nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
nitsuja wrote:
You mean while inside or just barely in a safe zone, or do you mean you found a bug that lets you bounce off walls without getting hit in other situations?
On the safe zones, both starting safe zone and middle of level ones.
nitsuja wrote:
And if you aren't taking any hits, then isn't aiming for in-game time the same thing as aiming for real-time, with the 100% completion restriction?
Well, i wait at the start of the level untill i have the best angle to start moving. I'm not sure right now, but it indeed seems to me that waiting at the start or waiting inside the level would lead to the same final real time, but there may be exceptions. Also, there are some places that its possible to get hit and finish the level with faster in-game time, but that breaks the star rule, and thats the only reason i chose no hits (because of the stars). So, it won't be fastest possible level times because i chose to get all stars.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
I've finished the 5 training levels, 3 grassland, and 2 ocean levels. Times were: Training: 1"56 2"10 3"16 2"71 1"88 Grassland: 5"91 11"06 13"21 Ocean: 5"55 10"31
Emulator Coder, Skilled player (1311)
Joined: 12/21/2004
Posts: 2687
Those are very good times, and you said you're not waiting at all (or at least not more than a second or so) at the start of the level to get them?
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
I wait at the start of the level, because the timer only starts after leaving the blue square. Sometimes i bump on the wall to help on the starting angle. And it takes more than 1 second for the helicopter to turn 180° (my guess is it takes 6 seconds for a full turn), so i may wait as long as 3 seconds at some places. But what i understand is that it doesn't affect total real time much because if i have to wait at start that's because i'd have to wait during the level later if i didnt wait at start.
Emulator Coder, Skilled player (1311)
Joined: 12/21/2004
Posts: 2687
Well, if you wait 3 seconds at the start and are only 1 second faster than if you had waited 0 seconds, then that means you got faster in-game time and slower real-time, I think. edit: My level times before without any waiting at the starts of levels or hitting walls were: Grassland: 6"88 12"13 13"81 Which are clearly worse than you did, but none of them by more than 3 seconds.
Former player
Joined: 7/29/2005
Posts: 459
Location: Brazil
is it the first non-NES TAS u are doing, foda?
<small>My big signature was cleared by admin; i should read <a href="http://tasvideos.org/ForumRules.html">forum rules</a>. But... who does?</small>
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
Hmmm.... nitsuja but i didn't have to wait too long yet, maybe once at one of the ocean levels. No, daniayaw, i've done TAS for mega drive (alex kidd, pit fighter and ecco the dolphin) and for SNES, with help of vatchern (battlemaniacs). Just finished ocean level 3, it was a hard one. The time was 22"08
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
finished 3 more levels: Training: 1"56 2"10 3"16 2"71 1"88 Grassland: 5"91 11"06 13"21 Ocean: 5"55 10"31 22"08 Jungle: 13"45 12"90 23"25 CakeLand 18"68 32"91 27"00 This is taking longer than i thought it'd take.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
hummmmmmmmm maybe i should do the chalenge puzzles first? lol :S only now i thought about it... when i finish normal game there will be an ending.. when chalenge puzzles are finished there's no ending.. EDIT: ok so i asked nitsuja about it and he suggest i try to hex-edit it later. So i started over, making the chalenge puzzles and after i finish it if i can't hex edit i'll just redo what i had done already... There are so many levels in this game O_O here are the times: 1-1: 1"86 || 2-1: 5"41 || 3-1: 4"93 1-2: 2"21 || 2-2: 3"65 || 3-2: 7"86 1-3: 2"61 || 2-3: 3"28 || 3-3: 6"63 1-4: 2"53 || 2-4: 3"78 || 3-4: 7"51 1-5: 3"10 || 2-5: 3"10 || 3-5: 5"31
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
01-1: 01"86 || 02-1: 05"41 || 03-1: 04"93 || 04-1: 04"30 || 05-1: 03"46 01-2: 02"21 || 02-2: 03"65 || 03-2: 07"86 || 04-2: 06"05 || 05-2: 01"56 01-3: 02"61 || 02-3: 03"28 || 03-3: 06"63 || 04-3: 04"05 || 05-3: 04"50 01-4: 02"53 || 02-4: 03"78 || 03-4: 07"51 || 04-4: 04"61 || 05-4: 03"65 01-5: 03"10 || 02-5: 03"10 || 03-5: 05"31 || 04-5: 02"91 || 05-5: 03"48 06-1: 03"45 || 07-1: 02"33 || 08-1: 03"08 || 09-1: 02"53 || 10-1: 04"73 || 11-1: 10"53 06-2: 02"58 || 07-2: 03"28 || 08-2: 04"15 || 09-2: 03"28 || 10-2: 03"10 || 11-2: 03"93 06-3: 02"50 || 07-3: 02"35 || 08-3: 02"70 || 09-3: 03"73 || 10-3: 02"38 || 11-3: 04"48 06-4: 01"73 || 07-4: 02"11 || 08-4: 09"50 || 09-4: 03"06 || 10-4: 05"86 || 11-4: 03"15 06-5: 02"41 || 07-5: 01"83 || 08-5: 08"20 || 09-5: 07"56 || 10-5: 03"91 || 11-5: 01"25 The reason 11-5 is 1"25 is one of the reasons that made me try to timeattack this game in the first place ;) I must say i'm impressed at this game. Its like that on every corner the programmers are talking to me "you can go there, just do this and that so you can barely pass" and most of the time it is possible to pass. It's like the real chalenge in this game is to make a TAS, not play in real time. The levels seem that they were made for a tool assisted time attack, but how if the game is old? I think it's related to how the levels might have been designed: Here is what i think, some levels were designed using a spinning helicopter to draw the walls. The person in charge of designing the level would just start the spinning and then start moving and drawing the walls. Then after it had drawn the basic tunnel, they would change it, add things, reshape the walls, etc. then position the starting and ending areas, and give a different starting angle. That's the only explanation i can think of, definitely what i saw happen here is no coincidence... Too many bare "can pass through" after carefull optimization.. So far the movie file has 9:10 minutes (at start of training level 1 now). There are 5 training levels then 33 (i think its 33) normal levels. That should make something around a 20 minutes movie tops... i hope.
