Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Another good level. ^_^
diggidoyo wrote:
I don't think there's a way to know for sure until we can watch the RAM address.
It should be possible to attach L Spiro's Memory Hacking Software ( http://memoryhacking.com/download.php ) to Dolphin to find any position/speed addresses.
Here's something I typed up on how to do it a while ago, I'm too lazy to type it up again specifically for a different scenario: http://tasvideos.org/forum/viewtopic.php?t=5444&start=80
The data type may be "byte" or "short" and you would want to use "greater than" and "less than" to tell it to search for bytes that have increased and decreased (speed/position.) It takes some trial and error.
Step 6 would be, move forward, look for values that have increased, etc.
If you're interested try it out. If you're not, it probably isn't hindering you too much at this point.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Bisqwit, adelikat, and Nach: - For creating and maintaining a great site.
Coders who maintain emulators: - These guys work so far behind the scenes that I probably don't even know of half of them.
Players, Supporters/Fans, Encoders, etc.
Specifically,
Fractalfusion: - For the immense amount of work he has done editing pages on the site.
Nitrodon: - For (probably unknowingly) teaching me a lot about game code and lua scripting while we worked on the 7th Saga run.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
The emulators on the Emulator Resources page are the ones with rerecording implemented. They tend to contain the most functionality.
http://tasvideos.org/EmulatorResources.html
The currently fastest run of Super Mario Bros would be listed on the Movies page.
http://tasvideos.org/Movies.html
NES Super Mario Bros (JPN/USA PRG0) in 04:57.31 by Lee (HappyLee).
Please do not submit anything slower than this in the any% category. Feel free to post any progress here for feedback. Please look around the site a bit before asking such obvious questions next time.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
This submission does not use the song Superman by Goldfinger in all levels.
No vote.
I already watched this on youtube. It was really good. I suppose I can give it a yes vote despite the previous horrendous oversight.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Awesome run yet again. The very first thing you did in the level was clever, and the health drops at the end were too good.
Just a random thought, consider changing weapons to the beat of the music at some point, perhaps?
I would speculate the RNG is either on a 256 or 65536 rotation. Maybe 256 if they programmed the game true to its roots. Commando Man's optimal pattern is probably (X/256), where X could possibly be around 4, but it would appear at seemingly random intervals over the course of the 256 rotation. If you want to check this, just check if 0 frames and 256 frames of waiting yield the same combination of events for some random actions.
I also wonder if subpixels have anything to do with the randomness of this game. By not moving at full speed, you can possibly get on a different subpixel, which may effect enemy behaviors or possibly the randomness. (I think you would've noticed any effects on random behavior by now, so I doubt it matters there.)
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Good job on getting 3 of the games to end at about the same time. It would be worth trying to find a longer route through SMB1 if possible in a future run of this, but this is definitely worth publishing as is.
For anyone who didn't notice, the published SMB3 time is 10:25.6.
This run expends less than 15 seconds to beat all the games.
Definitely a yes vote.
(Does anyone have a comment on whether the official encode should go black or play the demo after SMB1 completes? I'm thinking the demo may be the better choice.)
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Brushy, how do you feel about the current quality of this year's April Fools submissions as of the time of your reply to this post.
Also, which is more legendary? -- Color a Dinosaur, or Desert Bus.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
I saw the ending coming after I read that there was an entertainment tradeoff. It is marginally novel to someone who hasn't played the game.
Wrong holiday.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
This following document suggests the following statistics:
http://www.nintendo.co.jp/ir/library/historical_data/pdf/consolidated_sales_e1012.pdf
54.9% of all NES systems were sold in the Americas
31.3% of all NES systems were sold in Japan
13.8% of all NES systems were sold outside of the Americas and Japan (not necessarily even PAL to my understanding.)
46.2% of all NES games sold were sold in the Americas
45.2% of all NES games sold were sold in Japan
8.6% of all NES games were sold outside of the Americas and Japan (not necessarily even PAL to my understanding.)
Japan has sold many more units of SNES software, but the Americas are winning in quantity in essentially every other Nintendo system category.
I don't think it is U.S.-centric to say that many fewer people played the PAL versions of NES games. Japanese and NTSC versions tend to play the same for a large portion games. English is a more widely spoken language. It makes sense to me to run the (U) version more often.
