kill you. You would have to avoid each question mark until they are all killed!
It's hard to look this good. My TAS projects