Joined: 8/9/2004
Posts: 5
I'm looking forward to the finished movie. Keep up the good work FODA :)
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
i'm obviously still working on this, but had a few other stuff to do lately so i didn't work much on it. I improved the last 6 on the list. The other ones are the same. Chalenge mode levels: 01-1: 01"86 || 02-1: 05"41 || 03-1: 04"93 || 04-1: 04"30 || 05-1: 03"46 01-2: 02"21 || 02-2: 03"65 || 03-2: 07"86 || 04-2: 06"05 || 05-2: 01"56 01-3: 02"61 || 02-3: 03"28 || 03-3: 06"63 || 04-3: 04"05 || 05-3: 04"50 01-4: 02"53 || 02-4: 03"78 || 03-4: 07"51 || 04-4: 04"61 || 05-4: 03"65 01-5: 03"10 || 02-5: 03"10 || 03-5: 05"31 || 04-5: 02"91 || 05-5: 03"48 06-1: 03"45 || 07-1: 02"33 || 08-1: 03"08 || 09-1: 02"53 || 10-1: 04"73 || 11-1: 10"53 06-2: 02"58 || 07-2: 03"28 || 08-2: 04"15 || 09-2: 03"28 || 10-2: 03"10 || 11-2: 03"93 06-3: 02"50 || 07-3: 02"35 || 08-3: 02"70 || 09-3: 03"73 || 10-3: 02"38 || 11-3: 04"48 06-4: 01"73 || 07-4: 02"11 || 08-4: 09"50 || 09-4: 03"06 || 10-4: 05"86 || 11-4: 03"15 06-5: 02"41 || 07-5: 01"83 || 08-5: 08"20 || 09-5: 07"56 || 10-5: 03"91 || 11-5: 01"25 Training levels: TR-1: 01"56 TR-2: 02"06 TR-3: 02"97 TR-4: 02"68 TR-5: 01"83 Normal mode levels: 01-1: 05"90 01-2: 10"90
Former player
Joined: 7/29/2005
Posts: 459
Location: Brazil
FODA wrote:
No, daniayaw, i've done TAS for mega drive (alex kidd, pit fighter and ecco the dolphin) and for SNES, with help of vatchern (battlemaniacs).
Aé, i forgot them.. i've watched them all XD arent u going to release WIPs for this TAS u r doing, foda?!
<small>My big signature was cleared by admin; i should read <a href="http://tasvideos.org/ForumRules.html">forum rules</a>. But... who does?</small>
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
I'm not gonna remake anything, so i don't see a reason to... i mean, i don't have much free time, so what i have is what you'll get : / i think it's good anyway.