I don't see any reason for the site to bother supporting minute differences in PAL categories for games without significant differences. The final run time will almost always be slightly different due to the way the game is modified for 50 fps vs 60 fps. This assumes the game was modified to play at roughly 6/5 speed per frame.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
DarkKobold wrote:
Touchme's Secret of Mana (U)
Mega Man 1, the new version that starts with Bomb Man
sgrunt wrote:
The finished Mega Man 10 run.
Hopefully, a Mega Man 9 run to go with it.
GC Sonic Adventure 2 Battle any% hopefully.
TheAxeMan's NES Crystalis
Kirkq's N64 Bomberman Hero. Yeah I'm looking forward to finishing my own run :P
I don't think Wind Waker any% will get finished any time soon. You can't heal out of a zombie hover by practical means to my knowledge unless the game boy advance tingle tuner is emulated.
EDIT:
Warepire wrote:
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
According to kuja killer, shouldn't you be shooting small objects as well? It seems to me you missed at least 2 at around 0:30 and then many more at 0:35 and 1:05.
Love the run, it just doesn't seem logically complete to me unless I am misunderstanding the RNG incrementing on small objects.
If my speculation is correct that this isn't optimal, now might be a good time to learn hex editing, as it should be a relatively quick fix.
Redo screen 1, hex screen 1 in.
Redo screen 5, hex screen 5 in. (don't kill the red tsuare stomp any more?)
Waste ~10 frames (if that) before the boss and hex then onward.
If it is an improvement, and it is possible to hex edit it I would suggest doing it. If it is not hexable, then it is probably okay to move on. This game sounds hexable based on your descriptions though. (Be sure to backup your progress, but I'm sure you know this.)
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
See also: 100x100 limit on avatar size. :P It actually took me a while to find that site page for reference, so it seems to be something one wouldn't come across right away.
EDIT: lol you fixed it before I even finished posting.
DJSecret wrote:
Really unoptimized, but I liked it regardless. Yes for me.
DJSecret wrote:
One thing that I noticed is that he broke consistent hopping speed in a level or two, decreasing his speed from 51 to something lower.
Also, *points at "lost frames" counts*
Er, does "really unoptimized" mean that like 20 frames were lost over the course of the run? It seems to me that you might be embellishing the fact that you can see extremely minor technical errors. I can find a few frames lost in many runs on the site, but I don't consider them "really unoptimized."
How many frames of improvement would you estimate are "known possible" over the course of the run?
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Pressing turbo start at the beginning is a bad habit, as you probably lose 1 frame half the time. I also think that maybe some shot opportunities are missed on bosses, but I do not know the game well enough to say for sure.
Just over 5k rerecords were used for 107k frames of gameplay. The average rerecord count of movies published on this site is around 20 per 1 second of gameplay. You rerecorded 3 times per one second of gameplay. This tells you that most TASers are recording 7 times more often than you. The rerecord count itself is not important, but it suggests that you may not be recording enough to produce a high quality, precise run.
I think pacifist is a bad goal because some normal enemies are required to be killed fairly often. I think it also takes away from the entertainment potential.
Alternating directions all the time is entertaining for a bit, but it gets old after a while. You have to be creative in a run like this. The run also doesn't do a great job with near death encounters. You need to be a pixel away from death instead of just arbitrarily weaving between bullets.
I think it is possible for this game to be published, but I think the entertainment value has to be better. Moving with the music, moving as close as possible to enemies, and killing a lot of enemies could all make this movie more entertaining. I think many expert TASers would have a hard time keeping this movie entertaining for 30 minutes, but I do think it is possible. Watch the NES Gradius run for some suggestions of what this movie should try to be like.
Like MESHUGGAH said, it should also be played on hardest difficulty. The rules say that runs should most often be played at the hardest difficulty. This is a case where a password is allowed to make the game harder and more entertaining.
MESHUGGAH wrote:
press Up on both controller at title screen and enter the S4FHFB password.
You guys had the right idea, but I don't think it is within the capabilities of most new TASers to present a run so highly dependent on entertainment value. I'd encourage you to work on a different game and in the future post your progress in the game's existing topic for feedback. I'm sorry, but I have to vote no.
(If someone could translate this it may be helpful to the authors.)
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
I'm curious.
How many distinct patterns does commando man have? If his optimal pattern doesn't appear for a 44 frame window, does this mean that the patterns stay constant for around a second or so?
