Posts for Kirkq


1 2
13 14 15
31 32
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I finished Cool Cave and beat Snowland. Cool Cave turned out really well. http://dehacked.2y.net/microstorage.php/info/408877005/Start%20Storm%20Valley.m64 neskamikaze, do you know the fastest way to beat the boss in Heaven Sky? I may have to dig around in MHS to find its hit points or something, because I can never really tell when I'm hitting it. I should be able to finish most of World 4 within a week or two. Hopefully this hits the workbench relatively soon (for real this time).
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Has anyone tried doing the LLL 8-coin puzzle with a BLJ off the right side of the bridge commonly used to BLJ? (Spur of the moment thought.)
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Could someone explain what seemingly random pauses do? (outside of P-BLJing obviously) I've seen these in a few freeruns now, and it doesn't make any sense to me.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
A huge congratulations on getting a stable rerecording DOS emulator. 1560 rerecords for ~16 minutes of gameplay is normally really low, but this is a limitation of the emulator right? This may be a problem in judging DOS runs at the moment. I played this game way back in the day . I'm actually surprised other people have played it. A fair job on entertainment. The routes chosen were generally good, but I feel overall that execution could have been a bit more precise and variable to increase the entertainment value. Good run. A bit slow paced unfortunately. I'm curious if people would find this more favorable if the execution were more precise, or if they would entirely owe it up to mediocre game choice When objectively compared to other situations of poor game choice, this is pretty borderline. I would like to see it published though, and possibly improved in entertainment in the future. Weak yes vote.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Mitjitsu wrote:
In order to justify accepting a new emu on the site there should at least be several games for the system that get people excited.
I don't see why that is necessary. I think it's actually a good thing to represent less common systems if the game(s) are good. The emulator has already been modified for TASing, so it would just be a waste to throw out that progress at this point. Wearing 3-D glasses to view a TAS is pretty novel.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
any% run is still in Cool Cave. I made that one jump and stopped working on it. Progress will resume in about 2 weeks after my finals, and hopefully it will hit the workbench towards the end of the summer.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
For your purposes no. It is easily the most random thing in the game, even simulating enemy movement would be painful on our part, so we brute force it. You may be able to strategically pause to get an enemy to divert its path, but it would only marginally improve the difficulty in avoiding enemies.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Progress will resume in about two weeks after I finish up finals. We will presumably be going with the Foma fight I charted. I started working on walking through Ligena sort of. A lot of people get in the way, so it is a bit of a hassle. Classes have definitely been keeping me busy though. This run will probably be finished about two months from now if all goes well.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
This is fairly old news. One point worth making: Quote metroidmaster88 (on my youtube video: http://www.youtube.com/watch?v=1RySTbjhJQY)
"Well i test out tryi skip last boss with every versions. Its look it working JPN version only. i did it very easily on JPN version, but i cannot make it in NTSC. Look out my vid where i show why its doesnt, work on NTSC. Even boss sometimes disappear you can,t pass it without taking death. Also this not work on PAL because flying enemys never appeared in last level on PAL even i haved some golds."
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Kumquat wrote:
For comparison, the currently known upper limit for quarter turns is 29.
Ah, I had an afterthought that something like this might be the case. Thanks for pointing that out. I still emphasize my opinion regarding automated solvers. This movie is clearly borderline. I don't think that there would be any inconsistencies with accepting it; I'm just not sure how much novelty and entertainment it actually contributes.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Thoughts: The fact that the 2x2x2 is one move away from solved is mildly humorous. The solution in the movie for a 3x3x3 is 20 moves. This site claims that any 3x3x3 cube can be solved in 22 moves or less. http://cubezzz.homelinux.org/drupal/?q=node/view/121 While watching the 4x4x4, I began to feel like the movie is just a demonstration of what many computer programs can already do in solving Rubik's Cubes, just played by a user. Here is a video of a computer solving a 3x3x3 at multiple speeds: http://www.youtube.com/watch?v=r1uc2Iteyak I don't really feel that this video contributes much more than that particular youtube video except for theoretically being played on a DS with a stylus. The concept is interesting, but I feel that these automated solvers have you beat on novelty. Meh. EDIT: I also enjoy how mz's MSPaint Cube has two cubes with one same color on two faces, making it extremely unsolvable. =p
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Don't do bonus levels, this 100% run is best defined by 'things the game keeps track of', as just stated. (I haven't gotten around to watching the new segment, I will eventually.)
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I use Mupen 64 rerecording v8 available here: http://code.google.com/p/mupen64-rr/downloads/list Video Plugin: Jabo's Direct3D8 1.6 Input Plugin: TAS Input Plugin 0.6 Sound Plugin: Jabo's DirectSound 1.6 RSP Plugin: RSP emulation Plugin You should not have to alter the file Captain Forehead posted. If I had to guess, the problem is the video plugin. Another possible solution may be to go into audio settings and disable 'sync sound'
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
AngerFist wrote:
To be honest, I think that pause trick looks really ugly and is a big turn-off.