Player (138)
Joined: 8/27/2004
Posts: 165
I just made a test WIP for Kururin Paradise (does anyone care?). Anyone want to see it? I'm about 1/3 of the way through the Adventure Mode. I'm sure none of you have played it, though, since it's only in Japanese, unlike Kuru Kuru Kururin. This is a total rough draft because some of the times in the small levels are actually *worse* than some of my record times in 100% speed... Ah well. I'm just setting a bar, really; if anyone wants, they can optimize more, but I'm fine with what I've got. When I finish it (should be Friday or Saturday, it's pretty quick) you can check it out, I guess. This is pretty easy for me because of my experience with this game. Maybe I'll make a finished game run in the future? By the way, FODA, this isn't the finished version, right? You're slower than the known records (compiled by Zanapher, at kurukuru.free.fr) in Challenge 1-1, 1-2, 4-3, 5-2, 5-4, 5-5, 6-2, 6-4, 6-5, 7-3, 7-4, 8-1, 8-2, 8-3, 9-3, 10-2, 10-4, 11-1; Training 1, 4, 5; and Adventure 1-1 and 1-2. At kurukuru.free.fr/videos, there are actually faster videos up for Challenge 1-2, 4-3, 5-2, 6-4, 7-3, 7-4, 8-2, 10-2, 10-4, 11-1 and Adventure 1-1. I don't see why these strategies shouldn't be replicable with slowdown and savestates... but whatever. I know from recent experience that, in small levels, it's often just as hard to get a near-perfect score with 25% speed and with 100% speed.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
there must be something i'm missing then. they are optimized to death using the knowledge i have, so i don't see how those times are possible. yes, this is my final version and those records highly discourage me to finish the video. I tried to dissect the elements that were key to getting a fast time on each stage (like starting the level on the last possible frame so that the rotation would be perfect etc) and i've defeinitely made turns on the last possible pixels, and still they got a better time. I guess i missed to realize, for example, that on the first chalenge level one should make an average of going up/down instead of going down the least at the start but then having to go up a little more at the end. it's hard to get a perfect score in any level length, the difference is that there are too many good players trying hard on the small levels, so its hard to match those.
Player (138)
Joined: 8/27/2004
Posts: 165
Aw, don't just give up. If you do, I'll probably end up doing this (after I do Kururin Paradise, of course), but that's still not a good attitude to take. I think you should finish your video with non-perfect optimization (run at 25%, savestate every 3 seconds or so) and then either start version 2 yourself or let someone else do it. Have some random tips: - Hitting the side of a level while in a heart zone can push you forward and will turn back your rotation a bit. - After the last turn, go all-out to the exit. - Try to navigate corners and wide turns in the wrong way, it might save time. - Hitting a spring either (a) turns you around or (b) speeds up your rotation, depending on how you hit it. Also, remember that if there's an item in a level, your time is definately going to be slower than the record, and you shouldn't worry about that. Also, take a look at these videos. They're the ones I have that beat or equal your TAS times: http://www.mzrg.com/filepost/kurukuru.zip (Please, try not to download this more than once, and only download it if you really want it, because I don't have ALL that much bandwidth and this is a 5MB file) EDIT: Oh yeah, forgot to tell you. If you want to optimize a level, you MUST play through it at least two or three times. Even if you start perfectly, and optimize perfectly, it may still be possible to get another frame or two out of it. Then there are little tricks that you barely see. On Challenge 1-2, for example, going through the top part level with the line is the safest (as your rotating stick is centered in the passageway), but you can only get 2"20 minimum with that. If you keep your stick one pixel lower, you have much less room for error, but you can get a 2"18.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
Like i said, optimizing is not a problem. I know how to optimize. The problem was that i couldn't 100% analyze in every level what would make it faster, i lacked the brain power to do it, and i certainly don't have time to try to do it ramdomly. Those ramdom tips are obvious and you can see them on my video.
Player (138)
Joined: 8/27/2004
Posts: 165
I can? Maybe when you release it. It doesn't look like you have yet. Anyway, I'm trying to make my own TAS thing (at least for the Challenge levels, maybe more if I actually finish that - you don't seem to be working on them, anyway, and I can improve most of them), and I am having a ton of trouble equaling your 2-1 record. I keep getting times in the 5"43-5"48 range. Is there something I'm missing? Can anyone see an obvious place where I can save a frame? My WIP is at http://www.mzrg.com/filepost/KK_v1_1.vbm.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
Player (138)
Joined: 8/27/2004
Posts: 165
Meh. Can't replicate it, can't equal it, can't hex it in. I give up. Go ahead with the final version, I'm not stopping you.
Player (138)
Joined: 8/27/2004
Posts: 165
If you ever have time to do a few levels, could we maybe do some kind of collaborative effort? You could do the levels I can't equal or beat your time on, and we could discuss strategies (or something). For example, if you did 2-1 on top of my WIP progress, I could continue with the levels that I can equal or beat the record in, and in the ones I can't you could step in. Who knows. Please respond if you think you will be interested now or in the future.
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
why can't it be hex-edited? the bad part is that i don't know much about hex-editing, let alone with vba files...
Player (206)
Joined: 5/29/2004
Posts: 5712
How about you both play each level at the same time and keep whichever is the faster play?
put yourself in my rocketpack if that poochie is one outrageous dude
Player (138)
Joined: 8/27/2004
Posts: 165
That's sort of what I was trying. On the levels where I can beat FODA, I plan to do so, but I suppose there must be more levels like this where I can't equal his time. I assume that, since FODA didn't want to restart the TAS, he probably won't want to redo every level, but maybe doing only a few would be more agreeable. I tried to hexedit a while ago and it didn't work, so I assumed that it wouldn't. But I'm going to try again, and if I can get it to work that would be great.