If you were speedrunning the game on console, and you had a consistently near perfect run, could you feasibly successfully gauge the wait in order to produce the right pattern?
Good work so far.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
MUGG wrote:
what about Mario Land 2
I think your comparison is flawed.
The any% 21 minute SML TAS uses the same trick as the two minute SML TAS except that it only uses it to clear levels. It probably isn't possible for a run to exist that "uses a game-breaking glitch to clear certain levels, but does not use the same glitch to clear the game." There is most likely room on the site for an any% run not using the glitch. If you made an optimized any% run that didn't use the pause glitch, I would vote yes. (I'm not sure if the halfway point glitch should be legal or not.) I think the judge made the correct decision with the movie based on the nature of the glitch.
Chrono Trigger on the other hand was somewhat wrongfully obsoleted. The previous run was extremely outdated by inichi's no save-glitch test run. If inichi had submitted his test run, I would've voted for it to take over the any% category and have the glitched category separate. I think if someone submitted an optimized no save-glitch run today it would be published separately though. I don't see any urgent need to repeal this decision though.
I don't think Super Mario World 100% will suffer the same fate. I think that the new run will either be published in its own category or not published at all.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
A run consisting of 20 minutes of powerup incrementation and 30 minutes of clearing levels in the same manner may not receive good feedback. In this case the proposed run sounds to me to be too repetitive to be entertaining. It is a novel concept for sure, but I probably don't really care to see every exit in the game glitched through by the same 2-3 methods.
The approximate purpose of the 96-exits run is to show off the whole game as fast as possible for entertainment. The purpose of your run seems to be "to be faster than the 96 exits run by not showing off the whole game."
If there really is some variance in the execution, maybe I would find it worth watching. Though if you just do the same thing ~106 times, and then similar things ~96 times, it won't be particularly entertaining.
I really have no basis on how varied the run will be. It could be worthwhile, it just depends on what the run requires.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Kitsune wrote:
You could call this game the Pokémon prototype, since I believe it came out before the Pokémon games.
According to Wikipedia:
Jade Cocoon: - JP 1998-12-03
Jade Cocoon: - NA 1999-07-29
Pokémon Red and Green: - JP February 27, 1996
Pokémon Red and Blue: - NA September 28, 1998
So yes it did come out in Japan around the same time Pokemon hit the U.S., but it did not come out first in either country.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
diggidoyo wrote:
such as the red Tsurare Stamp before Commando Man. I can easily pass under him without stopping but I chose to halt for 5 frames mid-air to kill him as fast as possible on my way through.
The choice you made in the posted run was not good in my opinion. It is a visible stop to kill an enemy for minimal entertainment gain. I had previously presumed it was necessary.
I also think grabbing the early extra life is unnecessary, because many can be manipulated from drops.
As for taking damage from Blade Man, I would try to think about how the whole run will look with or without Mr. Perfect. It really depends if this creates any visually suboptimal spots, like waiting for enemies to move. Make sure you can get it after beat call and spike protect though.
Also, I think if you had started with Pump Man you might've had to waste a bit of time taking damage to get by enemies. I'm not sure really.
I'm leaning towards agreeing that you should go for Mr. Perfect.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
So is the in-game time route actually different from the real time route or not?
Does the Rush Coil jump at 1:25 (the last one) actually get you high enough to benefit from going above the sandstorm? If so, is it possible to land and shoot the mettaur, and then rush coil a bit earlier? The rush coil right before the last one doesn't seem to get the boost, and it looks as though the last one would if it occurred before the sandstorm ended. If possible it looks to me like it would save a few tenths.
Also, I assume pausing can't be used to effect the randomness at all for the boss room?
Overall good run though. Style is important; shooting things with various weapons can add a lot of entertainment.
Experienced Forum User, Published Author, Player
(208)
Joined: 7/7/2006
Posts: 798
Location: US
Tangentially related point:
If a game (maybe NES) has default 3 lives, but you can turn it down to 1 life in an options menu, do the rules request that you play at one life? Is that "hardest difficulty"? Alternatively, I'm sure we would frown if it was turned up to 5 lives.
If we relate this to an arcade game, the player starts with one coin's worth of lives, should that be able to be increased?
I still do feel that both sides of the argument in this thread are quite valid.