Disagree. Maybe for the 2 seconds it is occurring it is a bit ugly, but the faster speeds only make for more ridiculous star routes in my opinion. The following example is a star where the new trick cuts out about 10 seconds of metal walking underwater. Looks much better to me. New: http://www.youtube.com/watch?v=mAxjqz0MtFo Old: http://www.youtube.com/watch?v=1tddRRj_TYM
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Kirkq (pages ago) wrote: I did a rough comparison [of time to exit kokiri forest] and the [method of megaflipping out of the lost woods] still seems to be about 10-15 seconds faster [than the superslide clip ocarina cutscene method] assuming the deku stick [and shield are] acquired in both cases. Someone please correct me if I'm wrong. Rupees still seem to be a pretty big concern in all cases.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Maybe this has already been done, but have you guys tried manipulating the position variables in MHS to determine exactly how high you have to be to make the ledge? It would also be useful to know exactly at what point during the megaflip Link is considered to be the highest. This may also help to determine exactly how close you have come on your best attempts as well. It is obviously ridiculously hard to optimize floating point values.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Fionorde - Instead of answering one game mechanics question at a time, I'd rather take the time to make a page detailing everything we know eventually. Though I/we may not get around to that for a while. The enemy attack type pattern should be determined only by the start of the game and changes per previous atttack used. Shyft - The run is expected to be completed in the 1:50-1:55 range. I just solidified a pretty good Foma fight, but we're going to have to fight the RNG pretty heavily, and I'm going to need to perfect the walking script from Pharano to Ligena. Currently the script gets stuck on a bad iteration for far too long. I need to modify it with a break condition for that particular run if it has tried too many times, so that I will be able to run large numbers of iterations at a time. EDIT: Buffed RNG up to Ligena, Foma should be dead soon: http://dehacked.2y.net/microstorage.php/info/991402492/To%20Ligena%204486.smv EDIT: I would estimate 2-3 months depending on Classes/Motivation/How badly the RNG decides to fight against us/etc.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
DemonStrate wrote:
but..., is it possible to bomb hover under the tower itself and come back up into the door?
Sounds like a job for MrGrunz. =p Though there is probably a small chance this wouldn't work out even if you got behind the door. If the run continues to decrease in time, maybe it will actually be consistent with current knowledge some day.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
We walk between cutscenes of 'Olvan recruited' and 'got Sky Rune', this is sort of expected by the game. We cannot walk through walls in that instance. We want to treat multiple tune-ups as a glitch that we will not be using, so what general limitations do we place on the scenario to limit the run to not get multiple tune-ups.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I do want to see this run as well, but I doubt you're going to see anyone working on a 70 star run until the 120 star run is up to par. Probably just the limitations of no BLJs and no HSWK will be sensible to most people. CCC-less is too arbitrary.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
What does everyone think we should do for the tune-up in this run? I'd be interested to hear any perspectives on this, even if you have no familiarity with the game. The video is here: http://www.youtube.com/watch?v=1oKoFCiHE-E The details are as follows: This run only gets one tune-up under our chosen circumstances. If done optimally, we pause between the second and third step and skip the dialog. This in turn opens up the possibility to get a second tune-up. My question is essentially, "Under what limitations should the tune-up be acquired?" Should we..? 1: Walk through walls, skip dialog, create scenario for 2nd tune-up and neglect it. 2: Walk through walls, do not skip dialog, another tune-up not available. 3: Walk during cutscene, but not through walls. 4: Do not walk during cutscene. Personally I think 3 is the most well defined, but 2 is reasonable. "Does not walk through walls" is an easy definition for "non-glitched" instead of "Does not get/allow multiple tune-ups", but the fact that we walk during the cutscene still puts us in "able to walk through walls" status. I got a decent Foma fight today. We will probably test a couple other possibilities and then finalize it.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
No. The RNG is at 7E006C and 7E006E. It is beyond your control. It is used to determine how to set such things as apprentice join values, apprentice locations, how often apprentice locations change, who is in Patrof. Once these things are set (randomly at certain points), they do not change until they 'expire' so to speak. You can observe what values these hold by testing from a save.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Observing 7E5120 and 7E5121 tells me that these are preserved on reset. I assume the emulator reset function is the same as power on/off. (You should pop open an emulator and familiarize yourself with the memory watch function.) Ideally a segmented run would be very conscious of these values, and also very conscious of upcoming enemy attack patterns perhaps as well.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
EDIT: It seems I was wrong. I just got lucky, lolz.
Post subject: --------------/o/
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Some things just weren't meant to be. ;_; EDIT: --------------/o/
1 2
13 14 15
31 32